From 553674974d98e5b93e7c280d526cc112fa6df6b2 Mon Sep 17 00:00:00 2001 From: cgnorthcutt Date: Sun, 2 Feb 2020 05:16:17 -0500 Subject: [PATCH] Add results training other methods on cifar10 --- .../.DS_Store | Bin 10244 -> 0 bytes .../.DS_Store | Bin 8196 -> 0 bytes .../.DS_Store | Bin 8196 -> 0 bytes .../.DS_Store | Bin 8196 -> 0 bytes .../.DS_Store | Bin 10244 -> 0 bytes .../coteaching/coteaching_results/out_0_2.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_0_4.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_0_6.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_2_2.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_2_4.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_2_6.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_4_2.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_4_4.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_4_6.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_6_2.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_6_4.log | 2041 +++++++++++++++ .../coteaching/coteaching_results/out_6_6.log | 2041 +++++++++++++++ .../coteaching_plus_results/out_0_2.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_0_4.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_0_6.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_2_2.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_2_4.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_2_6.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_4_2.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_4_4.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_4_6.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_6_2.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_6_4.log | 2015 +++++++++++++++ .../coteaching_plus_results/out_6_6.log | 2015 +++++++++++++++ .../sceloss/sceloss_results/out_0_2.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_0_4.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_0_6.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_2_2.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_2_4.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_2_6.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_4_2.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_4_4.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_4_6.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_6_2.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_6_4.log | 2227 +++++++++++++++++ .../sceloss/sceloss_results/out_6_6.log | 2227 +++++++++++++++++ 41 files changed, 75396 insertions(+) delete mode 100644 cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_0/.DS_Store delete mode 100644 cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_2/.DS_Store delete mode 100644 cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_4__noise_amount__0_4/.DS_Store delete mode 100644 cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_6__noise_amount__0_4/.DS_Store delete mode 100644 cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_6__noise_amount__0_6/.DS_Store create mode 100644 other_methods/coteaching/coteaching_results/out_0_2.log create mode 100644 other_methods/coteaching/coteaching_results/out_0_4.log create mode 100644 other_methods/coteaching/coteaching_results/out_0_6.log create mode 100644 other_methods/coteaching/coteaching_results/out_2_2.log create mode 100644 other_methods/coteaching/coteaching_results/out_2_4.log create mode 100644 other_methods/coteaching/coteaching_results/out_2_6.log create mode 100644 other_methods/coteaching/coteaching_results/out_4_2.log create mode 100644 other_methods/coteaching/coteaching_results/out_4_4.log create mode 100644 other_methods/coteaching/coteaching_results/out_4_6.log create mode 100644 other_methods/coteaching/coteaching_results/out_6_2.log create mode 100644 other_methods/coteaching/coteaching_results/out_6_4.log create mode 100644 other_methods/coteaching/coteaching_results/out_6_6.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_0_2.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_0_4.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_0_6.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_2_2.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_2_4.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_2_6.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_4_2.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_4_4.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_4_6.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_6_2.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_6_4.log create mode 100644 other_methods/coteaching_plus/coteaching_plus_results/out_6_6.log create mode 100644 other_methods/sceloss/sceloss_results/out_0_2.log create mode 100644 other_methods/sceloss/sceloss_results/out_0_4.log create mode 100644 other_methods/sceloss/sceloss_results/out_0_6.log create mode 100644 other_methods/sceloss/sceloss_results/out_2_2.log create mode 100644 other_methods/sceloss/sceloss_results/out_2_4.log create mode 100644 other_methods/sceloss/sceloss_results/out_2_6.log create mode 100644 other_methods/sceloss/sceloss_results/out_4_2.log create mode 100644 other_methods/sceloss/sceloss_results/out_4_4.log create mode 100644 other_methods/sceloss/sceloss_results/out_4_6.log create mode 100644 other_methods/sceloss/sceloss_results/out_6_2.log create mode 100644 other_methods/sceloss/sceloss_results/out_6_4.log create mode 100644 other_methods/sceloss/sceloss_results/out_6_6.log diff --git a/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_0/.DS_Store b/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_0/.DS_Store deleted file mode 100644 index 2f73bcbe5d7376be07b17d88a37dde6a97ad46b6..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 10244 zcmeHM&ubGw6n@)Q-PnqSh896&FM?u`O?%OU5Yu?@CPY-wl5T!9T}gIBei+jj$w>tj zL_AneD)>*-lQ(Z3z4az|^58$9-<#Pc8?#B1DiX>J%)Z&#_ig6;W_EVp3lV9tY9>Nt z5Rs0`c6J7bfWp_g_LK>)=M=04JW+`vWKxnGa(ld0hf+W(pcGIFCR^E8(O`IBDK5YhHw+iC+iK5bmT!rBz3zfKs4W0gl~QC`~r4lhqZ! z+f=4HV);D!Q(lx%eKoP+))Ez~QZ2`<(rrGwW>q#) zUi-T1=G$#6b;HiqlC$9z*Qr>xlPw9boyo%G&Q06NxV4yDbuwjfpM<;+3WfAga(1E7 zST+|8v$<$AB9Y~#MI*9gn$2cNUpO7Pbakyz-)uZ+K71_&)x1rKj{hO>*7&HK&qWex zw@~B}J}I2#Z@w{^_ktUbd-JYu&+ojfZ-2TxPNU0=@#u;homu6XMH!P>y_YCCdDqEd zMErEj;7RA5!NxP6q$1k<^k*sKUza?5l)k?Hl|t0I_;X9FlmY&oxf{kmK+U;+;|2@f zFLT(*EV}P!asJ2iFJLhw($DGXukYcHuLP%|(H(}PqXYPc2k=ef9gxJkz=h9-pYi;o zx5y>a&{#j`7e`M-&%F_xN5FSj?*|{idyw9ecM;6f-K;-6+hd-6NI&P*$2;XA=D7>q zVLUoIfbZT~Ghc^F-HhMg8fuxBFARPWZ|odq5CeJ{T*t8?FW*TP@iucOnyxL<&7psR{fKq;UUPzoFn1tujAgJZ!f z{r~?zAZ!&4rGQc(NC9mso`}bAWjBrWrxWMe4b(MMF6`HmRzPqN>3CR?j>ie7t<8 diff --git a/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_2/.DS_Store b/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_0__noise_amount__0_2/.DS_Store deleted file mode 100644 index b60504ed3beda2a851e4369fb456223ef7ce4cce..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 8196 zcmeHMzi-n(6n@tR>Zn4EMAUX*@WPhCZs(#j1P~j@2nIl@9f!oo@tMRS4IwB}`xg`m zl@a~`#LULV&cI)QSn0^dyB|&BHf^00AtLTcci-jizVp5BPIrC|0FXkUV6ppUh%jci!HdQ36>u?{x3Q%P zb;NC_-3%R{I>`9qgud5|4C5P_n@?T1xS(m;qIOApY}IJnitMP>EZB{Q?AIMTsyVWJ zk5Ye7ww?Q4rCrKouhSs1y`a+IWY4W&mq!o0z@_a1Z3S+Vk7J33bULl2OPQ5!ciqV8 zMlYv#O>=!Mr<-er(d(tPm8<5)?X7y}MfYXz)xL0;$U_5`jtKBJx%Yn6w92#|vJsYd z4xS}v)@iXQb<^Jc`o7os^m%g9o6la}w^TQn%i3yp~EOes-M>wVH-Mp<2T45Irx)yxuwZdWOy;%u%0ze)y4O zcf{C3k0)o3@%L=4nZE%vVjl0l`Iu;(C3n>4Z0}m~bsq4CUQfdxn1-m5JBzj5;S z{~7g8OJOqMt2!h;?0}4x_9bOUiRQWAo%&tjP5)(ci$RB;&))?Gt8UM%zQpOGmr0EB2q(h<$j`G zBFZB(Ub}+g4~d_1Zb}zy%XOdvpJQIQXCKmnjPTSy+E^w*^4xlir*ho8PR+Z zPbyReK94H>f8pKuH_*aXb5Du1)Xf*T(AYw}!J+*QC9#o4=RaZw-f? zvS?IW)ZSjUXqslyidP4rI5XGPd}+Ag_hsE0c5YYo%U?fkFJ#>smY>r7R_nge@<_x` z`~=71DUJnS60@jtTSPBXKD_HI!iQ6j<@%BfH;%4}zZkD^go2UP&x^*VQ;+BR*8O3h z&xa&&;LJL3(X70e@v2_!JV{H$40+|!{OAKlX m3LTCEbvO<@{=*P;4_%q#TTBe11?k^^2$1)`HeQNz?Z6@QUQP)B diff --git a/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_6__noise_amount__0_4/.DS_Store b/cifar10/cifar10_noisy_labels__frac_zero_noise_rates__0_6__noise_amount__0_4/.DS_Store deleted file mode 100644 index a44399987a6b5620a9aaaaba150de43789b063d5..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 8196 zcmeHMzi-n(6n@tRa8x0JsM-#IywIV_;3NwRQV3ySt7HTNpu|o>V&wR%Ur6w#$Z z3=BnT*!e@)SdfqyiufCtSr8KMevmjcO`Q~>M%Qo zTq&Rw_%9V;`-6^&VOC;KAm2JL;~@ZI46C_e8RG!X(IsXj_5^Y&jwzxCBB?}%7(~#q zU1!}9vl4p(1s#Z>1CcWm8KDq9JMwhi9Y|K7%9R32frARL*!=*V$}G{k&jwl9`SLwFvrhAQ+od)#w2iMjdp~#IU!R6uJ_g&b4euJ~WVSrP z-x9d!wG9;e)J8nsPJfz)8}|8Q|3~`LTkdn@x$AW3^(_1`eotCw;kS>Tm+W@am#g)` z_FbnL7{)$XSd5)IeMZx?CGD*Cz^u`>8CXH9nX?-A*wv;L)NEP4O{qI5TlO8N(k>*H zZ%{9=9Iw*gWT#xgF8A*`UYWLYwB?nXd>m6W#N%--UP!EVIvFFW8{MSdNu@IBq@GF} zMzbV;z46Xr|r!CG~BRKpWj@rtVjKA$hpH>&B7m} z*2Rf?0!5D>;=b?*_XV~~23TfwE^LkH@N)ef>M-+IVO`Ge7W;CaJjQs#J>+kUH%2#f zSJ|8@uqf_(jm7^r5C8r@r{1aCRthKuq9{OOc{88Gp1)t8{v@$jTgOzy#EkXp3FH*a vcsP#8!*RsHABI@gu`A=!O~oMSd>DABAfIq#+V+w6;||6rQOZOCbHRGl8v#Z(M=Ct z1@)jJ2%ZHm9>gD@H*X$91pfd*^rm-x^P}k`)pQdf6q$jUH`$%q4#OsQAWg-8MFlT03NJB02QcVPZdQ#5l{pa0YyL&*oy@4JDV15 zoLki@0*Zhl&?Z2i4?0$AGr~sPvPTCtECoP2hL>5vKIQ>TqYIl6HsY2^aZa8+5FsTZ z#UNac_jNiPZARFLTP_E}k*g?HAVjP|NsuLid3+o*VF2X3hv8pbCwGCHJcteak*qD%4jdr%wA`XqCJ|g^5c}0f$^TU-BHkp7m>? z&=MU5JzZG9u_H#MGO?&n7_R{yeZ%_53cG~J+&mS$dUi705ircGBX3%Tw zPy32l@m$|qLj|Qb)gwmo;9{iM;NzI0Q_Z_Je$IV~I^b?@-08=mjMoe7_eUNuK(QBG z_O}7Qp13(D0LRwLh^Hf5d2hrVR)DzjGOAvCP?B_jY8^cp-5rvXp8Pc(@ z%Rg@H9)*%xMkJzecf1sl2NfAoY*7cM&BhnR6k98U`dprgA5UVzGOC})AAgKEs@VQJ zV9{s-nCZFse8hAKH{fO5fb;Q$-$?mzb%S=I`DjQFiyXru{d}zc73C;$yB2VC<2OrBA^H;0z)P+%Kho|{D1N9 zzyA+;Rmz uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0146, Loss2: 0.0154, Pure Ratio1: 10.1120, Pure Ratio2 10.0800 +Epoch [2/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0127, Loss2: 0.0131, Pure Ratio1: 10.4320, Pure Ratio2 10.4320 +Epoch [2/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0124, Loss2: 0.0132, Pure Ratio1: 10.3840, Pure Ratio2 10.3840 +Epoch [2/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 37.5000, Loss1: 0.0127, Loss2: 0.0131, Pure Ratio1: 10.3600, Pure Ratio2 10.3680 +Epoch [2/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.0312, Loss1: 0.0138, Loss2: 0.0147, Pure Ratio1: 10.2816, Pure Ratio2 10.2880 +Epoch [2/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 43.7500, Loss1: 0.0115, Loss2: 0.0125, Pure Ratio1: 10.2400, Pure Ratio2 10.2373 +Epoch [2/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0119, Loss2: 0.0122, Pure Ratio1: 10.1806, Pure Ratio2 10.1623 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 55.4988 % Model2 56.0497 %, Pure Ratio 1 10.1538 %, Pure Ratio 2 10.1313 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0100, Loss2: 0.0103, Pure Ratio1: 9.2131, Pure Ratio2 9.2623 +Epoch [3/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0111, Loss2: 0.0116, Pure Ratio1: 9.6393, Pure Ratio2 9.6639 +Epoch [3/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0119, Loss2: 0.0116, Pure Ratio1: 9.8306, Pure Ratio2 9.8689 +Epoch [3/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0109, Loss2: 0.0110, Pure Ratio1: 9.9549, Pure Ratio2 9.9590 +Epoch [3/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0118, Loss2: 0.0122, Pure Ratio1: 10.0754, Pure Ratio2 10.0754 +Epoch [3/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0108, Loss2: 0.0110, Pure Ratio1: 10.2186, Pure Ratio2 10.2213 +Epoch [3/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0105, Loss2: 0.0104, Pure Ratio1: 10.2365, Pure Ratio2 10.2295 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 64.7035 % Model2 62.3498 %, Pure Ratio 1 10.1450 %, Pure Ratio 2 10.1324 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0087, Loss2: 0.0089, Pure Ratio1: 9.9160, Pure Ratio2 9.9832 +Epoch [4/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0125, Loss2: 0.0125, Pure Ratio1: 10.0840, Pure Ratio2 10.1261 +Epoch [4/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0097, Loss2: 0.0096, Pure Ratio1: 10.0896, Pure Ratio2 10.1401 +Epoch [4/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0096, Loss2: 0.0091, Pure Ratio1: 10.0084, Pure Ratio2 10.0378 +Epoch [4/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0088, Loss2: 0.0086, Pure Ratio1: 10.0269, Pure Ratio2 10.0303 +Epoch [4/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0071, Loss2: 0.0073, Pure Ratio1: 10.0840, Pure Ratio2 10.0756 +Epoch [4/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0076, Loss2: 0.0076, Pure Ratio1: 10.1104, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 69.0304 % Model2 68.8001 %, Pure Ratio 1 10.1099 %, Pure Ratio 2 10.1013 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0102, Pure Ratio1: 10.0517, Pure Ratio2 10.1034 +Epoch [5/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0083, Loss2: 0.0079, Pure Ratio1: 10.1638, Pure Ratio2 10.2155 +Epoch [5/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0087, Loss2: 0.0082, Pure Ratio1: 10.2414, Pure Ratio2 10.2644 +Epoch [5/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0082, Loss2: 0.0078, Pure Ratio1: 10.2328, Pure Ratio2 10.2155 +Epoch [5/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0090, Loss2: 0.0085, Pure Ratio1: 10.2310, Pure Ratio2 10.2414 +Epoch [5/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0084, Loss2: 0.0079, Pure Ratio1: 10.0920, Pure Ratio2 10.1293 +Epoch [5/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0080, Loss2: 0.0074, Pure Ratio1: 10.1995, Pure Ratio2 10.2094 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 72.6062 % Model2 71.7047 %, Pure Ratio 1 10.1304 %, Pure Ratio 2 10.1326 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0089, Loss2: 0.0085, Pure Ratio1: 10.1770, Pure Ratio2 10.0708 +Epoch [6/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0078, Loss2: 0.0079, Pure Ratio1: 10.0885, Pure Ratio2 10.0973 +Epoch [6/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0058, Loss2: 0.0059, Pure Ratio1: 10.1298, Pure Ratio2 10.1593 +Epoch [6/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0060, Loss2: 0.0063, Pure Ratio1: 10.1283, Pure Ratio2 10.1106 +Epoch [6/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0082, Loss2: 0.0083, Pure Ratio1: 10.1274, Pure Ratio2 10.1168 +Epoch [6/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0070, Loss2: 0.0068, Pure Ratio1: 10.0383, Pure Ratio2 10.0354 +Epoch [6/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.0061, Loss2: 0.0053, Pure Ratio1: 10.0430, Pure Ratio2 10.0632 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 75.6110 % Model2 74.3790 %, Pure Ratio 1 10.0862 %, Pure Ratio 2 10.1157 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0062, Loss2: 0.0065, Pure Ratio1: 10.4364, Pure Ratio2 10.4364 +Epoch [7/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0055, Loss2: 0.0055, Pure Ratio1: 10.4182, Pure Ratio2 10.3636 +Epoch [7/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0049, Loss2: 0.0052, Pure Ratio1: 10.6606, Pure Ratio2 10.6121 +Epoch [7/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0067, Loss2: 0.0069, Pure Ratio1: 10.2773, Pure Ratio2 10.2364 +Epoch [7/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0071, Loss2: 0.0070, Pure Ratio1: 10.3018, Pure Ratio2 10.2764 +Epoch [7/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0052, Loss2: 0.0052, Pure Ratio1: 10.2394, Pure Ratio2 10.2000 +Epoch [7/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0096, Loss2: 0.0095, Pure Ratio1: 10.1532, Pure Ratio2 10.1377 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 75.2003 % Model2 75.1502 %, Pure Ratio 1 10.1189 %, Pure Ratio 2 10.1119 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0049, Loss2: 0.0047, Pure Ratio1: 10.5741, Pure Ratio2 10.7037 +Epoch [8/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0063, Loss2: 0.0061, Pure Ratio1: 10.0648, Pure Ratio2 10.1481 +Epoch [8/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0055, Loss2: 0.0052, Pure Ratio1: 9.9321, Pure Ratio2 9.9877 +Epoch [8/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0043, Loss2: 0.0038, Pure Ratio1: 9.8148, Pure Ratio2 9.8750 +Epoch [8/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.0039, Loss2: 0.0041, Pure Ratio1: 9.9963, Pure Ratio2 10.0296 +Epoch [8/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0061, Loss2: 0.0063, Pure Ratio1: 9.9846, Pure Ratio2 10.0093 +Epoch [8/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0057, Loss2: 0.0059, Pure Ratio1: 10.0476, Pure Ratio2 10.0582 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 77.9647 % Model2 76.7728 %, Pure Ratio 1 10.0404 %, Pure Ratio 2 10.0617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0043, Loss2: 0.0040, Pure Ratio1: 9.8857, Pure Ratio2 9.8286 +Epoch [9/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0043, Loss2: 0.0038, Pure Ratio1: 9.9429, Pure Ratio2 9.8095 +Epoch [9/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0041, Loss2: 0.0028, Pure Ratio1: 10.0254, Pure Ratio2 9.8794 +Epoch [9/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0042, Loss2: 0.0048, Pure Ratio1: 9.9238, Pure Ratio2 9.8000 +Epoch [9/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0056, Loss2: 0.0053, Pure Ratio1: 10.1219, Pure Ratio2 10.0267 +Epoch [9/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0050, Loss2: 0.0045, Pure Ratio1: 10.1429, Pure Ratio2 10.0635 +Epoch [9/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0057, Loss2: 0.0047, Pure Ratio1: 10.1088, Pure Ratio2 10.0435 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 78.9663 % Model2 78.5657 %, Pure Ratio 1 10.1392 %, Pure Ratio 2 10.0659 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0046, Loss2: 0.0049, Pure Ratio1: 10.6275, Pure Ratio2 10.9020 +Epoch [10/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0031, Loss2: 0.0026, Pure Ratio1: 10.4412, Pure Ratio2 10.5882 +Epoch [10/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0044, Loss2: 0.0041, Pure Ratio1: 10.1046, Pure Ratio2 10.1830 +Epoch [10/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0040, Loss2: 0.0042, Pure Ratio1: 10.0441, Pure Ratio2 10.0882 +Epoch [10/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0029, Loss2: 0.0026, Pure Ratio1: 9.9804, Pure Ratio2 10.0196 +Epoch [10/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0044, Loss2: 0.0040, Pure Ratio1: 10.0490, Pure Ratio2 10.0980 +Epoch [10/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0037, Loss2: 0.0031, Pure Ratio1: 10.0756, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 77.3337 % Model2 78.8061 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.0021, Loss2: 0.0027, Pure Ratio1: 9.7647, Pure Ratio2 9.9020 +Epoch [11/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0029, Loss2: 0.0031, Pure Ratio1: 10.0686, Pure Ratio2 10.1471 +Epoch [11/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0038, Loss2: 0.0046, Pure Ratio1: 10.0065, Pure Ratio2 10.1111 +Epoch [11/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0035, Loss2: 0.0034, Pure Ratio1: 9.9559, Pure Ratio2 10.0686 +Epoch [11/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0050, Loss2: 0.0040, Pure Ratio1: 10.0235, Pure Ratio2 10.1020 +Epoch [11/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0042, Loss2: 0.0045, Pure Ratio1: 10.0229, Pure Ratio2 10.1176 +Epoch [11/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 9.9888, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 80.3986 % Model2 79.9880 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0980 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0025, Loss2: 0.0027, Pure Ratio1: 9.6863, Pure Ratio2 9.5686 +Epoch [12/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 10.1961, Pure Ratio2 10.1569 +Epoch [12/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0023, Pure Ratio1: 10.2810, Pure Ratio2 10.2288 +Epoch [12/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0025, Loss2: 0.0022, Pure Ratio1: 10.2353, Pure Ratio2 10.2108 +Epoch [12/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.0029, Loss2: 0.0036, Pure Ratio1: 10.1137, Pure Ratio2 10.0824 +Epoch [12/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0021, Loss2: 0.0024, Pure Ratio1: 9.9608, Pure Ratio2 9.9052 +Epoch [12/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0042, Loss2: 0.0038, Pure Ratio1: 10.0448, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 81.0697 % Model2 81.4503 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0028, Loss2: 0.0023, Pure Ratio1: 10.1176, Pure Ratio2 9.8824 +Epoch [13/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0028, Loss2: 0.0029, Pure Ratio1: 10.5000, Pure Ratio2 10.3922 +Epoch [13/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0019, Loss2: 0.0016, Pure Ratio1: 9.9216, Pure Ratio2 9.8497 +Epoch [13/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 77.3438, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 9.8284, Pure Ratio2 9.7598 +Epoch [13/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0024, Loss2: 0.0030, Pure Ratio1: 9.9255, Pure Ratio2 9.8824 +Epoch [13/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0032, Loss2: 0.0034, Pure Ratio1: 9.9673, Pure Ratio2 9.9608 +Epoch [13/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 10.0280, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 81.3401 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0012, Pure Ratio1: 10.2353, Pure Ratio2 10.3137 +Epoch [14/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0025, Loss2: 0.0027, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [14/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0020, Loss2: 0.0020, Pure Ratio1: 9.8301, Pure Ratio2 9.8497 +Epoch [14/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0016, Loss2: 0.0018, Pure Ratio1: 9.9755, Pure Ratio2 9.9412 +Epoch [14/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0030, Loss2: 0.0033, Pure Ratio1: 9.9686, Pure Ratio2 9.9451 +Epoch [14/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 70.3125, Loss1: 0.0019, Loss2: 0.0029, Pure Ratio1: 9.8301, Pure Ratio2 9.8039 +Epoch [14/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.0033, Loss2: 0.0025, Pure Ratio1: 9.9804, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 81.5605 % Model2 82.0012 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 82.0312, Loss1: 0.0021, Loss2: 0.0014, Pure Ratio1: 9.0000, Pure Ratio2 9.2549 +Epoch [15/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.6275, Pure Ratio2 9.6667 +Epoch [15/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 9.6340, Pure Ratio2 9.6732 +Epoch [15/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0015, Pure Ratio1: 9.9118, Pure Ratio2 9.9412 +Epoch [15/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0037, Loss2: 0.0025, Pure Ratio1: 9.9373, Pure Ratio2 9.9647 +Epoch [15/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0030, Loss2: 0.0033, Pure Ratio1: 9.9771, Pure Ratio2 10.0033 +Epoch [15/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.0020, Loss2: 0.0024, Pure Ratio1: 10.0252, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 81.8409 % Model2 82.1815 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0035, Loss2: 0.0018, Pure Ratio1: 10.3529, Pure Ratio2 10.2941 +Epoch [16/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0021, Loss2: 0.0020, Pure Ratio1: 9.9020, Pure Ratio2 9.8922 +Epoch [16/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0023, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [16/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0017, Pure Ratio1: 10.1275, Pure Ratio2 10.1127 +Epoch [16/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0019, Pure Ratio1: 10.1176, Pure Ratio2 10.0863 +Epoch [16/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0037, Loss2: 0.0041, Pure Ratio1: 10.1275, Pure Ratio2 10.0817 +Epoch [16/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 10.0644, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 81.7808 % Model2 81.9411 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0024, Pure Ratio1: 9.9804, Pure Ratio2 9.8039 +Epoch [17/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.9608, Pure Ratio2 9.8627 +Epoch [17/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 80.4688, Loss1: 0.0025, Loss2: 0.0018, Pure Ratio1: 10.1503, Pure Ratio2 10.0523 +Epoch [17/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 10.1471, Pure Ratio2 10.1029 +Epoch [17/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0023, Pure Ratio1: 10.0392, Pure Ratio2 10.0235 +Epoch [17/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0020, Loss2: 0.0019, Pure Ratio1: 10.0033, Pure Ratio2 9.9902 +Epoch [17/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0032, Loss2: 0.0034, Pure Ratio1: 9.9552, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 81.3602 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [18/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0021, Loss2: 0.0020, Pure Ratio1: 10.0588, Pure Ratio2 10.1275 +Epoch [18/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0025, Pure Ratio1: 10.0000, Pure Ratio2 10.0719 +Epoch [18/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 9.9804, Pure Ratio2 10.0147 +Epoch [18/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 10.0078, Pure Ratio2 10.0353 +Epoch [18/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.0013, Loss2: 0.0020, Pure Ratio1: 10.0882, Pure Ratio2 10.0719 +Epoch [18/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 72.6562, Loss1: 0.0049, Loss2: 0.0040, Pure Ratio1: 10.0616, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 81.9611 % Model2 82.0513 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.5490, Pure Ratio2 9.4902 +Epoch [19/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0026, Loss2: 0.0036, Pure Ratio1: 9.5784, Pure Ratio2 9.6078 +Epoch [19/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 76.5625, Loss1: 0.0028, Loss2: 0.0020, Pure Ratio1: 9.7190, Pure Ratio2 9.7451 +Epoch [19/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.7941, Pure Ratio2 9.8186 +Epoch [19/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 9.7961, Pure Ratio2 9.8275 +Epoch [19/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 9.8627, Pure Ratio2 9.8856 +Epoch [19/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.9608, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 81.6707 % Model2 81.8009 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [20/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.9412, Pure Ratio2 9.8922 +Epoch [20/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0011, Loss2: 0.0016, Pure Ratio1: 10.0327, Pure Ratio2 9.9869 +Epoch [20/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 10.0245, Pure Ratio2 9.9412 +Epoch [20/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.9765, Pure Ratio2 9.9216 +Epoch [20/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0817, Pure Ratio2 10.0327 +Epoch [20/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0017, Loss2: 0.0018, Pure Ratio1: 10.0280, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 80.0280 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.8039, Pure Ratio2 9.8039 +Epoch [21/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.6275, Pure Ratio2 9.5784 +Epoch [21/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.0029, Loss2: 0.0019, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [21/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 10.0392, Pure Ratio2 10.0245 +Epoch [21/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 10.0196, Pure Ratio2 10.0196 +Epoch [21/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.9739, Pure Ratio2 9.9673 +Epoch [21/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.9468, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 81.9211 % Model2 82.3317 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6275, Pure Ratio2 9.6667 +Epoch [22/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.9706 +Epoch [22/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8758, Pure Ratio2 9.8954 +Epoch [22/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.8775, Pure Ratio2 9.8922 +Epoch [22/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 10.0196 +Epoch [22/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0012, Pure Ratio1: 9.9967, Pure Ratio2 9.9967 +Epoch [22/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0018, Pure Ratio1: 10.0336, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 81.0397 % Model2 82.0913 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [23/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0020, Loss2: 0.0011, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [23/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0025, Pure Ratio1: 10.1046, Pure Ratio2 10.0915 +Epoch [23/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0020, Loss2: 0.0013, Pure Ratio1: 10.0833, Pure Ratio2 10.0196 +Epoch [23/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 10.0588, Pure Ratio2 10.0118 +Epoch [23/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.1569, Pure Ratio2 10.0882 +Epoch [23/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.1120, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 81.5705 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.8627 +Epoch [24/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 9.9706 +Epoch [24/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.9542, Pure Ratio2 10.0392 +Epoch [24/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0016, Pure Ratio1: 10.0833, Pure Ratio2 10.1569 +Epoch [24/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0824, Pure Ratio2 10.0824 +Epoch [24/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.9869, Pure Ratio2 9.9935 +Epoch [24/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0005, Pure Ratio1: 9.9832, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 81.4503 % Model2 81.4403 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.4706, Pure Ratio2 10.6471 +Epoch [25/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.2549, Pure Ratio2 10.3235 +Epoch [25/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0020, Loss2: 0.0015, Pure Ratio1: 10.1830, Pure Ratio2 10.1634 +Epoch [25/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 10.0147, Pure Ratio2 10.0294 +Epoch [25/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 10.0706, Pure Ratio2 10.0667 +Epoch [25/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.0229, Pure Ratio2 10.0458 +Epoch [25/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.0420, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 80.6190 % Model2 80.4187 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.6667, Pure Ratio2 10.6275 +Epoch [26/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.7157, Pure Ratio2 10.6176 +Epoch [26/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.4967, Pure Ratio2 10.3725 +Epoch [26/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0011, Loss2: 0.0017, Pure Ratio1: 10.3824, Pure Ratio2 10.2990 +Epoch [26/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 10.1922, Pure Ratio2 10.1059 +Epoch [26/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0015, Pure Ratio1: 10.1536, Pure Ratio2 10.1046 +Epoch [26/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.1317, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 81.4704 % Model2 81.2600 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.4118, Pure Ratio2 10.4902 +Epoch [27/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.1667, Pure Ratio2 10.2255 +Epoch [27/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 10.0980, Pure Ratio2 10.1307 +Epoch [27/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0018, Loss2: 0.0026, Pure Ratio1: 9.9902, Pure Ratio2 10.0343 +Epoch [27/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0549, Pure Ratio2 10.1137 +Epoch [27/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 10.0490, Pure Ratio2 10.1046 +Epoch [27/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.1120, Pure Ratio2 10.1681 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 81.2500 % Model2 80.9796 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.5686, Pure Ratio2 9.4314 +Epoch [28/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.6471, Pure Ratio2 9.5588 +Epoch [28/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.7843 +Epoch [28/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8627, Pure Ratio2 9.8922 +Epoch [28/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9490, Pure Ratio2 9.9373 +Epoch [28/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.9837, Pure Ratio2 9.9477 +Epoch [28/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9972, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 81.3902 % Model2 81.7608 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.6078, Pure Ratio2 9.6863 +Epoch [29/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0034, Loss2: 0.0043, Pure Ratio1: 10.0588, Pure Ratio2 10.0490 +Epoch [29/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1046, Pure Ratio2 10.0523 +Epoch [29/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 10.0049, Pure Ratio2 9.9804 +Epoch [29/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0471, Pure Ratio2 10.0275 +Epoch [29/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 10.0033 +Epoch [29/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.0023, Loss2: 0.0020, Pure Ratio1: 9.9524, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 80.7192 % Model2 81.6607 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.8431 +Epoch [30/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8725 +Epoch [30/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0523 +Epoch [30/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9657, Pure Ratio2 9.8971 +Epoch [30/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0007, Loss2: 0.0016, Pure Ratio1: 9.8392, Pure Ratio2 9.8039 +Epoch [30/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.9216, Pure Ratio2 9.9052 +Epoch [30/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.0336, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 81.4503 % Model2 81.7909 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 9.8627 +Epoch [31/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.9216, Pure Ratio2 9.7745 +Epoch [31/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0073, Loss2: 0.0049, Pure Ratio1: 10.0784, Pure Ratio2 9.9673 +Epoch [31/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.3971, Pure Ratio2 10.2745 +Epoch [31/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.3882, Pure Ratio2 10.3098 +Epoch [31/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.2222, Pure Ratio2 10.1765 +Epoch [31/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0013, Pure Ratio1: 10.1064, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 81.3502 % Model2 81.3502 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.7843, Pure Ratio2 10.6275 +Epoch [32/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.1078, Pure Ratio2 10.0980 +Epoch [32/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0654 +Epoch [32/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.2206, Pure Ratio2 10.2549 +Epoch [32/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.1333, Pure Ratio2 10.1804 +Epoch [32/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0024, Loss2: 0.0021, Pure Ratio1: 10.0784, Pure Ratio2 10.1275 +Epoch [32/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0532, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 81.9511 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 10.0980 +Epoch [33/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9412, Pure Ratio2 9.9510 +Epoch [33/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 10.0784, Pure Ratio2 10.0392 +Epoch [33/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0015, Pure Ratio1: 10.2108, Pure Ratio2 10.1373 +Epoch [33/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.1412, Pure Ratio2 10.0941 +Epoch [33/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 10.0000, Pure Ratio2 9.9739 +Epoch [33/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 9.9300, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 80.7091 % Model2 80.9495 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.0980 +Epoch [34/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 10.1961, Pure Ratio2 10.1471 +Epoch [34/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.2810, Pure Ratio2 10.2222 +Epoch [34/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.3235, Pure Ratio2 10.2598 +Epoch [34/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.1490, Pure Ratio2 10.1608 +Epoch [34/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 10.1078, Pure Ratio2 10.0882 +Epoch [34/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 10.0728, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 80.4387 % Model2 81.3502 %, Pure Ratio 1 10.1081 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.2157, Pure Ratio2 9.3137 +Epoch [35/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0490 +Epoch [35/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0392 +Epoch [35/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8971, Pure Ratio2 10.0343 +Epoch [35/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8980, Pure Ratio2 9.9882 +Epoch [35/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.9706, Pure Ratio2 10.0327 +Epoch [35/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0308, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 80.9595 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.3529, Pure Ratio2 9.4510 +Epoch [36/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.7255 +Epoch [36/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.9673, Pure Ratio2 9.9935 +Epoch [36/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0245, Pure Ratio2 10.0049 +Epoch [36/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1333, Pure Ratio2 10.1294 +Epoch [36/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 10.1732, Pure Ratio2 10.1699 +Epoch [36/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0644, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 81.3602 % Model2 80.9696 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [37/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9902, Pure Ratio2 9.9118 +Epoch [37/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [37/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.2255 +Epoch [37/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0017, Pure Ratio1: 10.0078, Pure Ratio2 10.0627 +Epoch [37/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0686, Pure Ratio2 10.0882 +Epoch [37/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9636, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 81.1699 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.6667, Pure Ratio2 9.5882 +Epoch [38/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.2059, Pure Ratio2 10.1863 +Epoch [38/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0327, Pure Ratio2 10.0523 +Epoch [38/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9216, Pure Ratio2 9.9265 +Epoch [38/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0314, Pure Ratio2 10.0392 +Epoch [38/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9837, Pure Ratio2 9.9902 +Epoch [38/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.9748, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 81.3502 % Model2 81.7208 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.4510, Pure Ratio2 9.4510 +Epoch [39/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.3333, Pure Ratio2 10.1765 +Epoch [39/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.1569, Pure Ratio2 10.0719 +Epoch [39/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 10.1618, Pure Ratio2 10.1176 +Epoch [39/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.0745 +Epoch [39/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.0686, Pure Ratio2 10.0556 +Epoch [39/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.0924, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 80.4387 % Model2 80.5589 %, Pure Ratio 1 10.0754 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2941, Pure Ratio2 10.4118 +Epoch [40/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.2157 +Epoch [40/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 10.1111, Pure Ratio2 10.2157 +Epoch [40/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0539 +Epoch [40/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8980, Pure Ratio2 9.9529 +Epoch [40/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9379, Pure Ratio2 9.9771 +Epoch [40/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 10.0280, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 80.9495 % Model2 81.5405 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.1373 +Epoch [41/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.2157, Pure Ratio2 10.1961 +Epoch [41/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.2418, Pure Ratio2 10.2092 +Epoch [41/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1225, Pure Ratio2 10.0833 +Epoch [41/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 10.1608, Pure Ratio2 10.1490 +Epoch [41/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1340 +Epoch [41/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 10.1317, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 79.2768 % Model2 79.3570 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.5294, Pure Ratio2 10.6275 +Epoch [42/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2255, Pure Ratio2 10.3039 +Epoch [42/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.2092, Pure Ratio2 10.2745 +Epoch [42/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.1667, Pure Ratio2 10.1765 +Epoch [42/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1961, Pure Ratio2 10.1725 +Epoch [42/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.1471, Pure Ratio2 10.0980 +Epoch [42/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0448, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 80.8494 % Model2 80.8594 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.0784 +Epoch [43/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0029, Loss2: 0.0023, Pure Ratio1: 10.1569, Pure Ratio2 10.0686 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.0131 +Epoch [43/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0294, Pure Ratio2 9.9559 +Epoch [43/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.1647, Pure Ratio2 10.1490 +Epoch [43/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0490, Pure Ratio2 10.0523 +Epoch [43/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.1092, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 81.1899 % Model2 81.7808 %, Pure Ratio 1 10.1332 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.2157, Pure Ratio2 9.3137 +Epoch [44/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6176, Pure Ratio2 9.7353 +Epoch [44/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8497, Pure Ratio2 9.8954 +Epoch [44/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9755, Pure Ratio2 10.0049 +Epoch [44/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9490, Pure Ratio2 9.9529 +Epoch [44/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 10.0065, Pure Ratio2 10.0425 +Epoch [44/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0504, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 81.5204 % Model2 81.5805 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.4118, Pure Ratio2 10.3922 +Epoch [45/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3235, Pure Ratio2 10.3627 +Epoch [45/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.2941, Pure Ratio2 10.2680 +Epoch [45/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1029, Pure Ratio2 10.0490 +Epoch [45/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1529, Pure Ratio2 10.1529 +Epoch [45/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.1046 +Epoch [45/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0476, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 79.3870 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.0000 +Epoch [46/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.3333, Pure Ratio2 10.2941 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.3333 +Epoch [46/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.1225 +Epoch [46/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.2235, Pure Ratio2 10.2431 +Epoch [46/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.1078, Pure Ratio2 10.1373 +Epoch [46/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.0812, Pure Ratio2 10.1092 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 81.0998 % Model2 80.9796 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.5686, Pure Ratio2 9.6275 +Epoch [47/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.5294, Pure Ratio2 9.6176 +Epoch [47/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.8301 +Epoch [47/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8333, Pure Ratio2 9.8284 +Epoch [47/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0157 +Epoch [47/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0556, Pure Ratio2 10.0490 +Epoch [47/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9860, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 81.7308 % Model2 81.8309 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [48/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7059 +Epoch [48/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6928, Pure Ratio2 9.8039 +Epoch [48/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7892, Pure Ratio2 9.9461 +Epoch [48/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8902, Pure Ratio2 10.0157 +Epoch [48/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9510, Pure Ratio2 10.0458 +Epoch [48/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9944, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 80.1783 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.3137 +Epoch [49/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.4020, Pure Ratio2 10.5294 +Epoch [49/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.3072, Pure Ratio2 10.3529 +Epoch [49/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.2598 +Epoch [49/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3647, Pure Ratio2 10.3961 +Epoch [49/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.1699, Pure Ratio2 10.1797 +Epoch [49/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1709, Pure Ratio2 10.1625 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 80.2985 % Model2 80.4087 %, Pure Ratio 1 10.1081 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.5098 +Epoch [50/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.1569, Pure Ratio2 9.9412 +Epoch [50/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1438, Pure Ratio2 9.9869 +Epoch [50/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.1863, Pure Ratio2 10.0882 +Epoch [50/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9961, Pure Ratio2 9.9255 +Epoch [50/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0458, Pure Ratio2 9.9739 +Epoch [50/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 80.6891 % Model2 80.8794 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.5882, Pure Ratio2 10.7843 +Epoch [51/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.3235, Pure Ratio2 10.3725 +Epoch [51/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1111, Pure Ratio2 10.1046 +Epoch [51/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9853, Pure Ratio2 10.0049 +Epoch [51/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1294, Pure Ratio2 10.1294 +Epoch [51/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [51/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0896, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 78.9263 % Model2 80.9295 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.5882 +Epoch [52/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0018, Pure Ratio1: 10.0196, Pure Ratio2 9.8333 +Epoch [52/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.7451 +Epoch [52/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9951, Pure Ratio2 9.8824 +Epoch [52/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8314 +Epoch [52/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.8987 +Epoch [52/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9972, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 80.2284 % Model2 81.2600 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.4118 +Epoch [53/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.1765 +Epoch [53/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0065, Pure Ratio2 10.0065 +Epoch [53/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8775, Pure Ratio2 9.9069 +Epoch [53/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8784, Pure Ratio2 9.9137 +Epoch [53/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9771, Pure Ratio2 10.0065 +Epoch [53/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0280, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 81.2400 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [54/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.0098 +Epoch [54/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.3007, Pure Ratio2 10.2026 +Epoch [54/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.1422, Pure Ratio2 10.1324 +Epoch [54/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1608, Pure Ratio2 10.1608 +Epoch [54/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0719, Pure Ratio2 10.0850 +Epoch [54/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0112, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 80.2484 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.4314 +Epoch [55/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1961, Pure Ratio2 10.3333 +Epoch [55/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0719, Pure Ratio2 10.1634 +Epoch [55/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0016, Pure Ratio1: 10.2059, Pure Ratio2 10.2402 +Epoch [55/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.3059, Pure Ratio2 10.3843 +Epoch [55/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1275 +Epoch [55/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1092, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 80.5990 % Model2 80.8994 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.2941 +Epoch [56/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.2255 +Epoch [56/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0261, Pure Ratio2 9.9739 +Epoch [56/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0147 +Epoch [56/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0627, Pure Ratio2 10.0627 +Epoch [56/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1405 +Epoch [56/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2269, Pure Ratio2 10.2353 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 80.6991 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.4902, Pure Ratio2 9.4902 +Epoch [57/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.6961 +Epoch [57/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.8170, Pure Ratio2 9.8301 +Epoch [57/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.8725, Pure Ratio2 9.8431 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 9.8667 +Epoch [57/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 9.8954 +Epoch [57/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9664, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 80.3085 % Model2 80.2183 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.7059, Pure Ratio2 9.9412 +Epoch [58/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.1078 +Epoch [58/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9477 +Epoch [58/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.9265, Pure Ratio2 9.9314 +Epoch [58/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9451, Pure Ratio2 9.9569 +Epoch [58/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.9510, Pure Ratio2 9.9542 +Epoch [58/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9580, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 79.2768 % Model2 79.6474 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.6667 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 10.1275 +Epoch [59/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.3137, Pure Ratio2 10.2222 +Epoch [59/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0003, Pure Ratio1: 10.1814, Pure Ratio2 10.0882 +Epoch [59/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0941 +Epoch [59/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.1340, Pure Ratio2 10.0980 +Epoch [59/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1148, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 80.4988 % Model2 80.9495 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4510, Pure Ratio2 9.3922 +Epoch [60/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.6863 +Epoch [60/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0523, Pure Ratio2 9.9673 +Epoch [60/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0049, Pure Ratio2 9.9608 +Epoch [60/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.0235, Pure Ratio2 9.9961 +Epoch [60/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1111 +Epoch [60/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0924, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 79.5773 % Model2 79.4772 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.7255, Pure Ratio2 10.6471 +Epoch [61/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.5784, Pure Ratio2 10.4706 +Epoch [61/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.4575, Pure Ratio2 10.3333 +Epoch [61/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.3235, Pure Ratio2 10.2451 +Epoch [61/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.1412, Pure Ratio2 10.0980 +Epoch [61/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.0327 +Epoch [61/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0840, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 79.2568 % Model2 81.1498 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.8235 +Epoch [62/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8039 +Epoch [62/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.9281 +Epoch [62/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 10.0490 +Epoch [62/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.1098, Pure Ratio2 10.1843 +Epoch [62/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 10.1438, Pure Ratio2 10.1863 +Epoch [62/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 80.2484 % Model2 80.3586 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.4902, Pure Ratio2 10.6275 +Epoch [63/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0882, Pure Ratio2 10.2353 +Epoch [63/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9673, Pure Ratio2 10.0980 +Epoch [63/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.2010 +Epoch [63/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1686, Pure Ratio2 10.2510 +Epoch [63/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.1928 +Epoch [63/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1289, Pure Ratio2 10.1653 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 81.5304 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0196, Pure Ratio2 10.3333 +Epoch [64/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.0882 +Epoch [64/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 9.9804 +Epoch [64/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0931, Pure Ratio2 10.0490 +Epoch [64/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1098, Pure Ratio2 10.0314 +Epoch [64/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0784, Pure Ratio2 10.0359 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1345, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 80.3185 % Model2 80.2885 %, Pure Ratio 1 10.1383 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.5490, Pure Ratio2 9.5686 +Epoch [65/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.8431 +Epoch [65/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8170 +Epoch [65/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9020 +Epoch [65/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0941, Pure Ratio2 10.0431 +Epoch [65/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0523, Pure Ratio2 10.0098 +Epoch [65/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0532, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 80.3185 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 11.3137, Pure Ratio2 10.7647 +Epoch [66/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4412, Pure Ratio2 10.2255 +Epoch [66/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3856, Pure Ratio2 10.2876 +Epoch [66/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9853 +Epoch [66/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.1333 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1046, Pure Ratio2 10.0098 +Epoch [66/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 80.3886 % Model2 80.5990 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [67/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0980 +Epoch [67/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0523 +Epoch [67/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2206, Pure Ratio2 10.2500 +Epoch [67/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.1451 +Epoch [67/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0915, Pure Ratio2 10.1046 +Epoch [67/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0672, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 80.5188 % Model2 80.3986 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0014, Pure Ratio1: 9.2941, Pure Ratio2 9.2157 +Epoch [68/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.2059, Pure Ratio2 10.0490 +Epoch [68/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.1307 +Epoch [68/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1667, Pure Ratio2 10.0588 +Epoch [68/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0980, Pure Ratio2 10.0392 +Epoch [68/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1340, Pure Ratio2 10.1046 +Epoch [68/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.1176, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 80.6891 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.5490, Pure Ratio2 10.4706 +Epoch [69/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [69/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 10.0392 +Epoch [69/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1618, Pure Ratio2 10.1275 +Epoch [69/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1529, Pure Ratio2 10.0863 +Epoch [69/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0523 +Epoch [69/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0448, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 80.4788 % Model2 80.6090 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.4706, Pure Ratio2 10.2941 +Epoch [70/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2451 +Epoch [70/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 9.9935 +Epoch [70/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9510 +Epoch [70/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9647 +Epoch [70/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0719 +Epoch [70/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0532, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 80.3886 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1176 +Epoch [71/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3529, Pure Ratio2 10.3431 +Epoch [71/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9673, Pure Ratio2 9.9020 +Epoch [71/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8971, Pure Ratio2 9.8725 +Epoch [71/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9333, Pure Ratio2 9.9020 +Epoch [71/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 9.9314 +Epoch [71/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0112, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 80.7091 % Model2 81.5905 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.5098, Pure Ratio2 10.6863 +Epoch [72/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.4706 +Epoch [72/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1242, Pure Ratio2 10.2614 +Epoch [72/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9755, Pure Ratio2 10.0392 +Epoch [72/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0510 +Epoch [72/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0261, Pure Ratio2 10.0359 +Epoch [72/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 80.0982 % Model2 81.0998 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0392 +Epoch [73/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9020 +Epoch [73/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0131 +Epoch [73/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0539, Pure Ratio2 10.1225 +Epoch [73/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0745, Pure Ratio2 10.0824 +Epoch [73/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0490 +Epoch [73/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0728, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 80.4587 % Model2 80.1983 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.2157 +Epoch [74/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 10.0980 +Epoch [74/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 10.0915 +Epoch [74/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.1275, Pure Ratio2 10.2451 +Epoch [74/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 10.1569 +Epoch [74/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.0915 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0616, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 80.5789 % Model2 80.7492 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.4510, Pure Ratio2 10.0980 +Epoch [75/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 9.9608 +Epoch [75/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2876, Pure Ratio2 9.9869 +Epoch [75/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3186, Pure Ratio2 10.0196 +Epoch [75/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.2431, Pure Ratio2 10.0392 +Epoch [75/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.0229 +Epoch [75/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.1282 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.1373, Pure Ratio2 9.9216 +Epoch [76/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2647, Pure Ratio2 10.0392 +Epoch [76/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1895, Pure Ratio2 9.9412 +Epoch [76/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9608 +Epoch [76/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1059, Pure Ratio2 9.9255 +Epoch [76/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.8137 +Epoch [76/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9692, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 80.3786 % Model2 80.8894 %, Pure Ratio 1 10.1081 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.1176 +Epoch [77/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 10.0196 +Epoch [77/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.9673 +Epoch [77/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9363, Pure Ratio2 10.0147 +Epoch [77/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1686 +Epoch [77/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.1111, Pure Ratio2 10.1830 +Epoch [77/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1401, Pure Ratio2 10.1709 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 80.8694 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.1106 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.5686 +Epoch [78/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [78/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.0065 +Epoch [78/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9657, Pure Ratio2 9.9216 +Epoch [78/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 9.9608 +Epoch [78/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0948, Pure Ratio2 10.0686 +Epoch [78/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0924, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 79.8878 % Model2 80.6390 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.5098 +Epoch [79/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.2353 +Epoch [79/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [79/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0686 +Epoch [79/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0667, Pure Ratio2 10.1020 +Epoch [79/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [79/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 79.6775 % Model2 80.5489 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6471, Pure Ratio2 10.4902 +Epoch [80/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.4020, Pure Ratio2 10.2549 +Epoch [80/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.0392 +Epoch [80/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0882, Pure Ratio2 10.0245 +Epoch [80/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0745, Pure Ratio2 10.0902 +Epoch [80/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0817 +Epoch [80/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0700, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 80.2384 % Model2 81.0296 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.9804 +Epoch [81/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.8824 +Epoch [81/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 10.0458 +Epoch [81/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9853, Pure Ratio2 10.0588 +Epoch [81/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0902, Pure Ratio2 10.1843 +Epoch [81/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0327, Pure Ratio2 10.1078 +Epoch [81/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9888, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 80.2284 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.5098, Pure Ratio2 10.1961 +Epoch [82/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.8725 +Epoch [82/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.7908 +Epoch [82/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9069, Pure Ratio2 9.8824 +Epoch [82/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.0353 +Epoch [82/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.1275 +Epoch [82/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1457, Pure Ratio2 10.1653 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.0982 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0009, Pure Ratio1: 10.1373, Pure Ratio2 10.3725 +Epoch [83/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 10.0196 +Epoch [83/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7712, Pure Ratio2 10.0065 +Epoch [83/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7206, Pure Ratio2 9.9608 +Epoch [83/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 10.0667 +Epoch [83/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 10.1536 +Epoch [83/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 10.0084, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 79.4271 % Model2 80.6190 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0196 +Epoch [84/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.1961, Pure Ratio2 10.2157 +Epoch [84/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0014, Pure Ratio1: 10.1111, Pure Ratio2 10.1046 +Epoch [84/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.1324 +Epoch [84/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.1490 +Epoch [84/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0784 +Epoch [84/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 80.1082 % Model2 80.5589 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.4510 +Epoch [85/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 10.0588 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 10.0392 +Epoch [85/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0245, Pure Ratio2 10.0049 +Epoch [85/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0706, Pure Ratio2 10.0392 +Epoch [85/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.9935, Pure Ratio2 9.9837 +Epoch [85/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0140, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 79.8778 % Model2 80.1082 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.9804 +Epoch [86/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.1863 +Epoch [86/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0392 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0980 +Epoch [86/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0118, Pure Ratio2 10.1294 +Epoch [86/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 10.0490 +Epoch [86/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9496, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 79.5673 % Model2 80.8393 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.4902, Pure Ratio2 9.5294 +Epoch [87/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9510 +Epoch [87/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9869 +Epoch [87/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 10.0637 +Epoch [87/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.0314 +Epoch [87/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.0229 +Epoch [87/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0952, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 80.2083 % Model2 80.9195 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.2745 +Epoch [88/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9902 +Epoch [88/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2222 +Epoch [88/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2402, Pure Ratio2 10.2206 +Epoch [88/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1294, Pure Ratio2 10.0824 +Epoch [88/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0098 +Epoch [88/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 81.1999 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2745 +Epoch [89/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 10.0196 +Epoch [89/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.2484 +Epoch [89/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.1618, Pure Ratio2 10.2402 +Epoch [89/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.2745 +Epoch [89/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1405, Pure Ratio2 10.2222 +Epoch [89/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0700, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 79.7476 % Model2 80.1883 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.5098, Pure Ratio2 10.6471 +Epoch [90/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0013, Loss2: 0.0001, Pure Ratio1: 10.2647, Pure Ratio2 10.3824 +Epoch [90/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4248, Pure Ratio2 10.4967 +Epoch [90/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.3088 +Epoch [90/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2314, Pure Ratio2 10.2745 +Epoch [90/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.1601 +Epoch [90/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0448, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 80.8594 % Model2 80.9495 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.1569 +Epoch [91/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3039, Pure Ratio2 10.1471 +Epoch [91/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1503, Pure Ratio2 10.0719 +Epoch [91/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1716, Pure Ratio2 10.1961 +Epoch [91/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.1529 +Epoch [91/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.0980 +Epoch [91/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0013, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2157 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 81.0096 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 9.8824 +Epoch [92/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.1667 +Epoch [92/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2876, Pure Ratio2 10.1765 +Epoch [92/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2206, Pure Ratio2 10.1618 +Epoch [92/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.0745 +Epoch [92/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 10.0686 +Epoch [92/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9636, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 80.0881 % Model2 80.8494 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.6863, Pure Ratio2 10.3333 +Epoch [93/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.0980 +Epoch [93/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.1765 +Epoch [93/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2990, Pure Ratio2 10.1520 +Epoch [93/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.0275 +Epoch [93/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 9.9575 +Epoch [93/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0532, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 80.4788 % Model2 80.7292 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [94/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.8627 +Epoch [94/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7386 +Epoch [94/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.8627 +Epoch [94/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 9.9255 +Epoch [94/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9542, Pure Ratio2 9.9771 +Epoch [94/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 80.7592 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.8431 +Epoch [95/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.0784 +Epoch [95/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0458, Pure Ratio2 10.0261 +Epoch [95/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0882, Pure Ratio2 10.1078 +Epoch [95/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1137, Pure Ratio2 10.0784 +Epoch [95/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2516, Pure Ratio2 10.2026 +Epoch [95/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1821, Pure Ratio2 10.1569 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 80.2784 %, Pure Ratio 1 10.1156 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.3137 +Epoch [96/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.7647 +Epoch [96/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0261, Pure Ratio2 9.8889 +Epoch [96/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9706, Pure Ratio2 9.9020 +Epoch [96/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0314, Pure Ratio2 10.0039 +Epoch [96/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0327 +Epoch [96/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 79.6675 % Model2 80.5088 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.2353 +Epoch [97/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1863 +Epoch [97/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.1699 +Epoch [97/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 10.0245 +Epoch [97/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0235 +Epoch [97/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0654 +Epoch [97/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1092 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 79.7376 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.3922 +Epoch [98/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3627, Pure Ratio2 10.1176 +Epoch [98/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 10.0654 +Epoch [98/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0441, Pure Ratio2 10.0049 +Epoch [98/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9843, Pure Ratio2 9.9412 +Epoch [98/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.0392 +Epoch [98/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0672, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 80.3886 % Model2 80.9896 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 9.8627 +Epoch [99/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.1078 +Epoch [99/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.2418, Pure Ratio2 10.0131 +Epoch [99/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.0588 +Epoch [99/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3451, Pure Ratio2 10.1216 +Epoch [99/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2712, Pure Ratio2 10.0490 +Epoch [99/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1877, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 80.3285 % Model2 80.7292 %, Pure Ratio 1 10.1936 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.3529 +Epoch [100/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8333 +Epoch [100/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8758 +Epoch [100/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7990, Pure Ratio2 9.7304 +Epoch [100/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.8314 +Epoch [100/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9967 +Epoch [100/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 79.9579 % Model2 81.0296 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [101/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7157, Pure Ratio2 9.6765 +Epoch [101/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1046 +Epoch [101/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0441, Pure Ratio2 10.1176 +Epoch [101/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0627 +Epoch [101/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 9.9837 +Epoch [101/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 79.7376 % Model2 80.1282 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.1765 +Epoch [102/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [102/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0719 +Epoch [102/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1422, Pure Ratio2 10.0833 +Epoch [102/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1098, Pure Ratio2 10.0275 +Epoch [102/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0425, Pure Ratio2 9.9510 +Epoch [102/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 80.4487 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6275, Pure Ratio2 9.8627 +Epoch [103/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7745, Pure Ratio2 9.8627 +Epoch [103/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9869, Pure Ratio2 10.1046 +Epoch [103/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1569 +Epoch [103/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0431, Pure Ratio2 10.0275 +Epoch [103/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1536, Pure Ratio2 10.1046 +Epoch [103/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0812, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 80.4788 % Model2 80.2584 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3922, Pure Ratio2 10.4314 +Epoch [104/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9804 +Epoch [104/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.1307 +Epoch [104/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1225, Pure Ratio2 10.1569 +Epoch [104/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1294 +Epoch [104/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1405, Pure Ratio2 10.1242 +Epoch [104/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1429, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 81.3201 % Model2 80.4688 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 9.9020 +Epoch [105/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.0784 +Epoch [105/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 9.9804 +Epoch [105/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8775 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1059, Pure Ratio2 10.0471 +Epoch [105/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1536, Pure Ratio2 10.0980 +Epoch [105/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1036, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 79.5373 % Model2 81.3401 %, Pure Ratio 1 10.1156 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.9216 +Epoch [106/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0392 +Epoch [106/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0523 +Epoch [106/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0980 +Epoch [106/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0471 +Epoch [106/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 10.1013 +Epoch [106/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1036, Pure Ratio2 10.1765 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 79.8878 % Model2 80.8393 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3725 +Epoch [107/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.6078 +Epoch [107/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.2614 +Epoch [107/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.3725 +Epoch [107/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2471, Pure Ratio2 10.3686 +Epoch [107/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.2712 +Epoch [107/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1793, Pure Ratio2 10.2241 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 80.6190 % Model2 80.7592 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7843 +Epoch [108/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6961 +Epoch [108/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5163, Pure Ratio2 9.5294 +Epoch [108/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6029, Pure Ratio2 9.6618 +Epoch [108/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.8431 +Epoch [108/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0261 +Epoch [108/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0560, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 80.9896 % Model2 81.0697 %, Pure Ratio 1 10.1156 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 11.1569, Pure Ratio2 10.9608 +Epoch [109/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.3922 +Epoch [109/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5229, Pure Ratio2 10.3922 +Epoch [109/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4020, Pure Ratio2 10.3235 +Epoch [109/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.2471 +Epoch [109/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2190, Pure Ratio2 10.2190 +Epoch [109/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1905, Pure Ratio2 10.1681 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 80.2885 % Model2 79.6675 %, Pure Ratio 1 10.1207 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.3725 +Epoch [110/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.4706 +Epoch [110/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1699, Pure Ratio2 10.2418 +Epoch [110/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1912, Pure Ratio2 10.3039 +Epoch [110/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1451, Pure Ratio2 10.2353 +Epoch [110/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.1863 +Epoch [110/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9972, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 80.9996 % Model2 80.9896 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.1483 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9216 +Epoch [111/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2941 +Epoch [111/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.1307 +Epoch [111/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.2157 +Epoch [111/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1843 +Epoch [111/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0850, Pure Ratio2 10.1405 +Epoch [111/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0616, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 80.8794 % Model2 81.3301 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.0196 +Epoch [112/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.1569 +Epoch [112/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3922 +Epoch [112/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3039, Pure Ratio2 10.4020 +Epoch [112/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2627, Pure Ratio2 10.3294 +Epoch [112/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1438 +Epoch [112/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 80.5789 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0980 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.5098 +Epoch [113/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1961 +Epoch [113/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 10.0588 +Epoch [113/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2402 +Epoch [113/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.1373 +Epoch [113/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.1242 +Epoch [113/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1569 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 80.3486 % Model2 81.2099 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.5686 +Epoch [114/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2353 +Epoch [114/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.2745 +Epoch [114/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.1618 +Epoch [114/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2196, Pure Ratio2 10.2392 +Epoch [114/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.1601 +Epoch [114/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1092, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 81.5204 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 10.0392 +Epoch [115/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1961 +Epoch [115/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.4248 +Epoch [115/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2451 +Epoch [115/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1725, Pure Ratio2 10.2118 +Epoch [115/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1340, Pure Ratio2 10.1732 +Epoch [115/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 80.8193 % Model2 80.9295 %, Pure Ratio 1 10.0754 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1961 +Epoch [116/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1667 +Epoch [116/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9608 +Epoch [116/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9706 +Epoch [116/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.0275 +Epoch [116/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1144, Pure Ratio2 10.1111 +Epoch [116/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 80.9295 % Model2 80.8093 %, Pure Ratio 1 10.1156 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.4314 +Epoch [117/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.3529 +Epoch [117/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.3595 +Epoch [117/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 10.2010 +Epoch [117/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8980, Pure Ratio2 10.0275 +Epoch [117/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8562, Pure Ratio2 9.9771 +Epoch [117/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.9595 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [118/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0000 +Epoch [118/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0588 +Epoch [118/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9853 +Epoch [118/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0314 +Epoch [118/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0915 +Epoch [118/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 80.4387 % Model2 80.9595 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1569 +Epoch [119/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4216, Pure Ratio2 10.3824 +Epoch [119/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0654 +Epoch [119/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 9.9559 +Epoch [119/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.9451 +Epoch [119/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0294 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 80.7091 % Model2 80.8393 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.2941 +Epoch [120/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4608, Pure Ratio2 10.3922 +Epoch [120/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5229, Pure Ratio2 10.4575 +Epoch [120/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2598, Pure Ratio2 10.2010 +Epoch [120/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [120/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.0327 +Epoch [120/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 80.4888 % Model2 81.1098 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.1765 +Epoch [121/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9510 +Epoch [121/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 9.9542 +Epoch [121/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1127 +Epoch [121/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9255, Pure Ratio2 9.8784 +Epoch [121/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9673 +Epoch [121/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 81.1298 % Model2 80.6691 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.1373 +Epoch [122/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.9118 +Epoch [122/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [122/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8873 +Epoch [122/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 10.0118 +Epoch [122/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 10.0425 +Epoch [122/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1148, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 81.4403 %, Pure Ratio 1 10.1559 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4118, Pure Ratio2 10.4314 +Epoch [123/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.4902 +Epoch [123/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4837, Pure Ratio2 10.5752 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3775, Pure Ratio2 10.3971 +Epoch [123/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2039, Pure Ratio2 10.2275 +Epoch [123/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1242 +Epoch [123/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1317, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 79.7977 % Model2 80.0982 %, Pure Ratio 1 10.1533 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6667, Pure Ratio2 10.5686 +Epoch [124/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1078 +Epoch [124/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1634 +Epoch [124/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2304, Pure Ratio2 10.1863 +Epoch [124/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.2000 +Epoch [124/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.1993 +Epoch [124/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.1653 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 80.7492 % Model2 80.9595 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.1257 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.1961 +Epoch [125/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.2941 +Epoch [125/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4379, Pure Ratio2 10.3529 +Epoch [125/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3186, Pure Ratio2 10.1569 +Epoch [125/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2118, Pure Ratio2 10.0902 +Epoch [125/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2712, Pure Ratio2 10.1732 +Epoch [125/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1345, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 80.5689 %, Pure Ratio 1 10.1559 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.0392 +Epoch [126/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.5882 +Epoch [126/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6471 +Epoch [126/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 9.9412 +Epoch [126/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9059, Pure Ratio2 9.8784 +Epoch [126/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9281 +Epoch [126/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 81.1098 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.5294 +Epoch [127/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1863 +Epoch [127/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.2745 +Epoch [127/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1520, Pure Ratio2 10.2892 +Epoch [127/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 10.0706 +Epoch [127/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.1307 +Epoch [127/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0252, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.8494 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1357 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6275 +Epoch [128/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [128/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0261 +Epoch [128/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.1814 +Epoch [128/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2471, Pure Ratio2 10.1922 +Epoch [128/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2516 +Epoch [128/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1989, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 80.9395 % Model2 81.1899 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.1765 +Epoch [129/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.5784 +Epoch [129/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.6667 +Epoch [129/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.0343 +Epoch [129/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.0667 +Epoch [129/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0948, Pure Ratio2 10.0098 +Epoch [129/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 79.9679 % Model2 79.9980 %, Pure Ratio 1 10.1307 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.7059 +Epoch [130/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.2549 +Epoch [130/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.2484 +Epoch [130/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [130/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1020 +Epoch [130/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.1176 +Epoch [130/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 10.1092 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 80.6991 % Model2 80.9295 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.1307 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [131/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.2843 +Epoch [131/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2222 +Epoch [131/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2304, Pure Ratio2 10.2696 +Epoch [131/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2431, Pure Ratio2 10.2471 +Epoch [131/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2451 +Epoch [131/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1008, Pure Ratio2 10.1317 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 81.2200 % Model2 81.1198 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [132/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1078 +Epoch [132/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [132/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9608 +Epoch [132/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9961 +Epoch [132/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.1046 +Epoch [132/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 80.7792 % Model2 81.2600 %, Pure Ratio 1 10.1131 %, Pure Ratio 2 10.1533 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.4706 +Epoch [133/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.4412 +Epoch [133/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0065 +Epoch [133/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.0784, Pure Ratio2 10.0098 +Epoch [133/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1686, Pure Ratio2 10.1373 +Epoch [133/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0490 +Epoch [133/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1513, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 79.7576 % Model2 80.0381 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.3529 +Epoch [134/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1863, Pure Ratio2 10.3725 +Epoch [134/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.0719 +Epoch [134/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0539 +Epoch [134/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0314, Pure Ratio2 10.0902 +Epoch [134/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 10.0327 +Epoch [134/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 80.5689 % Model2 81.5605 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.2745 +Epoch [135/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.3431 +Epoch [135/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2876, Pure Ratio2 10.2484 +Epoch [135/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1618, Pure Ratio2 10.0882 +Epoch [135/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1882, Pure Ratio2 10.0980 +Epoch [135/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.1144 +Epoch [135/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1317, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 80.5489 % Model2 80.6390 %, Pure Ratio 1 10.1483 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8235, Pure Ratio2 10.7451 +Epoch [136/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0294 +Epoch [136/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.8039 +Epoch [136/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8922 +Epoch [136/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9843, Pure Ratio2 9.9137 +Epoch [136/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 9.9935 +Epoch [136/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 80.7692 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1257 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.3137 +Epoch [137/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.4020 +Epoch [137/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.2549 +Epoch [137/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.1618 +Epoch [137/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1216 +Epoch [137/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1307, Pure Ratio2 10.1863 +Epoch [137/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 10.2129 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 80.0781 % Model2 81.1699 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.1408 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2157 +Epoch [138/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8922 +Epoch [138/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0131 +Epoch [138/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0980 +Epoch [138/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9255, Pure Ratio2 10.0471 +Epoch [138/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0882 +Epoch [138/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0056, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 80.0581 % Model2 80.9295 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.1332 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.3137 +Epoch [139/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 9.9706 +Epoch [139/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8366 +Epoch [139/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 9.9363 +Epoch [139/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.1176 +Epoch [139/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.0654 +Epoch [139/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1709, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 80.3085 % Model2 80.8293 %, Pure Ratio 1 10.1282 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0000 +Epoch [140/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9314 +Epoch [140/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0261 +Epoch [140/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8039 +Epoch [140/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9647 +Epoch [140/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.1176 +Epoch [140/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 81.0797 % Model2 81.2099 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7843 +Epoch [141/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9902 +Epoch [141/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0784 +Epoch [141/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 10.0588 +Epoch [141/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 10.0000 +Epoch [141/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 10.0098 +Epoch [141/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0924, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 80.1783 % Model2 81.1799 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [142/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5980 +Epoch [142/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9608 +Epoch [142/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9853 +Epoch [142/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 10.0235 +Epoch [142/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 10.0000 +Epoch [142/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8936, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 80.7893 % Model2 81.6006 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.2353 +Epoch [143/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0098 +Epoch [143/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9216 +Epoch [143/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.1078 +Epoch [143/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2039, Pure Ratio2 10.2000 +Epoch [143/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1078 +Epoch [143/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0364, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 81.0397 % Model2 81.2901 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.3333 +Epoch [144/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0882 +Epoch [144/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9739 +Epoch [144/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 10.1520, Pure Ratio2 10.1225 +Epoch [144/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 10.1882, Pure Ratio2 10.1647 +Epoch [144/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1013 +Epoch [144/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1036, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 80.8293 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.5784 +Epoch [145/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8562 +Epoch [145/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1225, Pure Ratio2 10.0980 +Epoch [145/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.0510 +Epoch [145/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.0359 +Epoch [145/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 80.3085 % Model2 81.1298 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.1508 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.9020 +Epoch [146/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.2059 +Epoch [146/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.1307 +Epoch [146/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0882 +Epoch [146/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0431, Pure Ratio2 10.1843 +Epoch [146/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.1242 +Epoch [146/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1765 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 80.8193 % Model2 81.2500 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3725 +Epoch [147/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3235 +Epoch [147/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3399, Pure Ratio2 10.3399 +Epoch [147/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2402, Pure Ratio2 10.1912 +Epoch [147/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2471, Pure Ratio2 10.1843 +Epoch [147/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.2810, Pure Ratio2 10.2418 +Epoch [147/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2409, Pure Ratio2 10.1821 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 81.1298 %, Pure Ratio 1 10.1684 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.4510 +Epoch [148/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7157 +Epoch [148/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.7712 +Epoch [148/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 9.8922 +Epoch [148/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1647, Pure Ratio2 10.0353 +Epoch [148/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.9281 +Epoch [148/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 80.4587 %, Pure Ratio 1 10.1383 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.3922, Pure Ratio2 9.6667 +Epoch [149/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 10.0196 +Epoch [149/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Epoch [149/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [149/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0196 +Epoch [149/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 10.0915 +Epoch [149/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.7492 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0980 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [150/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3824, Pure Ratio2 10.2745 +Epoch [150/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0588 +Epoch [150/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1520, Pure Ratio2 10.1078 +Epoch [150/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1294, Pure Ratio2 10.0902 +Epoch [150/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2124, Pure Ratio2 10.1797 +Epoch [150/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1625, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 81.2700 % Model2 81.3802 %, Pure Ratio 1 10.1709 %, Pure Ratio 2 10.1458 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.3725 +Epoch [151/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 10.1373 +Epoch [151/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.4837 +Epoch [151/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2990 +Epoch [151/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1490, Pure Ratio2 10.2588 +Epoch [151/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1536, Pure Ratio2 10.2157 +Epoch [151/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2101, Pure Ratio2 10.2857 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 81.0497 % Model2 81.4303 %, Pure Ratio 1 10.1257 %, Pure Ratio 2 10.1835 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 10.1176 +Epoch [152/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [152/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0915 +Epoch [152/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.9265 +Epoch [152/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [152/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 9.9641 +Epoch [152/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 80.7392 % Model2 80.7692 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.2157 +Epoch [153/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1863 +Epoch [153/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.1699 +Epoch [153/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.0784 +Epoch [153/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0627, Pure Ratio2 10.1882 +Epoch [153/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1536, Pure Ratio2 10.2386 +Epoch [153/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1232, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 80.3986 % Model2 80.4688 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.5294 +Epoch [154/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3235, Pure Ratio2 10.3431 +Epoch [154/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.3072 +Epoch [154/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1029 +Epoch [154/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0471, Pure Ratio2 10.0510 +Epoch [154/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0850 +Epoch [154/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 81.3902 % Model2 81.2300 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0588 +Epoch [155/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1471 +Epoch [155/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0327 +Epoch [155/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.1520 +Epoch [155/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1647 +Epoch [155/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2418, Pure Ratio2 10.2451 +Epoch [155/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2913, Pure Ratio2 10.2521 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 81.1599 %, Pure Ratio 1 10.1584 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 9.8235 +Epoch [156/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8235 +Epoch [156/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9739 +Epoch [156/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.1225 +Epoch [156/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2588, Pure Ratio2 10.1373 +Epoch [156/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.1405 +Epoch [156/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 80.9095 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7059 +Epoch [157/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [157/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [157/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.1373 +Epoch [157/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.1137 +Epoch [157/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.0458 +Epoch [157/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 81.4203 % Model2 81.3201 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.5490 +Epoch [158/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2353 +Epoch [158/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9739 +Epoch [158/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8676 +Epoch [158/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8314 +Epoch [158/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.8824 +Epoch [158/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 80.6290 % Model2 80.8594 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.6667 +Epoch [159/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8137 +Epoch [159/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8235 +Epoch [159/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9118 +Epoch [159/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.9725 +Epoch [159/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.0556 +Epoch [159/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 80.8694 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5882, Pure Ratio2 10.6078 +Epoch [160/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3431, Pure Ratio2 10.2549 +Epoch [160/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4248, Pure Ratio2 10.3464 +Epoch [160/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2304 +Epoch [160/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3490, Pure Ratio2 10.3098 +Epoch [160/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2582, Pure Ratio2 10.2190 +Epoch [160/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 81.3902 % Model2 81.0196 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.4314 +Epoch [161/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3824, Pure Ratio2 10.3824 +Epoch [161/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0915 +Epoch [161/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1422, Pure Ratio2 10.1373 +Epoch [161/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2588, Pure Ratio2 10.2627 +Epoch [161/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.1732 +Epoch [161/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1345, Pure Ratio2 10.1849 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 80.8694 % Model2 80.6490 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.8431 +Epoch [162/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.3529 +Epoch [162/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1895 +Epoch [162/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0637 +Epoch [162/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0667 +Epoch [162/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.0359 +Epoch [162/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 80.8994 % Model2 81.0497 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0980 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0588 +Epoch [163/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.0588 +Epoch [163/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0327 +Epoch [163/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.9657 +Epoch [163/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9686 +Epoch [163/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1209 +Epoch [163/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 81.2600 % Model2 81.0597 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0392 +Epoch [164/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.9412 +Epoch [164/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0000 +Epoch [164/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9657 +Epoch [164/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0980 +Epoch [164/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.1144 +Epoch [164/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 81.2901 % Model2 81.0998 %, Pure Ratio 1 10.1282 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6275 +Epoch [165/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8137 +Epoch [165/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7516 +Epoch [165/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8873 +Epoch [165/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 10.0314 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9804 +Epoch [165/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 81.0397 % Model2 81.2400 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.0784 +Epoch [166/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 9.9216 +Epoch [166/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.8889 +Epoch [166/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2010, Pure Ratio2 10.0686 +Epoch [166/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1059, Pure Ratio2 9.9725 +Epoch [166/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.0000 +Epoch [166/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1457, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 81.2700 % Model2 81.1298 %, Pure Ratio 1 10.1332 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.4510 +Epoch [167/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [167/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 9.9477 +Epoch [167/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9412 +Epoch [167/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0941 +Epoch [167/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.0980 +Epoch [167/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2101, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 80.5288 %, Pure Ratio 1 10.1533 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5686, Pure Ratio2 10.5098 +Epoch [168/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4216, Pure Ratio2 10.4216 +Epoch [168/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3464 +Epoch [168/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.3431 +Epoch [168/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3176, Pure Ratio2 10.2902 +Epoch [168/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4085, Pure Ratio2 10.4020 +Epoch [168/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1933 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.7292 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [169/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9706 +Epoch [169/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1634 +Epoch [169/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0343 +Epoch [169/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0706, Pure Ratio2 10.0510 +Epoch [169/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.0654 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 80.5789 % Model2 81.5505 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [170/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1078 +Epoch [170/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0196 +Epoch [170/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1520, Pure Ratio2 10.1667 +Epoch [170/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0196 +Epoch [170/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0000 +Epoch [170/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1148, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 81.2500 % Model2 81.2600 %, Pure Ratio 1 10.1131 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0588 +Epoch [171/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 9.9804 +Epoch [171/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.1111 +Epoch [171/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1324 +Epoch [171/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1725, Pure Ratio2 10.0471 +Epoch [171/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 9.9869 +Epoch [171/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2017, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 80.9295 % Model2 81.0998 %, Pure Ratio 1 10.2187 %, Pure Ratio 2 10.0603 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.1765 +Epoch [172/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7451, Pure Ratio2 10.5098 +Epoch [172/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5556, Pure Ratio2 10.3529 +Epoch [172/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4069, Pure Ratio2 10.1961 +Epoch [172/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4431, Pure Ratio2 10.2392 +Epoch [172/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3007, Pure Ratio2 10.1144 +Epoch [172/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2325, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 80.4487 % Model2 81.2700 %, Pure Ratio 1 10.2036 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6471 +Epoch [173/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7941 +Epoch [173/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.8039 +Epoch [173/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9902 +Epoch [173/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.9373 +Epoch [173/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9739 +Epoch [173/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 81.1298 % Model2 81.4303 %, Pure Ratio 1 10.1131 %, Pure Ratio 2 10.1408 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0392 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.2353 +Epoch [174/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4248, Pure Ratio2 10.2157 +Epoch [174/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.4216, Pure Ratio2 10.2598 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.2471 +Epoch [174/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2092, Pure Ratio2 10.1373 +Epoch [174/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1036, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 80.2784 % Model2 81.1899 %, Pure Ratio 1 10.1709 %, Pure Ratio 2 10.1584 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9216 +Epoch [175/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8235 +Epoch [175/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8039 +Epoch [175/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8578 +Epoch [175/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9804 +Epoch [175/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1536, Pure Ratio2 10.0621 +Epoch [175/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1737, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 80.8293 % Model2 81.5004 %, Pure Ratio 1 10.1684 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.8824 +Epoch [176/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0784 +Epoch [176/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0458 +Epoch [176/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2402, Pure Ratio2 10.1716 +Epoch [176/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.1333 +Epoch [176/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 9.9608 +Epoch [176/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.1905, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 81.4002 % Model2 81.7007 %, Pure Ratio 1 10.1785 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.2157 +Epoch [177/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.5784 +Epoch [177/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6928 +Epoch [177/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0245 +Epoch [177/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.9529 +Epoch [177/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.0359 +Epoch [177/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 81.1699 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.2745 +Epoch [178/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.3039 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.2092 +Epoch [178/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.1961 +Epoch [178/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1412, Pure Ratio2 10.0863 +Epoch [178/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.0065 +Epoch [178/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1429, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 80.6991 % Model2 81.3401 %, Pure Ratio 1 10.1081 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8235 +Epoch [179/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8137 +Epoch [179/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [179/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.1127 +Epoch [179/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1216, Pure Ratio2 10.1686 +Epoch [179/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1797, Pure Ratio2 10.2941 +Epoch [179/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 80.4988 % Model2 81.0697 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [180/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.3529 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.2418 +Epoch [180/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.1912 +Epoch [180/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1529 +Epoch [180/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [180/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0756, Pure Ratio2 10.1681 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 80.8494 % Model2 81.6907 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.1408 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3922 +Epoch [181/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5392, Pure Ratio2 10.4412 +Epoch [181/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.3987 +Epoch [181/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2010, Pure Ratio2 10.0931 +Epoch [181/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2471, Pure Ratio2 10.2118 +Epoch [181/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1144, Pure Ratio2 10.0850 +Epoch [181/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 80.5489 % Model2 81.4103 %, Pure Ratio 1 10.1810 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3922 +Epoch [182/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0588 +Epoch [182/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0654 +Epoch [182/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1569 +Epoch [182/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0745 +Epoch [182/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9967 +Epoch [182/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 81.4503 % Model2 81.3001 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9608 +Epoch [183/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.9314 +Epoch [183/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.1176 +Epoch [183/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.2696 +Epoch [183/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.1529 +Epoch [183/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0752 +Epoch [183/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1345, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 80.9495 % Model2 81.3702 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2941 +Epoch [184/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9902 +Epoch [184/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9412 +Epoch [184/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0294 +Epoch [184/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1608, Pure Ratio2 10.1020 +Epoch [184/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2680, Pure Ratio2 10.2288 +Epoch [184/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2213, Pure Ratio2 10.1681 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 80.2284 % Model2 81.0597 %, Pure Ratio 1 10.1408 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.8235 +Epoch [185/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8922 +Epoch [185/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 10.0261 +Epoch [185/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.1176 +Epoch [185/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0824 +Epoch [185/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0654 +Epoch [185/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 80.3185 % Model2 81.1899 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [186/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.5980 +Epoch [186/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8235 +Epoch [186/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9216 +Epoch [186/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 10.0392 +Epoch [186/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0915 +Epoch [186/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1148 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 80.8994 % Model2 80.9095 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.3333 +Epoch [187/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.5000 +Epoch [187/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.1307 +Epoch [187/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0196 +Epoch [187/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.2314 +Epoch [187/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2092 +Epoch [187/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1541 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 81.6607 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8039, Pure Ratio2 10.4314 +Epoch [188/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5784, Pure Ratio2 10.2745 +Epoch [188/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3399, Pure Ratio2 10.1503 +Epoch [188/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [188/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 9.9373 +Epoch [188/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0261 +Epoch [188/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.9796 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3137 +Epoch [189/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2255 +Epoch [189/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Epoch [189/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 10.0049 +Epoch [189/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9647 +Epoch [189/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0621 +Epoch [189/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2045, Pure Ratio2 10.1849 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 81.3101 % Model2 80.8093 %, Pure Ratio 1 10.1810 %, Pure Ratio 2 10.1735 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.2745 +Epoch [190/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.3627 +Epoch [190/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8954 +Epoch [190/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.9804 +Epoch [190/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.9608 +Epoch [190/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9542 +Epoch [190/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0924, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 81.2700 % Model2 81.0096 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1373 +Epoch [191/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7451 +Epoch [191/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0719 +Epoch [191/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [191/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0157, Pure Ratio2 9.9804 +Epoch [191/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0621 +Epoch [191/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 80.9595 % Model2 81.3802 %, Pure Ratio 1 10.1684 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7059, Pure Ratio2 10.5294 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.4902 +Epoch [192/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5033, Pure Ratio2 10.3856 +Epoch [192/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2696 +Epoch [192/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1608, Pure Ratio2 10.1294 +Epoch [192/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1667 +Epoch [192/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2185, Pure Ratio2 10.1765 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 81.3802 % Model2 81.5004 %, Pure Ratio 1 10.1533 %, Pure Ratio 2 10.1332 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7647, Pure Ratio2 10.5686 +Epoch [193/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.4314 +Epoch [193/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6013, Pure Ratio2 10.4444 +Epoch [193/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3431, Pure Ratio2 10.1765 +Epoch [193/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4235, Pure Ratio2 10.2863 +Epoch [193/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3203, Pure Ratio2 10.1536 +Epoch [193/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2437, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 81.5304 % Model2 81.2800 %, Pure Ratio 1 10.2011 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0392 +Epoch [194/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1569 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2614, Pure Ratio2 10.2941 +Epoch [194/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2108, Pure Ratio2 10.2353 +Epoch [194/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1922 +Epoch [194/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.2353 +Epoch [194/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 81.4804 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.8039 +Epoch [195/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5784 +Epoch [195/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8693 +Epoch [195/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.9902 +Epoch [195/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.7843 +Epoch [195/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.8922 +Epoch [195/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1008, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 81.4103 % Model2 81.6607 %, Pure Ratio 1 10.1760 %, Pure Ratio 2 10.1709 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8431, Pure Ratio2 10.6078 +Epoch [196/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3431, Pure Ratio2 10.3725 +Epoch [196/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3660, Pure Ratio2 10.3725 +Epoch [196/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.2549 +Epoch [196/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2706, Pure Ratio2 10.2706 +Epoch [196/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2190, Pure Ratio2 10.1928 +Epoch [196/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.1092 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 80.7893 % Model2 81.5104 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.1106 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9412 +Epoch [197/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2451 +Epoch [197/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1895 +Epoch [197/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.2255 +Epoch [197/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1412, Pure Ratio2 10.2039 +Epoch [197/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1993 +Epoch [197/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 81.0697 % Model2 81.5505 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1659 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3333 +Epoch [198/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4902 +Epoch [198/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2484 +Epoch [198/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0245 +Epoch [198/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.9176 +Epoch [198/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9542 +Epoch [198/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0280, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 80.5589 % Model2 81.7007 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.4902 +Epoch [199/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3137 +Epoch [199/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [199/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [199/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9882 +Epoch [199/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9575 +Epoch [199/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 81.6406 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6863, Pure Ratio2 10.3922 +Epoch [200/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3627, Pure Ratio2 10.2255 +Epoch [200/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2810, Pure Ratio2 10.1373 +Epoch [200/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2598, Pure Ratio2 10.1716 +Epoch [200/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1529, Pure Ratio2 10.0745 +Epoch [200/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.1209 +Epoch [200/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 80.9796 % Model2 81.1599 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0151 % diff --git a/other_methods/coteaching/coteaching_results/out_0_4.log b/other_methods/coteaching/coteaching_results/out_0_4.log new file mode 100644 index 0000000..992ea2b --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_0_4.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 21.0938, Training Accuracy2: 21.8750, Loss1: 0.0165, Loss2: 0.0167, Pure Ratio1: 9.9360, Pure Ratio2 9.9680 +Epoch [2/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 36.7188, Loss1: 0.0154, Loss2: 0.0152, Pure Ratio1: 10.1120, Pure Ratio2 10.0880 +Epoch [2/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 29.6875, Loss1: 0.0158, Loss2: 0.0164, Pure Ratio1: 9.9680, Pure Ratio2 9.9467 +Epoch [2/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0163, Loss2: 0.0157, Pure Ratio1: 10.1040, Pure Ratio2 10.1080 +Epoch [2/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0157, Loss2: 0.0154, Pure Ratio1: 10.0992, Pure Ratio2 10.0896 +Epoch [2/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0144, Loss2: 0.0150, Pure Ratio1: 10.0827, Pure Ratio2 10.0667 +Epoch [2/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0147, Loss2: 0.0150, Pure Ratio1: 9.9360, Pure Ratio2 9.9131 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 46.6647 % Model2 48.3373 %, Pure Ratio 1 9.9364 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 43.7500, Loss1: 0.0141, Loss2: 0.0134, Pure Ratio1: 9.2295, Pure Ratio2 9.1967 +Epoch [3/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0140, Pure Ratio1: 9.5574, Pure Ratio2 9.5738 +Epoch [3/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0140, Loss2: 0.0136, Pure Ratio1: 9.7760, Pure Ratio2 9.7814 +Epoch [3/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0137, Loss2: 0.0134, Pure Ratio1: 9.8156, Pure Ratio2 9.8361 +Epoch [3/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0145, Loss2: 0.0142, Pure Ratio1: 9.8590, Pure Ratio2 9.8918 +Epoch [3/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0132, Loss2: 0.0130, Pure Ratio1: 9.9098, Pure Ratio2 9.9372 +Epoch [3/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0137, Loss2: 0.0137, Pure Ratio1: 9.9157, Pure Ratio2 9.9297 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 53.8762 % Model2 57.0913 %, Pure Ratio 1 9.9201 %, Pure Ratio 2 9.9222 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0130, Loss2: 0.0135, Pure Ratio1: 9.2773, Pure Ratio2 9.1933 +Epoch [4/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0147, Loss2: 0.0144, Pure Ratio1: 9.5714, Pure Ratio2 9.5210 +Epoch [4/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0132, Loss2: 0.0130, Pure Ratio1: 9.7143, Pure Ratio2 9.6919 +Epoch [4/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0128, Loss2: 0.0126, Pure Ratio1: 9.6387, Pure Ratio2 9.6218 +Epoch [4/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0134, Loss2: 0.0134, Pure Ratio1: 9.7076, Pure Ratio2 9.6672 +Epoch [4/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0123, Loss2: 0.0130, Pure Ratio1: 9.8179, Pure Ratio2 9.7759 +Epoch [4/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0114, Loss2: 0.0114, Pure Ratio1: 9.8535, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 62.7604 % Model2 61.7488 %, Pure Ratio 1 9.8880 %, Pure Ratio 2 9.8686 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0143, Loss2: 0.0144, Pure Ratio1: 10.1034, Pure Ratio2 10.0690 +Epoch [5/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0115, Loss2: 0.0113, Pure Ratio1: 10.1983, Pure Ratio2 10.2241 +Epoch [5/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0130, Loss2: 0.0134, Pure Ratio1: 10.1092, Pure Ratio2 10.0862 +Epoch [5/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0122, Loss2: 0.0117, Pure Ratio1: 9.9052, Pure Ratio2 9.9138 +Epoch [5/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0110, Loss2: 0.0105, Pure Ratio1: 10.0621, Pure Ratio2 10.0414 +Epoch [5/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0122, Loss2: 0.0117, Pure Ratio1: 10.0029, Pure Ratio2 9.9971 +Epoch [5/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0125, Loss2: 0.0122, Pure Ratio1: 9.9384, Pure Ratio2 9.9360 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 60.8073 % Model2 63.4716 %, Pure Ratio 1 9.8696 %, Pure Ratio 2 9.8762 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 29.6875, Loss1: 0.0152, Loss2: 0.0153, Pure Ratio1: 10.0354, Pure Ratio2 9.9292 +Epoch [6/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0137, Pure Ratio1: 10.0796, Pure Ratio2 10.0708 +Epoch [6/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0119, Loss2: 0.0120, Pure Ratio1: 9.9705, Pure Ratio2 9.9823 +Epoch [6/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0125, Loss2: 0.0123, Pure Ratio1: 9.9381, Pure Ratio2 9.9381 +Epoch [6/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0127, Loss2: 0.0127, Pure Ratio1: 9.8442, Pure Ratio2 9.8407 +Epoch [6/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0126, Loss2: 0.0126, Pure Ratio1: 9.7876, Pure Ratio2 9.8171 +Epoch [6/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0108, Loss2: 0.0112, Pure Ratio1: 9.8255, Pure Ratio2 9.8382 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 64.1526 % Model2 61.5585 %, Pure Ratio 1 9.8797 %, Pure Ratio 2 9.8979 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0115, Loss2: 0.0119, Pure Ratio1: 10.2545, Pure Ratio2 10.1455 +Epoch [7/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0120, Loss2: 0.0118, Pure Ratio1: 10.0455, Pure Ratio2 10.0273 +Epoch [7/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0125, Loss2: 0.0124, Pure Ratio1: 10.0970, Pure Ratio2 10.1030 +Epoch [7/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0103, Loss2: 0.0105, Pure Ratio1: 9.9364, Pure Ratio2 9.9545 +Epoch [7/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0109, Loss2: 0.0118, Pure Ratio1: 9.9636, Pure Ratio2 9.9709 +Epoch [7/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0104, Loss2: 0.0104, Pure Ratio1: 9.9152, Pure Ratio2 9.9606 +Epoch [7/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0119, Loss2: 0.0118, Pure Ratio1: 9.9039, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 69.5913 % Model2 69.7115 %, Pure Ratio 1 9.8578 %, Pure Ratio 2 9.9207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0097, Loss2: 0.0094, Pure Ratio1: 10.0000, Pure Ratio2 10.2593 +Epoch [8/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0103, Loss2: 0.0102, Pure Ratio1: 9.6667, Pure Ratio2 9.9074 +Epoch [8/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0107, Loss2: 0.0105, Pure Ratio1: 9.5988, Pure Ratio2 9.7593 +Epoch [8/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0096, Loss2: 0.0096, Pure Ratio1: 9.4954, Pure Ratio2 9.5926 +Epoch [8/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0108, Loss2: 0.0113, Pure Ratio1: 9.6926, Pure Ratio2 9.7667 +Epoch [8/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0103, Loss2: 0.0097, Pure Ratio1: 9.8086, Pure Ratio2 9.8395 +Epoch [8/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0111, Loss2: 0.0104, Pure Ratio1: 9.7778, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 70.9936 % Model2 70.8133 %, Pure Ratio 1 9.8504 %, Pure Ratio 2 9.8670 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0088, Loss2: 0.0086, Pure Ratio1: 9.5810, Pure Ratio2 9.6381 +Epoch [9/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0100, Pure Ratio1: 9.6000, Pure Ratio2 9.6476 +Epoch [9/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0077, Loss2: 0.0080, Pure Ratio1: 9.6889, Pure Ratio2 9.7333 +Epoch [9/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0109, Loss2: 0.0113, Pure Ratio1: 9.6810, Pure Ratio2 9.7000 +Epoch [9/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0110, Loss2: 0.0106, Pure Ratio1: 9.8971, Pure Ratio2 9.9238 +Epoch [9/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0098, Loss2: 0.0099, Pure Ratio1: 9.8921, Pure Ratio2 9.9079 +Epoch [9/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0106, Loss2: 0.0112, Pure Ratio1: 9.9102, Pure Ratio2 9.9156 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 71.4844 % Model2 72.2055 %, Pure Ratio 1 9.9292 %, Pure Ratio 2 9.9438 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0089, Loss2: 0.0096, Pure Ratio1: 10.0980, Pure Ratio2 10.2549 +Epoch [10/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0092, Loss2: 0.0094, Pure Ratio1: 10.3431, Pure Ratio2 10.3039 +Epoch [10/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0112, Loss2: 0.0103, Pure Ratio1: 9.9477, Pure Ratio2 9.9281 +Epoch [10/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0108, Loss2: 0.0111, Pure Ratio1: 9.9069, Pure Ratio2 9.8431 +Epoch [10/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0099, Loss2: 0.0097, Pure Ratio1: 9.9451, Pure Ratio2 9.8863 +Epoch [10/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0120, Loss2: 0.0120, Pure Ratio1: 9.8758, Pure Ratio2 9.8203 +Epoch [10/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0090, Loss2: 0.0085, Pure Ratio1: 9.8964, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 68.8401 % Model2 68.8802 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0079, Loss2: 0.0079, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [11/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0101, Loss2: 0.0094, Pure Ratio1: 9.9510, Pure Ratio2 9.9020 +Epoch [11/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0095, Loss2: 0.0097, Pure Ratio1: 9.9477, Pure Ratio2 9.8954 +Epoch [11/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0093, Loss2: 0.0086, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [11/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0087, Loss2: 0.0091, Pure Ratio1: 9.8353, Pure Ratio2 9.8196 +Epoch [11/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0100, Loss2: 0.0100, Pure Ratio1: 9.8660, Pure Ratio2 9.8497 +Epoch [11/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0099, Loss2: 0.0105, Pure Ratio1: 9.8431, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 73.3574 % Model2 73.5276 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0091, Loss2: 0.0087, Pure Ratio1: 9.4314, Pure Ratio2 9.2941 +Epoch [12/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0072, Loss2: 0.0076, Pure Ratio1: 9.9608, Pure Ratio2 9.9314 +Epoch [12/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0060, Loss2: 0.0061, Pure Ratio1: 9.9739, Pure Ratio2 9.9477 +Epoch [12/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0075, Loss2: 0.0071, Pure Ratio1: 9.8824, Pure Ratio2 9.8775 +Epoch [12/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0095, Loss2: 0.0092, Pure Ratio1: 9.8588, Pure Ratio2 9.8235 +Epoch [12/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0109, Loss2: 0.0097, Pure Ratio1: 9.8235, Pure Ratio2 9.8203 +Epoch [12/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0070, Loss2: 0.0066, Pure Ratio1: 9.8599, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 71.1739 % Model2 70.5929 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0089, Loss2: 0.0089, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [13/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0090, Loss2: 0.0094, Pure Ratio1: 10.2255, Pure Ratio2 10.1765 +Epoch [13/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0075, Loss2: 0.0073, Pure Ratio1: 9.7778, Pure Ratio2 9.7451 +Epoch [13/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0082, Loss2: 0.0079, Pure Ratio1: 9.6912, Pure Ratio2 9.6912 +Epoch [13/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0106, Loss2: 0.0111, Pure Ratio1: 9.8549, Pure Ratio2 9.8196 +Epoch [13/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0089, Loss2: 0.0097, Pure Ratio1: 9.9706, Pure Ratio2 9.9150 +Epoch [13/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0108, Loss2: 0.0107, Pure Ratio1: 9.9496, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 70.7933 % Model2 71.5144 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0073, Loss2: 0.0073, Pure Ratio1: 9.6863, Pure Ratio2 9.8039 +Epoch [14/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0106, Loss2: 0.0113, Pure Ratio1: 9.5294, Pure Ratio2 9.5784 +Epoch [14/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0089, Loss2: 0.0101, Pure Ratio1: 9.6797, Pure Ratio2 9.7124 +Epoch [14/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0091, Loss2: 0.0083, Pure Ratio1: 9.8284, Pure Ratio2 9.8922 +Epoch [14/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0084, Loss2: 0.0087, Pure Ratio1: 9.7882, Pure Ratio2 9.8235 +Epoch [14/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0105, Loss2: 0.0101, Pure Ratio1: 9.7190, Pure Ratio2 9.7386 +Epoch [14/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 49.2188, Loss1: 0.0080, Loss2: 0.0083, Pure Ratio1: 9.8039, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 73.6979 % Model2 73.0769 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0069, Loss2: 0.0072, Pure Ratio1: 9.0980, Pure Ratio2 9.0784 +Epoch [15/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0080, Loss2: 0.0076, Pure Ratio1: 9.5294, Pure Ratio2 9.5392 +Epoch [15/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0090, Loss2: 0.0089, Pure Ratio1: 9.7582, Pure Ratio2 9.7647 +Epoch [15/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0100, Loss2: 0.0102, Pure Ratio1: 9.9265, Pure Ratio2 9.9363 +Epoch [15/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0088, Loss2: 0.0086, Pure Ratio1: 9.8431, Pure Ratio2 9.8784 +Epoch [15/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0064, Loss2: 0.0073, Pure Ratio1: 9.9150, Pure Ratio2 9.9346 +Epoch [15/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0080, Loss2: 0.0079, Pure Ratio1: 9.8908, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 71.7248 % Model2 73.1070 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0102, Loss2: 0.0095, Pure Ratio1: 9.4510, Pure Ratio2 9.5294 +Epoch [16/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.0000, Loss1: 0.0078, Loss2: 0.0088, Pure Ratio1: 9.2549, Pure Ratio2 9.2255 +Epoch [16/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0090, Loss2: 0.0095, Pure Ratio1: 9.6340, Pure Ratio2 9.6275 +Epoch [16/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0057, Loss2: 0.0070, Pure Ratio1: 9.7549, Pure Ratio2 9.7647 +Epoch [16/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0066, Loss2: 0.0072, Pure Ratio1: 9.7804, Pure Ratio2 9.8078 +Epoch [16/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0119, Loss2: 0.0098, Pure Ratio1: 9.7712, Pure Ratio2 9.7941 +Epoch [16/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0078, Loss2: 0.0091, Pure Ratio1: 9.8179, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 70.2624 % Model2 69.7015 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0055, Loss2: 0.0062, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [17/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0053, Loss2: 0.0050, Pure Ratio1: 9.9412, Pure Ratio2 9.8824 +Epoch [17/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0061, Loss2: 0.0061, Pure Ratio1: 9.9739, Pure Ratio2 9.9869 +Epoch [17/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0099, Loss2: 0.0107, Pure Ratio1: 9.9020, Pure Ratio2 9.9755 +Epoch [17/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0064, Loss2: 0.0059, Pure Ratio1: 9.9255, Pure Ratio2 9.9804 +Epoch [17/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0082, Loss2: 0.0075, Pure Ratio1: 9.9641, Pure Ratio2 9.9837 +Epoch [17/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0087, Loss2: 0.0093, Pure Ratio1: 9.8375, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 69.1506 % Model2 69.0004 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0049, Loss2: 0.0045, Pure Ratio1: 9.8235, Pure Ratio2 9.7647 +Epoch [18/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0082, Loss2: 0.0091, Pure Ratio1: 9.8235, Pure Ratio2 9.7059 +Epoch [18/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0080, Loss2: 0.0075, Pure Ratio1: 9.9935, Pure Ratio2 9.9216 +Epoch [18/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0058, Loss2: 0.0062, Pure Ratio1: 9.7892, Pure Ratio2 9.7451 +Epoch [18/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0079, Loss2: 0.0076, Pure Ratio1: 9.7765, Pure Ratio2 9.7529 +Epoch [18/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0060, Loss2: 0.0066, Pure Ratio1: 9.7843, Pure Ratio2 9.7484 +Epoch [18/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0069, Loss2: 0.0070, Pure Ratio1: 9.8739, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 69.9720 % Model2 70.4327 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0055, Loss2: 0.0057, Pure Ratio1: 9.5882, Pure Ratio2 9.5882 +Epoch [19/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0077, Loss2: 0.0079, Pure Ratio1: 9.4608, Pure Ratio2 9.5392 +Epoch [19/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0072, Loss2: 0.0064, Pure Ratio1: 9.5098, Pure Ratio2 9.5229 +Epoch [19/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0051, Loss2: 0.0054, Pure Ratio1: 9.5245, Pure Ratio2 9.4951 +Epoch [19/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0064, Loss2: 0.0077, Pure Ratio1: 9.5961, Pure Ratio2 9.5882 +Epoch [19/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0079, Loss2: 0.0087, Pure Ratio1: 9.6340, Pure Ratio2 9.6111 +Epoch [19/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0059, Loss2: 0.0061, Pure Ratio1: 9.7759, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 67.4279 % Model2 66.5264 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0039, Loss2: 0.0040, Pure Ratio1: 10.1765, Pure Ratio2 10.2549 +Epoch [20/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0038, Loss2: 0.0041, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Epoch [20/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0066, Loss2: 0.0061, Pure Ratio1: 10.1046, Pure Ratio2 10.0850 +Epoch [20/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0050, Loss2: 0.0060, Pure Ratio1: 10.0833, Pure Ratio2 10.0735 +Epoch [20/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0074, Loss2: 0.0085, Pure Ratio1: 10.0549, Pure Ratio2 10.0235 +Epoch [20/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0045, Loss2: 0.0042, Pure Ratio1: 10.0654, Pure Ratio2 10.0163 +Epoch [20/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0053, Loss2: 0.0059, Pure Ratio1: 9.8852, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 67.1775 % Model2 67.6883 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0047, Loss2: 0.0046, Pure Ratio1: 10.2549, Pure Ratio2 10.1569 +Epoch [21/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0035, Loss2: 0.0036, Pure Ratio1: 9.8137, Pure Ratio2 9.7353 +Epoch [21/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0049, Loss2: 0.0049, Pure Ratio1: 9.9542, Pure Ratio2 9.8301 +Epoch [21/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0048, Loss2: 0.0048, Pure Ratio1: 10.1324, Pure Ratio2 10.0343 +Epoch [21/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0053, Loss2: 0.0039, Pure Ratio1: 10.0667, Pure Ratio2 10.0275 +Epoch [21/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0052, Loss2: 0.0055, Pure Ratio1: 10.0000, Pure Ratio2 9.9542 +Epoch [21/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0056, Loss2: 0.0057, Pure Ratio1: 9.8908, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 67.1975 % Model2 66.5264 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0077, Loss2: 0.0073, Pure Ratio1: 9.3333, Pure Ratio2 9.2745 +Epoch [22/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0043, Loss2: 0.0048, Pure Ratio1: 9.7157, Pure Ratio2 9.6471 +Epoch [22/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0030, Loss2: 0.0036, Pure Ratio1: 9.7516, Pure Ratio2 9.6863 +Epoch [22/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0046, Loss2: 0.0059, Pure Ratio1: 9.8284, Pure Ratio2 9.7990 +Epoch [22/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0065, Loss2: 0.0058, Pure Ratio1: 9.9255, Pure Ratio2 9.8706 +Epoch [22/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0055, Loss2: 0.0045, Pure Ratio1: 9.8725, Pure Ratio2 9.7974 +Epoch [22/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0046, Loss2: 0.0050, Pure Ratio1: 9.9384, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 67.1875 % Model2 65.2945 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0031, Loss2: 0.0033, Pure Ratio1: 9.2941, Pure Ratio2 9.3529 +Epoch [23/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 9.5882, Pure Ratio2 9.5980 +Epoch [23/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0054, Loss2: 0.0049, Pure Ratio1: 9.7320, Pure Ratio2 9.7778 +Epoch [23/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 60.9375, Loss1: 0.0052, Loss2: 0.0058, Pure Ratio1: 9.8725, Pure Ratio2 9.8873 +Epoch [23/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0036, Loss2: 0.0029, Pure Ratio1: 9.8667, Pure Ratio2 9.8627 +Epoch [23/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0034, Loss2: 0.0041, Pure Ratio1: 9.9183, Pure Ratio2 9.9052 +Epoch [23/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0035, Loss2: 0.0038, Pure Ratio1: 9.9356, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 65.6150 % Model2 67.1675 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.0029, Loss2: 0.0026, Pure Ratio1: 9.7647, Pure Ratio2 9.6667 +Epoch [24/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0044, Loss2: 0.0036, Pure Ratio1: 9.6569, Pure Ratio2 9.5784 +Epoch [24/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0023, Pure Ratio1: 9.9673, Pure Ratio2 9.8889 +Epoch [24/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0036, Loss2: 0.0038, Pure Ratio1: 9.9706, Pure Ratio2 9.8971 +Epoch [24/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0029, Loss2: 0.0033, Pure Ratio1: 9.9255, Pure Ratio2 9.8667 +Epoch [24/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0031, Loss2: 0.0042, Pure Ratio1: 9.8333, Pure Ratio2 9.7908 +Epoch [24/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0029, Loss2: 0.0040, Pure Ratio1: 9.8431, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 65.5950 % Model2 65.9655 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.0026, Loss2: 0.0030, Pure Ratio1: 10.3333, Pure Ratio2 10.2549 +Epoch [25/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.0037, Loss2: 0.0028, Pure Ratio1: 10.1373, Pure Ratio2 10.0882 +Epoch [25/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.1875, Loss1: 0.0047, Loss2: 0.0043, Pure Ratio1: 10.3203, Pure Ratio2 10.2745 +Epoch [25/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 9.9608, Pure Ratio2 9.9363 +Epoch [25/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0036, Loss2: 0.0036, Pure Ratio1: 9.9412, Pure Ratio2 9.9373 +Epoch [25/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0069, Loss2: 0.0056, Pure Ratio1: 9.7941, Pure Ratio2 9.8137 +Epoch [25/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0027, Loss2: 0.0032, Pure Ratio1: 9.8319, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 63.5917 % Model2 62.7905 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.7059, Pure Ratio2 9.7255 +Epoch [26/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0019, Loss2: 0.0025, Pure Ratio1: 9.8333, Pure Ratio2 9.8627 +Epoch [26/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0020, Loss2: 0.0018, Pure Ratio1: 9.7843, Pure Ratio2 9.8497 +Epoch [26/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0021, Loss2: 0.0017, Pure Ratio1: 9.8971, Pure Ratio2 9.9510 +Epoch [26/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0033, Loss2: 0.0034, Pure Ratio1: 9.8275, Pure Ratio2 9.8784 +Epoch [26/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0028, Loss2: 0.0038, Pure Ratio1: 9.9216, Pure Ratio2 9.9673 +Epoch [26/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0029, Loss2: 0.0041, Pure Ratio1: 9.8908, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 64.4231 % Model2 65.3045 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0011, Pure Ratio1: 9.6667, Pure Ratio2 9.5490 +Epoch [27/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.0023, Loss2: 0.0021, Pure Ratio1: 9.6569, Pure Ratio2 9.6176 +Epoch [27/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0022, Loss2: 0.0018, Pure Ratio1: 9.6078, Pure Ratio2 9.5621 +Epoch [27/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 80.4688, Loss1: 0.0026, Loss2: 0.0019, Pure Ratio1: 9.7108, Pure Ratio2 9.6324 +Epoch [27/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0016, Loss2: 0.0022, Pure Ratio1: 9.7882, Pure Ratio2 9.7098 +Epoch [27/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0022, Loss2: 0.0022, Pure Ratio1: 9.8922, Pure Ratio2 9.8235 +Epoch [27/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.0938, Loss1: 0.0018, Loss2: 0.0029, Pure Ratio1: 9.9860, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 61.8890 % Model2 60.9575 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0021, Loss2: 0.0020, Pure Ratio1: 9.5294, Pure Ratio2 9.2941 +Epoch [28/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0012, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [28/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0015, Loss2: 0.0030, Pure Ratio1: 10.0719, Pure Ratio2 10.1046 +Epoch [28/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.0017, Loss2: 0.0027, Pure Ratio1: 10.0392, Pure Ratio2 10.0441 +Epoch [28/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0021, Loss2: 0.0032, Pure Ratio1: 9.8510, Pure Ratio2 9.8549 +Epoch [28/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0017, Pure Ratio1: 9.8431, Pure Ratio2 9.8660 +Epoch [28/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0025, Loss2: 0.0026, Pure Ratio1: 9.8796, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 65.4046 % Model2 64.5433 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0016, Loss2: 0.0020, Pure Ratio1: 9.3137, Pure Ratio2 9.3922 +Epoch [29/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0027, Loss2: 0.0023, Pure Ratio1: 9.8922, Pure Ratio2 9.9216 +Epoch [29/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0020, Loss2: 0.0018, Pure Ratio1: 9.9935, Pure Ratio2 9.9935 +Epoch [29/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0031, Loss2: 0.0034, Pure Ratio1: 9.7941, Pure Ratio2 9.8039 +Epoch [29/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0022, Loss2: 0.0025, Pure Ratio1: 9.7412, Pure Ratio2 9.7176 +Epoch [29/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0028, Loss2: 0.0028, Pure Ratio1: 9.7614, Pure Ratio2 9.7614 +Epoch [29/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.7812, Loss1: 0.0016, Loss2: 0.0024, Pure Ratio1: 9.8263, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 63.7821 % Model2 63.9924 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 9.8039, Pure Ratio2 9.8627 +Epoch [30/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0020, Pure Ratio1: 9.8431, Pure Ratio2 9.8137 +Epoch [30/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0015, Pure Ratio1: 9.8235, Pure Ratio2 9.8301 +Epoch [30/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0021, Loss2: 0.0008, Pure Ratio1: 9.7598, Pure Ratio2 9.7745 +Epoch [30/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 82.0312, Loss1: 0.0031, Loss2: 0.0014, Pure Ratio1: 9.7294, Pure Ratio2 9.7176 +Epoch [30/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 83.5938, Loss1: 0.0019, Loss2: 0.0010, Pure Ratio1: 9.7876, Pure Ratio2 9.8072 +Epoch [30/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.7955, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 62.3698 % Model2 62.8105 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.8627, Pure Ratio2 9.8235 +Epoch [31/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.0011, Loss2: 0.0016, Pure Ratio1: 9.6078, Pure Ratio2 9.6078 +Epoch [31/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0012, Pure Ratio1: 9.6928, Pure Ratio2 9.6144 +Epoch [31/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0017, Pure Ratio1: 10.0931, Pure Ratio2 9.9902 +Epoch [31/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0016, Pure Ratio1: 10.1137, Pure Ratio2 10.0157 +Epoch [31/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0019, Pure Ratio1: 10.0163, Pure Ratio2 9.9444 +Epoch [31/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 9.9972, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 62.4700 % Model2 63.9523 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 10.1176, Pure Ratio2 9.8627 +Epoch [32/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0027, Loss2: 0.0026, Pure Ratio1: 9.7353, Pure Ratio2 9.6373 +Epoch [32/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0018, Pure Ratio1: 9.8758, Pure Ratio2 9.8039 +Epoch [32/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0005, Loss2: 0.0014, Pure Ratio1: 9.9363, Pure Ratio2 9.9069 +Epoch [32/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0014, Loss2: 0.0015, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [32/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 9.9248, Pure Ratio2 9.9118 +Epoch [32/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0029, Loss2: 0.0036, Pure Ratio1: 9.9244, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 63.2612 % Model2 61.8289 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0022, Loss2: 0.0018, Pure Ratio1: 8.8627, Pure Ratio2 8.9608 +Epoch [33/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.2157, Pure Ratio2 9.1275 +Epoch [33/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 9.4183, Pure Ratio2 9.3856 +Epoch [33/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.6127, Pure Ratio2 9.5931 +Epoch [33/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0019, Pure Ratio1: 9.7451, Pure Ratio2 9.6863 +Epoch [33/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0012, Pure Ratio1: 9.7680, Pure Ratio2 9.7059 +Epoch [33/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.7927, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 64.2027 % Model2 61.6687 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.0022, Loss2: 0.0017, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [34/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0014, Loss2: 0.0025, Pure Ratio1: 9.7843, Pure Ratio2 9.8235 +Epoch [34/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.0020, Loss2: 0.0015, Pure Ratio1: 10.0261, Pure Ratio2 10.0065 +Epoch [34/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0020, Pure Ratio1: 9.9069, Pure Ratio2 9.8382 +Epoch [34/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0021, Pure Ratio1: 9.8235, Pure Ratio2 9.7608 +Epoch [34/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 9.8595, Pure Ratio2 9.8235 +Epoch [34/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0020, Pure Ratio1: 9.8655, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 63.0609 % Model2 61.7087 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 9.8235, Pure Ratio2 9.9412 +Epoch [35/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.3824, Pure Ratio2 10.4118 +Epoch [35/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0027, Loss2: 0.0017, Pure Ratio1: 10.3072, Pure Ratio2 10.3464 +Epoch [35/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 10.0098, Pure Ratio2 10.0833 +Epoch [35/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.0026, Loss2: 0.0010, Pure Ratio1: 9.9255, Pure Ratio2 9.9843 +Epoch [35/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0015, Pure Ratio1: 9.9510, Pure Ratio2 10.0033 +Epoch [35/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.9496, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 63.1210 % Model2 62.2596 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.7059, Pure Ratio2 9.5882 +Epoch [36/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 10.0980, Pure Ratio2 9.8725 +Epoch [36/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 10.2614, Pure Ratio2 10.1569 +Epoch [36/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.0014, Loss2: 0.0009, Pure Ratio1: 10.1667, Pure Ratio2 10.0882 +Epoch [36/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 10.0784, Pure Ratio2 9.9961 +Epoch [36/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.0065, Pure Ratio2 9.9216 +Epoch [36/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0023, Loss2: 0.0012, Pure Ratio1: 9.8936, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 61.4183 % Model2 62.7003 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.7059, Pure Ratio2 10.5882 +Epoch [37/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.9706, Pure Ratio2 9.8333 +Epoch [37/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0013, Loss2: 0.0018, Pure Ratio1: 9.9085, Pure Ratio2 9.8235 +Epoch [37/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.0490, Pure Ratio2 9.9853 +Epoch [37/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.8941, Pure Ratio2 9.8275 +Epoch [37/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.9542, Pure Ratio2 9.8889 +Epoch [37/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9048, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 61.1679 % Model2 61.2780 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.5686, Pure Ratio2 9.5294 +Epoch [38/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8529, Pure Ratio2 9.8725 +Epoch [38/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.7778, Pure Ratio2 9.8301 +Epoch [38/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.7500, Pure Ratio2 9.7892 +Epoch [38/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0012, Pure Ratio1: 9.8392, Pure Ratio2 9.8824 +Epoch [38/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.7353, Pure Ratio2 9.7745 +Epoch [38/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7703, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 62.6502 % Model2 62.5300 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.4706, Pure Ratio2 9.6471 +Epoch [39/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0013, Pure Ratio1: 9.9706, Pure Ratio2 10.1275 +Epoch [39/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 10.0392, Pure Ratio2 10.2222 +Epoch [39/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0015, Loss2: 0.0014, Pure Ratio1: 9.9363, Pure Ratio2 10.0441 +Epoch [39/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9412, Pure Ratio2 10.0314 +Epoch [39/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 9.8170, Pure Ratio2 9.9150 +Epoch [39/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.7927, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 62.5601 % Model2 62.4399 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0020, Loss2: 0.0009, Pure Ratio1: 10.2941, Pure Ratio2 10.2157 +Epoch [40/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 9.9804 +Epoch [40/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 10.0261, Pure Ratio2 10.0261 +Epoch [40/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0017, Loss2: 0.0008, Pure Ratio1: 9.8824, Pure Ratio2 9.8824 +Epoch [40/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.7333, Pure Ratio2 9.7216 +Epoch [40/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 83.5938, Loss1: 0.0017, Loss2: 0.0008, Pure Ratio1: 9.6993, Pure Ratio2 9.7092 +Epoch [40/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.7815, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 61.7688 % Model2 62.0493 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [41/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [41/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.7843, Pure Ratio2 9.7843 +Epoch [41/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.6765, Pure Ratio2 9.6912 +Epoch [41/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7922, Pure Ratio2 9.7412 +Epoch [41/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8203, Pure Ratio2 9.7680 +Epoch [41/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0016, Pure Ratio1: 9.8880, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 61.3582 % Model2 61.0978 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.3333, Pure Ratio2 9.2549 +Epoch [42/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.7549, Pure Ratio2 9.7255 +Epoch [42/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0020, Pure Ratio1: 9.7974, Pure Ratio2 9.7843 +Epoch [42/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7941, Pure Ratio2 9.8186 +Epoch [42/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.0003, Loss2: 0.0019, Pure Ratio1: 9.8353, Pure Ratio2 9.8471 +Epoch [42/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0028, Pure Ratio1: 9.8039, Pure Ratio2 9.8105 +Epoch [42/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7703, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 62.0292 % Model2 60.7372 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.7451 +Epoch [43/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.2255 +Epoch [43/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0523, Pure Ratio2 10.1634 +Epoch [43/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0017, Pure Ratio1: 10.0000, Pure Ratio2 10.0539 +Epoch [43/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 10.0471, Pure Ratio2 10.0902 +Epoch [43/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0013, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.9216 +Epoch [43/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.8627, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 62.2997 % Model2 60.7572 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0014, Pure Ratio1: 9.1765, Pure Ratio2 9.1373 +Epoch [44/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.5294, Pure Ratio2 9.6176 +Epoch [44/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.7059 +Epoch [44/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6569, Pure Ratio2 9.7255 +Epoch [44/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.7333, Pure Ratio2 9.7412 +Epoch [44/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.8170, Pure Ratio2 9.8464 +Epoch [44/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.8571, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 62.8305 % Model2 61.5885 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [45/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.8137, Pure Ratio2 9.7549 +Epoch [45/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.7974, Pure Ratio2 9.7582 +Epoch [45/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.7010, Pure Ratio2 9.6667 +Epoch [45/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.7216, Pure Ratio2 9.6863 +Epoch [45/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.7974 +Epoch [45/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0013, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 62.6803 % Model2 63.1110 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.9804, Pure Ratio2 10.1569 +Epoch [46/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.8725 +Epoch [46/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9542, Pure Ratio2 9.9085 +Epoch [46/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9167, Pure Ratio2 9.8284 +Epoch [46/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0157, Pure Ratio2 9.9098 +Epoch [46/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0030, Loss2: 0.0013, Pure Ratio1: 9.9412, Pure Ratio2 9.8170 +Epoch [46/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0007, Pure Ratio1: 9.9132, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 62.1394 % Model2 61.4183 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.2549, Pure Ratio2 9.2353 +Epoch [47/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6176, Pure Ratio2 9.6176 +Epoch [47/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9216, Pure Ratio2 9.8889 +Epoch [47/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.7451, Pure Ratio2 9.7549 +Epoch [47/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7882, Pure Ratio2 9.8235 +Epoch [47/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.8562, Pure Ratio2 9.8824 +Epoch [47/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8123, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 62.2596 % Model2 61.2981 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.4118, Pure Ratio2 9.2745 +Epoch [48/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.5392, Pure Ratio2 9.4608 +Epoch [48/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.7516, Pure Ratio2 9.6209 +Epoch [48/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0025, Pure Ratio1: 9.7598, Pure Ratio2 9.6520 +Epoch [48/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.8471, Pure Ratio2 9.7804 +Epoch [48/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0017, Pure Ratio1: 9.8824, Pure Ratio2 9.8203 +Epoch [48/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0016, Pure Ratio1: 9.8487, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 61.8089 % Model2 61.8490 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 9.8235 +Epoch [49/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.1275, Pure Ratio2 10.0098 +Epoch [49/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8170, Pure Ratio2 9.8562 +Epoch [49/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8627, Pure Ratio2 9.9118 +Epoch [49/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8784, Pure Ratio2 9.9020 +Epoch [49/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7288, Pure Ratio2 9.7582 +Epoch [49/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8291, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 61.0978 % Model2 62.5701 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [50/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.7647, Pure Ratio2 9.7941 +Epoch [50/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7386, Pure Ratio2 9.6536 +Epoch [50/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0015, Pure Ratio1: 9.7696, Pure Ratio2 9.7500 +Epoch [50/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.6275, Pure Ratio2 9.5765 +Epoch [50/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.5980, Pure Ratio2 9.5458 +Epoch [50/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.7143, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 62.2196 % Model2 61.3782 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.2745, Pure Ratio2 10.3529 +Epoch [51/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.8333, Pure Ratio2 9.8137 +Epoch [51/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7124, Pure Ratio2 9.6536 +Epoch [51/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7696, Pure Ratio2 9.7206 +Epoch [51/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.8039, Pure Ratio2 9.8000 +Epoch [51/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.8595, Pure Ratio2 9.8758 +Epoch [51/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8599, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 61.5986 % Model2 60.6270 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.6471, Pure Ratio2 9.6863 +Epoch [52/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.9118, Pure Ratio2 9.8039 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [52/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8284, Pure Ratio2 9.8235 +Epoch [52/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.7765, Pure Ratio2 9.7804 +Epoch [52/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7712, Pure Ratio2 9.7549 +Epoch [52/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.8123, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 60.4067 % Model2 61.5184 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 10.2941 +Epoch [53/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 10.0490 +Epoch [53/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6732, Pure Ratio2 9.7843 +Epoch [53/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.5980, Pure Ratio2 9.7304 +Epoch [53/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.5725, Pure Ratio2 9.7098 +Epoch [53/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.6405, Pure Ratio2 9.7876 +Epoch [53/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.6919, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 61.6286 % Model2 61.0176 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [54/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.8725 +Epoch [54/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.1569, Pure Ratio2 10.0261 +Epoch [54/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9755, Pure Ratio2 9.8775 +Epoch [54/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8902, Pure Ratio2 9.7843 +Epoch [54/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8954, Pure Ratio2 9.8072 +Epoch [54/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8515, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 61.3682 % Model2 60.1863 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.2549, Pure Ratio2 9.5686 +Epoch [55/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.3431, Pure Ratio2 9.5980 +Epoch [55/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0014, Loss2: 0.0009, Pure Ratio1: 9.5556, Pure Ratio2 9.7712 +Epoch [55/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9363, Pure Ratio2 10.1765 +Epoch [55/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.9922, Pure Ratio2 10.1725 +Epoch [55/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.7680, Pure Ratio2 9.9183 +Epoch [55/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8431, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 60.2063 % Model2 61.4383 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.6863, Pure Ratio2 9.7451 +Epoch [56/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.9706, Pure Ratio2 9.9804 +Epoch [56/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 10.0588, Pure Ratio2 9.9869 +Epoch [56/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 9.9510 +Epoch [56/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0431, Pure Ratio2 9.9216 +Epoch [56/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 10.0163, Pure Ratio2 9.9412 +Epoch [56/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.9748, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 61.6587 % Model2 60.3966 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.4314, Pure Ratio2 9.3922 +Epoch [57/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7549, Pure Ratio2 9.7451 +Epoch [57/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [57/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7353, Pure Ratio2 9.7892 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.8980, Pure Ratio2 9.9255 +Epoch [57/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.8137 +Epoch [57/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7563, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 61.8089 % Model2 61.2780 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.4902, Pure Ratio2 9.4902 +Epoch [58/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8529, Pure Ratio2 9.7255 +Epoch [58/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8431, Pure Ratio2 9.7647 +Epoch [58/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [58/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.8980, Pure Ratio2 9.8667 +Epoch [58/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.8333 +Epoch [58/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8067, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 60.9375 % Model2 61.0477 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.1961 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8235, Pure Ratio2 9.7941 +Epoch [59/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [59/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.7794, Pure Ratio2 9.7941 +Epoch [59/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.7529 +Epoch [59/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.7484, Pure Ratio2 9.7745 +Epoch [59/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.7563, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 62.5901 % Model2 61.1078 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.3725 +Epoch [60/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8922, Pure Ratio2 9.8039 +Epoch [60/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9150, Pure Ratio2 9.8758 +Epoch [60/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7745, Pure Ratio2 9.7941 +Epoch [60/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7804, Pure Ratio2 9.7882 +Epoch [60/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7778, Pure Ratio2 9.7843 +Epoch [60/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8571, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 60.2965 % Model2 61.8389 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.9216 +Epoch [61/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0098, Pure Ratio2 10.1078 +Epoch [61/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 9.9673 +Epoch [61/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.9510, Pure Ratio2 9.9363 +Epoch [61/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.7255 +Epoch [61/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.7484, Pure Ratio2 9.6993 +Epoch [61/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.8459, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 61.5184 % Model2 62.1294 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.8431 +Epoch [62/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.7745, Pure Ratio2 9.8137 +Epoch [62/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.8824 +Epoch [62/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8529, Pure Ratio2 9.8627 +Epoch [62/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 91.4062, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.9569, Pure Ratio2 9.9176 +Epoch [62/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.9248 +Epoch [62/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8263, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 62.0192 % Model2 60.5168 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.8627, Pure Ratio2 10.6275 +Epoch [63/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.3235 +Epoch [63/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.8824 +Epoch [63/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8137 +Epoch [63/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9451, Pure Ratio2 9.8471 +Epoch [63/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9052, Pure Ratio2 9.8105 +Epoch [63/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.9636, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 61.0978 % Model2 60.8273 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 10.0000 +Epoch [64/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.6765, Pure Ratio2 9.8137 +Epoch [64/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.5033, Pure Ratio2 9.6144 +Epoch [64/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.5882, Pure Ratio2 9.7157 +Epoch [64/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7412, Pure Ratio2 9.8157 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7059, Pure Ratio2 9.7680 +Epoch [64/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 61.4984 % Model2 61.7588 %, Pure Ratio 1 9.8014 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [65/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.7745 +Epoch [65/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8562, Pure Ratio2 9.7451 +Epoch [65/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8088, Pure Ratio2 9.7451 +Epoch [65/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.8235 +Epoch [65/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8497, Pure Ratio2 9.8170 +Epoch [65/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.8964, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 60.9776 % Model2 60.5068 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.6667, Pure Ratio2 10.4118 +Epoch [66/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.4608, Pure Ratio2 10.3824 +Epoch [66/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.2614, Pure Ratio2 10.2353 +Epoch [66/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.9265 +Epoch [66/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 9.9765 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.9216, Pure Ratio2 9.8954 +Epoch [66/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8992, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 60.9575 % Model2 60.6270 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4902, Pure Ratio2 9.5882 +Epoch [67/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.6569, Pure Ratio2 9.5882 +Epoch [67/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7712, Pure Ratio2 9.7124 +Epoch [67/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0032, Loss2: 0.0046, Pure Ratio1: 9.9167, Pure Ratio2 9.9216 +Epoch [67/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7922, Pure Ratio2 9.7843 +Epoch [67/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.7222, Pure Ratio2 9.7222 +Epoch [67/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.7367, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 61.5385 % Model2 61.8590 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [68/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.8529 +Epoch [68/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.7778, Pure Ratio2 9.8954 +Epoch [68/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6765, Pure Ratio2 9.7598 +Epoch [68/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8314, Pure Ratio2 9.8941 +Epoch [68/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.8333, Pure Ratio2 9.8954 +Epoch [68/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8375, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 61.7488 % Model2 61.6086 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 10.0980 +Epoch [69/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0004, Pure Ratio1: 9.8529, Pure Ratio2 10.0196 +Epoch [69/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 9.9542 +Epoch [69/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9265, Pure Ratio2 9.8971 +Epoch [69/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7686, Pure Ratio2 9.7922 +Epoch [69/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.6797, Pure Ratio2 9.7092 +Epoch [69/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.6919, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 61.4283 % Model2 61.8490 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.2353, Pure Ratio2 9.3922 +Epoch [70/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.8137 +Epoch [70/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7255, Pure Ratio2 9.7582 +Epoch [70/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7206, Pure Ratio2 9.7353 +Epoch [70/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7333, Pure Ratio2 9.7294 +Epoch [70/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7288, Pure Ratio2 9.7255 +Epoch [70/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8263, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 62.4599 % Model2 62.3097 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [71/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 10.3725, Pure Ratio2 10.4804 +Epoch [71/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.0915, Pure Ratio2 10.0784 +Epoch [71/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.1176, Pure Ratio2 10.0343 +Epoch [71/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0078, Pure Ratio2 9.9529 +Epoch [71/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8954, Pure Ratio2 9.8725 +Epoch [71/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9076, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 61.3882 % Model2 61.5485 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.1765, Pure Ratio2 10.2157 +Epoch [72/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.3725, Pure Ratio2 10.3529 +Epoch [72/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3856, Pure Ratio2 10.3268 +Epoch [72/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0833, Pure Ratio2 10.0343 +Epoch [72/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.9176 +Epoch [72/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9085, Pure Ratio2 9.9020 +Epoch [72/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.8067, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 61.3482 % Model2 59.8357 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.5490, Pure Ratio2 9.6275 +Epoch [73/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6667 +Epoch [73/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.9346 +Epoch [73/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7402, Pure Ratio2 9.8284 +Epoch [73/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8196, Pure Ratio2 9.8824 +Epoch [73/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0011, Pure Ratio1: 9.7941, Pure Ratio2 9.8137 +Epoch [73/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 90.6250, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.8347, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 61.8089 % Model2 61.5184 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.1961, Pure Ratio2 9.9216 +Epoch [74/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7255, Pure Ratio2 9.7353 +Epoch [74/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9542, Pure Ratio2 9.9085 +Epoch [74/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9902, Pure Ratio2 9.9559 +Epoch [74/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 9.8706 +Epoch [74/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [74/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8487, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 62.2596 % Model2 61.1478 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9608 +Epoch [75/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.1961 +Epoch [75/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.1503 +Epoch [75/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [75/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0745, Pure Ratio2 10.1059 +Epoch [75/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0016, Pure Ratio1: 10.0327, Pure Ratio2 10.0719 +Epoch [75/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8487, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 61.0978 % Model2 61.6887 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [76/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [76/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9346, Pure Ratio2 9.8301 +Epoch [76/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8137 +Epoch [76/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.7255, Pure Ratio2 9.6745 +Epoch [76/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0001, Pure Ratio1: 9.6013, Pure Ratio2 9.5882 +Epoch [76/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.6303, Pure Ratio2 9.6275 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 60.3966 % Model2 60.7171 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.9216 +Epoch [77/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.7451 +Epoch [77/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6732, Pure Ratio2 9.7582 +Epoch [77/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.7402, Pure Ratio2 9.7941 +Epoch [77/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8863 +Epoch [77/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.8464, Pure Ratio2 9.8235 +Epoch [77/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8936, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 61.6587 % Model2 61.8590 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 8.9804, Pure Ratio2 9.0392 +Epoch [78/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.2549, Pure Ratio2 9.2843 +Epoch [78/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.4706, Pure Ratio2 9.5163 +Epoch [78/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.5098, Pure Ratio2 9.4706 +Epoch [78/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.5882, Pure Ratio2 9.6039 +Epoch [78/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6340, Pure Ratio2 9.6536 +Epoch [78/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.6695, Pure Ratio2 9.6667 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 62.0593 % Model2 61.8289 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0017, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [79/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.6765 +Epoch [79/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.7712 +Epoch [79/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.6569, Pure Ratio2 9.7500 +Epoch [79/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.6549, Pure Ratio2 9.7843 +Epoch [79/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6765, Pure Ratio2 9.8072 +Epoch [79/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7087, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 61.6086 % Model2 59.7356 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 9.8039 +Epoch [80/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.5588, Pure Ratio2 9.5588 +Epoch [80/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.5163 +Epoch [80/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5441, Pure Ratio2 9.6176 +Epoch [80/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7412, Pure Ratio2 9.8118 +Epoch [80/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.8529 +Epoch [80/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7563, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 60.8874 % Model2 62.7504 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.7059, Pure Ratio2 10.2157 +Epoch [81/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1471 +Epoch [81/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0131, Pure Ratio2 10.0850 +Epoch [81/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7402, Pure Ratio2 9.8088 +Epoch [81/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.9373 +Epoch [81/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.8562 +Epoch [81/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7731, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 61.6186 % Model2 61.5284 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.2745, Pure Ratio2 9.5098 +Epoch [82/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.6471, Pure Ratio2 9.7255 +Epoch [82/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6536 +Epoch [82/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0003, Pure Ratio1: 9.7157, Pure Ratio2 9.6569 +Epoch [82/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.8706 +Epoch [82/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.8660 +Epoch [82/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9440, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 61.3882 % Model2 62.5601 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 10.6275, Pure Ratio2 10.5882 +Epoch [83/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.2745, Pure Ratio2 10.2059 +Epoch [83/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 9.9477 +Epoch [83/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7696, Pure Ratio2 9.7451 +Epoch [83/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Epoch [83/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8529, Pure Ratio2 9.8170 +Epoch [83/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8543, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 61.9491 % Model2 61.0877 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.3333, Pure Ratio2 9.2941 +Epoch [84/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.5000, Pure Ratio2 9.6471 +Epoch [84/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0001, Pure Ratio1: 9.3399, Pure Ratio2 9.4510 +Epoch [84/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5343, Pure Ratio2 9.5245 +Epoch [84/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6157, Pure Ratio2 9.5961 +Epoch [84/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.7320, Pure Ratio2 9.6993 +Epoch [84/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 61.5385 % Model2 61.0477 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.0392 +Epoch [85/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.6765 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7516 +Epoch [85/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8186, Pure Ratio2 9.7598 +Epoch [85/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8980, Pure Ratio2 9.8588 +Epoch [85/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8497, Pure Ratio2 9.8529 +Epoch [85/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8123, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 62.4499 % Model2 61.6987 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.6275 +Epoch [86/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.8235 +Epoch [86/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5752, Pure Ratio2 9.6601 +Epoch [86/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7745, Pure Ratio2 9.8627 +Epoch [86/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8902 +Epoch [86/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7320, Pure Ratio2 9.7157 +Epoch [86/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7395, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 61.8490 % Model2 61.2881 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6078, Pure Ratio2 9.6275 +Epoch [87/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.9510 +Epoch [87/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9150, Pure Ratio2 9.9216 +Epoch [87/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.8725 +Epoch [87/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8078, Pure Ratio2 9.8784 +Epoch [87/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7582, Pure Ratio2 9.8693 +Epoch [87/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8095, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 62.0893 % Model2 61.3582 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.2745, Pure Ratio2 10.4706 +Epoch [88/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.0098 +Epoch [88/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.0915 +Epoch [88/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9559, Pure Ratio2 9.9020 +Epoch [88/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8706, Pure Ratio2 9.8392 +Epoch [88/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7680, Pure Ratio2 9.7190 +Epoch [88/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0014, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.6919 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 61.6587 % Model2 60.1863 %, Pure Ratio 1 9.8014 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.3725 +Epoch [89/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.5000, Pure Ratio2 9.4902 +Epoch [89/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9935, Pure Ratio2 9.9935 +Epoch [89/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0003, Pure Ratio1: 10.0147, Pure Ratio2 9.9755 +Epoch [89/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9020 +Epoch [89/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8954, Pure Ratio2 9.8268 +Epoch [89/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8319, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 61.9992 % Model2 61.0978 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.4314, Pure Ratio2 10.4706 +Epoch [90/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.1667, Pure Ratio2 10.1765 +Epoch [90/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0065, Pure Ratio2 10.1503 +Epoch [90/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9559, Pure Ratio2 10.0882 +Epoch [90/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9608, Pure Ratio2 10.0745 +Epoch [90/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 9.9902 +Epoch [90/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7871, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 62.1194 % Model2 61.4483 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Epoch [91/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.8922 +Epoch [91/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.8889, Pure Ratio2 9.8235 +Epoch [91/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [91/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.6980, Pure Ratio2 9.6902 +Epoch [91/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.6961 +Epoch [91/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8739, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 61.2179 % Model2 60.9976 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [92/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.6961 +Epoch [92/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9216 +Epoch [92/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8333 +Epoch [92/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7922, Pure Ratio2 9.7725 +Epoch [92/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8693, Pure Ratio2 9.8235 +Epoch [92/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0001, Pure Ratio1: 9.8739, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 61.2580 % Model2 61.5084 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [93/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6275 +Epoch [93/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.5294, Pure Ratio2 9.6275 +Epoch [93/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.6814 +Epoch [93/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6000, Pure Ratio2 9.6902 +Epoch [93/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.5850, Pure Ratio2 9.6830 +Epoch [93/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.6947, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 62.0793 % Model2 59.9760 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.3725, Pure Ratio2 9.3922 +Epoch [94/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4020, Pure Ratio2 9.3627 +Epoch [94/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.2876, Pure Ratio2 9.3203 +Epoch [94/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.5245, Pure Ratio2 9.5539 +Epoch [94/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5765, Pure Ratio2 9.6039 +Epoch [94/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6013, Pure Ratio2 9.6307 +Epoch [94/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.7759, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 61.8089 % Model2 61.0677 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.8431 +Epoch [95/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.2255 +Epoch [95/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 9.9935 +Epoch [95/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.0490 +Epoch [95/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0078, Pure Ratio2 9.9608 +Epoch [95/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0359, Pure Ratio2 9.9346 +Epoch [95/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9440, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 61.5585 % Model2 62.2296 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.7059 +Epoch [96/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6667 +Epoch [96/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.6340 +Epoch [96/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.6275 +Epoch [96/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7216, Pure Ratio2 9.7961 +Epoch [96/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6993, Pure Ratio2 9.8105 +Epoch [96/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7815, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 61.3482 % Model2 61.6086 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0015, Pure Ratio1: 9.4902, Pure Ratio2 9.2941 +Epoch [97/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6078, Pure Ratio2 9.3431 +Epoch [97/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.6471 +Epoch [97/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7402, Pure Ratio2 9.6912 +Epoch [97/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7020, Pure Ratio2 9.6941 +Epoch [97/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7157, Pure Ratio2 9.7222 +Epoch [97/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7395, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 62.0994 % Model2 60.1562 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.4706 +Epoch [98/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.7843 +Epoch [98/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7843, Pure Ratio2 9.7778 +Epoch [98/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7892, Pure Ratio2 9.8922 +Epoch [98/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.8078 +Epoch [98/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9183 +Epoch [98/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8179, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 62.7704 % Model2 62.0793 %, Pure Ratio 1 9.7587 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [99/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 10.0980 +Epoch [99/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 10.1438 +Epoch [99/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 10.1520 +Epoch [99/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9922, Pure Ratio2 10.1608 +Epoch [99/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8562, Pure Ratio2 10.0425 +Epoch [99/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8347, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 61.1679 % Model2 60.8373 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0021, Pure Ratio1: 9.4706, Pure Ratio2 9.4118 +Epoch [100/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0490 +Epoch [100/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8954, Pure Ratio2 9.8758 +Epoch [100/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6765, Pure Ratio2 9.6814 +Epoch [100/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6510, Pure Ratio2 9.6157 +Epoch [100/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7353, Pure Ratio2 9.7026 +Epoch [100/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 62.5801 % Model2 61.4183 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5490 +Epoch [101/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7941 +Epoch [101/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9216, Pure Ratio2 10.0196 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.9412 +Epoch [101/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8000, Pure Ratio2 9.8745 +Epoch [101/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.7843 +Epoch [101/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8347, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 61.4283 % Model2 59.8558 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8922, Pure Ratio2 9.7941 +Epoch [102/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8562 +Epoch [102/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9363, Pure Ratio2 9.8284 +Epoch [102/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8941, Pure Ratio2 9.8431 +Epoch [102/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.7974 +Epoch [102/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9384, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 61.4383 % Model2 62.9307 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.3137 +Epoch [103/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.7059 +Epoch [103/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8170 +Epoch [103/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0539, Pure Ratio2 9.8922 +Epoch [103/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 9.7529 +Epoch [103/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8889, Pure Ratio2 9.7712 +Epoch [103/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0001, Pure Ratio1: 9.8347, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 61.8089 % Model2 61.4383 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Epoch [104/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5294, Pure Ratio2 9.4902 +Epoch [104/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8366, Pure Ratio2 9.8235 +Epoch [104/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8922 +Epoch [104/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9216 +Epoch [104/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9346 +Epoch [104/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 61.1178 % Model2 62.0393 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [105/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.6863 +Epoch [105/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.7451 +Epoch [105/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7206 +Epoch [105/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8118, Pure Ratio2 9.7608 +Epoch [105/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8464, Pure Ratio2 9.8105 +Epoch [105/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 63.5016 % Model2 61.2280 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.8235 +Epoch [106/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8824 +Epoch [106/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.7190 +Epoch [106/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8088, Pure Ratio2 9.7892 +Epoch [106/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8667, Pure Ratio2 9.8510 +Epoch [106/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 9.8922 +Epoch [106/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 62.6402 % Model2 61.6787 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3137, Pure Ratio2 10.4902 +Epoch [107/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2451, Pure Ratio2 10.5000 +Epoch [107/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0261 +Epoch [107/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [107/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0431, Pure Ratio2 10.0706 +Epoch [107/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9575 +Epoch [107/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8683, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 62.2196 % Model2 61.6486 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7662 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1176, Pure Ratio2 9.0784 +Epoch [108/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3431, Pure Ratio2 9.3039 +Epoch [108/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.3595, Pure Ratio2 9.3333 +Epoch [108/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.4951, Pure Ratio2 9.4412 +Epoch [108/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5725 +Epoch [108/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.7222 +Epoch [108/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8291, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 62.2596 % Model2 62.2396 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.6863, Pure Ratio2 10.8235 +Epoch [109/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3235, Pure Ratio2 10.4314 +Epoch [109/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3856, Pure Ratio2 10.5098 +Epoch [109/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2402, Pure Ratio2 10.3039 +Epoch [109/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.1373 +Epoch [109/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9935, Pure Ratio2 10.1111 +Epoch [109/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8992, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 61.9892 % Model2 61.0677 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.4706 +Epoch [110/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.1569 +Epoch [110/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7124, Pure Ratio2 9.8889 +Epoch [110/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0000 +Epoch [110/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8745, Pure Ratio2 10.0431 +Epoch [110/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0163 +Epoch [110/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.7787, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 62.5901 % Model2 60.9675 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6667 +Epoch [111/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 10.1176, Pure Ratio2 9.9902 +Epoch [111/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.8889, Pure Ratio2 9.7778 +Epoch [111/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8529, Pure Ratio2 9.7108 +Epoch [111/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8314, Pure Ratio2 9.7059 +Epoch [111/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9085, Pure Ratio2 9.8072 +Epoch [111/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8711, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 61.8790 % Model2 61.5585 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0588 +Epoch [112/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.9412 +Epoch [112/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0458, Pure Ratio2 10.1438 +Epoch [112/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1667 +Epoch [112/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9765, Pure Ratio2 10.0471 +Epoch [112/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9314 +Epoch [112/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8515, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 61.4083 % Model2 62.1494 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.2745, Pure Ratio2 10.2157 +Epoch [113/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.3137, Pure Ratio2 10.1863 +Epoch [113/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7516 +Epoch [113/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7990, Pure Ratio2 9.7892 +Epoch [113/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7922, Pure Ratio2 9.7922 +Epoch [113/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.7712 +Epoch [113/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 60.7973 % Model2 61.3782 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.2745 +Epoch [114/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.1373 +Epoch [114/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2222, Pure Ratio2 10.1242 +Epoch [114/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2402, Pure Ratio2 10.1127 +Epoch [114/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2588, Pure Ratio2 10.1137 +Epoch [114/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9739 +Epoch [114/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9720, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 61.7788 % Model2 61.5885 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1569 +Epoch [115/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9020 +Epoch [115/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 94.5312, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0131 +Epoch [115/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9314 +Epoch [115/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8314 +Epoch [115/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8072 +Epoch [115/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 62.3998 % Model2 60.4267 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 9.9216 +Epoch [116/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9216 +Epoch [116/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7059 +Epoch [116/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [116/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8118, Pure Ratio2 9.7569 +Epoch [116/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.7712 +Epoch [116/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0014, Pure Ratio1: 9.8543, Pure Ratio2 9.7367 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 61.4083 % Model2 61.4083 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.3922 +Epoch [117/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6961, Pure Ratio2 9.7647 +Epoch [117/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.7778 +Epoch [117/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7745 +Epoch [117/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6745, Pure Ratio2 9.6549 +Epoch [117/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7810, Pure Ratio2 9.7680 +Epoch [117/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7899, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 61.5385 % Model2 62.5501 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 10.0196 +Epoch [118/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9706 +Epoch [118/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0021, Loss2: 0.0029, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [118/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1127, Pure Ratio2 10.1520 +Epoch [118/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0471, Pure Ratio2 10.0784 +Epoch [118/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [118/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 60.9275 % Model2 62.0693 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.2745 +Epoch [119/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.3235 +Epoch [119/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.0719 +Epoch [119/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0147, Pure Ratio2 10.0343 +Epoch [119/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0017, Pure Ratio1: 9.9765, Pure Ratio2 10.0588 +Epoch [119/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.9804 +Epoch [119/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8207, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 61.7588 % Model2 61.8389 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.7647 +Epoch [120/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6373 +Epoch [120/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.7647 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7402, Pure Ratio2 9.7304 +Epoch [120/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7529, Pure Ratio2 9.7176 +Epoch [120/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7353 +Epoch [120/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7983, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 60.0160 % Model2 60.9776 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7451 +Epoch [121/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8824 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.6667 +Epoch [121/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8480, Pure Ratio2 9.8480 +Epoch [121/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.7490 +Epoch [121/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8072 +Epoch [121/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8067, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 61.2480 % Model2 60.6070 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7059 +Epoch [122/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.7549 +Epoch [122/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 9.9673 +Epoch [122/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [122/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7922, Pure Ratio2 9.7765 +Epoch [122/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7222, Pure Ratio2 9.7353 +Epoch [122/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 61.3281 % Model2 61.2380 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.8824 +Epoch [123/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0294 +Epoch [123/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.0458, Pure Ratio2 9.9346 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0098 +Epoch [123/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8314 +Epoch [123/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7582 +Epoch [123/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 61.1078 % Model2 62.1294 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8627 +Epoch [124/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7157 +Epoch [124/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7582, Pure Ratio2 9.8627 +Epoch [124/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9902 +Epoch [124/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8706, Pure Ratio2 9.9333 +Epoch [124/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9052, Pure Ratio2 9.9542 +Epoch [124/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8599, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 61.3482 % Model2 60.8874 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4902, Pure Ratio2 9.4118 +Epoch [125/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.6765 +Epoch [125/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 9.8431 +Epoch [125/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.8039 +Epoch [125/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9451, Pure Ratio2 9.8588 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.8824 +Epoch [125/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7899, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 61.7288 % Model2 61.1679 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.2157 +Epoch [126/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4020, Pure Ratio2 9.3137 +Epoch [126/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.2222 +Epoch [126/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4314, Pure Ratio2 9.4314 +Epoch [126/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6510, Pure Ratio2 9.5922 +Epoch [126/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.6503 +Epoch [126/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 62.7704 % Model2 61.8790 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0784 +Epoch [127/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [127/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8235 +Epoch [127/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.7941 +Epoch [127/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.7412 +Epoch [127/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8105 +Epoch [127/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8207, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 63.4615 % Model2 62.3097 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.5882 +Epoch [128/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8529 +Epoch [128/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8431 +Epoch [128/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0539 +Epoch [128/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0018, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [128/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0261, Pure Ratio2 10.0261 +Epoch [128/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8655, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 60.4367 % Model2 60.5669 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0000 +Epoch [129/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6078 +Epoch [129/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6601, Pure Ratio2 9.7059 +Epoch [129/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.8088 +Epoch [129/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6824, Pure Ratio2 9.7137 +Epoch [129/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.7386 +Epoch [129/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7619, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 61.1879 % Model2 61.2981 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0010, Pure Ratio1: 10.4706, Pure Ratio2 10.4510 +Epoch [130/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8922 +Epoch [130/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6405, Pure Ratio2 9.6275 +Epoch [130/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4755, Pure Ratio2 9.4510 +Epoch [130/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6353, Pure Ratio2 9.6667 +Epoch [130/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6405, Pure Ratio2 9.6699 +Epoch [130/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7535, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 60.9475 % Model2 61.8590 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8235 +Epoch [131/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7353 +Epoch [131/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7190, Pure Ratio2 9.7778 +Epoch [131/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0024, Pure Ratio1: 9.8382, Pure Ratio2 9.9167 +Epoch [131/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8157, Pure Ratio2 9.8863 +Epoch [131/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.9575 +Epoch [131/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 62.1194 % Model2 62.1795 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.0980, Pure Ratio2 9.2745 +Epoch [132/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.4608 +Epoch [132/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4575, Pure Ratio2 9.5817 +Epoch [132/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5049, Pure Ratio2 9.5931 +Epoch [132/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4392, Pure Ratio2 9.5098 +Epoch [132/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5425, Pure Ratio2 9.6078 +Epoch [132/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6751, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 62.4499 % Model2 61.5685 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.7059 +Epoch [133/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9118, Pure Ratio2 9.8627 +Epoch [133/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8954 +Epoch [133/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8578 +Epoch [133/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9725 +Epoch [133/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.7810 +Epoch [133/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 61.7688 % Model2 62.2897 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8235 +Epoch [134/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1176 +Epoch [134/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7320, Pure Ratio2 9.7843 +Epoch [134/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7451 +Epoch [134/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8000, Pure Ratio2 9.7804 +Epoch [134/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.7974, Pure Ratio2 9.8072 +Epoch [134/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8347, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 60.9976 % Model2 63.0308 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.5882 +Epoch [135/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.2157 +Epoch [135/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1046 +Epoch [135/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.9706 +Epoch [135/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8510, Pure Ratio2 9.8627 +Epoch [135/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9183, Pure Ratio2 9.9281 +Epoch [135/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 62.0593 % Model2 61.8690 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.7255 +Epoch [136/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.6373 +Epoch [136/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.5752 +Epoch [136/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.5294 +Epoch [136/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.5529 +Epoch [136/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6765, Pure Ratio2 9.6961 +Epoch [136/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7871, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 62.1695 % Model2 61.6987 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.8824 +Epoch [137/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0490, Pure Ratio2 10.1471 +Epoch [137/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0196 +Epoch [137/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.9020 +Epoch [137/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 9.7176, Pure Ratio2 9.8549 +Epoch [137/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.9477 +Epoch [137/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8347, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 61.8389 % Model2 62.5100 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0196 +Epoch [138/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8137, Pure Ratio2 9.9314 +Epoch [138/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7712 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.7010, Pure Ratio2 9.8039 +Epoch [138/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7490, Pure Ratio2 9.8471 +Epoch [138/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8039 +Epoch [138/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7675, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 61.8189 % Model2 60.8574 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9608 +Epoch [139/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.7843 +Epoch [139/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4771, Pure Ratio2 9.7320 +Epoch [139/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5147, Pure Ratio2 9.7451 +Epoch [139/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7765 +Epoch [139/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5654, Pure Ratio2 9.6634 +Epoch [139/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7815, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 62.2997 % Model2 60.7772 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0023, Pure Ratio1: 9.5686, Pure Ratio2 9.3725 +Epoch [140/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.5490 +Epoch [140/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8366 +Epoch [140/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.6471 +Epoch [140/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8275, Pure Ratio2 9.7647 +Epoch [140/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.7582 +Epoch [140/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7675, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 62.7204 % Model2 62.0292 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [141/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7157 +Epoch [141/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8627 +Epoch [141/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7402, Pure Ratio2 9.7108 +Epoch [141/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.7059 +Epoch [141/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8203, Pure Ratio2 9.7745 +Epoch [141/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 63.0108 % Model2 62.1194 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [142/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.6961 +Epoch [142/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.7908 +Epoch [142/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8186 +Epoch [142/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8314, Pure Ratio2 9.7373 +Epoch [142/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8562, Pure Ratio2 9.7647 +Epoch [142/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8515, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 60.8874 % Model2 61.3582 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.3529 +Epoch [143/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5000, Pure Ratio2 9.5686 +Epoch [143/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5882 +Epoch [143/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.7402 +Epoch [143/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8667 +Epoch [143/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.7516 +Epoch [143/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 62.4299 % Model2 61.4884 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8431 +Epoch [144/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 10.0490 +Epoch [144/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.9673 +Epoch [144/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0588 +Epoch [144/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 9.9882 +Epoch [144/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8791 +Epoch [144/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 63.1310 % Model2 62.3698 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.1373, Pure Ratio2 9.1961 +Epoch [145/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.5294 +Epoch [145/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4837, Pure Ratio2 9.5229 +Epoch [145/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5637, Pure Ratio2 9.6716 +Epoch [145/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6431, Pure Ratio2 9.7686 +Epoch [145/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.7647 +Epoch [145/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 62.8906 % Model2 62.0693 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0784 +Epoch [146/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.1569 +Epoch [146/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.1242 +Epoch [146/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9657 +Epoch [146/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 9.9294 +Epoch [146/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7712 +Epoch [146/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7871, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 63.0208 % Model2 62.3798 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 10.0196 +Epoch [147/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7843 +Epoch [147/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.8366 +Epoch [147/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6912 +Epoch [147/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6549, Pure Ratio2 9.6314 +Epoch [147/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.7222, Pure Ratio2 9.6928 +Epoch [147/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7815, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 61.6286 % Model2 62.6102 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6667 +Epoch [148/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0016, Loss2: 0.0001, Pure Ratio1: 9.5784, Pure Ratio2 9.6078 +Epoch [148/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5948, Pure Ratio2 9.7255 +Epoch [148/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6716, Pure Ratio2 9.8039 +Epoch [148/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8314, Pure Ratio2 9.8980 +Epoch [148/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.8562 +Epoch [148/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7731, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 62.1695 % Model2 62.4399 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.4902 +Epoch [149/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.7157 +Epoch [149/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8366 +Epoch [149/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.8971 +Epoch [149/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8745 +Epoch [149/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8268, Pure Ratio2 9.7974 +Epoch [149/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 62.1394 % Model2 61.3281 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.2157 +Epoch [150/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.1275 +Epoch [150/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7386, Pure Ratio2 9.9216 +Epoch [150/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.8529 +Epoch [150/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0016, Pure Ratio1: 9.6941, Pure Ratio2 9.7725 +Epoch [150/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.8105 +Epoch [150/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6779, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 61.8389 % Model2 61.2079 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0784 +Epoch [151/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7157 +Epoch [151/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0719, Pure Ratio2 9.9935 +Epoch [151/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.9216 +Epoch [151/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.8235 +Epoch [151/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0033, Loss2: 0.0048, Pure Ratio1: 9.7320, Pure Ratio2 9.8333 +Epoch [151/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8123, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 62.7204 % Model2 60.8974 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.2157 +Epoch [152/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.8922 +Epoch [152/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9281 +Epoch [152/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7745, Pure Ratio2 9.7010 +Epoch [152/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6588 +Epoch [152/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7157 +Epoch [152/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8263, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 62.4900 % Model2 62.2596 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2941 +Epoch [153/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0588 +Epoch [153/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [153/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 10.0539 +Epoch [153/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9098 +Epoch [153/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9085 +Epoch [153/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 62.1795 % Model2 63.4415 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.4314 +Epoch [154/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1275 +Epoch [154/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1111 +Epoch [154/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9706 +Epoch [154/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8980, Pure Ratio2 9.9059 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7222, Pure Ratio2 9.7288 +Epoch [154/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 61.2881 % Model2 61.6086 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.4510 +Epoch [155/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6373 +Epoch [155/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.6013 +Epoch [155/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.8039 +Epoch [155/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.8471 +Epoch [155/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8856 +Epoch [155/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8515, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 64.0124 % Model2 62.1094 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5294, Pure Ratio2 9.6471 +Epoch [156/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8725 +Epoch [156/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.8758 +Epoch [156/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0637, Pure Ratio2 10.0637 +Epoch [156/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9608 +Epoch [156/200], Iter [300/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9771, Pure Ratio2 9.8954 +Epoch [156/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 63.0208 % Model2 62.3197 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8235 +Epoch [157/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4608 +Epoch [157/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6863 +Epoch [157/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.8137 +Epoch [157/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 9.8667 +Epoch [157/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.8595 +Epoch [157/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 61.3081 % Model2 61.5184 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0392 +Epoch [158/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [158/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9804 +Epoch [158/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [158/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.8353 +Epoch [158/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.7582 +Epoch [158/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7955, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 62.4499 % Model2 61.8189 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8627 +Epoch [159/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6667 +Epoch [159/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.7516 +Epoch [159/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7892 +Epoch [159/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8667, Pure Ratio2 9.9059 +Epoch [159/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 9.9869 +Epoch [159/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8683, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 62.1795 % Model2 61.8790 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.7255 +Epoch [160/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2843 +Epoch [160/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1307 +Epoch [160/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.0931 +Epoch [160/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.9882 +Epoch [160/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8725 +Epoch [160/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 62.7304 % Model2 61.6086 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4510, Pure Ratio2 9.3725 +Epoch [161/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7059 +Epoch [161/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.4837 +Epoch [161/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6176 +Epoch [161/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.8157 +Epoch [161/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8007 +Epoch [161/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8487, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 62.0593 % Model2 61.3281 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.3725 +Epoch [162/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9412 +Epoch [162/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8562 +Epoch [162/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9461 +Epoch [162/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9843 +Epoch [162/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.9804 +Epoch [162/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8711, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 62.2196 % Model2 62.2596 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.3529 +Epoch [163/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.6863 +Epoch [163/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.7582 +Epoch [163/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6324, Pure Ratio2 9.7549 +Epoch [163/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6235, Pure Ratio2 9.7569 +Epoch [163/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.8725 +Epoch [163/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 61.8790 % Model2 60.9175 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8824 +Epoch [164/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8039 +Epoch [164/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7320, Pure Ratio2 9.8039 +Epoch [164/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7794, Pure Ratio2 9.7696 +Epoch [164/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8549 +Epoch [164/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.7516 +Epoch [164/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6947, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 62.9808 % Model2 62.1094 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9020 +Epoch [165/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7941 +Epoch [165/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7908, Pure Ratio2 9.6928 +Epoch [165/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.6863 +Epoch [165/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6784 +Epoch [165/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.7320 +Epoch [165/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 61.7788 % Model2 61.1478 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Epoch [166/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1275 +Epoch [166/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.8562 +Epoch [166/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 9.9706 +Epoch [166/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 9.8588 +Epoch [166/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.8562 +Epoch [166/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9384, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 62.2196 % Model2 61.9992 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0196 +Epoch [167/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.6765 +Epoch [167/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6405, Pure Ratio2 9.6601 +Epoch [167/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.6176 +Epoch [167/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8784 +Epoch [167/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8954 +Epoch [167/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8683, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 62.2095 % Model2 62.3297 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [168/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.7353 +Epoch [168/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.8301 +Epoch [168/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0098 +Epoch [168/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.9647 +Epoch [168/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0458 +Epoch [168/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 63.0609 % Model2 62.6903 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8235 +Epoch [169/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9608 +Epoch [169/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9804 +Epoch [169/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8480 +Epoch [169/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8706 +Epoch [169/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.8627 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9272, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 62.5000 % Model2 62.2696 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.2745 +Epoch [170/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.5980 +Epoch [170/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6013 +Epoch [170/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7500, Pure Ratio2 9.7794 +Epoch [170/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7020, Pure Ratio2 9.7843 +Epoch [170/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6895, Pure Ratio2 9.7712 +Epoch [170/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7815, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 62.2396 % Model2 61.6486 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.9412 +Epoch [171/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6765 +Epoch [171/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.8497 +Epoch [171/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8725 +Epoch [171/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8980, Pure Ratio2 9.9216 +Epoch [171/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.8791 +Epoch [171/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8908, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 62.7804 % Model2 61.1378 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 10.2549 +Epoch [172/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2549 +Epoch [172/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0523 +Epoch [172/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.8676 +Epoch [172/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.8980 +Epoch [172/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8660 +Epoch [172/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8179, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 61.8990 % Model2 62.3397 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.4706 +Epoch [173/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.5980 +Epoch [173/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8170 +Epoch [173/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8382 +Epoch [173/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7373, Pure Ratio2 9.6902 +Epoch [173/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [173/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7955, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 62.1394 % Model2 61.7288 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.2745 +Epoch [174/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.7255 +Epoch [174/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.8366 +Epoch [174/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.9118 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8510, Pure Ratio2 9.9176 +Epoch [174/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8529 +Epoch [174/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 62.5701 % Model2 61.3181 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.6863 +Epoch [175/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.9902 +Epoch [175/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.9542 +Epoch [175/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7108, Pure Ratio2 9.9363 +Epoch [175/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6980, Pure Ratio2 9.8392 +Epoch [175/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7614, Pure Ratio2 9.9020 +Epoch [175/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 63.3714 % Model2 62.5200 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [176/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.7157 +Epoch [176/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7908 +Epoch [176/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 9.9118 +Epoch [176/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.7686 +Epoch [176/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7549 +Epoch [176/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 63.3514 % Model2 62.3698 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0196, Pure Ratio2 9.0000 +Epoch [177/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.5784 +Epoch [177/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.6797 +Epoch [177/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.7549 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6431, Pure Ratio2 9.5725 +Epoch [177/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7026, Pure Ratio2 9.6144 +Epoch [177/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6807, Pure Ratio2 9.6303 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 63.6518 % Model2 62.9307 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2353, Pure Ratio2 9.6863 +Epoch [178/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.9118 +Epoch [178/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [178/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.8971 +Epoch [178/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.8588 +Epoch [178/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7157 +Epoch [178/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 63.1010 % Model2 61.9491 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.5490 +Epoch [179/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3824, Pure Ratio2 9.4510 +Epoch [179/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.6667 +Epoch [179/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.7059 +Epoch [179/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.9098 +Epoch [179/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9967 +Epoch [179/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8179, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 62.2696 % Model2 62.9507 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.2157 +Epoch [180/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0098 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 9.9673 +Epoch [180/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8627 +Epoch [180/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8000 +Epoch [180/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8235 +Epoch [180/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 62.1595 % Model2 62.4099 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.5098 +Epoch [181/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9510 +Epoch [181/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.9346 +Epoch [181/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8529 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.9608 +Epoch [181/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8595 +Epoch [181/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 62.2095 % Model2 61.8790 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.5098 +Epoch [182/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0098 +Epoch [182/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.9020 +Epoch [182/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.8627 +Epoch [182/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7569 +Epoch [182/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6503, Pure Ratio2 9.6601 +Epoch [182/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7479, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 62.8606 % Model2 62.6202 %, Pure Ratio 1 9.7360 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6471 +Epoch [183/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.9118 +Epoch [183/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8954 +Epoch [183/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0441, Pure Ratio2 10.0931 +Epoch [183/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 10.0000 +Epoch [183/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9510 +Epoch [183/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 62.5300 % Model2 61.9591 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [184/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7549 +Epoch [184/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.9085 +Epoch [184/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7206, Pure Ratio2 9.8627 +Epoch [184/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8706, Pure Ratio2 9.9569 +Epoch [184/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0131 +Epoch [184/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 62.1895 % Model2 61.5184 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6078 +Epoch [185/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.4118 +Epoch [185/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.5686 +Epoch [185/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.7353 +Epoch [185/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.6902 +Epoch [185/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7614 +Epoch [185/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 63.0208 % Model2 62.0593 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.7511 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0392, Pure Ratio2 9.2157 +Epoch [186/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1765, Pure Ratio2 9.2549 +Epoch [186/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4837, Pure Ratio2 9.5817 +Epoch [186/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6127, Pure Ratio2 9.7157 +Epoch [186/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8706 +Epoch [186/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.8856 +Epoch [186/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8683, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 62.7704 % Model2 62.2997 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.1765 +Epoch [187/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.4706 +Epoch [187/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 10.0196 +Epoch [187/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9363 +Epoch [187/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9137 +Epoch [187/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8954 +Epoch [187/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8403, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 62.6002 % Model2 62.2296 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0065, Loss2: 0.0059, Pure Ratio1: 10.5294, Pure Ratio2 10.6471 +Epoch [188/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1569 +Epoch [188/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0327 +Epoch [188/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8137 +Epoch [188/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.8941 +Epoch [188/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.9052 +Epoch [188/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 62.1194 % Model2 61.9391 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.8235 +Epoch [189/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7745 +Epoch [189/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.8170 +Epoch [189/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7647 +Epoch [189/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.7922 +Epoch [189/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7092, Pure Ratio2 9.7974 +Epoch [189/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8011, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 62.0994 % Model2 61.9792 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0784, Pure Ratio2 9.2353 +Epoch [190/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.3529, Pure Ratio2 9.3824 +Epoch [190/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5948 +Epoch [190/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.6373 +Epoch [190/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6980, Pure Ratio2 9.5961 +Epoch [190/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7288, Pure Ratio2 9.6699 +Epoch [190/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 62.9107 % Model2 62.0893 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 10.0980 +Epoch [191/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.7647 +Epoch [191/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9935 +Epoch [191/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9363 +Epoch [191/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.8941 +Epoch [191/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.9510 +Epoch [191/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 62.9006 % Model2 62.6202 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3922 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1471 +Epoch [192/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.1111 +Epoch [192/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8922 +Epoch [192/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7765, Pure Ratio2 9.7333 +Epoch [192/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.8464 +Epoch [192/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 62.9407 % Model2 62.1595 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8627 +Epoch [193/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.7745 +Epoch [193/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.8039 +Epoch [193/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7304 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7961 +Epoch [193/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7026, Pure Ratio2 9.7549 +Epoch [193/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 62.7103 % Model2 62.4499 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.5490 +Epoch [194/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.9314 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7974 +Epoch [194/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6814, Pure Ratio2 9.7353 +Epoch [194/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7373, Pure Ratio2 9.7961 +Epoch [194/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8824 +Epoch [194/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7171, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 62.6402 % Model2 62.4099 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.0000 +Epoch [195/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3627, Pure Ratio2 9.3431 +Epoch [195/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4379, Pure Ratio2 9.4379 +Epoch [195/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6029, Pure Ratio2 9.6422 +Epoch [195/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5020, Pure Ratio2 9.5333 +Epoch [195/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5980 +Epoch [195/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 62.4499 % Model2 62.4399 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.5490 +Epoch [196/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9118 +Epoch [196/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9608 +Epoch [196/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9118 +Epoch [196/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.8784 +Epoch [196/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8497 +Epoch [196/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 63.0008 % Model2 62.0793 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0000 +Epoch [197/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 9.9902 +Epoch [197/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8562 +Epoch [197/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7108 +Epoch [197/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.7765 +Epoch [197/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7974 +Epoch [197/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8711, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 62.5200 % Model2 62.0292 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.9412 +Epoch [198/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.8137 +Epoch [198/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8039 +Epoch [198/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7794 +Epoch [198/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.8275 +Epoch [198/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7712 +Epoch [198/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7367, Pure Ratio2 9.6975 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 62.4499 % Model2 62.4499 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4902 +Epoch [199/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.6765 +Epoch [199/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.6732 +Epoch [199/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5735, Pure Ratio2 9.5686 +Epoch [199/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6627, Pure Ratio2 9.6000 +Epoch [199/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7255 +Epoch [199/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.8543, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 63.3413 % Model2 62.0893 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.7255 +Epoch [200/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.4314 +Epoch [200/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.1699 +Epoch [200/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.0882 +Epoch [200/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9490 +Epoch [200/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9085 +Epoch [200/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 62.9307 % Model2 62.5801 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.7888 % diff --git a/other_methods/coteaching/coteaching_results/out_0_6.log b/other_methods/coteaching/coteaching_results/out_0_6.log new file mode 100644 index 0000000..c123676 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_0_6.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 19.5312, Training Accuracy2: 21.8750, Loss1: 0.0178, Loss2: 0.0178, Pure Ratio1: 9.7920, Pure Ratio2 9.7600 +Epoch [2/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0167, Loss2: 0.0165, Pure Ratio1: 9.8160, Pure Ratio2 9.7920 +Epoch [2/200], Iter [150/390] Training Accuracy1: 21.0938, Training Accuracy2: 19.5312, Loss1: 0.0176, Loss2: 0.0176, Pure Ratio1: 9.9200, Pure Ratio2 9.8987 +Epoch [2/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 21.0938, Loss1: 0.0169, Loss2: 0.0169, Pure Ratio1: 9.9720, Pure Ratio2 9.9520 +Epoch [2/200], Iter [250/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.8750, Loss1: 0.0172, Loss2: 0.0171, Pure Ratio1: 9.9744, Pure Ratio2 9.9712 +Epoch [2/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0164, Loss2: 0.0165, Pure Ratio1: 9.9813, Pure Ratio2 9.9813 +Epoch [2/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0161, Loss2: 0.0162, Pure Ratio1: 9.9931, Pure Ratio2 9.9954 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 19.1406 % Model2 21.2139 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 21.8750, Loss1: 0.0172, Loss2: 0.0171, Pure Ratio1: 9.3607, Pure Ratio2 9.3770 +Epoch [3/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 24.2188, Loss1: 0.0168, Loss2: 0.0169, Pure Ratio1: 9.9098, Pure Ratio2 9.9426 +Epoch [3/200], Iter [150/390] Training Accuracy1: 23.4375, Training Accuracy2: 23.4375, Loss1: 0.0167, Loss2: 0.0167, Pure Ratio1: 10.1311, Pure Ratio2 10.1694 +Epoch [3/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 21.8750, Loss1: 0.0164, Loss2: 0.0165, Pure Ratio1: 10.0492, Pure Ratio2 10.0615 +Epoch [3/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0165, Loss2: 0.0163, Pure Ratio1: 10.1344, Pure Ratio2 10.1475 +Epoch [3/200], Iter [300/390] Training Accuracy1: 25.0000, Training Accuracy2: 23.4375, Loss1: 0.0169, Loss2: 0.0167, Pure Ratio1: 10.1120, Pure Ratio2 10.1148 +Epoch [3/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 24.2188, Loss1: 0.0170, Loss2: 0.0172, Pure Ratio1: 10.0679, Pure Ratio2 10.0679 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 19.9519 % Model2 19.6214 %, Pure Ratio 1 10.0147 %, Pure Ratio 2 10.0168 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.0938, Loss1: 0.0171, Loss2: 0.0178, Pure Ratio1: 9.5966, Pure Ratio2 9.7311 +Epoch [4/200], Iter [100/390] Training Accuracy1: 19.5312, Training Accuracy2: 21.8750, Loss1: 0.0175, Loss2: 0.0174, Pure Ratio1: 9.6218, Pure Ratio2 9.7143 +Epoch [4/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0161, Loss2: 0.0160, Pure Ratio1: 9.8768, Pure Ratio2 9.9048 +Epoch [4/200], Iter [200/390] Training Accuracy1: 16.4062, Training Accuracy2: 15.6250, Loss1: 0.0182, Loss2: 0.0179, Pure Ratio1: 9.7185, Pure Ratio2 9.7605 +Epoch [4/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 22.6562, Loss1: 0.0170, Loss2: 0.0170, Pure Ratio1: 9.8689, Pure Ratio2 9.8958 +Epoch [4/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0168, Loss2: 0.0171, Pure Ratio1: 9.8739, Pure Ratio2 9.9188 +Epoch [4/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0164, Loss2: 0.0166, Pure Ratio1: 9.9904, Pure Ratio2 10.0096 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 25.6811 % Model2 24.9599 %, Pure Ratio 1 9.9828 %, Pure Ratio 2 9.9978 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 29.6875, Loss1: 0.0168, Loss2: 0.0168, Pure Ratio1: 10.2586, Pure Ratio2 10.1207 +Epoch [5/200], Iter [100/390] Training Accuracy1: 25.0000, Training Accuracy2: 27.3438, Loss1: 0.0165, Loss2: 0.0166, Pure Ratio1: 10.4655, Pure Ratio2 10.4483 +Epoch [5/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 21.0938, Loss1: 0.0168, Loss2: 0.0176, Pure Ratio1: 10.4943, Pure Ratio2 10.4943 +Epoch [5/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0163, Loss2: 0.0161, Pure Ratio1: 10.2672, Pure Ratio2 10.2845 +Epoch [5/200], Iter [250/390] Training Accuracy1: 21.8750, Training Accuracy2: 25.7812, Loss1: 0.0170, Loss2: 0.0175, Pure Ratio1: 10.3000, Pure Ratio2 10.3207 +Epoch [5/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 23.4375, Loss1: 0.0161, Loss2: 0.0163, Pure Ratio1: 10.1983, Pure Ratio2 10.2126 +Epoch [5/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0169, Loss2: 0.0166, Pure Ratio1: 10.0640, Pure Ratio2 10.0764 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 28.3554 % Model2 29.3570 %, Pure Ratio 1 10.0133 %, Pure Ratio 2 10.0088 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 21.0938, Loss1: 0.0173, Loss2: 0.0175, Pure Ratio1: 10.8142, Pure Ratio2 10.8142 +Epoch [6/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 32.0312, Loss1: 0.0171, Loss2: 0.0167, Pure Ratio1: 10.4425, Pure Ratio2 10.4248 +Epoch [6/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0163, Loss2: 0.0157, Pure Ratio1: 10.1180, Pure Ratio2 10.1239 +Epoch [6/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 27.3438, Loss1: 0.0165, Loss2: 0.0167, Pure Ratio1: 10.0664, Pure Ratio2 10.0885 +Epoch [6/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 33.5938, Loss1: 0.0151, Loss2: 0.0153, Pure Ratio1: 10.0354, Pure Ratio2 10.0602 +Epoch [6/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0168, Loss2: 0.0167, Pure Ratio1: 9.9705, Pure Ratio2 10.0000 +Epoch [6/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0155, Pure Ratio1: 9.9494, Pure Ratio2 9.9595 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 25.1402 % Model2 25.1502 %, Pure Ratio 1 9.9614 %, Pure Ratio 2 9.9841 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0165, Loss2: 0.0166, Pure Ratio1: 9.8909, Pure Ratio2 9.8000 +Epoch [7/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 24.2188, Loss1: 0.0156, Loss2: 0.0158, Pure Ratio1: 9.7364, Pure Ratio2 9.7545 +Epoch [7/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 26.5625, Loss1: 0.0161, Loss2: 0.0159, Pure Ratio1: 9.9939, Pure Ratio2 9.9515 +Epoch [7/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0145, Loss2: 0.0140, Pure Ratio1: 9.9591, Pure Ratio2 9.9045 +Epoch [7/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0147, Loss2: 0.0143, Pure Ratio1: 9.9491, Pure Ratio2 9.9127 +Epoch [7/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0151, Loss2: 0.0149, Pure Ratio1: 9.9212, Pure Ratio2 9.8848 +Epoch [7/200], Iter [350/390] Training Accuracy1: 25.7812, Training Accuracy2: 27.3438, Loss1: 0.0167, Loss2: 0.0165, Pure Ratio1: 9.9662, Pure Ratio2 9.9325 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 31.3401 % Model2 30.7592 %, Pure Ratio 1 9.9860 %, Pure Ratio 2 9.9510 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.0312, Loss1: 0.0141, Loss2: 0.0145, Pure Ratio1: 10.1481, Pure Ratio2 10.1111 +Epoch [8/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0160, Loss2: 0.0155, Pure Ratio1: 9.6019, Pure Ratio2 9.6296 +Epoch [8/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0145, Loss2: 0.0144, Pure Ratio1: 9.5556, Pure Ratio2 9.5494 +Epoch [8/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0143, Loss2: 0.0145, Pure Ratio1: 9.5880, Pure Ratio2 9.5833 +Epoch [8/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 27.3438, Loss1: 0.0159, Loss2: 0.0154, Pure Ratio1: 9.7778, Pure Ratio2 9.7889 +Epoch [8/200], Iter [300/390] Training Accuracy1: 22.6562, Training Accuracy2: 29.6875, Loss1: 0.0170, Loss2: 0.0168, Pure Ratio1: 9.7253, Pure Ratio2 9.7593 +Epoch [8/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0159, Loss2: 0.0159, Pure Ratio1: 9.8545, Pure Ratio2 9.8889 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 28.7760 % Model2 28.8662 %, Pure Ratio 1 9.9668 %, Pure Ratio 2 9.9905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0141, Loss2: 0.0134, Pure Ratio1: 9.9238, Pure Ratio2 10.0381 +Epoch [9/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0138, Loss2: 0.0139, Pure Ratio1: 9.9143, Pure Ratio2 9.9810 +Epoch [9/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0142, Loss2: 0.0145, Pure Ratio1: 9.8921, Pure Ratio2 9.8603 +Epoch [9/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0172, Loss2: 0.0165, Pure Ratio1: 9.9381, Pure Ratio2 9.9238 +Epoch [9/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.0000, Loss1: 0.0158, Loss2: 0.0161, Pure Ratio1: 9.9886, Pure Ratio2 10.0038 +Epoch [9/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0147, Loss2: 0.0149, Pure Ratio1: 9.9333, Pure Ratio2 9.9556 +Epoch [9/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0142, Loss2: 0.0141, Pure Ratio1: 9.9891, Pure Ratio2 9.9973 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 31.4503 % Model2 30.2083 %, Pure Ratio 1 9.9756 %, Pure Ratio 2 10.0024 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 27.3438, Loss1: 0.0156, Loss2: 0.0158, Pure Ratio1: 9.8431, Pure Ratio2 9.7647 +Epoch [10/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.1250, Loss1: 0.0145, Loss2: 0.0148, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [10/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0155, Loss2: 0.0153, Pure Ratio1: 10.1111, Pure Ratio2 10.1111 +Epoch [10/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0147, Loss2: 0.0145, Pure Ratio1: 10.0735, Pure Ratio2 10.0931 +Epoch [10/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 28.9062, Loss1: 0.0149, Loss2: 0.0153, Pure Ratio1: 10.0510, Pure Ratio2 10.0275 +Epoch [10/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 26.5625, Loss1: 0.0155, Loss2: 0.0159, Pure Ratio1: 10.0065, Pure Ratio2 9.9902 +Epoch [10/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0151, Loss2: 0.0149, Pure Ratio1: 9.9916, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 32.9928 % Model2 33.8341 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0137, Loss2: 0.0137, Pure Ratio1: 10.3529, Pure Ratio2 10.2157 +Epoch [11/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0148, Loss2: 0.0151, Pure Ratio1: 10.1176, Pure Ratio2 10.0098 +Epoch [11/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 30.4688, Loss1: 0.0159, Loss2: 0.0161, Pure Ratio1: 10.0327, Pure Ratio2 10.0000 +Epoch [11/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0142, Loss2: 0.0147, Pure Ratio1: 9.9118, Pure Ratio2 9.8627 +Epoch [11/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0151, Loss2: 0.0156, Pure Ratio1: 9.9490, Pure Ratio2 9.9451 +Epoch [11/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0150, Loss2: 0.0151, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [11/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 34.3750, Loss1: 0.0142, Loss2: 0.0138, Pure Ratio1: 9.9552, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 32.4619 % Model2 33.2732 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0140, Loss2: 0.0136, Pure Ratio1: 9.8627, Pure Ratio2 9.9020 +Epoch [12/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 29.6875, Loss1: 0.0147, Loss2: 0.0146, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [12/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0125, Loss2: 0.0127, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Epoch [12/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0131, Loss2: 0.0130, Pure Ratio1: 9.8235, Pure Ratio2 9.7794 +Epoch [12/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0149, Pure Ratio1: 9.7961, Pure Ratio2 9.7804 +Epoch [12/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0150, Pure Ratio1: 9.8856, Pure Ratio2 9.8693 +Epoch [12/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0146, Loss2: 0.0142, Pure Ratio1: 9.8599, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 34.4050 % Model2 32.8726 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0145, Loss2: 0.0141, Pure Ratio1: 10.8824, Pure Ratio2 10.9216 +Epoch [13/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0153, Loss2: 0.0154, Pure Ratio1: 10.3137, Pure Ratio2 10.4118 +Epoch [13/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0144, Loss2: 0.0138, Pure Ratio1: 9.8301, Pure Ratio2 9.9281 +Epoch [13/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0145, Loss2: 0.0139, Pure Ratio1: 9.9412, Pure Ratio2 10.0343 +Epoch [13/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.0000, Loss1: 0.0168, Loss2: 0.0168, Pure Ratio1: 9.8039, Pure Ratio2 9.8980 +Epoch [13/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 30.4688, Loss1: 0.0148, Loss2: 0.0149, Pure Ratio1: 9.8889, Pure Ratio2 9.9673 +Epoch [13/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0138, Loss2: 0.0148, Pure Ratio1: 9.9076, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 34.1146 % Model2 35.6771 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0136, Loss2: 0.0138, Pure Ratio1: 9.9020, Pure Ratio2 10.0196 +Epoch [14/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0146, Loss2: 0.0157, Pure Ratio1: 9.5000, Pure Ratio2 9.5980 +Epoch [14/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 24.2188, Loss1: 0.0149, Loss2: 0.0154, Pure Ratio1: 9.6471, Pure Ratio2 9.7320 +Epoch [14/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0135, Loss2: 0.0140, Pure Ratio1: 9.8382, Pure Ratio2 9.9461 +Epoch [14/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0145, Pure Ratio1: 9.8235, Pure Ratio2 9.9020 +Epoch [14/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.8438, Loss1: 0.0149, Loss2: 0.0139, Pure Ratio1: 9.7745, Pure Ratio2 9.8235 +Epoch [14/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0134, Loss2: 0.0136, Pure Ratio1: 9.8515, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 33.9343 % Model2 33.6739 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 31.2500, Loss1: 0.0141, Loss2: 0.0148, Pure Ratio1: 9.0784, Pure Ratio2 9.3137 +Epoch [15/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0134, Loss2: 0.0132, Pure Ratio1: 9.6765, Pure Ratio2 9.7843 +Epoch [15/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.0312, Loss1: 0.0151, Loss2: 0.0153, Pure Ratio1: 9.9739, Pure Ratio2 10.0458 +Epoch [15/200], Iter [200/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.7812, Loss1: 0.0171, Loss2: 0.0162, Pure Ratio1: 10.0343, Pure Ratio2 10.0490 +Epoch [15/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0155, Loss2: 0.0150, Pure Ratio1: 9.9647, Pure Ratio2 9.9882 +Epoch [15/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0136, Pure Ratio1: 9.9967, Pure Ratio2 9.9837 +Epoch [15/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0140, Loss2: 0.0139, Pure Ratio1: 9.9076, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 34.1246 % Model2 34.1546 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0139, Loss2: 0.0144, Pure Ratio1: 9.5294, Pure Ratio2 9.4510 +Epoch [16/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 25.0000, Loss1: 0.0158, Loss2: 0.0160, Pure Ratio1: 9.6863, Pure Ratio2 9.7255 +Epoch [16/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0140, Loss2: 0.0150, Pure Ratio1: 9.6209, Pure Ratio2 9.6340 +Epoch [16/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0136, Loss2: 0.0132, Pure Ratio1: 9.7598, Pure Ratio2 9.7794 +Epoch [16/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0182, Loss2: 0.0172, Pure Ratio1: 9.7882, Pure Ratio2 9.8118 +Epoch [16/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0155, Loss2: 0.0154, Pure Ratio1: 9.8072, Pure Ratio2 9.8268 +Epoch [16/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 29.6875, Loss1: 0.0150, Loss2: 0.0154, Pure Ratio1: 9.9412, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 34.6154 % Model2 34.2147 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0132, Loss2: 0.0136, Pure Ratio1: 9.6863, Pure Ratio2 9.4706 +Epoch [17/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0137, Loss2: 0.0139, Pure Ratio1: 9.8725, Pure Ratio2 9.8431 +Epoch [17/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0141, Loss2: 0.0136, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [17/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0131, Loss2: 0.0129, Pure Ratio1: 10.0784, Pure Ratio2 10.0931 +Epoch [17/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 35.9375, Loss1: 0.0125, Loss2: 0.0123, Pure Ratio1: 10.0353, Pure Ratio2 10.0471 +Epoch [17/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0150, Loss2: 0.0148, Pure Ratio1: 10.0490, Pure Ratio2 9.9935 +Epoch [17/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0148, Loss2: 0.0149, Pure Ratio1: 9.9860, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 33.4836 % Model2 32.8826 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0112, Loss2: 0.0108, Pure Ratio1: 9.1176, Pure Ratio2 9.2745 +Epoch [18/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0145, Loss2: 0.0145, Pure Ratio1: 9.4216, Pure Ratio2 9.5490 +Epoch [18/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0156, Loss2: 0.0154, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [18/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0135, Loss2: 0.0143, Pure Ratio1: 9.7696, Pure Ratio2 9.8088 +Epoch [18/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0151, Loss2: 0.0146, Pure Ratio1: 9.7412, Pure Ratio2 9.8196 +Epoch [18/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0136, Loss2: 0.0135, Pure Ratio1: 9.7647, Pure Ratio2 9.8333 +Epoch [18/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0158, Loss2: 0.0157, Pure Ratio1: 9.8599, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 32.6122 % Model2 32.4119 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0128, Loss2: 0.0126, Pure Ratio1: 9.7255, Pure Ratio2 9.7843 +Epoch [19/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0123, Loss2: 0.0133, Pure Ratio1: 9.6471, Pure Ratio2 9.5588 +Epoch [19/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 43.7500, Loss1: 0.0114, Loss2: 0.0115, Pure Ratio1: 9.7843, Pure Ratio2 9.7974 +Epoch [19/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0139, Loss2: 0.0142, Pure Ratio1: 9.7500, Pure Ratio2 9.7843 +Epoch [19/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0118, Loss2: 0.0121, Pure Ratio1: 9.8549, Pure Ratio2 9.8824 +Epoch [19/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0135, Loss2: 0.0132, Pure Ratio1: 9.9183, Pure Ratio2 9.9444 +Epoch [19/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 35.1562, Loss1: 0.0127, Loss2: 0.0135, Pure Ratio1: 9.9664, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 35.3566 % Model2 34.4451 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0102, Loss2: 0.0108, Pure Ratio1: 9.4902, Pure Ratio2 9.4706 +Epoch [20/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0111, Loss2: 0.0118, Pure Ratio1: 9.8039, Pure Ratio2 9.8137 +Epoch [20/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0098, Loss2: 0.0102, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [20/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0147, Loss2: 0.0142, Pure Ratio1: 9.8971, Pure Ratio2 9.9314 +Epoch [20/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0128, Loss2: 0.0130, Pure Ratio1: 9.9569, Pure Ratio2 9.9843 +Epoch [20/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0132, Loss2: 0.0126, Pure Ratio1: 9.9935, Pure Ratio2 9.9869 +Epoch [20/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0118, Loss2: 0.0122, Pure Ratio1: 9.9300, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 32.2015 % Model2 33.0629 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0107, Loss2: 0.0106, Pure Ratio1: 9.4902, Pure Ratio2 9.6275 +Epoch [21/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0099, Loss2: 0.0106, Pure Ratio1: 9.8039, Pure Ratio2 9.8725 +Epoch [21/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0122, Loss2: 0.0130, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [21/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0113, Loss2: 0.0122, Pure Ratio1: 10.0637, Pure Ratio2 10.0245 +Epoch [21/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0133, Loss2: 0.0131, Pure Ratio1: 10.0196, Pure Ratio2 10.0157 +Epoch [21/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0113, Loss2: 0.0121, Pure Ratio1: 9.9771, Pure Ratio2 9.9804 +Epoch [21/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 39.8438, Loss1: 0.0117, Loss2: 0.0119, Pure Ratio1: 10.0028, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 33.5136 % Model2 33.6238 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0097, Loss2: 0.0104, Pure Ratio1: 9.2353, Pure Ratio2 9.0980 +Epoch [22/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0124, Loss2: 0.0129, Pure Ratio1: 9.5294, Pure Ratio2 9.5196 +Epoch [22/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 43.7500, Loss1: 0.0129, Loss2: 0.0117, Pure Ratio1: 9.4575, Pure Ratio2 9.5033 +Epoch [22/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.1875, Loss1: 0.0103, Loss2: 0.0115, Pure Ratio1: 9.5833, Pure Ratio2 9.6078 +Epoch [22/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0109, Loss2: 0.0116, Pure Ratio1: 9.6706, Pure Ratio2 9.6745 +Epoch [22/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0128, Loss2: 0.0122, Pure Ratio1: 9.7549, Pure Ratio2 9.7484 +Epoch [22/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0111, Loss2: 0.0109, Pure Ratio1: 9.8683, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 33.6138 % Model2 34.6254 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0095, Loss2: 0.0106, Pure Ratio1: 10.0196, Pure Ratio2 9.7843 +Epoch [23/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0112, Loss2: 0.0112, Pure Ratio1: 9.9608, Pure Ratio2 9.7647 +Epoch [23/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0117, Loss2: 0.0113, Pure Ratio1: 9.8889, Pure Ratio2 9.7516 +Epoch [23/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0117, Loss2: 0.0114, Pure Ratio1: 9.8431, Pure Ratio2 9.7500 +Epoch [23/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0085, Loss2: 0.0093, Pure Ratio1: 9.8235, Pure Ratio2 9.7529 +Epoch [23/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0103, Loss2: 0.0099, Pure Ratio1: 9.8333, Pure Ratio2 9.7908 +Epoch [23/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0107, Loss2: 0.0120, Pure Ratio1: 9.9580, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 33.3333 % Model2 33.4235 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 52.3438, Loss1: 0.0076, Loss2: 0.0084, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [24/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0098, Loss2: 0.0109, Pure Ratio1: 9.6275, Pure Ratio2 9.5784 +Epoch [24/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0083, Loss2: 0.0081, Pure Ratio1: 9.6993, Pure Ratio2 9.6863 +Epoch [24/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0094, Loss2: 0.0101, Pure Ratio1: 9.8627, Pure Ratio2 9.7941 +Epoch [24/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0101, Loss2: 0.0100, Pure Ratio1: 9.8510, Pure Ratio2 9.7882 +Epoch [24/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0091, Loss2: 0.0095, Pure Ratio1: 9.7745, Pure Ratio2 9.7516 +Epoch [24/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0096, Loss2: 0.0104, Pure Ratio1: 9.8515, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 34.3850 % Model2 30.8794 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0095, Loss2: 0.0084, Pure Ratio1: 9.6078, Pure Ratio2 9.5882 +Epoch [25/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0075, Loss2: 0.0081, Pure Ratio1: 9.9118, Pure Ratio2 9.9510 +Epoch [25/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0087, Loss2: 0.0103, Pure Ratio1: 9.9673, Pure Ratio2 10.0000 +Epoch [25/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0099, Loss2: 0.0103, Pure Ratio1: 9.9020, Pure Ratio2 9.9559 +Epoch [25/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0079, Loss2: 0.0079, Pure Ratio1: 9.9137, Pure Ratio2 9.9490 +Epoch [25/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0114, Pure Ratio1: 9.8399, Pure Ratio2 9.8824 +Epoch [25/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0076, Loss2: 0.0090, Pure Ratio1: 9.8235, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 30.4988 % Model2 32.5621 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0069, Loss2: 0.0082, Pure Ratio1: 9.6863, Pure Ratio2 9.4314 +Epoch [26/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0071, Loss2: 0.0074, Pure Ratio1: 9.6078, Pure Ratio2 9.5588 +Epoch [26/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0064, Loss2: 0.0066, Pure Ratio1: 9.7908, Pure Ratio2 9.7974 +Epoch [26/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0070, Loss2: 0.0074, Pure Ratio1: 10.0000, Pure Ratio2 9.9706 +Epoch [26/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0083, Loss2: 0.0076, Pure Ratio1: 9.8431, Pure Ratio2 9.8314 +Epoch [26/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0061, Loss2: 0.0074, Pure Ratio1: 9.9085, Pure Ratio2 9.9085 +Epoch [26/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0080, Loss2: 0.0079, Pure Ratio1: 9.8263, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 32.2917 % Model2 33.8942 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0064, Loss2: 0.0067, Pure Ratio1: 9.7059, Pure Ratio2 9.8824 +Epoch [27/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0072, Loss2: 0.0075, Pure Ratio1: 9.6471, Pure Ratio2 9.7255 +Epoch [27/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.9688, Loss1: 0.0053, Loss2: 0.0058, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [27/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0068, Loss2: 0.0063, Pure Ratio1: 10.0147, Pure Ratio2 10.0294 +Epoch [27/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0062, Loss2: 0.0063, Pure Ratio1: 9.8902, Pure Ratio2 9.8980 +Epoch [27/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0074, Loss2: 0.0082, Pure Ratio1: 9.8954, Pure Ratio2 9.9248 +Epoch [27/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 59.3750, Loss1: 0.0058, Loss2: 0.0067, Pure Ratio1: 9.9188, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 31.6206 % Model2 30.3986 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0048, Loss2: 0.0058, Pure Ratio1: 9.8431, Pure Ratio2 10.0784 +Epoch [28/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0056, Loss2: 0.0054, Pure Ratio1: 9.5392, Pure Ratio2 9.8431 +Epoch [28/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0049, Loss2: 0.0063, Pure Ratio1: 9.9412, Pure Ratio2 10.1569 +Epoch [28/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0055, Loss2: 0.0067, Pure Ratio1: 10.0686, Pure Ratio2 10.2402 +Epoch [28/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0054, Loss2: 0.0056, Pure Ratio1: 9.7333, Pure Ratio2 9.8980 +Epoch [28/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0066, Loss2: 0.0058, Pure Ratio1: 9.8072, Pure Ratio2 9.8954 +Epoch [28/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0066, Loss2: 0.0074, Pure Ratio1: 9.8852, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 30.9696 % Model2 29.9179 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0044, Loss2: 0.0060, Pure Ratio1: 10.2549, Pure Ratio2 10.1961 +Epoch [29/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0045, Loss2: 0.0052, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [29/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 72.6562, Loss1: 0.0050, Loss2: 0.0043, Pure Ratio1: 10.2353, Pure Ratio2 10.1830 +Epoch [29/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0063, Loss2: 0.0067, Pure Ratio1: 10.2451, Pure Ratio2 10.1863 +Epoch [29/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0061, Loss2: 0.0066, Pure Ratio1: 10.1373, Pure Ratio2 10.0941 +Epoch [29/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0053, Loss2: 0.0066, Pure Ratio1: 9.9967, Pure Ratio2 9.9804 +Epoch [29/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0051, Loss2: 0.0056, Pure Ratio1: 9.9384, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 29.9279 % Model2 29.9980 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0051, Loss2: 0.0057, Pure Ratio1: 9.5294, Pure Ratio2 9.4902 +Epoch [30/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0058, Loss2: 0.0051, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Epoch [30/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.0037, Loss2: 0.0050, Pure Ratio1: 9.9281, Pure Ratio2 9.9346 +Epoch [30/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.0046, Loss2: 0.0036, Pure Ratio1: 9.8922, Pure Ratio2 9.9167 +Epoch [30/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0040, Loss2: 0.0045, Pure Ratio1: 9.8392, Pure Ratio2 9.8627 +Epoch [30/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 54.6875, Loss1: 0.0061, Loss2: 0.0082, Pure Ratio1: 9.9216, Pure Ratio2 9.9412 +Epoch [30/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0041, Loss2: 0.0049, Pure Ratio1: 9.8908, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 29.9179 % Model2 29.3570 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0032, Loss2: 0.0040, Pure Ratio1: 10.2353, Pure Ratio2 10.0588 +Epoch [31/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0047, Loss2: 0.0052, Pure Ratio1: 9.9314, Pure Ratio2 9.8725 +Epoch [31/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0043, Loss2: 0.0056, Pure Ratio1: 9.9281, Pure Ratio2 9.9020 +Epoch [31/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0048, Loss2: 0.0034, Pure Ratio1: 10.0784, Pure Ratio2 10.0441 +Epoch [31/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0037, Loss2: 0.0042, Pure Ratio1: 9.9922, Pure Ratio2 9.9647 +Epoch [31/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0045, Loss2: 0.0063, Pure Ratio1: 9.9967, Pure Ratio2 9.9510 +Epoch [31/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0040, Loss2: 0.0042, Pure Ratio1: 10.0420, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 32.6122 % Model2 32.6322 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0029, Loss2: 0.0036, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [32/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0025, Loss2: 0.0036, Pure Ratio1: 9.7451, Pure Ratio2 9.8824 +Epoch [32/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0030, Loss2: 0.0029, Pure Ratio1: 10.0261, Pure Ratio2 10.0392 +Epoch [32/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0025, Loss2: 0.0030, Pure Ratio1: 10.1127, Pure Ratio2 10.0931 +Epoch [32/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 77.3438, Loss1: 0.0038, Loss2: 0.0031, Pure Ratio1: 10.0196, Pure Ratio2 10.0118 +Epoch [32/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0044, Loss2: 0.0055, Pure Ratio1: 9.9542, Pure Ratio2 9.9248 +Epoch [32/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0044, Loss2: 0.0044, Pure Ratio1: 9.9720, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 28.5457 % Model2 28.5958 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0045, Loss2: 0.0042, Pure Ratio1: 9.7255, Pure Ratio2 9.5294 +Epoch [33/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 77.3438, Loss1: 0.0020, Loss2: 0.0025, Pure Ratio1: 9.9216, Pure Ratio2 9.7549 +Epoch [33/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0035, Loss2: 0.0027, Pure Ratio1: 9.7908, Pure Ratio2 9.7386 +Epoch [33/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0023, Loss2: 0.0029, Pure Ratio1: 9.9461, Pure Ratio2 9.9020 +Epoch [33/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0029, Loss2: 0.0032, Pure Ratio1: 9.9059, Pure Ratio2 9.8784 +Epoch [33/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 72.6562, Loss1: 0.0028, Loss2: 0.0035, Pure Ratio1: 9.9673, Pure Ratio2 9.9248 +Epoch [33/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0044, Loss2: 0.0033, Pure Ratio1: 9.9188, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 30.7192 % Model2 30.5889 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 76.5625, Loss1: 0.0017, Loss2: 0.0022, Pure Ratio1: 10.3725, Pure Ratio2 10.4118 +Epoch [34/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0021, Loss2: 0.0019, Pure Ratio1: 9.9412, Pure Ratio2 9.8922 +Epoch [34/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 74.2188, Loss1: 0.0028, Loss2: 0.0025, Pure Ratio1: 9.9150, Pure Ratio2 9.9020 +Epoch [34/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0026, Loss2: 0.0035, Pure Ratio1: 9.8824, Pure Ratio2 9.8284 +Epoch [34/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 73.4375, Loss1: 0.0026, Loss2: 0.0041, Pure Ratio1: 9.8471, Pure Ratio2 9.8078 +Epoch [34/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0059, Loss2: 0.0073, Pure Ratio1: 9.8595, Pure Ratio2 9.8137 +Epoch [34/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0045, Loss2: 0.0040, Pure Ratio1: 9.9888, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 30.7792 % Model2 29.5974 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0019, Loss2: 0.0021, Pure Ratio1: 9.2157, Pure Ratio2 9.3725 +Epoch [35/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.8725, Pure Ratio2 9.8725 +Epoch [35/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 79.6875, Loss1: 0.0033, Loss2: 0.0022, Pure Ratio1: 9.8105, Pure Ratio2 9.8889 +Epoch [35/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0022, Loss2: 0.0026, Pure Ratio1: 9.8382, Pure Ratio2 9.9118 +Epoch [35/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 81.2500, Loss1: 0.0017, Loss2: 0.0025, Pure Ratio1: 9.9020, Pure Ratio2 9.9882 +Epoch [35/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0039, Pure Ratio1: 10.0033, Pure Ratio2 10.0915 +Epoch [35/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0019, Pure Ratio1: 10.0084, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 29.0865 % Model2 30.8293 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0024, Loss2: 0.0017, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [36/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0034, Loss2: 0.0025, Pure Ratio1: 9.8725, Pure Ratio2 9.8137 +Epoch [36/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0016, Loss2: 0.0027, Pure Ratio1: 9.9739, Pure Ratio2 9.9608 +Epoch [36/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.0022, Loss2: 0.0031, Pure Ratio1: 9.9706, Pure Ratio2 9.9608 +Epoch [36/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0021, Loss2: 0.0019, Pure Ratio1: 9.9725, Pure Ratio2 9.9961 +Epoch [36/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.0027, Loss2: 0.0020, Pure Ratio1: 9.8758, Pure Ratio2 9.9052 +Epoch [36/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 88.2812, Loss1: 0.0021, Loss2: 0.0014, Pure Ratio1: 9.9244, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 30.1983 % Model2 29.4671 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 9.7647, Pure Ratio2 9.7647 +Epoch [37/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.7843, Pure Ratio2 9.8922 +Epoch [37/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [37/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0010, Pure Ratio1: 10.1078, Pure Ratio2 10.1029 +Epoch [37/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 73.4375, Loss1: 0.0013, Loss2: 0.0024, Pure Ratio1: 9.9373, Pure Ratio2 9.9176 +Epoch [37/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0018, Loss2: 0.0024, Pure Ratio1: 9.9935, Pure Ratio2 9.9935 +Epoch [37/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0021, Pure Ratio1: 9.9048, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 30.3686 % Model2 32.2917 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0042, Loss2: 0.0031, Pure Ratio1: 9.7059, Pure Ratio2 9.7255 +Epoch [38/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 10.0098, Pure Ratio2 10.0294 +Epoch [38/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.7647, Pure Ratio2 9.7908 +Epoch [38/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.7353, Pure Ratio2 9.7647 +Epoch [38/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0014, Loss2: 0.0015, Pure Ratio1: 9.9098, Pure Ratio2 9.9255 +Epoch [38/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0022, Loss2: 0.0021, Pure Ratio1: 9.9314, Pure Ratio2 9.9248 +Epoch [38/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0056, Loss2: 0.0043, Pure Ratio1: 9.8599, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 28.8662 % Model2 30.0481 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0019, Pure Ratio1: 10.0392, Pure Ratio2 10.0588 +Epoch [39/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.0392, Pure Ratio2 10.0294 +Epoch [39/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0022, Loss2: 0.0015, Pure Ratio1: 10.1830, Pure Ratio2 10.1307 +Epoch [39/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 10.0098, Pure Ratio2 9.9510 +Epoch [39/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0016, Pure Ratio1: 10.0510, Pure Ratio2 10.0235 +Epoch [39/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0041, Loss2: 0.0050, Pure Ratio1: 9.9608, Pure Ratio2 9.9510 +Epoch [39/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.9076, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 29.7576 % Model2 28.9964 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0024, Loss2: 0.0024, Pure Ratio1: 10.5686, Pure Ratio2 10.5490 +Epoch [40/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0019, Pure Ratio1: 10.2255, Pure Ratio2 10.2157 +Epoch [40/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0026, Loss2: 0.0031, Pure Ratio1: 10.0588, Pure Ratio2 10.0000 +Epoch [40/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0023, Loss2: 0.0016, Pure Ratio1: 9.8431, Pure Ratio2 9.7745 +Epoch [40/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0021, Pure Ratio1: 9.7882, Pure Ratio2 9.7255 +Epoch [40/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.0013, Loss2: 0.0018, Pure Ratio1: 9.7124, Pure Ratio2 9.6601 +Epoch [40/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.0026, Loss2: 0.0016, Pure Ratio1: 9.9580, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 29.3470 % Model2 29.6474 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0020, Loss2: 0.0015, Pure Ratio1: 9.4118, Pure Ratio2 9.8627 +Epoch [41/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.7812, Loss1: 0.0012, Loss2: 0.0025, Pure Ratio1: 9.5588, Pure Ratio2 9.8333 +Epoch [41/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 9.6013, Pure Ratio2 9.7255 +Epoch [41/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.6716, Pure Ratio2 9.8039 +Epoch [41/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.7059, Pure Ratio2 9.8000 +Epoch [41/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.8203, Pure Ratio2 9.9216 +Epoch [41/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0015, Loss2: 0.0011, Pure Ratio1: 9.8515, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 31.5705 % Model2 30.0681 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.0392, Pure Ratio2 9.9216 +Epoch [42/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0023, Pure Ratio1: 10.0392, Pure Ratio2 9.9510 +Epoch [42/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 9.9412, Pure Ratio2 9.9281 +Epoch [42/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 10.1029, Pure Ratio2 10.0000 +Epoch [42/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0020, Pure Ratio1: 10.1490, Pure Ratio2 10.0431 +Epoch [42/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 10.1111, Pure Ratio2 10.0229 +Epoch [42/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 10.0280, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 29.7676 % Model2 31.2400 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0020, Pure Ratio1: 9.4706, Pure Ratio2 9.4510 +Epoch [43/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.6373, Pure Ratio2 9.6078 +Epoch [43/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0017, Pure Ratio1: 9.8366, Pure Ratio2 9.8366 +Epoch [43/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 10.0147, Pure Ratio2 9.9853 +Epoch [43/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.8438, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 10.1176, Pure Ratio2 10.0902 +Epoch [43/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0014, Loss2: 0.0024, Pure Ratio1: 10.0327, Pure Ratio2 10.0229 +Epoch [43/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0013, Pure Ratio1: 9.9944, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 28.3253 % Model2 30.6490 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [44/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0035, Loss2: 0.0020, Pure Ratio1: 10.0098, Pure Ratio2 9.9804 +Epoch [44/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.0023, Loss2: 0.0015, Pure Ratio1: 9.9673, Pure Ratio2 9.9085 +Epoch [44/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0033, Pure Ratio1: 9.9314, Pure Ratio2 9.8824 +Epoch [44/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.9608, Pure Ratio2 9.9098 +Epoch [44/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0024, Loss2: 0.0010, Pure Ratio1: 9.9248, Pure Ratio2 9.8497 +Epoch [44/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.9692, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 29.0865 % Model2 31.1999 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 10.3529, Pure Ratio2 10.3137 +Epoch [45/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 10.0686, Pure Ratio2 10.1176 +Epoch [45/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.9020, Pure Ratio2 9.9542 +Epoch [45/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 9.7549, Pure Ratio2 9.8529 +Epoch [45/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.7843, Pure Ratio2 9.8549 +Epoch [45/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 9.8235, Pure Ratio2 9.8366 +Epoch [45/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.8627, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 32.2616 % Model2 29.3069 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0017, Pure Ratio1: 9.4314, Pure Ratio2 9.3922 +Epoch [46/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.6961, Pure Ratio2 9.5882 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 9.8170, Pure Ratio2 9.8039 +Epoch [46/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0016, Loss2: 0.0013, Pure Ratio1: 10.0049, Pure Ratio2 10.0147 +Epoch [46/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9490, Pure Ratio2 10.0000 +Epoch [46/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0014, Pure Ratio1: 10.0425, Pure Ratio2 10.0556 +Epoch [46/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0011, Loss2: 0.0017, Pure Ratio1: 9.9636, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 28.6659 % Model2 30.7492 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.4118, Pure Ratio2 9.6078 +Epoch [47/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.4902, Pure Ratio2 9.6176 +Epoch [47/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 9.7778, Pure Ratio2 9.9085 +Epoch [47/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.7990, Pure Ratio2 9.8578 +Epoch [47/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.8235, Pure Ratio2 9.9216 +Epoch [47/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0023, Pure Ratio1: 9.8529, Pure Ratio2 9.9641 +Epoch [47/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 9.8964, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 31.6106 % Model2 29.8878 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.3137, Pure Ratio2 9.2157 +Epoch [48/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.5294, Pure Ratio2 9.4118 +Epoch [48/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0015, Loss2: 0.0011, Pure Ratio1: 9.8235, Pure Ratio2 9.7386 +Epoch [48/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.8480, Pure Ratio2 9.7402 +Epoch [48/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0016, Pure Ratio1: 9.9804, Pure Ratio2 9.8667 +Epoch [48/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.9902, Pure Ratio2 9.8824 +Epoch [48/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0019, Loss2: 0.0013, Pure Ratio1: 10.0056, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 31.7208 % Model2 31.9611 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 10.5490, Pure Ratio2 10.6078 +Epoch [49/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 10.4314, Pure Ratio2 10.3235 +Epoch [49/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 83.5938, Loss1: 0.0020, Loss2: 0.0004, Pure Ratio1: 10.4248, Pure Ratio2 10.3137 +Epoch [49/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 10.3039, Pure Ratio2 10.2010 +Epoch [49/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0018, Pure Ratio1: 10.2627, Pure Ratio2 10.1412 +Epoch [49/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0016, Loss2: 0.0011, Pure Ratio1: 10.0784, Pure Ratio2 9.9673 +Epoch [49/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 10.0952, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 30.7993 % Model2 29.9980 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 9.9804 +Epoch [50/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0021, Pure Ratio1: 10.0000, Pure Ratio2 9.9314 +Epoch [50/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.9869, Pure Ratio2 9.9477 +Epoch [50/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.9804, Pure Ratio2 9.8922 +Epoch [50/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.7608, Pure Ratio2 9.6588 +Epoch [50/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.8072, Pure Ratio2 9.7288 +Epoch [50/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0032, Loss2: 0.0041, Pure Ratio1: 9.8655, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 30.7492 % Model2 30.3085 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.0392, Pure Ratio2 9.9608 +Epoch [51/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 10.0882, Pure Ratio2 10.0784 +Epoch [51/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 10.0458, Pure Ratio2 10.1373 +Epoch [51/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9902, Pure Ratio2 10.0735 +Epoch [51/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0021, Loss2: 0.0008, Pure Ratio1: 9.8275, Pure Ratio2 9.9176 +Epoch [51/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.8595, Pure Ratio2 9.9248 +Epoch [51/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0014, Pure Ratio1: 9.8908, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 30.0180 % Model2 32.6322 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8039, Pure Ratio2 9.9804 +Epoch [52/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.9314, Pure Ratio2 9.9412 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0018, Loss2: 0.0013, Pure Ratio1: 9.8562, Pure Ratio2 9.8170 +Epoch [52/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.6275, Pure Ratio2 9.6520 +Epoch [52/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 9.7412, Pure Ratio2 9.7804 +Epoch [52/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0012, Pure Ratio1: 9.7516, Pure Ratio2 9.7680 +Epoch [52/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.8655, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 31.2700 % Model2 31.3301 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.5490, Pure Ratio2 9.5882 +Epoch [53/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.1275, Pure Ratio2 10.1471 +Epoch [53/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 10.1438, Pure Ratio2 10.1569 +Epoch [53/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 10.1716, Pure Ratio2 10.1520 +Epoch [53/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0510, Pure Ratio2 9.9922 +Epoch [53/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9902, Pure Ratio2 9.9314 +Epoch [53/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0014, Loss2: 0.0013, Pure Ratio1: 9.9916, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 29.9579 % Model2 30.0180 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [54/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 9.8529, Pure Ratio2 9.8922 +Epoch [54/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0032, Loss2: 0.0014, Pure Ratio1: 9.8366, Pure Ratio2 9.8366 +Epoch [54/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0022, Loss2: 0.0008, Pure Ratio1: 9.8235, Pure Ratio2 9.7353 +Epoch [54/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.8078, Pure Ratio2 9.7529 +Epoch [54/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.8758, Pure Ratio2 9.8235 +Epoch [54/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0028, Pure Ratio1: 9.9160, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 29.4671 % Model2 30.4387 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.1569, Pure Ratio2 9.0980 +Epoch [55/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.0980, Pure Ratio2 8.9412 +Epoch [55/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.4771, Pure Ratio2 9.3464 +Epoch [55/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.9657, Pure Ratio2 9.8333 +Epoch [55/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 10.1529, Pure Ratio2 10.0588 +Epoch [55/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0014, Loss2: 0.0008, Pure Ratio1: 10.0261, Pure Ratio2 9.9706 +Epoch [55/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.9356, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 28.6458 % Model2 30.2985 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0010, Pure Ratio1: 9.5098, Pure Ratio2 9.6275 +Epoch [56/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0019, Loss2: 0.0006, Pure Ratio1: 9.6667, Pure Ratio2 9.8627 +Epoch [56/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.9346, Pure Ratio2 10.0784 +Epoch [56/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8382, Pure Ratio2 10.0049 +Epoch [56/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0014, Loss2: 0.0010, Pure Ratio1: 9.9451, Pure Ratio2 10.0706 +Epoch [56/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.9739, Pure Ratio2 10.0784 +Epoch [56/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 9.9636, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 31.3101 % Model2 31.4203 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0013, Loss2: 0.0020, Pure Ratio1: 9.5294, Pure Ratio2 9.8235 +Epoch [57/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0005, Pure Ratio1: 9.7941, Pure Ratio2 9.9216 +Epoch [57/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.1176 +Epoch [57/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.9853, Pure Ratio2 10.0490 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.0000, Pure Ratio2 10.0941 +Epoch [57/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.9837, Pure Ratio2 10.0261 +Epoch [57/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9356, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 29.0865 % Model2 30.2183 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.6667, Pure Ratio2 9.7255 +Epoch [58/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.8627, Pure Ratio2 10.0784 +Epoch [58/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9673, Pure Ratio2 10.0719 +Epoch [58/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.9951 +Epoch [58/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0021, Pure Ratio1: 9.8941, Pure Ratio2 9.9686 +Epoch [58/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.8301, Pure Ratio2 9.9118 +Epoch [58/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8039, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 29.1967 % Model2 30.5789 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0022, Pure Ratio1: 10.3922, Pure Ratio2 10.3333 +Epoch [59/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.1667, Pure Ratio2 10.1961 +Epoch [59/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.2092, Pure Ratio2 10.2222 +Epoch [59/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0028, Loss2: 0.0023, Pure Ratio1: 10.1618, Pure Ratio2 10.1716 +Epoch [59/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 10.0078, Pure Ratio2 10.0431 +Epoch [59/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0019, Pure Ratio1: 9.9412, Pure Ratio2 10.0131 +Epoch [59/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.8431, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 29.1066 % Model2 30.3486 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0012, Pure Ratio1: 8.8824, Pure Ratio2 9.1176 +Epoch [60/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [60/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8758, Pure Ratio2 9.9608 +Epoch [60/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.9363 +Epoch [60/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.1098, Pure Ratio2 10.1608 +Epoch [60/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0015, Loss2: 0.0009, Pure Ratio1: 9.9967, Pure Ratio2 10.0261 +Epoch [60/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9636, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 30.5188 % Model2 30.0481 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.1569, Pure Ratio2 10.1569 +Epoch [61/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.3137, Pure Ratio2 10.4314 +Epoch [61/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.1438 +Epoch [61/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9167, Pure Ratio2 10.1422 +Epoch [61/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8510, Pure Ratio2 10.0706 +Epoch [61/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 10.0458 +Epoch [61/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0018, Pure Ratio1: 9.8768, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 30.8994 % Model2 30.3185 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 10.0392 +Epoch [62/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [62/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8105, Pure Ratio2 9.7974 +Epoch [62/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9265, Pure Ratio2 9.8529 +Epoch [62/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0235, Pure Ratio2 10.0314 +Epoch [62/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 10.0000 +Epoch [62/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.8880, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 30.0080 % Model2 29.6274 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.2353, Pure Ratio2 10.0980 +Epoch [63/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [63/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7516, Pure Ratio2 9.6863 +Epoch [63/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.0735, Pure Ratio2 10.0392 +Epoch [63/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.1098, Pure Ratio2 10.1255 +Epoch [63/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.9837, Pure Ratio2 9.9412 +Epoch [63/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 9.9608, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 29.6374 % Model2 29.1767 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0013, Pure Ratio1: 9.2353, Pure Ratio2 9.1765 +Epoch [64/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.5980, Pure Ratio2 9.5588 +Epoch [64/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.6667, Pure Ratio2 9.6601 +Epoch [64/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8382, Pure Ratio2 9.8039 +Epoch [64/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9843, Pure Ratio2 9.9765 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9837, Pure Ratio2 9.9608 +Epoch [64/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0014, Pure Ratio1: 9.9524, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 29.4772 % Model2 29.2768 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.2353, Pure Ratio2 9.8824 +Epoch [65/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.6863 +Epoch [65/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.7190, Pure Ratio2 9.5752 +Epoch [65/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 9.8333, Pure Ratio2 9.7255 +Epoch [65/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.7137 +Epoch [65/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8399, Pure Ratio2 9.7876 +Epoch [65/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9552, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 30.3886 % Model2 30.2584 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 10.3725, Pure Ratio2 10.3922 +Epoch [66/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [66/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0023, Loss2: 0.0022, Pure Ratio1: 10.0523, Pure Ratio2 10.0588 +Epoch [66/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8529, Pure Ratio2 9.8480 +Epoch [66/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.9176, Pure Ratio2 9.8235 +Epoch [66/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.9771, Pure Ratio2 9.9216 +Epoch [66/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9776, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 31.7107 % Model2 29.2067 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 9.6078, Pure Ratio2 9.4706 +Epoch [67/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.6569, Pure Ratio2 9.6176 +Epoch [67/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8562 +Epoch [67/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.0833, Pure Ratio2 10.0735 +Epoch [67/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 9.9647 +Epoch [67/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 10.0000, Pure Ratio2 10.0033 +Epoch [67/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.9776, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 29.8578 % Model2 29.7576 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.5686, Pure Ratio2 9.5686 +Epoch [68/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6176, Pure Ratio2 9.7157 +Epoch [68/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.6405 +Epoch [68/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.4020, Pure Ratio2 9.4902 +Epoch [68/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0012, Pure Ratio1: 9.5412, Pure Ratio2 9.6078 +Epoch [68/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.7418, Pure Ratio2 9.7549 +Epoch [68/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8543, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 30.3486 % Model2 30.2584 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.3725, Pure Ratio2 9.3333 +Epoch [69/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5196, Pure Ratio2 9.4804 +Epoch [69/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0006, Pure Ratio1: 9.6209, Pure Ratio2 9.6275 +Epoch [69/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7108, Pure Ratio2 9.6912 +Epoch [69/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.7020, Pure Ratio2 9.7176 +Epoch [69/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0014, Pure Ratio1: 9.8399, Pure Ratio2 9.8105 +Epoch [69/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.9076, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 29.8177 % Model2 28.1751 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.5294, Pure Ratio2 10.6667 +Epoch [70/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.2941, Pure Ratio2 10.3039 +Epoch [70/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0004, Pure Ratio1: 10.1373, Pure Ratio2 10.1242 +Epoch [70/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 10.0490, Pure Ratio2 10.0980 +Epoch [70/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0157, Pure Ratio2 10.0549 +Epoch [70/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 10.0458 +Epoch [70/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9664, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 31.0196 % Model2 30.8494 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0980, Pure Ratio2 10.3922 +Epoch [71/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 10.2059 +Epoch [71/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0392, Pure Ratio2 10.1111 +Epoch [71/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0147, Pure Ratio2 10.1569 +Epoch [71/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0392, Pure Ratio2 10.1608 +Epoch [71/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8693, Pure Ratio2 9.9673 +Epoch [71/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8655, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 30.5589 % Model2 29.9679 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.2353, Pure Ratio2 9.4902 +Epoch [72/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8431, Pure Ratio2 10.0490 +Epoch [72/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.0588 +Epoch [72/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0022, Loss2: 0.0008, Pure Ratio1: 10.0784, Pure Ratio2 10.1275 +Epoch [72/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 10.1569, Pure Ratio2 10.1765 +Epoch [72/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0065, Pure Ratio2 10.0588 +Epoch [72/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.9636, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 30.2484 % Model2 29.3870 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0784, Pure Ratio2 10.2157 +Epoch [73/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.1765 +Epoch [73/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8954, Pure Ratio2 10.0392 +Epoch [73/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8676, Pure Ratio2 9.9706 +Epoch [73/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.9961 +Epoch [73/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9183, Pure Ratio2 9.9706 +Epoch [73/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0013, Pure Ratio1: 9.9104, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 29.9179 % Model2 30.0381 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [74/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.7745 +Epoch [74/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.1830, Pure Ratio2 10.0458 +Epoch [74/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.1912, Pure Ratio2 10.0147 +Epoch [74/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.1373, Pure Ratio2 9.9490 +Epoch [74/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0490, Pure Ratio2 9.9314 +Epoch [74/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9272, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 29.1567 % Model2 30.0481 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.8235 +Epoch [75/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.1078 +Epoch [75/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0016, Pure Ratio1: 10.0392, Pure Ratio2 10.0131 +Epoch [75/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0441, Pure Ratio2 9.9706 +Epoch [75/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.1059, Pure Ratio2 9.9843 +Epoch [75/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0261, Pure Ratio2 9.9118 +Epoch [75/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0336, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 30.5088 % Model2 29.7276 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0392, Pure Ratio2 10.1961 +Epoch [76/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.6961, Pure Ratio2 9.6961 +Epoch [76/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7516, Pure Ratio2 9.7190 +Epoch [76/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8382, Pure Ratio2 9.8431 +Epoch [76/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8863, Pure Ratio2 9.8275 +Epoch [76/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.8268, Pure Ratio2 9.7745 +Epoch [76/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9188, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 30.7893 % Model2 31.2099 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.4314, Pure Ratio2 10.3922 +Epoch [77/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9216, Pure Ratio2 9.9804 +Epoch [77/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.8366 +Epoch [77/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 9.9412 +Epoch [77/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9098, Pure Ratio2 9.8863 +Epoch [77/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9150, Pure Ratio2 9.8693 +Epoch [77/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.9860, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 30.2985 % Model2 30.1082 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.3137, Pure Ratio2 9.5294 +Epoch [78/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.7157, Pure Ratio2 9.9412 +Epoch [78/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7843, Pure Ratio2 10.0131 +Epoch [78/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0012, Pure Ratio1: 9.7353, Pure Ratio2 9.9510 +Epoch [78/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.7294, Pure Ratio2 9.9059 +Epoch [78/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.8301, Pure Ratio2 9.9412 +Epoch [78/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.8487, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 29.4171 % Model2 30.7792 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.6471, Pure Ratio2 10.6078 +Epoch [79/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.4314, Pure Ratio2 10.3922 +Epoch [79/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 10.4706, Pure Ratio2 10.4902 +Epoch [79/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.1667, Pure Ratio2 10.1176 +Epoch [79/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0824, Pure Ratio2 10.0314 +Epoch [79/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.9869, Pure Ratio2 10.0098 +Epoch [79/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9692, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 30.7292 % Model2 29.4571 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8627, Pure Ratio2 9.7451 +Epoch [80/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8922, Pure Ratio2 9.8039 +Epoch [80/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [80/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.6814, Pure Ratio2 9.6078 +Epoch [80/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9333, Pure Ratio2 9.8627 +Epoch [80/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9641, Pure Ratio2 9.9641 +Epoch [80/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9524, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 30.7091 % Model2 29.4071 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 9.8824 +Epoch [81/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0015, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.2157 +Epoch [81/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.0850, Pure Ratio2 10.0915 +Epoch [81/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 10.0000, Pure Ratio2 9.9559 +Epoch [81/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0017, Pure Ratio1: 9.9922, Pure Ratio2 9.9922 +Epoch [81/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9052, Pure Ratio2 9.9183 +Epoch [81/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9104, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 30.1182 % Model2 29.6575 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 9.9020 +Epoch [82/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8137, Pure Ratio2 9.6471 +Epoch [82/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7320, Pure Ratio2 9.5817 +Epoch [82/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.7696 +Epoch [82/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0235, Pure Ratio2 9.8588 +Epoch [82/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9052, Pure Ratio2 9.7353 +Epoch [82/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 10.0532, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 29.4872 % Model2 29.8377 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.5098, Pure Ratio2 10.6667 +Epoch [83/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.3431, Pure Ratio2 10.4314 +Epoch [83/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0261, Pure Ratio2 10.1046 +Epoch [83/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [83/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 9.9765, Pure Ratio2 9.9804 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [83/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.9440, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 29.5673 % Model2 29.6575 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 9.8235 +Epoch [84/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.0098, Pure Ratio2 9.9902 +Epoch [84/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [84/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.0147 +Epoch [84/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.0157, Pure Ratio2 9.9686 +Epoch [84/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0002, Pure Ratio1: 9.9641, Pure Ratio2 9.9281 +Epoch [84/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0057, Loss2: 0.0064, Pure Ratio1: 9.9300, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 31.0296 % Model2 31.2099 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 10.1569 +Epoch [85/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9510, Pure Ratio2 10.2059 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 10.2353 +Epoch [85/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.0294, Pure Ratio2 10.1618 +Epoch [85/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.0392, Pure Ratio2 10.1804 +Epoch [85/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8987, Pure Ratio2 10.0523 +Epoch [85/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.8039, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 30.4688 % Model2 30.4788 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.0784, Pure Ratio2 9.7843 +Epoch [86/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0882, Pure Ratio2 9.8039 +Epoch [86/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0392, Pure Ratio2 9.9085 +Epoch [86/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 9.9265 +Epoch [86/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1804, Pure Ratio2 10.0314 +Epoch [86/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.0065, Pure Ratio2 9.8725 +Epoch [86/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0336, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 30.6691 % Model2 27.9647 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.3725, Pure Ratio2 9.4706 +Epoch [87/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.5392 +Epoch [87/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8562 +Epoch [87/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9559, Pure Ratio2 10.0000 +Epoch [87/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8667, Pure Ratio2 9.9059 +Epoch [87/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.9739, Pure Ratio2 9.9673 +Epoch [87/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0252, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 28.9964 % Model2 29.6775 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [88/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.9118, Pure Ratio2 9.8431 +Epoch [88/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.9281, Pure Ratio2 9.8954 +Epoch [88/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9216, Pure Ratio2 9.9167 +Epoch [88/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8941, Pure Ratio2 9.9608 +Epoch [88/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9444, Pure Ratio2 10.0458 +Epoch [88/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 30.9595 % Model2 30.0381 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.1569 +Epoch [89/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6961, Pure Ratio2 9.7745 +Epoch [89/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0031, Loss2: 0.0025, Pure Ratio1: 9.9085, Pure Ratio2 9.9869 +Epoch [89/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.1324 +Epoch [89/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9373, Pure Ratio2 10.0510 +Epoch [89/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8856, Pure Ratio2 10.0000 +Epoch [89/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8179, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 29.6374 % Model2 30.3886 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.1765 +Epoch [90/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.5392, Pure Ratio2 10.5000 +Epoch [90/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.2418, Pure Ratio2 10.2810 +Epoch [90/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9461, Pure Ratio2 10.0392 +Epoch [90/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0035, Loss2: 0.0047, Pure Ratio1: 10.0431, Pure Ratio2 10.1098 +Epoch [90/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 10.0654 +Epoch [90/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9832, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 29.5072 % Model2 29.8878 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.4902 +Epoch [91/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8333, Pure Ratio2 9.7549 +Epoch [91/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 10.0915, Pure Ratio2 10.0327 +Epoch [91/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.1373, Pure Ratio2 10.0441 +Epoch [91/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1412, Pure Ratio2 10.0667 +Epoch [91/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.1242, Pure Ratio2 10.0915 +Epoch [91/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0504, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 30.3285 % Model2 29.7075 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.7647 +Epoch [92/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.6176 +Epoch [92/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0523, Pure Ratio2 9.9608 +Epoch [92/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9412 +Epoch [92/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 10.0431, Pure Ratio2 10.0353 +Epoch [92/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0000, Pure Ratio2 10.0392 +Epoch [92/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9300, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 28.9263 % Model2 28.8962 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9020, Pure Ratio2 9.7843 +Epoch [93/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8333, Pure Ratio2 9.6863 +Epoch [93/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 9.9542 +Epoch [93/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9951, Pure Ratio2 9.9608 +Epoch [93/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0941, Pure Ratio2 10.0667 +Epoch [93/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0020, Loss2: 0.0002, Pure Ratio1: 10.0621, Pure Ratio2 10.0000 +Epoch [93/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0084, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 31.2600 % Model2 30.0180 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.9020 +Epoch [94/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7059, Pure Ratio2 9.8431 +Epoch [94/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.6732, Pure Ratio2 9.8301 +Epoch [94/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6912, Pure Ratio2 9.8971 +Epoch [94/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.6118, Pure Ratio2 9.8627 +Epoch [94/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7157, Pure Ratio2 9.9248 +Epoch [94/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.7955, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 28.9463 % Model2 28.7760 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.1569, Pure Ratio2 10.4314 +Epoch [95/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.3039 +Epoch [95/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 10.0654 +Epoch [95/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.9657, Pure Ratio2 10.0000 +Epoch [95/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 10.0157 +Epoch [95/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.9248, Pure Ratio2 9.9477 +Epoch [95/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 30.7292 % Model2 29.5673 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6275, Pure Ratio2 9.6863 +Epoch [96/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 9.7353 +Epoch [96/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.8039 +Epoch [96/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.8382 +Epoch [96/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0157, Pure Ratio2 9.8667 +Epoch [96/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.8856 +Epoch [96/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0013, Pure Ratio1: 10.0364, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 29.6975 % Model2 29.4271 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.3333, Pure Ratio2 10.1373 +Epoch [97/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.1078 +Epoch [97/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0719, Pure Ratio2 10.0065 +Epoch [97/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 9.9657 +Epoch [97/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0275, Pure Ratio2 9.9294 +Epoch [97/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 9.8562 +Epoch [97/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8936, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 30.6190 % Model2 30.1482 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [98/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.7549 +Epoch [98/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7712, Pure Ratio2 9.8366 +Epoch [98/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7598, Pure Ratio2 9.7549 +Epoch [98/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7412, Pure Ratio2 9.7137 +Epoch [98/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8562 +Epoch [98/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8739, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 29.7977 % Model2 29.5573 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.4706 +Epoch [99/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.8922 +Epoch [99/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0001, Loss2: 0.0013, Pure Ratio1: 9.9412, Pure Ratio2 9.9869 +Epoch [99/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0931, Pure Ratio2 10.1029 +Epoch [99/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.1137, Pure Ratio2 10.0471 +Epoch [99/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 9.9967 +Epoch [99/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9356, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 29.4571 % Model2 30.5288 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 10.1765 +Epoch [100/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 10.0294 +Epoch [100/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.9608 +Epoch [100/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8676, Pure Ratio2 10.0147 +Epoch [100/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 9.7961, Pure Ratio2 9.9294 +Epoch [100/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7810, Pure Ratio2 9.8889 +Epoch [100/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8683, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 30.9896 % Model2 29.7476 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.5686 +Epoch [101/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.6765, Pure Ratio2 9.6667 +Epoch [101/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9673, Pure Ratio2 10.0915 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 10.0637 +Epoch [101/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9569, Pure Ratio2 10.0784 +Epoch [101/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.8660, Pure Ratio2 9.9902 +Epoch [101/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8515, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 31.0897 % Model2 30.2284 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.2549 +Epoch [102/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8431 +Epoch [102/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6993, Pure Ratio2 9.7516 +Epoch [102/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8676, Pure Ratio2 9.9461 +Epoch [102/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9176, Pure Ratio2 10.0235 +Epoch [102/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7908, Pure Ratio2 9.8824 +Epoch [102/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8543, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 29.4671 % Model2 28.7260 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.7843 +Epoch [103/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.9510 +Epoch [103/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.2745 +Epoch [103/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0637, Pure Ratio2 10.1569 +Epoch [103/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9059, Pure Ratio2 9.9333 +Epoch [103/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9739 +Epoch [103/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.8683, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 29.6575 % Model2 31.5705 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4118, Pure Ratio2 9.3137 +Epoch [104/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.2353, Pure Ratio2 9.2353 +Epoch [104/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4837, Pure Ratio2 9.5229 +Epoch [104/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7745 +Epoch [104/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8000, Pure Ratio2 9.8353 +Epoch [104/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6928, Pure Ratio2 9.7451 +Epoch [104/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8151, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 29.3970 % Model2 30.4888 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [105/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 10.0294 +Epoch [105/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1438, Pure Ratio2 10.1569 +Epoch [105/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9167, Pure Ratio2 9.9265 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9961, Pure Ratio2 9.9725 +Epoch [105/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.9444 +Epoch [105/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8908, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 29.8377 % Model2 29.2568 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [106/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.7157, Pure Ratio2 9.7843 +Epoch [106/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.8170 +Epoch [106/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7500, Pure Ratio2 9.8088 +Epoch [106/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.8745 +Epoch [106/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8170, Pure Ratio2 9.9379 +Epoch [106/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8599, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 29.7175 % Model2 30.2784 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [107/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3431, Pure Ratio2 10.2843 +Epoch [107/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 10.0850 +Epoch [107/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1520 +Epoch [107/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 10.1255 +Epoch [107/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 10.0000 +Epoch [107/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9468, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 30.0982 % Model2 28.4655 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 8.9608, Pure Ratio2 9.0784 +Epoch [108/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.3039, Pure Ratio2 9.4804 +Epoch [108/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4641, Pure Ratio2 9.5948 +Epoch [108/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7402 +Epoch [108/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8314, Pure Ratio2 9.8627 +Epoch [108/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.9216 +Epoch [108/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8179, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 29.7877 % Model2 29.9880 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.4510, Pure Ratio2 10.7647 +Epoch [109/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.3725 +Epoch [109/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.3725, Pure Ratio2 10.3856 +Epoch [109/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.2059 +Epoch [109/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1412, Pure Ratio2 10.1804 +Epoch [109/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.1242 +Epoch [109/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 9.9720, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 29.4872 % Model2 30.2284 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0011, Pure Ratio1: 10.0392, Pure Ratio2 9.9804 +Epoch [110/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 10.1667 +Epoch [110/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9608 +Epoch [110/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.9559 +Epoch [110/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9373, Pure Ratio2 9.9451 +Epoch [110/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9085, Pure Ratio2 9.8987 +Epoch [110/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8908, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 30.4387 % Model2 29.4872 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [111/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0490 +Epoch [111/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 9.9739, Pure Ratio2 10.1242 +Epoch [111/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0294, Pure Ratio2 10.1176 +Epoch [111/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0157, Pure Ratio2 10.0863 +Epoch [111/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1438 +Epoch [111/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9524, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 30.1482 % Model2 29.3570 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.3725 +Epoch [112/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.0882 +Epoch [112/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3007, Pure Ratio2 10.2157 +Epoch [112/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0441, Pure Ratio2 9.9755 +Epoch [112/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1412, Pure Ratio2 10.0471 +Epoch [112/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9020 +Epoch [112/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 30.5589 % Model2 29.3970 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.5098, Pure Ratio2 10.3922 +Epoch [113/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.3039 +Epoch [113/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7582, Pure Ratio2 9.7974 +Epoch [113/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8137, Pure Ratio2 9.8529 +Epoch [113/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8745, Pure Ratio2 9.8745 +Epoch [113/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9477 +Epoch [113/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 31.0897 % Model2 29.2969 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.3725 +Epoch [114/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8039 +Epoch [114/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 9.9346 +Epoch [114/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1422, Pure Ratio2 10.0392 +Epoch [114/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0667, Pure Ratio2 9.9882 +Epoch [114/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0752, Pure Ratio2 9.9869 +Epoch [114/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0140, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 30.3686 % Model2 29.5773 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.2745 +Epoch [115/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8529 +Epoch [115/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0458, Pure Ratio2 10.1111 +Epoch [115/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 10.0294 +Epoch [115/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0118, Pure Ratio2 10.0196 +Epoch [115/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9575, Pure Ratio2 9.9706 +Epoch [115/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9916, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 29.7776 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.9804 +Epoch [116/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 10.1569 +Epoch [116/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 10.0654 +Epoch [116/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9755, Pure Ratio2 10.0000 +Epoch [116/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 10.0000 +Epoch [116/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8333, Pure Ratio2 9.8562 +Epoch [116/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8992, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 31.2500 % Model2 29.8978 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.0784, Pure Ratio2 8.9608 +Epoch [117/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.5784, Pure Ratio2 9.5294 +Epoch [117/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.6928 +Epoch [117/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8578, Pure Ratio2 9.7010 +Epoch [117/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9098, Pure Ratio2 9.7255 +Epoch [117/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.8105 +Epoch [117/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9776, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 30.2684 % Model2 30.5589 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5490, Pure Ratio2 10.4706 +Epoch [118/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.4118, Pure Ratio2 10.4510 +Epoch [118/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.4118 +Epoch [118/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1716, Pure Ratio2 10.1618 +Epoch [118/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.1412, Pure Ratio2 10.1451 +Epoch [118/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 10.0784 +Epoch [118/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0168, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 30.0681 % Model2 29.2668 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [119/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.1961 +Epoch [119/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0458 +Epoch [119/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9363, Pure Ratio2 9.9265 +Epoch [119/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 9.9804 +Epoch [119/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9248, Pure Ratio2 9.9641 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9076, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 28.8862 % Model2 29.6875 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.0588 +Epoch [120/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9706 +Epoch [120/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.9412 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8873 +Epoch [120/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8980, Pure Ratio2 9.8627 +Epoch [120/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9641 +Epoch [120/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 30.7492 % Model2 30.0180 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.4314 +Epoch [121/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0490 +Epoch [121/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7908 +Epoch [121/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6814, Pure Ratio2 9.7500 +Epoch [121/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8275, Pure Ratio2 9.8510 +Epoch [121/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8660, Pure Ratio2 9.8987 +Epoch [121/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9076, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 30.6691 % Model2 30.2284 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [122/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 10.0392 +Epoch [122/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0654, Pure Ratio2 10.1111 +Epoch [122/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9559, Pure Ratio2 10.0000 +Epoch [122/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9137, Pure Ratio2 9.9686 +Epoch [122/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.9281 +Epoch [122/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.9636, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 30.6891 % Model2 30.4888 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9608 +Epoch [123/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 10.0000 +Epoch [123/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.9281 +Epoch [123/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.8333 +Epoch [123/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8078, Pure Ratio2 9.8235 +Epoch [123/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8105, Pure Ratio2 9.7974 +Epoch [123/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8908, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 30.3886 % Model2 31.1899 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.9804 +Epoch [124/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.9706 +Epoch [124/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0588 +Epoch [124/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1667, Pure Ratio2 10.2353 +Epoch [124/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.1059 +Epoch [124/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9804 +Epoch [124/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9076, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 29.9579 % Model2 30.1783 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 96.8750, Training Accuracy2: 96.0938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.9216 +Epoch [125/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.7255 +Epoch [125/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9739, Pure Ratio2 9.8954 +Epoch [125/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.8186 +Epoch [125/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9725, Pure Ratio2 9.8902 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9052 +Epoch [125/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9580, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 30.7592 % Model2 30.6390 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.7059 +Epoch [126/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.4412 +Epoch [126/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.2876, Pure Ratio2 9.3987 +Epoch [126/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5931, Pure Ratio2 9.7108 +Epoch [126/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7373, Pure Ratio2 9.8863 +Epoch [126/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.8562 +Epoch [126/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8011, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 30.0080 % Model2 29.1166 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 10.1373 +Epoch [127/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.5784 +Epoch [127/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6209, Pure Ratio2 9.6928 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.8480 +Epoch [127/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7725, Pure Ratio2 9.8275 +Epoch [127/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7614, Pure Ratio2 9.8170 +Epoch [127/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0013, Loss2: 0.0003, Pure Ratio1: 9.8291, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 29.8478 % Model2 29.6575 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.3529, Pure Ratio2 9.3137 +Epoch [128/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0196 +Epoch [128/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 9.9739 +Epoch [128/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.1029 +Epoch [128/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 10.0196 +Epoch [128/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 10.1895 +Epoch [128/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9748, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 30.7993 % Model2 30.4587 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0784 +Epoch [129/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9412 +Epoch [129/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7974 +Epoch [129/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8333 +Epoch [129/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.8431 +Epoch [129/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.8170 +Epoch [129/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 30.3385 % Model2 29.6374 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.4902 +Epoch [130/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.1078 +Epoch [130/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9346 +Epoch [130/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.8284 +Epoch [130/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9059, Pure Ratio2 9.9765 +Epoch [130/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 10.0425 +Epoch [130/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 29.6174 % Model2 28.9864 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7059 +Epoch [131/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8922 +Epoch [131/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8562 +Epoch [131/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 9.9167 +Epoch [131/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.8039 +Epoch [131/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.8497 +Epoch [131/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 31.0196 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5882, Pure Ratio2 10.4706 +Epoch [132/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.3627, Pure Ratio2 10.3922 +Epoch [132/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0915, Pure Ratio2 10.0850 +Epoch [132/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9461, Pure Ratio2 9.9657 +Epoch [132/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 9.9333 +Epoch [132/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9608 +Epoch [132/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9692, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 30.3786 % Model2 30.1883 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8627 +Epoch [133/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [133/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.6993 +Epoch [133/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8873, Pure Ratio2 9.8088 +Epoch [133/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.7922 +Epoch [133/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8039 +Epoch [133/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8992, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 31.0597 % Model2 29.6174 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.3333, Pure Ratio2 10.5490 +Epoch [134/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.3824, Pure Ratio2 10.3824 +Epoch [134/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0654 +Epoch [134/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0882 +Epoch [134/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0824, Pure Ratio2 10.1255 +Epoch [134/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8889, Pure Ratio2 9.9477 +Epoch [134/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8964, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 29.4772 % Model2 30.4387 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8431 +Epoch [135/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.0098 +Epoch [135/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.1111 +Epoch [135/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.8284 +Epoch [135/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9843, Pure Ratio2 9.8510 +Epoch [135/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 9.8987 +Epoch [135/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9776, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 29.9479 % Model2 29.8377 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 9.8824 +Epoch [136/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8824 +Epoch [136/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9542 +Epoch [136/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9510 +Epoch [136/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.8157 +Epoch [136/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8007, Pure Ratio2 9.8824 +Epoch [136/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 30.9996 % Model2 30.4087 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.8235 +Epoch [137/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 10.0392 +Epoch [137/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9542 +Epoch [137/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8186 +Epoch [137/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8471, Pure Ratio2 9.8392 +Epoch [137/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.8725 +Epoch [137/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9720, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 29.5172 % Model2 31.3802 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.5294 +Epoch [138/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.2549 +Epoch [138/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9739, Pure Ratio2 9.9869 +Epoch [138/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 10.0980 +Epoch [138/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0013, Pure Ratio1: 9.8941, Pure Ratio2 10.0196 +Epoch [138/200], Iter [300/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.9379 +Epoch [138/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 30.3285 % Model2 30.1082 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5882 +Epoch [139/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7157 +Epoch [139/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6732, Pure Ratio2 9.6405 +Epoch [139/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6225, Pure Ratio2 9.5539 +Epoch [139/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7922, Pure Ratio2 9.6902 +Epoch [139/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.7516 +Epoch [139/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 30.2885 % Model2 30.1583 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1765, Pure Ratio2 9.1373 +Epoch [140/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.9804 +Epoch [140/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.8758 +Epoch [140/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6814, Pure Ratio2 9.8186 +Epoch [140/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7882, Pure Ratio2 9.8824 +Epoch [140/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7810 +Epoch [140/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7787, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 30.0381 % Model2 29.7576 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.7451 +Epoch [141/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8627 +Epoch [141/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 10.0588 +Epoch [141/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8971 +Epoch [141/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.0118 +Epoch [141/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0163 +Epoch [141/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0224, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 29.3369 % Model2 29.7977 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.5490 +Epoch [142/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.6569 +Epoch [142/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [142/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7892, Pure Ratio2 9.8627 +Epoch [142/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.8314 +Epoch [142/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.8497 +Epoch [142/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8347, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 30.5689 % Model2 29.3169 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.5294 +Epoch [143/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9706, Pure Ratio2 9.9118 +Epoch [143/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.5948 +Epoch [143/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.8284 +Epoch [143/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9137 +Epoch [143/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.8954 +Epoch [143/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9468, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 29.5773 % Model2 29.6575 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.0196 +Epoch [144/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 9.8627 +Epoch [144/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8758, Pure Ratio2 9.6013 +Epoch [144/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.7255 +Epoch [144/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.7922 +Epoch [144/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8987, Pure Ratio2 9.7876 +Epoch [144/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9524, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 29.8778 % Model2 29.8277 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7843 +Epoch [145/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7549 +Epoch [145/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0327, Pure Ratio2 9.9020 +Epoch [145/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.0000 +Epoch [145/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0510, Pure Ratio2 9.9961 +Epoch [145/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.8562 +Epoch [145/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 30.9696 % Model2 30.9395 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.3725, Pure Ratio2 9.4510 +Epoch [146/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.8137 +Epoch [146/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7582, Pure Ratio2 9.8693 +Epoch [146/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8971, Pure Ratio2 9.9657 +Epoch [146/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.9412 +Epoch [146/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8203 +Epoch [146/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9356, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 31.2800 % Model2 30.2985 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2549 +Epoch [147/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 9.9020 +Epoch [147/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9085 +Epoch [147/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.7598 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.8275, Pure Ratio2 9.7765 +Epoch [147/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8595 +Epoch [147/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8543, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 30.5188 % Model2 30.5389 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.6667 +Epoch [148/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.5784 +Epoch [148/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.6209 +Epoch [148/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.6863 +Epoch [148/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.7059 +Epoch [148/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.7092 +Epoch [148/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9748, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 30.7592 % Model2 30.4587 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.4314 +Epoch [149/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0022, Loss2: 0.0027, Pure Ratio1: 9.8725, Pure Ratio2 9.8922 +Epoch [149/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.7712 +Epoch [149/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 10.0000 +Epoch [149/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9569 +Epoch [149/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.9052 +Epoch [149/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8599, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 29.6474 % Model2 29.8177 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.4902, Pure Ratio2 9.7255 +Epoch [150/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 10.1471 +Epoch [150/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8693, Pure Ratio2 9.9412 +Epoch [150/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8578, Pure Ratio2 9.9657 +Epoch [150/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8196, Pure Ratio2 9.8902 +Epoch [150/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9118, Pure Ratio2 9.9314 +Epoch [150/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 30.1783 % Model2 31.0897 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [151/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.5980 +Epoch [151/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.6340 +Epoch [151/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.5980 +Epoch [151/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6941 +Epoch [151/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7157 +Epoch [151/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8459, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 30.9696 % Model2 30.6891 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.6667 +Epoch [152/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8333 +Epoch [152/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.8889 +Epoch [152/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.7794 +Epoch [152/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8000, Pure Ratio2 9.7294 +Epoch [152/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7908, Pure Ratio2 9.7516 +Epoch [152/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8711, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 30.2885 % Model2 29.2268 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.3725 +Epoch [153/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 10.3039, Pure Ratio2 10.4706 +Epoch [153/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.2614 +Epoch [153/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1127 +Epoch [153/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0471 +Epoch [153/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 10.0000 +Epoch [153/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 30.2484 % Model2 30.3986 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0588 +Epoch [154/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1569 +Epoch [154/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [154/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7500, Pure Ratio2 9.8529 +Epoch [154/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8275, Pure Ratio2 9.9176 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.8954 +Epoch [154/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8459, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 30.7592 % Model2 31.6206 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [155/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [155/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9150 +Epoch [155/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 9.9069 +Epoch [155/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 9.9451 +Epoch [155/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.9118 +Epoch [155/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 29.6775 % Model2 30.9996 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [156/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [156/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9216 +Epoch [156/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.8725 +Epoch [156/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0019, Loss2: 0.0022, Pure Ratio1: 9.9020, Pure Ratio2 9.9294 +Epoch [156/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8889 +Epoch [156/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8992, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 29.9579 % Model2 29.2268 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.5686 +Epoch [157/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.2059 +Epoch [157/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9412 +Epoch [157/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9510 +Epoch [157/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8706 +Epoch [157/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8987, Pure Ratio2 9.9183 +Epoch [157/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8655, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 29.5473 % Model2 29.8778 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6863 +Epoch [158/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1667 +Epoch [158/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.9085 +Epoch [158/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8578 +Epoch [158/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9569, Pure Ratio2 10.0000 +Epoch [158/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9641, Pure Ratio2 9.9804 +Epoch [158/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 30.4487 % Model2 30.6691 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.2549 +Epoch [159/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.4608 +Epoch [159/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.7124 +Epoch [159/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.8578 +Epoch [159/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9098, Pure Ratio2 9.8431 +Epoch [159/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9967 +Epoch [159/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 30.3786 % Model2 31.0597 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6471, Pure Ratio2 10.9020 +Epoch [160/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3627, Pure Ratio2 10.4706 +Epoch [160/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3203 +Epoch [160/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.1275 +Epoch [160/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9529 +Epoch [160/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8693 +Epoch [160/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 30.6090 % Model2 29.7977 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7647 +Epoch [161/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8039 +Epoch [161/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.7516 +Epoch [161/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.7206 +Epoch [161/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.9176 +Epoch [161/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9052, Pure Ratio2 9.8529 +Epoch [161/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 30.6190 % Model2 29.9579 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1176 +Epoch [162/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [162/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2222 +Epoch [162/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8971, Pure Ratio2 9.9559 +Epoch [162/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0588 +Epoch [162/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9771, Pure Ratio2 9.9869 +Epoch [162/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9272, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 29.9179 % Model2 29.5573 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.2745 +Epoch [163/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [163/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8889 +Epoch [163/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9608 +Epoch [163/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9765 +Epoch [163/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9542 +Epoch [163/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 29.7776 % Model2 30.7692 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.2353 +Epoch [164/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8529 +Epoch [164/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.9477 +Epoch [164/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.9363 +Epoch [164/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.9569 +Epoch [164/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7974, Pure Ratio2 9.9085 +Epoch [164/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7675, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 29.6074 % Model2 30.6991 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.5490 +Epoch [165/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4412, Pure Ratio2 9.4020 +Epoch [165/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.6732 +Epoch [165/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.7353 +Epoch [165/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.7647 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8039 +Epoch [165/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 30.1082 % Model2 30.3385 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7059 +Epoch [166/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8922 +Epoch [166/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.0000 +Epoch [166/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [166/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8980, Pure Ratio2 9.8588 +Epoch [166/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8497 +Epoch [166/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 30.1683 % Model2 29.4271 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.0588 +Epoch [167/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9118 +Epoch [167/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9085 +Epoch [167/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9314, Pure Ratio2 9.8480 +Epoch [167/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9569, Pure Ratio2 9.9020 +Epoch [167/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9641 +Epoch [167/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9972, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 29.4872 % Model2 29.8878 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.5098 +Epoch [168/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.6078 +Epoch [168/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.6601 +Epoch [168/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.8676 +Epoch [168/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9373 +Epoch [168/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9902 +Epoch [168/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 30.8694 % Model2 29.5873 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.8824 +Epoch [169/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8333 +Epoch [169/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8889 +Epoch [169/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8480 +Epoch [169/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7490, Pure Ratio2 9.7961 +Epoch [169/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.8464 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 30.4587 % Model2 30.3085 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.2157 +Epoch [170/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.5588 +Epoch [170/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.6340 +Epoch [170/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5588 +Epoch [170/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.7686 +Epoch [170/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8627 +Epoch [170/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 30.7492 % Model2 29.4671 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 9.9804 +Epoch [171/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 9.9804 +Epoch [171/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 9.9935 +Epoch [171/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1912, Pure Ratio2 10.1275 +Epoch [171/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1255, Pure Ratio2 10.1451 +Epoch [171/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 10.0784 +Epoch [171/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 30.4287 % Model2 30.4487 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.3137 +Epoch [172/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9706 +Epoch [172/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9281 +Epoch [172/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9608 +Epoch [172/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9922 +Epoch [172/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 9.9379 +Epoch [172/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 29.7276 % Model2 29.6675 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8627 +Epoch [173/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8039 +Epoch [173/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.8824 +Epoch [173/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8578, Pure Ratio2 9.9657 +Epoch [173/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7216, Pure Ratio2 9.8078 +Epoch [173/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.8203 +Epoch [173/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8151, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 30.5288 % Model2 29.6474 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1373 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.0196 +Epoch [174/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.8039 +Epoch [174/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0147 +Epoch [174/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.9373 +Epoch [174/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9542 +Epoch [174/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8291, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 30.3085 % Model2 29.7676 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.6275 +Epoch [175/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.7941 +Epoch [175/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7516 +Epoch [175/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7108 +Epoch [175/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.7255, Pure Ratio2 9.7490 +Epoch [175/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8529 +Epoch [175/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8543, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 30.2384 % Model2 30.3786 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.4706 +Epoch [176/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7941 +Epoch [176/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9020 +Epoch [176/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9657 +Epoch [176/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9686 +Epoch [176/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.8954 +Epoch [176/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 29.9780 % Model2 30.4888 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.5882 +Epoch [177/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7353 +Epoch [177/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9020 +Epoch [177/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.9118 +Epoch [177/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [177/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.8399 +Epoch [177/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8403, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 30.5188 % Model2 29.9079 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8627 +Epoch [178/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.0490 +Epoch [178/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 10.0131 +Epoch [178/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 9.8676 +Epoch [178/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.8314 +Epoch [178/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.8072 +Epoch [178/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7675 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 30.7692 % Model2 30.3486 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.2745 +Epoch [179/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 9.9412 +Epoch [179/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 9.9085 +Epoch [179/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8137 +Epoch [179/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 9.9373 +Epoch [179/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8824 +Epoch [179/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 31.1599 % Model2 30.3686 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5686 +Epoch [180/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7745 +Epoch [180/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8693 +Epoch [180/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.7353 +Epoch [180/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.7804 +Epoch [180/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8399, Pure Ratio2 9.7843 +Epoch [180/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 30.6190 % Model2 30.1182 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Epoch [181/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7843 +Epoch [181/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0915 +Epoch [181/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.9755 +Epoch [181/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 10.0353 +Epoch [181/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8856 +Epoch [181/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 30.5990 % Model2 29.8878 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [182/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0098 +Epoch [182/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8758 +Epoch [182/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8725 +Epoch [182/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8706 +Epoch [182/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7484, Pure Ratio2 9.7418 +Epoch [182/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 29.8778 % Model2 30.1382 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [183/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0588 +Epoch [183/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.1046 +Epoch [183/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.0882 +Epoch [183/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9725 +Epoch [183/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.9608 +Epoch [183/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 30.5288 % Model2 29.9179 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.8627 +Epoch [184/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.5784 +Epoch [184/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9281 +Epoch [184/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8284 +Epoch [184/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7176, Pure Ratio2 9.7059 +Epoch [184/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8301 +Epoch [184/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 31.0196 % Model2 30.5789 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 9.8627 +Epoch [185/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6961 +Epoch [185/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8889 +Epoch [185/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9118 +Epoch [185/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7961 +Epoch [185/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8268 +Epoch [185/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8011, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 30.2784 % Model2 29.7576 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.2549 +Epoch [186/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.4608 +Epoch [186/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6340 +Epoch [186/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6618 +Epoch [186/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.7882 +Epoch [186/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.9020 +Epoch [186/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 30.5188 % Model2 30.3285 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.3725 +Epoch [187/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 10.0392 +Epoch [187/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.9739 +Epoch [187/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9412 +Epoch [187/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 10.0510 +Epoch [187/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.9412 +Epoch [187/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8515, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 30.5088 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.3137 +Epoch [188/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.3824 +Epoch [188/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0784 +Epoch [188/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.0294 +Epoch [188/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0863 +Epoch [188/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1340, Pure Ratio2 10.0196 +Epoch [188/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0728, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 30.4587 % Model2 29.5172 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.8431 +Epoch [189/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.8333 +Epoch [189/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.0523 +Epoch [189/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.0539 +Epoch [189/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.8314 +Epoch [189/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8268 +Epoch [189/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 30.1983 % Model2 30.4187 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0196 +Epoch [190/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.7745 +Epoch [190/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 9.8235 +Epoch [190/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.7990 +Epoch [190/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.8353 +Epoch [190/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7516 +Epoch [190/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 30.6390 % Model2 30.3185 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0392 +Epoch [191/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.9118 +Epoch [191/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 10.0719 +Epoch [191/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9902 +Epoch [191/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9647 +Epoch [191/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0425 +Epoch [191/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 30.4888 % Model2 30.2784 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.4118 +Epoch [192/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2941 +Epoch [192/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0196 +Epoch [192/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 9.9510 +Epoch [192/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.8431 +Epoch [192/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9837 +Epoch [192/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 30.2985 % Model2 30.8494 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.6275 +Epoch [193/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2745 +Epoch [193/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1046 +Epoch [193/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0294 +Epoch [193/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0627, Pure Ratio2 10.0745 +Epoch [193/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9020 +Epoch [193/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 30.2484 % Model2 30.1382 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2157 +Epoch [194/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0490 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0000 +Epoch [194/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9657 +Epoch [194/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9804 +Epoch [194/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9641 +Epoch [194/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8655, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 30.1382 % Model2 31.0697 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0392 +Epoch [195/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.6765 +Epoch [195/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8366 +Epoch [195/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.8775 +Epoch [195/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8706, Pure Ratio2 9.7843 +Epoch [195/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8301 +Epoch [195/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9356, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 30.2183 % Model2 30.8894 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.4314 +Epoch [196/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0000 +Epoch [196/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 9.9869 +Epoch [196/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.8725 +Epoch [196/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 9.8745 +Epoch [196/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 9.8660 +Epoch [196/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0252, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 30.6791 % Model2 30.7392 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 10.0588 +Epoch [197/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.1275 +Epoch [197/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 10.0131 +Epoch [197/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9657 +Epoch [197/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.1176 +Epoch [197/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8987, Pure Ratio2 10.0588 +Epoch [197/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 30.6791 % Model2 30.6591 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.2941 +Epoch [198/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.5588 +Epoch [198/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5621, Pure Ratio2 9.6013 +Epoch [198/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5637, Pure Ratio2 9.5637 +Epoch [198/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6588 +Epoch [198/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.7647 +Epoch [198/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 30.2784 % Model2 29.8878 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [199/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9608 +Epoch [199/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8301 +Epoch [199/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.5833 +Epoch [199/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7020, Pure Ratio2 9.7725 +Epoch [199/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8693 +Epoch [199/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8796, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 30.2284 % Model2 30.3886 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7255 +Epoch [200/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8725 +Epoch [200/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.9542 +Epoch [200/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.8873 +Epoch [200/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.9255 +Epoch [200/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9314 +Epoch [200/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8543, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 30.4688 % Model2 30.3686 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9623 % diff --git a/other_methods/coteaching/coteaching_results/out_2_2.log b/other_methods/coteaching/coteaching_results/out_2_2.log new file mode 100644 index 0000000..f4b5a49 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_2_2.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0158, Loss2: 0.0157, Pure Ratio1: 10.1600, Pure Ratio2 10.1440 +Epoch [2/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0140, Loss2: 0.0140, Pure Ratio1: 10.1760, Pure Ratio2 10.2400 +Epoch [2/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 37.5000, Loss1: 0.0127, Loss2: 0.0129, Pure Ratio1: 10.1600, Pure Ratio2 10.2240 +Epoch [2/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0127, Loss2: 0.0131, Pure Ratio1: 10.1960, Pure Ratio2 10.2320 +Epoch [2/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0149, Loss2: 0.0148, Pure Ratio1: 10.1600, Pure Ratio2 10.1856 +Epoch [2/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0117, Loss2: 0.0118, Pure Ratio1: 10.0800, Pure Ratio2 10.1093 +Epoch [2/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0122, Loss2: 0.0119, Pure Ratio1: 10.0709, Pure Ratio2 10.0937 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 57.2516 % Model2 55.1583 %, Pure Ratio 1 10.0615 %, Pure Ratio 2 10.0821 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0110, Loss2: 0.0109, Pure Ratio1: 9.4754, Pure Ratio2 9.5246 +Epoch [3/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0110, Loss2: 0.0112, Pure Ratio1: 9.5738, Pure Ratio2 9.5574 +Epoch [3/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0114, Loss2: 0.0119, Pure Ratio1: 9.8743, Pure Ratio2 9.8798 +Epoch [3/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0115, Loss2: 0.0114, Pure Ratio1: 9.9631, Pure Ratio2 9.9508 +Epoch [3/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 39.8438, Loss1: 0.0125, Loss2: 0.0127, Pure Ratio1: 10.0525, Pure Ratio2 10.0525 +Epoch [3/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0113, Loss2: 0.0110, Pure Ratio1: 10.1421, Pure Ratio2 10.1175 +Epoch [3/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0102, Loss2: 0.0107, Pure Ratio1: 10.1265, Pure Ratio2 10.1007 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 65.8854 % Model2 65.4247 %, Pure Ratio 1 10.0715 %, Pure Ratio 2 10.0652 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 54.6875, Loss1: 0.0088, Loss2: 0.0096, Pure Ratio1: 9.6639, Pure Ratio2 9.7983 +Epoch [4/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0120, Loss2: 0.0131, Pure Ratio1: 9.9328, Pure Ratio2 10.0000 +Epoch [4/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0105, Loss2: 0.0109, Pure Ratio1: 10.0224, Pure Ratio2 10.0784 +Epoch [4/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0092, Loss2: 0.0092, Pure Ratio1: 10.0168, Pure Ratio2 10.0336 +Epoch [4/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 54.6875, Loss1: 0.0086, Loss2: 0.0087, Pure Ratio1: 10.0000, Pure Ratio2 10.0235 +Epoch [4/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0071, Loss2: 0.0072, Pure Ratio1: 10.0280, Pure Ratio2 10.0336 +Epoch [4/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0079, Loss2: 0.0078, Pure Ratio1: 10.0864, Pure Ratio2 10.0936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 68.8101 % Model2 68.7200 %, Pure Ratio 1 10.0819 %, Pure Ratio 2 10.0991 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0100, Loss2: 0.0096, Pure Ratio1: 10.0862, Pure Ratio2 10.0172 +Epoch [5/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0084, Loss2: 0.0081, Pure Ratio1: 10.1034, Pure Ratio2 10.0948 +Epoch [5/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0083, Loss2: 0.0087, Pure Ratio1: 10.2701, Pure Ratio2 10.2701 +Epoch [5/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0074, Loss2: 0.0078, Pure Ratio1: 10.2802, Pure Ratio2 10.3276 +Epoch [5/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0088, Loss2: 0.0087, Pure Ratio1: 10.2690, Pure Ratio2 10.3379 +Epoch [5/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0076, Loss2: 0.0079, Pure Ratio1: 10.0632, Pure Ratio2 10.1149 +Epoch [5/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0092, Loss2: 0.0086, Pure Ratio1: 10.1453, Pure Ratio2 10.1897 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 73.9884 % Model2 72.2256 %, Pure Ratio 1 10.0597 %, Pure Ratio 2 10.0973 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0092, Loss2: 0.0097, Pure Ratio1: 10.0708, Pure Ratio2 9.9469 +Epoch [6/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0086, Loss2: 0.0087, Pure Ratio1: 10.1770, Pure Ratio2 10.0973 +Epoch [6/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0059, Loss2: 0.0060, Pure Ratio1: 10.2006, Pure Ratio2 10.1711 +Epoch [6/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0067, Loss2: 0.0072, Pure Ratio1: 10.0575, Pure Ratio2 10.0531 +Epoch [6/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0095, Loss2: 0.0100, Pure Ratio1: 10.0885, Pure Ratio2 10.1027 +Epoch [6/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0070, Loss2: 0.0073, Pure Ratio1: 10.0000, Pure Ratio2 10.0147 +Epoch [6/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0047, Loss2: 0.0046, Pure Ratio1: 10.0379, Pure Ratio2 10.0582 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 75.7212 % Model2 74.5192 %, Pure Ratio 1 10.0567 %, Pure Ratio 2 10.0794 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0065, Loss2: 0.0063, Pure Ratio1: 10.4000, Pure Ratio2 10.3273 +Epoch [7/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0060, Loss2: 0.0053, Pure Ratio1: 10.2455, Pure Ratio2 10.2364 +Epoch [7/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0063, Loss2: 0.0060, Pure Ratio1: 10.5394, Pure Ratio2 10.5394 +Epoch [7/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0064, Loss2: 0.0068, Pure Ratio1: 10.1773, Pure Ratio2 10.1864 +Epoch [7/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0055, Loss2: 0.0057, Pure Ratio1: 10.2145, Pure Ratio2 10.2291 +Epoch [7/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0058, Loss2: 0.0053, Pure Ratio1: 10.1515, Pure Ratio2 10.1515 +Epoch [7/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0097, Loss2: 0.0083, Pure Ratio1: 10.1143, Pure Ratio2 10.1065 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 76.4123 % Model2 76.7929 %, Pure Ratio 1 10.0839 %, Pure Ratio 2 10.0862 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0046, Loss2: 0.0042, Pure Ratio1: 10.3148, Pure Ratio2 10.5000 +Epoch [8/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0053, Loss2: 0.0050, Pure Ratio1: 10.0833, Pure Ratio2 10.1852 +Epoch [8/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0047, Loss2: 0.0046, Pure Ratio1: 9.9198, Pure Ratio2 10.0123 +Epoch [8/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0051, Loss2: 0.0046, Pure Ratio1: 9.8009, Pure Ratio2 9.8704 +Epoch [8/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0051, Loss2: 0.0044, Pure Ratio1: 9.9815, Pure Ratio2 10.0259 +Epoch [8/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0049, Loss2: 0.0052, Pure Ratio1: 10.0062, Pure Ratio2 10.0000 +Epoch [8/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0047, Loss2: 0.0058, Pure Ratio1: 10.0317, Pure Ratio2 10.0265 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 78.1651 % Model2 79.1567 %, Pure Ratio 1 10.0332 %, Pure Ratio 2 10.0594 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0044, Loss2: 0.0035, Pure Ratio1: 9.7524, Pure Ratio2 9.6952 +Epoch [9/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0040, Loss2: 0.0037, Pure Ratio1: 9.6667, Pure Ratio2 9.6286 +Epoch [9/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0030, Loss2: 0.0027, Pure Ratio1: 9.8222, Pure Ratio2 9.7778 +Epoch [9/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 64.0625, Loss1: 0.0037, Loss2: 0.0048, Pure Ratio1: 9.8333, Pure Ratio2 9.7714 +Epoch [9/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0040, Loss2: 0.0052, Pure Ratio1: 10.0686, Pure Ratio2 10.0038 +Epoch [9/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0047, Loss2: 0.0050, Pure Ratio1: 10.0794, Pure Ratio2 10.0540 +Epoch [9/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0047, Loss2: 0.0043, Pure Ratio1: 10.0517, Pure Ratio2 10.0272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 75.6911 % Model2 78.7460 %, Pure Ratio 1 10.0879 %, Pure Ratio 2 10.0781 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0050, Loss2: 0.0056, Pure Ratio1: 10.6471, Pure Ratio2 10.9216 +Epoch [10/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0031, Loss2: 0.0024, Pure Ratio1: 10.3529, Pure Ratio2 10.5294 +Epoch [10/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0034, Loss2: 0.0036, Pure Ratio1: 9.9608, Pure Ratio2 10.0261 +Epoch [10/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0042, Loss2: 0.0043, Pure Ratio1: 9.9706, Pure Ratio2 10.0294 +Epoch [10/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0029, Loss2: 0.0028, Pure Ratio1: 9.9765, Pure Ratio2 10.0314 +Epoch [10/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0039, Loss2: 0.0039, Pure Ratio1: 10.0163, Pure Ratio2 10.0882 +Epoch [10/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0033, Loss2: 0.0030, Pure Ratio1: 10.0588, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 77.9347 % Model2 77.7644 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0024, Loss2: 0.0020, Pure Ratio1: 9.5882, Pure Ratio2 9.8039 +Epoch [11/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0038, Loss2: 0.0029, Pure Ratio1: 9.9118, Pure Ratio2 10.0294 +Epoch [11/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0040, Loss2: 0.0043, Pure Ratio1: 9.9150, Pure Ratio2 10.0000 +Epoch [11/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0044, Loss2: 0.0046, Pure Ratio1: 9.8333, Pure Ratio2 9.8922 +Epoch [11/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0042, Loss2: 0.0041, Pure Ratio1: 9.9686, Pure Ratio2 9.9608 +Epoch [11/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0041, Loss2: 0.0043, Pure Ratio1: 9.9902, Pure Ratio2 9.9739 +Epoch [11/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0042, Loss2: 0.0041, Pure Ratio1: 10.0084, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 80.2083 % Model2 80.3586 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.0026, Loss2: 0.0037, Pure Ratio1: 9.7647, Pure Ratio2 9.6275 +Epoch [12/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0027, Loss2: 0.0023, Pure Ratio1: 10.2745, Pure Ratio2 10.2941 +Epoch [12/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0019, Loss2: 0.0022, Pure Ratio1: 10.2876, Pure Ratio2 10.2876 +Epoch [12/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 10.2010, Pure Ratio2 10.2598 +Epoch [12/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0029, Loss2: 0.0033, Pure Ratio1: 10.1059, Pure Ratio2 10.1294 +Epoch [12/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0032, Loss2: 0.0031, Pure Ratio1: 9.9150, Pure Ratio2 9.9510 +Epoch [12/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0023, Loss2: 0.0025, Pure Ratio1: 10.0140, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 79.0865 % Model2 80.2584 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0026, Pure Ratio1: 10.1373, Pure Ratio2 10.1765 +Epoch [13/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0025, Loss2: 0.0029, Pure Ratio1: 10.5098, Pure Ratio2 10.5686 +Epoch [13/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0015, Pure Ratio1: 9.8954, Pure Ratio2 9.9869 +Epoch [13/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0019, Pure Ratio1: 9.8382, Pure Ratio2 9.9118 +Epoch [13/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.0033, Loss2: 0.0027, Pure Ratio1: 9.8902, Pure Ratio2 9.9529 +Epoch [13/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0028, Loss2: 0.0029, Pure Ratio1: 9.9183, Pure Ratio2 10.0033 +Epoch [13/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0028, Loss2: 0.0031, Pure Ratio1: 9.9552, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 81.3702 % Model2 81.8009 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0022, Loss2: 0.0021, Pure Ratio1: 10.1765, Pure Ratio2 10.3333 +Epoch [14/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0023, Pure Ratio1: 9.9608, Pure Ratio2 10.0784 +Epoch [14/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0029, Loss2: 0.0025, Pure Ratio1: 9.9281, Pure Ratio2 9.9477 +Epoch [14/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0011, Pure Ratio1: 9.9902, Pure Ratio2 10.0049 +Epoch [14/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 10.0039, Pure Ratio2 10.0039 +Epoch [14/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0030, Loss2: 0.0034, Pure Ratio1: 9.8725, Pure Ratio2 9.8399 +Epoch [14/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0026, Loss2: 0.0025, Pure Ratio1: 10.0420, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 81.7708 % Model2 81.3802 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 9.2157, Pure Ratio2 9.1961 +Epoch [15/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.6863, Pure Ratio2 9.6961 +Epoch [15/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0020, Loss2: 0.0026, Pure Ratio1: 9.7124, Pure Ratio2 9.6601 +Epoch [15/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0017, Pure Ratio1: 10.0343, Pure Ratio2 10.0049 +Epoch [15/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0035, Loss2: 0.0030, Pure Ratio1: 10.0039, Pure Ratio2 9.9882 +Epoch [15/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0020, Loss2: 0.0023, Pure Ratio1: 10.0359, Pure Ratio2 10.0000 +Epoch [15/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0025, Pure Ratio1: 10.1008, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 81.7107 % Model2 81.5104 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0029, Loss2: 0.0023, Pure Ratio1: 10.5294, Pure Ratio2 10.4314 +Epoch [16/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.0023, Loss2: 0.0020, Pure Ratio1: 10.0588, Pure Ratio2 9.9412 +Epoch [16/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.0026, Loss2: 0.0020, Pure Ratio1: 10.1961, Pure Ratio2 10.0850 +Epoch [16/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 10.1716, Pure Ratio2 10.1176 +Epoch [16/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0013, Loss2: 0.0016, Pure Ratio1: 10.2039, Pure Ratio2 10.1529 +Epoch [16/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0029, Loss2: 0.0019, Pure Ratio1: 10.1699, Pure Ratio2 10.1144 +Epoch [16/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0018, Pure Ratio1: 10.1092, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 81.6406 % Model2 81.5004 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0015, Loss2: 0.0019, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [17/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9608, Pure Ratio2 10.0490 +Epoch [17/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 10.1046, Pure Ratio2 10.1503 +Epoch [17/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.0018, Loss2: 0.0023, Pure Ratio1: 10.1520, Pure Ratio2 10.1912 +Epoch [17/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.7812, Loss1: 0.0015, Loss2: 0.0026, Pure Ratio1: 10.0784, Pure Ratio2 10.1294 +Epoch [17/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0027, Pure Ratio1: 10.0359, Pure Ratio2 10.0588 +Epoch [17/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0018, Loss2: 0.0020, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 80.9796 % Model2 82.0513 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [18/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0022, Pure Ratio1: 10.1471, Pure Ratio2 10.1275 +Epoch [18/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 10.0000, Pure Ratio2 10.0131 +Epoch [18/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 80.4688, Loss1: 0.0015, Loss2: 0.0009, Pure Ratio1: 9.9951, Pure Ratio2 10.0441 +Epoch [18/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0016, Loss2: 0.0016, Pure Ratio1: 10.0275, Pure Ratio2 10.0392 +Epoch [18/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0015, Loss2: 0.0018, Pure Ratio1: 10.0523, Pure Ratio2 10.0621 +Epoch [18/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0050, Loss2: 0.0041, Pure Ratio1: 10.0616, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 81.5304 % Model2 81.2500 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0014, Pure Ratio1: 9.7059, Pure Ratio2 9.6471 +Epoch [19/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0034, Loss2: 0.0023, Pure Ratio1: 9.5686, Pure Ratio2 9.6765 +Epoch [19/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.8693, Pure Ratio2 9.8497 +Epoch [19/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0012, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [19/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.8941, Pure Ratio2 9.8784 +Epoch [19/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 74.2188, Loss1: 0.0010, Loss2: 0.0020, Pure Ratio1: 9.9052, Pure Ratio2 9.9216 +Epoch [19/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 9.9860, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 82.5020 % Model2 81.7808 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 10.2745, Pure Ratio2 10.2157 +Epoch [20/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [20/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0015, Loss2: 0.0024, Pure Ratio1: 10.0131, Pure Ratio2 10.0392 +Epoch [20/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9755, Pure Ratio2 10.0441 +Epoch [20/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.7812, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 9.9765, Pure Ratio2 10.0745 +Epoch [20/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 10.0458, Pure Ratio2 10.1471 +Epoch [20/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9944, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 81.1298 % Model2 80.5589 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0019, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [21/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.6078, Pure Ratio2 9.5882 +Epoch [21/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0025, Loss2: 0.0020, Pure Ratio1: 9.9739, Pure Ratio2 9.9085 +Epoch [21/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.7812, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 10.0735, Pure Ratio2 10.0343 +Epoch [21/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 10.0353, Pure Ratio2 10.0275 +Epoch [21/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.9510, Pure Ratio2 9.9314 +Epoch [21/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.9692, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 80.8293 % Model2 81.5304 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.4510, Pure Ratio2 9.4902 +Epoch [22/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.7353 +Epoch [22/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.7582, Pure Ratio2 9.7451 +Epoch [22/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8480, Pure Ratio2 9.7990 +Epoch [22/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 10.0510, Pure Ratio2 9.9686 +Epoch [22/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 10.0458, Pure Ratio2 9.9739 +Epoch [22/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 10.0784, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 81.9311 % Model2 82.4419 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 9.9608, Pure Ratio2 10.0588 +Epoch [23/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0020, Pure Ratio1: 9.8333, Pure Ratio2 9.9216 +Epoch [23/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.0028, Loss2: 0.0027, Pure Ratio1: 10.0131, Pure Ratio2 10.0784 +Epoch [23/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 10.0588, Pure Ratio2 10.1078 +Epoch [23/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 10.0510, Pure Ratio2 10.0510 +Epoch [23/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0817, Pure Ratio2 10.0882 +Epoch [23/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0616, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 82.1414 % Model2 80.9796 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0196, Pure Ratio2 9.8627 +Epoch [24/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [24/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0654, Pure Ratio2 10.0980 +Epoch [24/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0018, Pure Ratio1: 10.0882, Pure Ratio2 10.1716 +Epoch [24/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0902, Pure Ratio2 10.1529 +Epoch [24/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.9902, Pure Ratio2 10.0327 +Epoch [24/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9916, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 82.6422 % Model2 81.6907 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 10.6078, Pure Ratio2 10.6863 +Epoch [25/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.2941, Pure Ratio2 10.3725 +Epoch [25/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0024, Pure Ratio1: 10.1046, Pure Ratio2 10.1961 +Epoch [25/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9755, Pure Ratio2 10.0098 +Epoch [25/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0020, Pure Ratio1: 9.9961, Pure Ratio2 10.0275 +Epoch [25/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.9739, Pure Ratio2 9.9804 +Epoch [25/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 10.0364, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 80.3285 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.3725, Pure Ratio2 10.5294 +Epoch [26/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.5000, Pure Ratio2 10.6471 +Epoch [26/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2614, Pure Ratio2 10.3529 +Epoch [26/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0017, Loss2: 0.0007, Pure Ratio1: 10.2353, Pure Ratio2 10.2647 +Epoch [26/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0745, Pure Ratio2 10.1176 +Epoch [26/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0019, Loss2: 0.0013, Pure Ratio1: 10.1013, Pure Ratio2 10.1078 +Epoch [26/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.1064, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 82.3718 % Model2 81.9010 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2353, Pure Ratio2 10.3922 +Epoch [27/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.1471, Pure Ratio2 10.2059 +Epoch [27/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 10.0980 +Epoch [27/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 10.0000, Pure Ratio2 10.0392 +Epoch [27/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0235, Pure Ratio2 10.0627 +Epoch [27/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 10.0817 +Epoch [27/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.0840, Pure Ratio2 10.1457 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 81.5004 % Model2 80.8494 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.6667 +Epoch [28/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.6961 +Epoch [28/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0007, Pure Ratio1: 9.8562, Pure Ratio2 9.8954 +Epoch [28/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 9.9216 +Epoch [28/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.0235, Pure Ratio2 9.9608 +Epoch [28/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 10.0196, Pure Ratio2 9.9739 +Epoch [28/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0014, Pure Ratio1: 10.0392, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 81.5104 % Model2 81.3001 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.6471 +Epoch [29/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0054, Loss2: 0.0057, Pure Ratio1: 10.0294, Pure Ratio2 9.8627 +Epoch [29/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0915, Pure Ratio2 10.0131 +Epoch [29/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.9804, Pure Ratio2 9.9902 +Epoch [29/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0019, Pure Ratio1: 10.0588, Pure Ratio2 10.0157 +Epoch [29/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.0098, Pure Ratio2 9.9837 +Epoch [29/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0011, Pure Ratio1: 9.9972, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 81.3902 % Model2 80.3385 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.8235 +Epoch [30/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.8824, Pure Ratio2 9.9118 +Epoch [30/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0523, Pure Ratio2 10.1046 +Epoch [30/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9216, Pure Ratio2 9.9804 +Epoch [30/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.7569, Pure Ratio2 9.8667 +Epoch [30/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8301, Pure Ratio2 9.9314 +Epoch [30/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 9.9300, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 82.8225 % Model2 82.2616 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.1332 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.0000 +Epoch [31/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0392, Pure Ratio2 9.9706 +Epoch [31/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0036, Loss2: 0.0025, Pure Ratio1: 10.1111, Pure Ratio2 10.0327 +Epoch [31/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.4314, Pure Ratio2 10.3284 +Epoch [31/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.3961, Pure Ratio2 10.3176 +Epoch [31/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.2647, Pure Ratio2 10.2026 +Epoch [31/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0008, Pure Ratio1: 10.1569, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 80.5188 % Model2 80.8794 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.6078, Pure Ratio2 10.5882 +Epoch [32/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0882, Pure Ratio2 10.0784 +Epoch [32/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1046, Pure Ratio2 10.0915 +Epoch [32/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.2941, Pure Ratio2 10.3088 +Epoch [32/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.2118, Pure Ratio2 10.2784 +Epoch [32/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0016, Loss2: 0.0014, Pure Ratio1: 10.1863, Pure Ratio2 10.2288 +Epoch [32/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.1120, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 80.6791 % Model2 81.1599 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.6863, Pure Ratio2 9.7451 +Epoch [33/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 9.6569, Pure Ratio2 9.7255 +Epoch [33/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8758, Pure Ratio2 9.9542 +Epoch [33/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.9902, Pure Ratio2 10.0392 +Epoch [33/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0078, Pure Ratio2 10.0275 +Epoch [33/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8791, Pure Ratio2 9.8987 +Epoch [33/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 81.7408 % Model2 80.8994 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 10.3137 +Epoch [34/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0882, Pure Ratio2 10.1863 +Epoch [34/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.1503, Pure Ratio2 10.1830 +Epoch [34/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2010, Pure Ratio2 10.1814 +Epoch [34/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 10.1059, Pure Ratio2 10.0667 +Epoch [34/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0015, Loss2: 0.0007, Pure Ratio1: 10.0784, Pure Ratio2 10.0196 +Epoch [34/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 10.0476, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 81.1198 % Model2 81.3001 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0603 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.3333, Pure Ratio2 9.4706 +Epoch [35/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 10.0490 +Epoch [35/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 10.0065 +Epoch [35/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9167, Pure Ratio2 9.9363 +Epoch [35/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8902, Pure Ratio2 9.8824 +Epoch [35/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0294, Pure Ratio2 9.9967 +Epoch [35/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 10.0700, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 80.9896 % Model2 81.3201 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.5686 +Epoch [36/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6961, Pure Ratio2 9.6471 +Epoch [36/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9346, Pure Ratio2 9.9412 +Epoch [36/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.9706, Pure Ratio2 9.9853 +Epoch [36/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 10.0627, Pure Ratio2 10.1059 +Epoch [36/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.1275, Pure Ratio2 10.1536 +Epoch [36/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0364, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 80.4888 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.1569, Pure Ratio2 10.2157 +Epoch [37/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9412, Pure Ratio2 9.9706 +Epoch [37/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.9085 +Epoch [37/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.1225, Pure Ratio2 10.1520 +Epoch [37/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9765, Pure Ratio2 10.0235 +Epoch [37/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0359, Pure Ratio2 10.0588 +Epoch [37/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.9440, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 80.8293 % Model2 81.2200 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.5686 +Epoch [38/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.2941, Pure Ratio2 10.2059 +Epoch [38/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0003, Pure Ratio1: 10.0523, Pure Ratio2 10.0065 +Epoch [38/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8971, Pure Ratio2 9.8775 +Epoch [38/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0471, Pure Ratio2 10.0000 +Epoch [38/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9673, Pure Ratio2 9.9608 +Epoch [38/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.9552, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 80.8994 % Model2 81.0397 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.3725, Pure Ratio2 9.2353 +Epoch [39/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0294 +Epoch [39/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9804 +Epoch [39/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 9.9951 +Epoch [39/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1059, Pure Ratio2 10.0039 +Epoch [39/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1013, Pure Ratio2 10.0065 +Epoch [39/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 10.0980, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 79.4772 % Model2 80.3686 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.4902, Pure Ratio2 10.5098 +Epoch [40/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.2941, Pure Ratio2 10.2941 +Epoch [40/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.2353, Pure Ratio2 10.3072 +Epoch [40/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0539, Pure Ratio2 10.1078 +Epoch [40/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9765, Pure Ratio2 10.0314 +Epoch [40/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [40/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0224, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 80.4688 % Model2 80.9195 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.3137 +Epoch [41/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.1667 +Epoch [41/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.2418, Pure Ratio2 10.1765 +Epoch [41/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0588 +Epoch [41/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.1020, Pure Ratio2 10.1451 +Epoch [41/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.1013 +Epoch [41/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0812, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.8494 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.4902, Pure Ratio2 10.6275 +Epoch [42/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1765, Pure Ratio2 10.3333 +Epoch [42/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.1111, Pure Ratio2 10.3203 +Epoch [42/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0980, Pure Ratio2 10.2059 +Epoch [42/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.1333, Pure Ratio2 10.2235 +Epoch [42/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0817, Pure Ratio2 10.1765 +Epoch [42/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9496, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 81.5705 % Model2 80.0881 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.1176 +Epoch [43/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0027, Loss2: 0.0033, Pure Ratio1: 10.1078, Pure Ratio2 10.0980 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [43/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9657, Pure Ratio2 10.0539 +Epoch [43/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1686, Pure Ratio2 10.1922 +Epoch [43/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.0425, Pure Ratio2 10.0948 +Epoch [43/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 10.0980, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 80.2885 % Model2 80.4087 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.1483 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.4510, Pure Ratio2 9.3333 +Epoch [44/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7941, Pure Ratio2 9.7157 +Epoch [44/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 9.9216 +Epoch [44/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0049 +Epoch [44/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0431, Pure Ratio2 9.9961 +Epoch [44/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.0817 +Epoch [44/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.1120, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 80.9195 % Model2 81.2800 %, Pure Ratio 1 10.1207 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.4314, Pure Ratio2 10.5490 +Epoch [45/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.3431, Pure Ratio2 10.4314 +Epoch [45/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 10.2941, Pure Ratio2 10.3529 +Epoch [45/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0833, Pure Ratio2 10.1029 +Epoch [45/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1804, Pure Ratio2 10.2353 +Epoch [45/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1307, Pure Ratio2 10.1634 +Epoch [45/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 10.0616, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 79.4171 % Model2 79.8277 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 10.1765, Pure Ratio2 10.0784 +Epoch [46/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.4412, Pure Ratio2 10.4216 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.3203, Pure Ratio2 10.3203 +Epoch [46/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1225, Pure Ratio2 10.1176 +Epoch [46/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [46/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.1275, Pure Ratio2 10.0882 +Epoch [46/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 10.0840, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 80.8994 % Model2 80.3786 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.4706, Pure Ratio2 9.4706 +Epoch [47/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 9.6471, Pure Ratio2 9.6471 +Epoch [47/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.7908, Pure Ratio2 9.7451 +Epoch [47/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 9.8627 +Epoch [47/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0078, Pure Ratio2 10.0627 +Epoch [47/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0915, Pure Ratio2 10.1340 +Epoch [47/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0364, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 80.8093 % Model2 80.3986 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.8039 +Epoch [48/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [48/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9608 +Epoch [48/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0343, Pure Ratio2 9.9804 +Epoch [48/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0275 +Epoch [48/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0654, Pure Ratio2 10.0131 +Epoch [48/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0924, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 80.3586 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.4706 +Epoch [49/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.3333, Pure Ratio2 10.4510 +Epoch [49/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.2941, Pure Ratio2 10.3987 +Epoch [49/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2402, Pure Ratio2 10.3333 +Epoch [49/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3176, Pure Ratio2 10.4549 +Epoch [49/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0621, Pure Ratio2 10.2124 +Epoch [49/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0756, Pure Ratio2 10.1821 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 79.7376 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.1106 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.7255, Pure Ratio2 9.9412 +Epoch [50/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0294, Pure Ratio2 10.2451 +Epoch [50/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.1307, Pure Ratio2 10.2222 +Epoch [50/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.2108 +Epoch [50/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9961, Pure Ratio2 10.0353 +Epoch [50/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0015, Pure Ratio1: 10.0359, Pure Ratio2 10.1046 +Epoch [50/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 10.0756, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 78.7861 % Model2 80.1282 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.6078, Pure Ratio2 10.7843 +Epoch [51/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.4020, Pure Ratio2 10.4412 +Epoch [51/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.0915, Pure Ratio2 10.1634 +Epoch [51/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9853 +Epoch [51/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0745, Pure Ratio2 10.1098 +Epoch [51/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0686, Pure Ratio2 10.1242 +Epoch [51/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0784, Pure Ratio2 10.1317 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 80.0381 % Model2 79.6575 %, Pure Ratio 1 10.0754 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8039, Pure Ratio2 10.1176 +Epoch [52/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.3039 +Epoch [52/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8954, Pure Ratio2 10.0980 +Epoch [52/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0245, Pure Ratio2 10.1961 +Epoch [52/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.1216 +Epoch [52/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9510, Pure Ratio2 10.1471 +Epoch [52/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9664, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 80.2183 % Model2 79.8978 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.1760 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.4706, Pure Ratio2 10.5882 +Epoch [53/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2157, Pure Ratio2 10.3235 +Epoch [53/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0719, Pure Ratio2 10.1765 +Epoch [53/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0343, Pure Ratio2 10.0833 +Epoch [53/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.9333 +Epoch [53/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0131, Pure Ratio2 10.0131 +Epoch [53/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0308, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 80.0581 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [54/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.0490 +Epoch [54/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.1895, Pure Ratio2 10.1961 +Epoch [54/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1275, Pure Ratio2 10.1471 +Epoch [54/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1804, Pure Ratio2 10.1490 +Epoch [54/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1144, Pure Ratio2 10.0654 +Epoch [54/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 80.2183 % Model2 78.7961 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.3333 +Epoch [55/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1765 +Epoch [55/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.0458 +Epoch [55/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1176 +Epoch [55/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.3255, Pure Ratio2 10.3490 +Epoch [55/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.1209 +Epoch [55/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.1485, Pure Ratio2 10.1653 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 81.2800 % Model2 80.2784 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1176 +Epoch [56/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1275, Pure Ratio2 10.0000 +Epoch [56/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 9.8693 +Epoch [56/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.0441, Pure Ratio2 9.9412 +Epoch [56/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1333, Pure Ratio2 10.0392 +Epoch [56/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.1340 +Epoch [56/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2577, Pure Ratio2 10.2185 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 80.4688 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7451 +Epoch [57/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9314 +Epoch [57/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9935, Pure Ratio2 10.0392 +Epoch [57/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0002, Pure Ratio1: 9.8775, Pure Ratio2 9.9216 +Epoch [57/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 9.9725 +Epoch [57/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9346, Pure Ratio2 9.9477 +Epoch [57/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0140, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 80.6090 % Model2 81.1398 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.5686, Pure Ratio2 9.6863 +Epoch [58/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9706, Pure Ratio2 10.0588 +Epoch [58/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.9346 +Epoch [58/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8971 +Epoch [58/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.9255 +Epoch [58/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 9.9641 +Epoch [58/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 80.1082 % Model2 80.0581 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 9.8235 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.1961 +Epoch [59/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.2876, Pure Ratio2 10.2614 +Epoch [59/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 10.0833 +Epoch [59/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9686, Pure Ratio2 9.9961 +Epoch [59/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.0556 +Epoch [59/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0084, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 80.3285 % Model2 80.4788 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.0784, Pure Ratio2 9.2353 +Epoch [60/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.6078, Pure Ratio2 9.6569 +Epoch [60/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9085 +Epoch [60/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0245, Pure Ratio2 9.9755 +Epoch [60/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0549, Pure Ratio2 9.9922 +Epoch [60/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1340, Pure Ratio2 10.0915 +Epoch [60/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.1232, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 80.2584 % Model2 79.9179 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.4510, Pure Ratio2 10.0784 +Epoch [61/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.4216, Pure Ratio2 10.2549 +Epoch [61/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3203, Pure Ratio2 10.2549 +Epoch [61/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2402, Pure Ratio2 10.2108 +Epoch [61/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1216, Pure Ratio2 10.0627 +Epoch [61/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 9.9967 +Epoch [61/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1232, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 79.7376 %, Pure Ratio 1 10.1131 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.8431 +Epoch [62/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 9.8922 +Epoch [62/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8105 +Epoch [62/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 10.0196, Pure Ratio2 9.9412 +Epoch [62/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1686, Pure Ratio2 10.0902 +Epoch [62/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.1863, Pure Ratio2 10.0948 +Epoch [62/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0015, Pure Ratio1: 10.1373, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 80.5589 % Model2 80.0481 %, Pure Ratio 1 10.1559 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 10.6471, Pure Ratio2 10.5882 +Epoch [63/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.4216, Pure Ratio2 10.2451 +Epoch [63/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1895, Pure Ratio2 10.1242 +Epoch [63/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.2451, Pure Ratio2 10.1765 +Epoch [63/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3294, Pure Ratio2 10.2549 +Epoch [63/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.2124, Pure Ratio2 10.2059 +Epoch [63/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1905, Pure Ratio2 10.1933 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 80.5389 % Model2 79.7676 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2941 +Epoch [64/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1471 +Epoch [64/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9281, Pure Ratio2 9.9935 +Epoch [64/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.9510 +Epoch [64/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9647, Pure Ratio2 9.9961 +Epoch [64/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0392 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0028, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 80.6290 % Model2 80.5389 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.7647 +Epoch [65/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0882 +Epoch [65/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.9869 +Epoch [65/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 10.0392 +Epoch [65/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0706, Pure Ratio2 10.1412 +Epoch [65/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0556, Pure Ratio2 10.1046 +Epoch [65/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0616, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 79.8377 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 11.0980, Pure Ratio2 10.8431 +Epoch [66/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.3431, Pure Ratio2 10.2745 +Epoch [66/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.4444, Pure Ratio2 10.3203 +Epoch [66/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.0392 +Epoch [66/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2588, Pure Ratio2 10.1843 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1830, Pure Ratio2 10.1242 +Epoch [66/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.2017, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 80.6791 % Model2 80.2885 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.9216 +Epoch [67/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.0784 +Epoch [67/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9477 +Epoch [67/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1814, Pure Ratio2 10.2108 +Epoch [67/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0314, Pure Ratio2 10.0510 +Epoch [67/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0654 +Epoch [67/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0084, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.3385 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 8.9412, Pure Ratio2 9.1569 +Epoch [68/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.9804 +Epoch [68/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0327 +Epoch [68/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.9755 +Epoch [68/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9490, Pure Ratio2 9.9843 +Epoch [68/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0359 +Epoch [68/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 10.0616, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 80.7893 % Model2 80.3986 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.8235, Pure Ratio2 10.7451 +Epoch [69/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.2451, Pure Ratio2 10.3039 +Epoch [69/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1242 +Epoch [69/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1078, Pure Ratio2 10.0539 +Epoch [69/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1216, Pure Ratio2 10.0588 +Epoch [69/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.0065 +Epoch [69/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0420, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 80.5288 % Model2 80.5088 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.4118 +Epoch [70/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.1863 +Epoch [70/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 10.0131, Pure Ratio2 10.0000 +Epoch [70/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0441, Pure Ratio2 10.0049 +Epoch [70/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.0784 +Epoch [70/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1438, Pure Ratio2 10.1275 +Epoch [70/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 10.1737, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 78.3353 % Model2 80.0481 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1176 +Epoch [71/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.4314, Pure Ratio2 10.3824 +Epoch [71/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 9.9739 +Epoch [71/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 9.9069 +Epoch [71/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9490, Pure Ratio2 9.9373 +Epoch [71/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.9477 +Epoch [71/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9776, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 79.3470 % Model2 79.8878 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0002, Pure Ratio1: 10.8431, Pure Ratio2 10.8824 +Epoch [72/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0023, Loss2: 0.0019, Pure Ratio1: 10.5784, Pure Ratio2 10.6765 +Epoch [72/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3791 +Epoch [72/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1324, Pure Ratio2 10.0931 +Epoch [72/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1686, Pure Ratio2 10.1333 +Epoch [72/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.0817 +Epoch [72/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0868, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 80.5689 % Model2 80.2684 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [73/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.2353 +Epoch [73/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.2941, Pure Ratio2 10.3072 +Epoch [73/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2647, Pure Ratio2 10.2451 +Epoch [73/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2235, Pure Ratio2 10.1961 +Epoch [73/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1046 +Epoch [73/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1877, Pure Ratio2 10.1148 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 79.1967 % Model2 79.3470 %, Pure Ratio 1 10.1332 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0015, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.3725 +Epoch [74/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 10.1765 +Epoch [74/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0327, Pure Ratio2 10.1242 +Epoch [74/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1569 +Epoch [74/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0941, Pure Ratio2 10.0902 +Epoch [74/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9967, Pure Ratio2 10.0294 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0280, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 79.5873 % Model2 80.2784 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.1081 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.2353 +Epoch [75/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1667 +Epoch [75/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.1373 +Epoch [75/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1912, Pure Ratio2 10.1324 +Epoch [75/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1686, Pure Ratio2 10.1490 +Epoch [75/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 10.1928, Pure Ratio2 10.1471 +Epoch [75/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1261, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 78.9163 % Model2 78.4756 %, Pure Ratio 1 10.1282 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.1176 +Epoch [76/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.1765, Pure Ratio2 10.2941 +Epoch [76/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1830 +Epoch [76/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1814 +Epoch [76/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0941 +Epoch [76/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9837, Pure Ratio2 9.9869 +Epoch [76/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9356, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 80.0280 % Model2 80.2284 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [77/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0294 +Epoch [77/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9739 +Epoch [77/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 10.0588 +Epoch [77/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1412, Pure Ratio2 10.1961 +Epoch [77/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1503, Pure Ratio2 10.1699 +Epoch [77/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1513, Pure Ratio2 10.1429 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 79.1767 % Model2 79.8578 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [78/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.7157 +Epoch [78/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0458, Pure Ratio2 9.9542 +Epoch [78/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9069 +Epoch [78/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9373 +Epoch [78/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 85.1562, Loss1: 0.0012, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0621 +Epoch [78/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.0868, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 79.7576 % Model2 79.7676 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6667, Pure Ratio2 10.4118 +Epoch [79/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.1765 +Epoch [79/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1634, Pure Ratio2 10.1438 +Epoch [79/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0686 +Epoch [79/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1137, Pure Ratio2 10.1373 +Epoch [79/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0719 +Epoch [79/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0700, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 78.9263 % Model2 79.0465 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1106 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6667, Pure Ratio2 10.5686 +Epoch [80/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2549 +Epoch [80/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0784 +Epoch [80/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 10.0588, Pure Ratio2 10.0049 +Epoch [80/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.2196, Pure Ratio2 10.1490 +Epoch [80/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1928, Pure Ratio2 10.1895 +Epoch [80/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 79.7276 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0588 +Epoch [81/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9216 +Epoch [81/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.1046 +Epoch [81/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0147, Pure Ratio2 10.0490 +Epoch [81/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0627, Pure Ratio2 10.1176 +Epoch [81/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0588 +Epoch [81/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9916, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 80.4587 % Model2 80.5589 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.5098, Pure Ratio2 10.2941 +Epoch [82/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.0588 +Epoch [82/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 10.0131 +Epoch [82/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9853, Pure Ratio2 10.0049 +Epoch [82/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.1176 +Epoch [82/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1307, Pure Ratio2 10.1928 +Epoch [82/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1541, Pure Ratio2 10.2045 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 80.2684 % Model2 79.6274 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1709 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.2941, Pure Ratio2 10.4118 +Epoch [83/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [83/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9608 +Epoch [83/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9167 +Epoch [83/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9451, Pure Ratio2 9.9765 +Epoch [83/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0948 +Epoch [83/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 79.2167 % Model2 80.2684 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9804 +Epoch [84/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.2255 +Epoch [84/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 10.1765 +Epoch [84/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.1667 +Epoch [84/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.1569 +Epoch [84/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0065, Pure Ratio2 10.1013 +Epoch [84/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9860, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 79.5974 % Model2 79.2668 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 10.0784 +Epoch [85/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8627 +Epoch [85/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8301, Pure Ratio2 9.9216 +Epoch [85/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9069, Pure Ratio2 9.9657 +Epoch [85/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0941 +Epoch [85/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0392 +Epoch [85/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9888, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 80.0982 % Model2 79.9479 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.1106 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9804 +Epoch [86/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0980 +Epoch [86/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 9.9608 +Epoch [86/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0735, Pure Ratio2 10.0539 +Epoch [86/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0471, Pure Ratio2 10.0510 +Epoch [86/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9608 +Epoch [86/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9832, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.7192 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.7451 +Epoch [87/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0294 +Epoch [87/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0392 +Epoch [87/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [87/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0235, Pure Ratio2 10.0745 +Epoch [87/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 10.0948 +Epoch [87/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0924, Pure Ratio2 10.1625 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 79.6975 % Model2 79.2568 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.1735 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1765 +Epoch [88/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9804 +Epoch [88/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.2941 +Epoch [88/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.2941 +Epoch [88/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 10.1333 +Epoch [88/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9641, Pure Ratio2 10.0621 +Epoch [88/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9328, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 79.8277 % Model2 80.4087 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [89/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8725 +Epoch [89/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1111 +Epoch [89/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0018, Pure Ratio1: 10.1912, Pure Ratio2 10.1863 +Epoch [89/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1961, Pure Ratio2 10.1137 +Epoch [89/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1209 +Epoch [89/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 80.1282 % Model2 79.8377 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.7843, Pure Ratio2 10.7843 +Epoch [90/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5392, Pure Ratio2 10.6176 +Epoch [90/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5817, Pure Ratio2 10.6275 +Epoch [90/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.4020, Pure Ratio2 10.3775 +Epoch [90/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3255, Pure Ratio2 10.3176 +Epoch [90/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1928, Pure Ratio2 10.1961 +Epoch [90/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1204, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 80.0982 % Model2 79.9079 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 9.9804 +Epoch [91/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2451, Pure Ratio2 10.2451 +Epoch [91/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [91/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1667, Pure Ratio2 10.1569 +Epoch [91/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1020, Pure Ratio2 10.0745 +Epoch [91/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0425 +Epoch [91/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1597, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 80.4788 % Model2 79.8978 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 10.0196 +Epoch [92/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.3333 +Epoch [92/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.3203 +Epoch [92/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.2157 +Epoch [92/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.1137 +Epoch [92/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0556 +Epoch [92/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9468, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 78.9964 % Model2 79.4271 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2745 +Epoch [93/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.0000 +Epoch [93/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.1111 +Epoch [93/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1422, Pure Ratio2 10.1471 +Epoch [93/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 10.0157 +Epoch [93/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9379, Pure Ratio2 9.9641 +Epoch [93/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 80.3886 % Model2 79.5172 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.7647 +Epoch [94/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.8824 +Epoch [94/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.7255 +Epoch [94/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.8873 +Epoch [94/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8784 +Epoch [94/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9150 +Epoch [94/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0504, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 80.0180 % Model2 79.8778 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.8431 +Epoch [95/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1667 +Epoch [95/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0065 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0882, Pure Ratio2 10.0637 +Epoch [95/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 10.0078 +Epoch [95/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1503 +Epoch [95/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1120, Pure Ratio2 10.1317 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 80.7192 % Model2 80.3285 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.3725 +Epoch [96/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7941 +Epoch [96/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.7516 +Epoch [96/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7990 +Epoch [96/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9059, Pure Ratio2 9.9137 +Epoch [96/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 10.0294 +Epoch [96/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0476, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 79.2869 % Model2 80.1382 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.0392 +Epoch [97/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9902 +Epoch [97/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.0719 +Epoch [97/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 9.9314 +Epoch [97/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1020, Pure Ratio2 9.9922 +Epoch [97/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 9.9869 +Epoch [97/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.1485, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 80.8293 % Model2 79.3670 %, Pure Ratio 1 10.1232 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.4902 +Epoch [98/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0000 +Epoch [98/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0719 +Epoch [98/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 9.9853 +Epoch [98/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0471, Pure Ratio2 10.0431 +Epoch [98/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0980 +Epoch [98/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1653, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 79.7075 % Model2 79.8478 %, Pure Ratio 1 10.1282 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0588 +Epoch [99/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1667, Pure Ratio2 10.1471 +Epoch [99/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1307 +Epoch [99/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.1275 +Epoch [99/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.2000 +Epoch [99/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.1471 +Epoch [99/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.7792 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6863 +Epoch [100/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0686 +Epoch [100/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0588 +Epoch [100/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8775 +Epoch [100/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9059 +Epoch [100/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0556 +Epoch [100/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 79.7376 % Model2 79.6675 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.1408 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.1373 +Epoch [101/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [101/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.3137, Pure Ratio2 10.2876 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2500, Pure Ratio2 10.2353 +Epoch [101/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1843, Pure Ratio2 10.1294 +Epoch [101/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0392 +Epoch [101/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2437, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 80.3586 % Model2 79.4471 %, Pure Ratio 1 10.1961 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.8431 +Epoch [102/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.1176 +Epoch [102/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1422 +Epoch [102/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.1216 +Epoch [102/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0686 +Epoch [102/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.1541 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 80.7192 % Model2 80.2083 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [103/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8333 +Epoch [103/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.1569 +Epoch [103/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2010, Pure Ratio2 10.2206 +Epoch [103/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 10.0941 +Epoch [103/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.1667 +Epoch [103/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0840, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 80.5088 % Model2 79.7075 %, Pure Ratio 1 10.1081 %, Pure Ratio 2 10.1609 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.1176 +Epoch [104/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.9216 +Epoch [104/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0458 +Epoch [104/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0735 +Epoch [104/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0667, Pure Ratio2 10.1451 +Epoch [104/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0817, Pure Ratio2 10.1275 +Epoch [104/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 79.9379 % Model2 79.8578 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.1357 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.0000 +Epoch [105/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [105/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0261 +Epoch [105/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9510 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.0980 +Epoch [105/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0719, Pure Ratio2 10.1405 +Epoch [105/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 79.2969 % Model2 79.8878 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Epoch [106/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0980 +Epoch [106/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.0523 +Epoch [106/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 10.0441 +Epoch [106/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 10.0039 +Epoch [106/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0131, Pure Ratio2 10.0654 +Epoch [106/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 80.5489 % Model2 80.4087 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.6078 +Epoch [107/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.5784 +Epoch [107/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.1503 +Epoch [107/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2598, Pure Ratio2 10.2402 +Epoch [107/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [107/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1895 +Epoch [107/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1289, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 79.8277 % Model2 79.9279 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.7843 +Epoch [108/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.4412, Pure Ratio2 9.5882 +Epoch [108/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3595, Pure Ratio2 9.5490 +Epoch [108/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.6127 +Epoch [108/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6784, Pure Ratio2 9.7882 +Epoch [108/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9869 +Epoch [108/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 79.9179 % Model2 80.1683 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 11.1569, Pure Ratio2 11.3529 +Epoch [109/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.4902 +Epoch [109/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3072, Pure Ratio2 10.3856 +Epoch [109/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2892, Pure Ratio2 10.3382 +Epoch [109/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.2353 +Epoch [109/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1340, Pure Ratio2 10.1601 +Epoch [109/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 79.9179 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2157, Pure Ratio2 10.0588 +Epoch [110/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.3039 +Epoch [110/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2810, Pure Ratio2 10.1699 +Epoch [110/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.1716 +Epoch [110/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0000, Pure Ratio1: 10.2118, Pure Ratio2 10.1882 +Epoch [110/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1601, Pure Ratio2 10.1569 +Epoch [110/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 79.3069 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0392 +Epoch [111/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.0588 +Epoch [111/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 9.9608 +Epoch [111/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.1078 +Epoch [111/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1922, Pure Ratio2 10.0824 +Epoch [111/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1797, Pure Ratio2 10.0588 +Epoch [111/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1597, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 79.9279 % Model2 80.8393 %, Pure Ratio 1 10.1357 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 9.9804 +Epoch [112/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.0588 +Epoch [112/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4248, Pure Ratio2 10.3399 +Epoch [112/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.3578 +Epoch [112/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3059, Pure Ratio2 10.2784 +Epoch [112/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1993, Pure Ratio2 10.1471 +Epoch [112/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1148, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 79.5673 % Model2 80.2484 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.2353 +Epoch [113/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9706 +Epoch [113/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9608 +Epoch [113/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.1716 +Epoch [113/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0902 +Epoch [113/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0882 +Epoch [113/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 10.2325, Pure Ratio2 10.2017 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 80.4187 % Model2 80.5990 %, Pure Ratio 1 10.1483 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.4510 +Epoch [114/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2451 +Epoch [114/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2810 +Epoch [114/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.1667 +Epoch [114/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3059, Pure Ratio2 10.2784 +Epoch [114/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2810, Pure Ratio2 10.2451 +Epoch [114/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.1653, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 80.7192 % Model2 80.4988 %, Pure Ratio 1 10.1181 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8431 +Epoch [115/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9902 +Epoch [115/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2222 +Epoch [115/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2206, Pure Ratio2 10.1765 +Epoch [115/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1608, Pure Ratio2 10.0667 +Epoch [115/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1438, Pure Ratio2 10.0621 +Epoch [115/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 79.6274 %, Pure Ratio 1 10.1332 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.4706 +Epoch [116/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.2059 +Epoch [116/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9608 +Epoch [116/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 10.0294 +Epoch [116/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 10.0588 +Epoch [116/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 10.1340 +Epoch [116/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9356, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 79.7576 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.1257 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.3529 +Epoch [117/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.2451 +Epoch [117/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2680, Pure Ratio2 10.3464 +Epoch [117/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1618, Pure Ratio2 10.2010 +Epoch [117/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 10.0078 +Epoch [117/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.9902 +Epoch [117/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0560, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 80.2784 % Model2 79.8978 %, Pure Ratio 1 10.0930 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.9020 +Epoch [118/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.1275 +Epoch [118/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.2092 +Epoch [118/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1569 +Epoch [118/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.2157 +Epoch [118/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.2908 +Epoch [118/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.2017 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 80.7792 % Model2 80.2083 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9412 +Epoch [119/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.3725 +Epoch [119/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1438 +Epoch [119/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9559 +Epoch [119/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [119/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.0719 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0420, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 79.2969 % Model2 79.6374 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.4706 +Epoch [120/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.5490 +Epoch [120/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.4183, Pure Ratio2 10.5229 +Epoch [120/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.3088 +Epoch [120/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 10.1098 +Epoch [120/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 10.1275 +Epoch [120/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0700, Pure Ratio2 10.1793 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 79.7977 % Model2 79.5172 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.1257 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2549 +Epoch [121/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.0784 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1242 +Epoch [121/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2941 +Epoch [121/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0627 +Epoch [121/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.1046 +Epoch [121/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0700, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 79.4071 % Model2 79.4571 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0784 +Epoch [122/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8824 +Epoch [122/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.9281 +Epoch [122/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7990, Pure Ratio2 9.8578 +Epoch [122/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [122/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9902 +Epoch [122/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0840, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 79.7175 %, Pure Ratio 1 10.1131 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0000 +Epoch [123/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.4118 +Epoch [123/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4575, Pure Ratio2 10.3333 +Epoch [123/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.2206 +Epoch [123/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2275, Pure Ratio2 10.1216 +Epoch [123/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1405, Pure Ratio2 10.0490 +Epoch [123/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1513, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 80.1983 % Model2 79.7776 %, Pure Ratio 1 10.1282 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5882, Pure Ratio2 10.6078 +Epoch [124/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.2157 +Epoch [124/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.2614 +Epoch [124/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2157 +Epoch [124/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2353 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1601, Pure Ratio2 10.2582 +Epoch [124/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.2073 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 80.1883 % Model2 80.7692 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.1634 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.6667 +Epoch [125/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4412, Pure Ratio2 10.6373 +Epoch [125/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3072, Pure Ratio2 10.4837 +Epoch [125/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.3578 +Epoch [125/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.2588 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1601, Pure Ratio2 10.2353 +Epoch [125/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 80.0982 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.8039 +Epoch [126/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.5588 +Epoch [126/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.6078 +Epoch [126/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9069 +Epoch [126/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8941, Pure Ratio2 9.8667 +Epoch [126/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9902 +Epoch [126/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9916, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 80.4187 % Model2 80.1883 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.5686, Pure Ratio2 10.8235 +Epoch [127/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3627 +Epoch [127/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3203, Pure Ratio2 10.3529 +Epoch [127/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3284, Pure Ratio2 10.3382 +Epoch [127/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0745, Pure Ratio2 10.0784 +Epoch [127/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.1307 +Epoch [127/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0840, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 79.7175 % Model2 80.4087 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [128/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1078 +Epoch [128/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0196 +Epoch [128/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2500, Pure Ratio2 10.1912 +Epoch [128/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.1843 +Epoch [128/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2908, Pure Ratio2 10.2876 +Epoch [128/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1793, Pure Ratio2 10.1961 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 81.1899 % Model2 80.7692 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 10.0196 +Epoch [129/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4020, Pure Ratio2 9.5686 +Epoch [129/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5948, Pure Ratio2 9.7124 +Epoch [129/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [129/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0667, Pure Ratio2 10.0784 +Epoch [129/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.0294 +Epoch [129/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 10.0392 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 79.6174 % Model2 79.9880 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7451, Pure Ratio2 10.6471 +Epoch [130/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4020, Pure Ratio2 10.2647 +Epoch [130/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3791, Pure Ratio2 10.1634 +Epoch [130/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0931, Pure Ratio2 9.9167 +Epoch [130/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0510 +Epoch [130/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.0882 +Epoch [130/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.9595 %, Pure Ratio 1 10.1559 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1176 +Epoch [131/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5098, Pure Ratio2 10.5098 +Epoch [131/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2876, Pure Ratio2 10.3725 +Epoch [131/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3578 +Epoch [131/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3216 +Epoch [131/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 10.2843, Pure Ratio2 10.2974 +Epoch [131/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1541, Pure Ratio2 10.1625 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 80.8494 % Model2 80.4888 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.5294 +Epoch [132/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Epoch [132/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.7516 +Epoch [132/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.7843 +Epoch [132/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.8078 +Epoch [132/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9281 +Epoch [132/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9496, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 80.3986 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8039, Pure Ratio2 10.6667 +Epoch [133/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.5196 +Epoch [133/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1373 +Epoch [133/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.0833 +Epoch [133/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0008, Pure Ratio1: 10.2000, Pure Ratio2 10.1490 +Epoch [133/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0196 +Epoch [133/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1345, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 79.6474 % Model2 80.8293 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.2353 +Epoch [134/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3627 +Epoch [134/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 10.0327 +Epoch [134/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9265, Pure Ratio2 9.9951 +Epoch [134/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0353 +Epoch [134/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9837 +Epoch [134/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 80.7592 % Model2 80.0982 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.1257 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.4510 +Epoch [135/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.4020 +Epoch [135/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3791, Pure Ratio2 10.3333 +Epoch [135/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1275 +Epoch [135/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.1686 +Epoch [135/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1536, Pure Ratio2 10.2157 +Epoch [135/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 79.9179 % Model2 80.4788 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.1559 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.5686 +Epoch [136/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0784 +Epoch [136/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8431 +Epoch [136/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8186, Pure Ratio2 9.9510 +Epoch [136/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8157, Pure Ratio2 9.9608 +Epoch [136/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [136/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 80.2183 % Model2 80.5990 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1765 +Epoch [137/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2059, Pure Ratio2 10.3431 +Epoch [137/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.2222 +Epoch [137/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.1176 +Epoch [137/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.1137 +Epoch [137/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1471 +Epoch [137/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 80.3886 % Model2 80.7292 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3137, Pure Ratio2 10.3333 +Epoch [138/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8824 +Epoch [138/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.8562 +Epoch [138/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.9118 +Epoch [138/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9412 +Epoch [138/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 9.9771 +Epoch [138/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1317, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 80.4087 % Model2 80.5389 %, Pure Ratio 1 10.1533 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6275, Pure Ratio2 10.6471 +Epoch [139/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8627 +Epoch [139/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7778 +Epoch [139/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9363 +Epoch [139/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1882, Pure Ratio2 10.1255 +Epoch [139/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0458 +Epoch [139/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 80.5589 % Model2 80.0681 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.2353 +Epoch [140/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [140/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0784 +Epoch [140/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.8039 +Epoch [140/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9490 +Epoch [140/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0654 +Epoch [140/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 80.2985 % Model2 80.9395 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9216 +Epoch [141/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.0098 +Epoch [141/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.1307 +Epoch [141/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 10.0049 +Epoch [141/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9725 +Epoch [141/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.9248 +Epoch [141/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 80.2885 % Model2 80.6390 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.9608 +Epoch [142/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.7549 +Epoch [142/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9085 +Epoch [142/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8922 +Epoch [142/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9294 +Epoch [142/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9641 +Epoch [142/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 81.1799 % Model2 80.4688 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.3529 +Epoch [143/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9706 +Epoch [143/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9150 +Epoch [143/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1716, Pure Ratio2 10.1373 +Epoch [143/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2471, Pure Ratio2 10.1373 +Epoch [143/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1993, Pure Ratio2 10.0915 +Epoch [143/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 80.9395 % Model2 81.3201 %, Pure Ratio 1 10.1634 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.6863 +Epoch [144/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2941 +Epoch [144/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.0784 +Epoch [144/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1569 +Epoch [144/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.2275, Pure Ratio2 10.1647 +Epoch [144/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2190, Pure Ratio2 10.1307 +Epoch [144/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1485, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 80.0781 % Model2 80.5689 %, Pure Ratio 1 10.1584 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.5686 +Epoch [145/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6176 +Epoch [145/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [145/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0784 +Epoch [145/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.0549 +Epoch [145/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0327 +Epoch [145/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 81.2700 % Model2 80.6190 %, Pure Ratio 1 10.1458 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9216 +Epoch [146/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.4314 +Epoch [146/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.2549 +Epoch [146/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.1078 +Epoch [146/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.2157 +Epoch [146/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.1895 +Epoch [146/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.1737 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.2284 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.1232 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2549 +Epoch [147/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1765 +Epoch [147/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [147/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0686 +Epoch [147/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0471, Pure Ratio2 10.0667 +Epoch [147/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.1569 +Epoch [147/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1232, Pure Ratio2 10.1737 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 80.7091 % Model2 81.2400 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.3529 +Epoch [148/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6765 +Epoch [148/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.7908 +Epoch [148/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.9069 +Epoch [148/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0157, Pure Ratio2 10.0275 +Epoch [148/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9510 +Epoch [148/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0392 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 80.6991 % Model2 80.5789 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [149/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9020 +Epoch [149/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0065 +Epoch [149/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0735, Pure Ratio2 10.0735 +Epoch [149/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 10.0471 +Epoch [149/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0016, Loss2: 0.0016, Pure Ratio1: 10.0948, Pure Ratio2 10.0882 +Epoch [149/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0896, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 81.2800 % Model2 80.6090 %, Pure Ratio 1 10.1207 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.6471 +Epoch [150/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6275, Pure Ratio2 10.5490 +Epoch [150/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2876, Pure Ratio2 10.1699 +Epoch [150/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.1716 +Epoch [150/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.1137 +Epoch [150/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1176 +Epoch [150/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1036, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 80.2584 % Model2 81.4103 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.9412 +Epoch [151/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8627 +Epoch [151/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2418 +Epoch [151/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1961 +Epoch [151/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1529 +Epoch [151/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.1013 +Epoch [151/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1120, Pure Ratio2 10.1625 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 80.4387 % Model2 80.7893 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.1373 +Epoch [152/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0882 +Epoch [152/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1503 +Epoch [152/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9461 +Epoch [152/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8275, Pure Ratio2 9.8235 +Epoch [152/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8889 +Epoch [152/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 80.6791 % Model2 80.4187 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9608 +Epoch [153/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2647 +Epoch [153/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.2549 +Epoch [153/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1127 +Epoch [153/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1843 +Epoch [153/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1732, Pure Ratio2 10.1993 +Epoch [153/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1289, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 80.8393 % Model2 80.7492 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5686, Pure Ratio2 10.5882 +Epoch [154/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3431 +Epoch [154/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.3333 +Epoch [154/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0343 +Epoch [154/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 10.0261 +Epoch [154/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 79.5673 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9804 +Epoch [155/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0980 +Epoch [155/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9477 +Epoch [155/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.1324 +Epoch [155/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.1686 +Epoch [155/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0032, Loss2: 0.0035, Pure Ratio1: 10.2647, Pure Ratio2 10.2320 +Epoch [155/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2689, Pure Ratio2 10.2717 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 81.4603 % Model2 80.2784 %, Pure Ratio 1 10.1332 %, Pure Ratio 2 10.1483 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.5686 +Epoch [156/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7353 +Epoch [156/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0065 +Epoch [156/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.2010 +Epoch [156/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1725, Pure Ratio2 10.2078 +Epoch [156/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1634 +Epoch [156/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 81.1398 % Model2 80.7592 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.7059 +Epoch [157/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8627 +Epoch [157/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8954 +Epoch [157/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.1520 +Epoch [157/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0667, Pure Ratio2 10.0863 +Epoch [157/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0556 +Epoch [157/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0252, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 80.7091 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.5098 +Epoch [158/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9902 +Epoch [158/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9281 +Epoch [158/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.8480 +Epoch [158/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.8157 +Epoch [158/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9379, Pure Ratio2 9.8889 +Epoch [158/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 80.6190 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9020 +Epoch [159/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0000 +Epoch [159/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8627 +Epoch [159/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9951 +Epoch [159/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0431, Pure Ratio2 9.9686 +Epoch [159/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0556 +Epoch [159/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 80.9495 % Model2 81.3201 %, Pure Ratio 1 10.1383 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7647, Pure Ratio2 10.6863 +Epoch [160/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4412, Pure Ratio2 10.4118 +Epoch [160/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4641, Pure Ratio2 10.4771 +Epoch [160/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3088, Pure Ratio2 10.3725 +Epoch [160/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2863, Pure Ratio2 10.3255 +Epoch [160/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2059, Pure Ratio2 10.2386 +Epoch [160/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 80.7492 % Model2 80.6591 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.1961 +Epoch [161/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3922 +Epoch [161/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.1307 +Epoch [161/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.1225 +Epoch [161/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3020, Pure Ratio2 10.2431 +Epoch [161/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.1438 +Epoch [161/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1849, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 81.1498 % Model2 80.8594 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7647, Pure Ratio2 10.6471 +Epoch [162/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0017, Pure Ratio1: 10.2157, Pure Ratio2 10.2157 +Epoch [162/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0850 +Epoch [162/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9951 +Epoch [162/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 10.0275 +Epoch [162/200], Iter [300/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9641 +Epoch [162/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 80.7692 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [163/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1667 +Epoch [163/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 10.1242 +Epoch [163/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 10.0882 +Epoch [163/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9961 +Epoch [163/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1144, Pure Ratio2 10.1373 +Epoch [163/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 80.8494 % Model2 80.8994 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7451 +Epoch [164/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8824 +Epoch [164/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9935 +Epoch [164/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9902 +Epoch [164/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.0627 +Epoch [164/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.1275 +Epoch [164/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.1317 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 80.9996 % Model2 81.2500 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.5294 +Epoch [165/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.6765 +Epoch [165/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.6797 +Epoch [165/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [165/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0000 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9085 +Epoch [165/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 80.1583 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5686, Pure Ratio2 10.6667 +Epoch [166/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1961 +Epoch [166/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9673 +Epoch [166/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1520, Pure Ratio2 10.0686 +Epoch [166/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9961 +Epoch [166/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0719 +Epoch [166/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 81.4603 % Model2 80.4287 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.4510 +Epoch [167/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1667 +Epoch [167/200], Iter [150/390] Training Accuracy1: 96.0938, Training Accuracy2: 96.0938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8301 +Epoch [167/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9265 +Epoch [167/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1059 +Epoch [167/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.1569 +Epoch [167/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1569 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 80.1182 % Model2 80.0280 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.1383 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.9216 +Epoch [168/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3627, Pure Ratio2 10.5686 +Epoch [168/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3464, Pure Ratio2 10.5229 +Epoch [168/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3971, Pure Ratio2 10.5196 +Epoch [168/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3490, Pure Ratio2 10.4078 +Epoch [168/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3791, Pure Ratio2 10.4771 +Epoch [168/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1625, Pure Ratio2 10.2549 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 80.6891 % Model2 80.0280 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.1765 +Epoch [169/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.2549 +Epoch [169/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.2484 +Epoch [169/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1176 +Epoch [169/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1765 +Epoch [169/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.1307 +Epoch [169/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0896, Pure Ratio2 10.1625 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 80.5188 % Model2 80.8193 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8039 +Epoch [170/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1569 +Epoch [170/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0458 +Epoch [170/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1225 +Epoch [170/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 9.9882 +Epoch [170/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0065 +Epoch [170/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1653, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 81.0797 %, Pure Ratio 1 10.1709 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7059 +Epoch [171/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.8529 +Epoch [171/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [171/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9216 +Epoch [171/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.8863 +Epoch [171/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.8922 +Epoch [171/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 81.1198 % Model2 80.8494 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9608 +Epoch [172/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3824, Pure Ratio2 10.4020 +Epoch [172/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.3987 +Epoch [172/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2500, Pure Ratio2 10.3186 +Epoch [172/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3373 +Epoch [172/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1797, Pure Ratio2 10.2255 +Epoch [172/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0952, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 80.9996 % Model2 79.9780 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.6667 +Epoch [173/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9608 +Epoch [173/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8758 +Epoch [173/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0098 +Epoch [173/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 9.9216 +Epoch [173/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9248 +Epoch [173/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 80.8894 %, Pure Ratio 1 10.1433 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6863, Pure Ratio2 10.6275 +Epoch [174/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.4216 +Epoch [174/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3268 +Epoch [174/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4020, Pure Ratio2 10.2843 +Epoch [174/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.2275 +Epoch [174/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1144 +Epoch [174/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 81.0697 % Model2 80.5990 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9608 +Epoch [175/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9804 +Epoch [175/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 10.0719 +Epoch [175/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0539 +Epoch [175/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 10.0745 +Epoch [175/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.1209 +Epoch [175/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0168, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 80.7292 % Model2 80.9996 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9020 +Epoch [176/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9706 +Epoch [176/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9542 +Epoch [176/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0196 +Epoch [176/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 10.0314 +Epoch [176/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8987, Pure Ratio2 9.8824 +Epoch [176/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0616, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 81.6506 % Model2 80.8393 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.2549 +Epoch [177/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.8235 +Epoch [177/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.8824 +Epoch [177/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.1569 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0000 +Epoch [177/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 10.1013 +Epoch [177/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 81.5304 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [178/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1078 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0719 +Epoch [178/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1176 +Epoch [178/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.0078 +Epoch [178/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9641 +Epoch [178/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0924, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 80.6691 % Model2 80.9095 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.9608 +Epoch [179/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.9510 +Epoch [179/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 10.0915 +Epoch [179/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 10.1863 +Epoch [179/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 10.2941 +Epoch [179/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.3399 +Epoch [179/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9748, Pure Ratio2 10.2045 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 80.7492 % Model2 80.9996 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.1885 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [180/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0098 +Epoch [180/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0523 +Epoch [180/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0098 +Epoch [180/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 10.0275 +Epoch [180/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 10.0556 +Epoch [180/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 81.5705 % Model2 81.2300 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4118 +Epoch [181/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3627, Pure Ratio2 10.3824 +Epoch [181/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.2941 +Epoch [181/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 10.0441 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 10.1059 +Epoch [181/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0000 +Epoch [181/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 81.4203 % Model2 81.3301 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7059, Pure Ratio2 10.6275 +Epoch [182/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4020, Pure Ratio2 10.4216 +Epoch [182/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2745 +Epoch [182/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.2843 +Epoch [182/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1059, Pure Ratio2 10.1451 +Epoch [182/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 10.0392 +Epoch [182/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.1204 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 81.0397 % Model2 81.9010 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9804 +Epoch [183/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8627 +Epoch [183/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.1046 +Epoch [183/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.2206 +Epoch [183/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.2000 +Epoch [183/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.1144 +Epoch [183/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0560, Pure Ratio2 10.2017 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 81.4904 % Model2 81.5104 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.1533 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.4902 +Epoch [184/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [184/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9346 +Epoch [184/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9412 +Epoch [184/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.0353 +Epoch [184/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1471 +Epoch [184/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1148, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 81.2200 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.8431 +Epoch [185/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1667 +Epoch [185/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.1830 +Epoch [185/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2549 +Epoch [185/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.1804 +Epoch [185/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.1307 +Epoch [185/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.1401 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 80.8594 % Model2 81.3502 %, Pure Ratio 1 10.0754 %, Pure Ratio 2 10.0980 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [186/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7745 +Epoch [186/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0261 +Epoch [186/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0686 +Epoch [186/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1176 +Epoch [186/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1732, Pure Ratio2 10.1863 +Epoch [186/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1849 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 81.0697 % Model2 80.8293 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.4314 +Epoch [187/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.5686 +Epoch [187/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.1046 +Epoch [187/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0441 +Epoch [187/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1529 +Epoch [187/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1340, Pure Ratio2 10.1405 +Epoch [187/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1036, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 80.6090 %, Pure Ratio 1 10.1106 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7843, Pure Ratio2 10.4706 +Epoch [188/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4804, Pure Ratio2 10.4118 +Epoch [188/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1895 +Epoch [188/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 9.9167 +Epoch [188/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.9098 +Epoch [188/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0523 +Epoch [188/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 81.4002 % Model2 81.5004 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3333 +Epoch [189/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0588 +Epoch [189/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.7451 +Epoch [189/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7500 +Epoch [189/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.7843 +Epoch [189/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 9.9118 +Epoch [189/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1345, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 81.6106 % Model2 81.4403 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.2353 +Epoch [190/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.2059 +Epoch [190/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7386 +Epoch [190/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8627 +Epoch [190/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9255 +Epoch [190/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9379 +Epoch [190/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1289, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 81.7508 % Model2 81.2099 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0784 +Epoch [191/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0196 +Epoch [191/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0915 +Epoch [191/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0147 +Epoch [191/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 10.0000 +Epoch [191/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.1307 +Epoch [191/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 81.8810 % Model2 80.9996 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.1357 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6667, Pure Ratio2 10.7255 +Epoch [192/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.3529 +Epoch [192/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.3007 +Epoch [192/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2843 +Epoch [192/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.2039 +Epoch [192/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3039 +Epoch [192/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0016, Loss2: 0.0010, Pure Ratio1: 10.1345, Pure Ratio2 10.2465 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 81.5104 % Model2 81.7208 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8431, Pure Ratio2 10.8824 +Epoch [193/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4706 +Epoch [193/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4183, Pure Ratio2 10.4248 +Epoch [193/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.1814 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2039, Pure Ratio2 10.3137 +Epoch [193/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.2092 +Epoch [193/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 81.2800 % Model2 81.6206 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.1181 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8627 +Epoch [194/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1078 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.2222 +Epoch [194/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1275 +Epoch [194/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1490 +Epoch [194/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1634 +Epoch [194/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0420, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 81.2600 % Model2 81.2800 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.8824 +Epoch [195/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4804, Pure Ratio2 9.7647 +Epoch [195/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.9542 +Epoch [195/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 10.0147 +Epoch [195/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6510, Pure Ratio2 9.8588 +Epoch [195/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.9314 +Epoch [195/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 81.2500 % Model2 81.1699 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.1584 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.4118 +Epoch [196/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.2255 +Epoch [196/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2026 +Epoch [196/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1324 +Epoch [196/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1843 +Epoch [196/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.1275 +Epoch [196/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 81.4002 % Model2 80.8994 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.2353 +Epoch [197/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.3431 +Epoch [197/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.2092 +Epoch [197/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1324 +Epoch [197/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.1216 +Epoch [197/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.1176 +Epoch [197/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.1457 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 81.8309 % Model2 81.4002 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.1433 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4902 +Epoch [198/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.3039 +Epoch [198/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.2092 +Epoch [198/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9706 +Epoch [198/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9569 +Epoch [198/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 10.0131 +Epoch [198/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 81.2600 % Model2 80.9495 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.4118 +Epoch [199/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.2059 +Epoch [199/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0261 +Epoch [199/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0245 +Epoch [199/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 10.0627 +Epoch [199/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0261 +Epoch [199/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 81.3301 % Model2 80.9696 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.8431 +Epoch [200/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.4216 +Epoch [200/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.2810 +Epoch [200/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.2794 +Epoch [200/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 10.2235 +Epoch [200/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0621, Pure Ratio2 10.2353 +Epoch [200/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.1877 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 81.1799 % Model2 81.3101 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.1232 % diff --git a/other_methods/coteaching/coteaching_results/out_2_4.log b/other_methods/coteaching/coteaching_results/out_2_4.log new file mode 100644 index 0000000..e3e3066 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_2_4.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 27.3438, Loss1: 0.0163, Loss2: 0.0166, Pure Ratio1: 9.8560, Pure Ratio2 9.8720 +Epoch [2/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0149, Loss2: 0.0149, Pure Ratio1: 9.9280, Pure Ratio2 9.9520 +Epoch [2/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 32.0312, Loss1: 0.0157, Loss2: 0.0156, Pure Ratio1: 10.0000, Pure Ratio2 10.0267 +Epoch [2/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.8125, Loss1: 0.0159, Loss2: 0.0155, Pure Ratio1: 10.0160, Pure Ratio2 10.0200 +Epoch [2/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 36.7188, Loss1: 0.0151, Loss2: 0.0153, Pure Ratio1: 10.0352, Pure Ratio2 10.0352 +Epoch [2/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0136, Loss2: 0.0136, Pure Ratio1: 9.9653, Pure Ratio2 9.9600 +Epoch [2/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0145, Loss2: 0.0148, Pure Ratio1: 9.8560, Pure Ratio2 9.8674 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 44.9018 % Model2 41.4764 %, Pure Ratio 1 9.8503 %, Pure Ratio 2 9.8626 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 36.7188, Loss1: 0.0131, Loss2: 0.0131, Pure Ratio1: 8.8525, Pure Ratio2 8.8852 +Epoch [3/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0141, Loss2: 0.0137, Pure Ratio1: 9.4180, Pure Ratio2 9.4918 +Epoch [3/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 35.1562, Loss1: 0.0129, Loss2: 0.0133, Pure Ratio1: 9.7158, Pure Ratio2 9.7213 +Epoch [3/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0128, Loss2: 0.0127, Pure Ratio1: 9.7582, Pure Ratio2 9.7500 +Epoch [3/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0137, Pure Ratio1: 9.8131, Pure Ratio2 9.7934 +Epoch [3/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0128, Loss2: 0.0126, Pure Ratio1: 9.9016, Pure Ratio2 9.8798 +Epoch [3/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0137, Loss2: 0.0133, Pure Ratio1: 9.9485, Pure Ratio2 9.9344 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 51.8329 % Model2 52.6542 %, Pure Ratio 1 9.8760 %, Pure Ratio 2 9.8676 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0129, Loss2: 0.0130, Pure Ratio1: 9.2773, Pure Ratio2 9.3950 +Epoch [4/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0141, Loss2: 0.0142, Pure Ratio1: 9.5126, Pure Ratio2 9.6134 +Epoch [4/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0126, Loss2: 0.0120, Pure Ratio1: 9.7031, Pure Ratio2 9.7815 +Epoch [4/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0125, Loss2: 0.0123, Pure Ratio1: 9.6555, Pure Ratio2 9.7059 +Epoch [4/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0127, Loss2: 0.0125, Pure Ratio1: 9.6672, Pure Ratio2 9.7076 +Epoch [4/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0112, Loss2: 0.0111, Pure Ratio1: 9.8235, Pure Ratio2 9.8543 +Epoch [4/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0109, Loss2: 0.0111, Pure Ratio1: 9.8247, Pure Ratio2 9.8583 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 58.1631 % Model2 54.3470 %, Pure Ratio 1 9.8686 %, Pure Ratio 2 9.9138 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0129, Loss2: 0.0130, Pure Ratio1: 10.1034, Pure Ratio2 10.1379 +Epoch [5/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0101, Loss2: 0.0107, Pure Ratio1: 10.0345, Pure Ratio2 10.0345 +Epoch [5/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0127, Loss2: 0.0122, Pure Ratio1: 10.1782, Pure Ratio2 10.2241 +Epoch [5/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0115, Loss2: 0.0110, Pure Ratio1: 9.9828, Pure Ratio2 10.0517 +Epoch [5/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0106, Loss2: 0.0105, Pure Ratio1: 10.0828, Pure Ratio2 10.1276 +Epoch [5/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0118, Loss2: 0.0112, Pure Ratio1: 9.9454, Pure Ratio2 9.9828 +Epoch [5/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0116, Loss2: 0.0121, Pure Ratio1: 9.8744, Pure Ratio2 9.9261 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 57.7524 % Model2 56.9111 %, Pure Ratio 1 9.8475 %, Pure Ratio 2 9.8961 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 34.3750, Loss1: 0.0140, Loss2: 0.0147, Pure Ratio1: 9.9823, Pure Ratio2 10.0885 +Epoch [6/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0124, Loss2: 0.0124, Pure Ratio1: 9.9735, Pure Ratio2 10.0177 +Epoch [6/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0105, Loss2: 0.0101, Pure Ratio1: 9.8525, Pure Ratio2 9.8820 +Epoch [6/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0119, Loss2: 0.0121, Pure Ratio1: 9.9159, Pure Ratio2 9.9690 +Epoch [6/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0111, Loss2: 0.0114, Pure Ratio1: 9.8655, Pure Ratio2 9.9221 +Epoch [6/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0114, Pure Ratio1: 9.8142, Pure Ratio2 9.8407 +Epoch [6/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0106, Loss2: 0.0108, Pure Ratio1: 9.7876, Pure Ratio2 9.8255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 62.9708 % Model2 60.3065 %, Pure Ratio 1 9.8207 %, Pure Ratio 2 9.8570 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0109, Loss2: 0.0106, Pure Ratio1: 9.8000, Pure Ratio2 9.8000 +Epoch [7/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0106, Loss2: 0.0105, Pure Ratio1: 9.8909, Pure Ratio2 9.8273 +Epoch [7/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0118, Loss2: 0.0118, Pure Ratio1: 9.8970, Pure Ratio2 9.8182 +Epoch [7/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0102, Loss2: 0.0108, Pure Ratio1: 9.7818, Pure Ratio2 9.7182 +Epoch [7/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0101, Loss2: 0.0108, Pure Ratio1: 9.8255, Pure Ratio2 9.8000 +Epoch [7/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0092, Loss2: 0.0092, Pure Ratio1: 9.8515, Pure Ratio2 9.8485 +Epoch [7/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.0625, Loss1: 0.0105, Loss2: 0.0109, Pure Ratio1: 9.9065, Pure Ratio2 9.9013 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 67.9287 % Model2 67.1875 %, Pure Ratio 1 9.8531 %, Pure Ratio 2 9.8625 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0084, Loss2: 0.0086, Pure Ratio1: 9.8519, Pure Ratio2 9.9259 +Epoch [8/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0102, Loss2: 0.0101, Pure Ratio1: 9.5833, Pure Ratio2 9.6759 +Epoch [8/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0112, Loss2: 0.0106, Pure Ratio1: 9.6543, Pure Ratio2 9.7099 +Epoch [8/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0089, Loss2: 0.0084, Pure Ratio1: 9.5417, Pure Ratio2 9.5694 +Epoch [8/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0107, Loss2: 0.0108, Pure Ratio1: 9.7778, Pure Ratio2 9.7778 +Epoch [8/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0090, Loss2: 0.0096, Pure Ratio1: 9.8981, Pure Ratio2 9.8981 +Epoch [8/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0111, Loss2: 0.0106, Pure Ratio1: 9.8307, Pure Ratio2 9.8704 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 68.1991 % Model2 68.7400 %, Pure Ratio 1 9.8623 %, Pure Ratio 2 9.8908 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0088, Loss2: 0.0080, Pure Ratio1: 9.2762, Pure Ratio2 9.1810 +Epoch [9/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0095, Loss2: 0.0086, Pure Ratio1: 9.4571, Pure Ratio2 9.4000 +Epoch [9/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 62.5000, Loss1: 0.0071, Loss2: 0.0066, Pure Ratio1: 9.5556, Pure Ratio2 9.5111 +Epoch [9/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0097, Loss2: 0.0098, Pure Ratio1: 9.7048, Pure Ratio2 9.6667 +Epoch [9/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0099, Loss2: 0.0099, Pure Ratio1: 9.8705, Pure Ratio2 9.8286 +Epoch [9/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0092, Loss2: 0.0087, Pure Ratio1: 9.8762, Pure Ratio2 9.8444 +Epoch [9/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0102, Loss2: 0.0104, Pure Ratio1: 9.8585, Pure Ratio2 9.8503 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 69.0905 % Model2 71.1939 %, Pure Ratio 1 9.8926 %, Pure Ratio 2 9.8632 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0085, Loss2: 0.0087, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [10/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0086, Loss2: 0.0087, Pure Ratio1: 10.1961, Pure Ratio2 10.2843 +Epoch [10/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0105, Loss2: 0.0103, Pure Ratio1: 9.9804, Pure Ratio2 9.9281 +Epoch [10/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0098, Loss2: 0.0094, Pure Ratio1: 10.0147, Pure Ratio2 9.9412 +Epoch [10/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0092, Loss2: 0.0091, Pure Ratio1: 10.0078, Pure Ratio2 9.9451 +Epoch [10/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0104, Loss2: 0.0099, Pure Ratio1: 9.9510, Pure Ratio2 9.8725 +Epoch [10/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0084, Loss2: 0.0086, Pure Ratio1: 9.9496, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 64.8938 % Model2 64.6835 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 61.7188, Loss1: 0.0076, Loss2: 0.0071, Pure Ratio1: 9.9412, Pure Ratio2 9.8235 +Epoch [11/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0091, Loss2: 0.0083, Pure Ratio1: 10.1176, Pure Ratio2 9.9804 +Epoch [11/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0082, Loss2: 0.0082, Pure Ratio1: 10.0458, Pure Ratio2 9.9739 +Epoch [11/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0074, Loss2: 0.0082, Pure Ratio1: 10.0049, Pure Ratio2 9.9461 +Epoch [11/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0078, Loss2: 0.0079, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Epoch [11/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0099, Loss2: 0.0103, Pure Ratio1: 9.9477, Pure Ratio2 9.8627 +Epoch [11/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0086, Loss2: 0.0087, Pure Ratio1: 9.9188, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 67.8185 % Model2 66.6066 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0083, Loss2: 0.0081, Pure Ratio1: 9.2353, Pure Ratio2 9.2745 +Epoch [12/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0076, Loss2: 0.0077, Pure Ratio1: 9.8922, Pure Ratio2 9.9118 +Epoch [12/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0066, Loss2: 0.0061, Pure Ratio1: 9.8105, Pure Ratio2 9.8562 +Epoch [12/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0064, Loss2: 0.0069, Pure Ratio1: 9.8529, Pure Ratio2 9.9118 +Epoch [12/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0078, Loss2: 0.0078, Pure Ratio1: 9.8118, Pure Ratio2 9.8627 +Epoch [12/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0099, Loss2: 0.0091, Pure Ratio1: 9.8464, Pure Ratio2 9.9052 +Epoch [12/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0055, Loss2: 0.0061, Pure Ratio1: 9.8711, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 67.4679 % Model2 67.2676 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0094, Loss2: 0.0087, Pure Ratio1: 9.6471, Pure Ratio2 9.6667 +Epoch [13/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0082, Loss2: 0.0080, Pure Ratio1: 10.0686, Pure Ratio2 10.0882 +Epoch [13/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0070, Loss2: 0.0069, Pure Ratio1: 9.7255, Pure Ratio2 9.7582 +Epoch [13/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0069, Loss2: 0.0067, Pure Ratio1: 9.7843, Pure Ratio2 9.7941 +Epoch [13/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0105, Loss2: 0.0105, Pure Ratio1: 9.8353, Pure Ratio2 9.8706 +Epoch [13/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0084, Loss2: 0.0080, Pure Ratio1: 9.9739, Pure Ratio2 10.0000 +Epoch [13/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0088, Loss2: 0.0089, Pure Ratio1: 9.9552, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 68.2993 % Model2 68.8201 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0069, Loss2: 0.0073, Pure Ratio1: 10.0784, Pure Ratio2 9.9412 +Epoch [14/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0106, Loss2: 0.0105, Pure Ratio1: 9.6373, Pure Ratio2 9.5294 +Epoch [14/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0083, Loss2: 0.0087, Pure Ratio1: 9.6797, Pure Ratio2 9.6471 +Epoch [14/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0092, Loss2: 0.0094, Pure Ratio1: 9.7402, Pure Ratio2 9.7157 +Epoch [14/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0070, Loss2: 0.0069, Pure Ratio1: 9.7725, Pure Ratio2 9.7333 +Epoch [14/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0086, Loss2: 0.0082, Pure Ratio1: 9.6928, Pure Ratio2 9.6503 +Epoch [14/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0074, Loss2: 0.0077, Pure Ratio1: 9.7927, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 70.7232 % Model2 70.3926 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0069, Loss2: 0.0072, Pure Ratio1: 9.2157, Pure Ratio2 9.2549 +Epoch [15/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0060, Loss2: 0.0060, Pure Ratio1: 9.4216, Pure Ratio2 9.4510 +Epoch [15/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0075, Loss2: 0.0088, Pure Ratio1: 9.7190, Pure Ratio2 9.7059 +Epoch [15/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0110, Loss2: 0.0108, Pure Ratio1: 9.9706, Pure Ratio2 9.9314 +Epoch [15/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0071, Loss2: 0.0078, Pure Ratio1: 9.9569, Pure Ratio2 9.9216 +Epoch [15/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0071, Loss2: 0.0068, Pure Ratio1: 10.0163, Pure Ratio2 9.9673 +Epoch [15/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0071, Loss2: 0.0071, Pure Ratio1: 10.0420, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 71.8550 % Model2 70.8133 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0080, Loss2: 0.0068, Pure Ratio1: 9.2941, Pure Ratio2 9.3137 +Epoch [16/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0078, Loss2: 0.0084, Pure Ratio1: 9.4118, Pure Ratio2 9.4510 +Epoch [16/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0090, Loss2: 0.0107, Pure Ratio1: 9.7451, Pure Ratio2 9.7778 +Epoch [16/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0062, Loss2: 0.0062, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [16/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0072, Loss2: 0.0070, Pure Ratio1: 9.9843, Pure Ratio2 9.9412 +Epoch [16/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0095, Loss2: 0.0094, Pure Ratio1: 9.9575, Pure Ratio2 9.8987 +Epoch [16/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0073, Loss2: 0.0064, Pure Ratio1: 9.8824, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 68.1591 % Model2 67.1374 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0066, Loss2: 0.0058, Pure Ratio1: 9.4902, Pure Ratio2 9.6471 +Epoch [17/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0052, Loss2: 0.0048, Pure Ratio1: 9.6765, Pure Ratio2 9.7549 +Epoch [17/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0046, Loss2: 0.0050, Pure Ratio1: 9.8627, Pure Ratio2 9.8693 +Epoch [17/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0079, Loss2: 0.0080, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [17/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0046, Loss2: 0.0043, Pure Ratio1: 9.9020, Pure Ratio2 9.8784 +Epoch [17/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.8125, Loss1: 0.0062, Loss2: 0.0074, Pure Ratio1: 9.9183, Pure Ratio2 9.8791 +Epoch [17/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0080, Loss2: 0.0073, Pure Ratio1: 9.8655, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 65.2444 % Model2 64.9239 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0045, Loss2: 0.0043, Pure Ratio1: 10.2157, Pure Ratio2 10.0588 +Epoch [18/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0065, Loss2: 0.0058, Pure Ratio1: 10.0294, Pure Ratio2 9.9412 +Epoch [18/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0078, Loss2: 0.0072, Pure Ratio1: 10.1438, Pure Ratio2 10.0458 +Epoch [18/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 63.2812, Loss1: 0.0044, Loss2: 0.0047, Pure Ratio1: 9.8922, Pure Ratio2 9.8333 +Epoch [18/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0080, Loss2: 0.0071, Pure Ratio1: 9.8902, Pure Ratio2 9.8549 +Epoch [18/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0064, Loss2: 0.0047, Pure Ratio1: 9.8529, Pure Ratio2 9.8333 +Epoch [18/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0057, Loss2: 0.0055, Pure Ratio1: 9.9272, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 65.9355 % Model2 67.6783 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0038, Loss2: 0.0044, Pure Ratio1: 9.4902, Pure Ratio2 9.5490 +Epoch [19/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0072, Loss2: 0.0078, Pure Ratio1: 9.4118, Pure Ratio2 9.4118 +Epoch [19/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.9688, Loss1: 0.0059, Loss2: 0.0048, Pure Ratio1: 9.4575, Pure Ratio2 9.4248 +Epoch [19/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0062, Loss2: 0.0055, Pure Ratio1: 9.6422, Pure Ratio2 9.5539 +Epoch [19/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0075, Loss2: 0.0075, Pure Ratio1: 9.6863, Pure Ratio2 9.5961 +Epoch [19/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 49.2188, Loss1: 0.0081, Loss2: 0.0100, Pure Ratio1: 9.7516, Pure Ratio2 9.6503 +Epoch [19/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0051, Loss2: 0.0048, Pure Ratio1: 9.9328, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 66.8269 % Model2 65.2945 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0037, Loss2: 0.0032, Pure Ratio1: 10.3725, Pure Ratio2 10.3137 +Epoch [20/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0040, Loss2: 0.0033, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [20/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0072, Loss2: 0.0053, Pure Ratio1: 10.0915, Pure Ratio2 10.0458 +Epoch [20/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0046, Loss2: 0.0048, Pure Ratio1: 10.0343, Pure Ratio2 9.9363 +Epoch [20/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0061, Loss2: 0.0061, Pure Ratio1: 9.9412, Pure Ratio2 9.8471 +Epoch [20/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0055, Loss2: 0.0062, Pure Ratio1: 9.9771, Pure Ratio2 9.9183 +Epoch [20/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0048, Loss2: 0.0058, Pure Ratio1: 9.8936, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 65.4848 % Model2 66.7668 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 66.4062, Loss1: 0.0029, Loss2: 0.0035, Pure Ratio1: 10.5294, Pure Ratio2 10.5098 +Epoch [21/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0034, Loss2: 0.0031, Pure Ratio1: 9.8627, Pure Ratio2 9.8922 +Epoch [21/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0057, Loss2: 0.0045, Pure Ratio1: 9.8693, Pure Ratio2 9.9281 +Epoch [21/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0036, Loss2: 0.0029, Pure Ratio1: 9.9951, Pure Ratio2 10.0735 +Epoch [21/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0056, Loss2: 0.0058, Pure Ratio1: 9.9922, Pure Ratio2 10.0549 +Epoch [21/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0058, Loss2: 0.0057, Pure Ratio1: 9.8758, Pure Ratio2 9.9575 +Epoch [21/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0055, Loss2: 0.0049, Pure Ratio1: 9.8235, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 63.5517 % Model2 64.2428 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0057, Loss2: 0.0047, Pure Ratio1: 9.1765, Pure Ratio2 9.1569 +Epoch [22/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0029, Loss2: 0.0029, Pure Ratio1: 9.6863, Pure Ratio2 9.6471 +Epoch [22/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.0036, Loss2: 0.0032, Pure Ratio1: 9.7974, Pure Ratio2 9.7386 +Epoch [22/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0043, Loss2: 0.0043, Pure Ratio1: 9.7255, Pure Ratio2 9.6961 +Epoch [22/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0067, Loss2: 0.0063, Pure Ratio1: 9.9137, Pure Ratio2 9.9098 +Epoch [22/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0040, Loss2: 0.0035, Pure Ratio1: 9.8333, Pure Ratio2 9.8072 +Epoch [22/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.0030, Loss2: 0.0033, Pure Ratio1: 9.8908, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 62.7704 % Model2 64.0024 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.9062, Loss1: 0.0031, Loss2: 0.0028, Pure Ratio1: 9.6275, Pure Ratio2 9.5098 +Epoch [23/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0028, Loss2: 0.0028, Pure Ratio1: 9.5882, Pure Ratio2 9.5490 +Epoch [23/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0048, Loss2: 0.0050, Pure Ratio1: 9.8170, Pure Ratio2 9.8170 +Epoch [23/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0044, Loss2: 0.0045, Pure Ratio1: 9.8627, Pure Ratio2 9.8235 +Epoch [23/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.0034, Loss2: 0.0023, Pure Ratio1: 9.8392, Pure Ratio2 9.8353 +Epoch [23/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 9.9837, Pure Ratio2 9.9444 +Epoch [23/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0033, Loss2: 0.0032, Pure Ratio1: 9.9356, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 64.0825 % Model2 64.1526 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0030, Loss2: 0.0029, Pure Ratio1: 10.1373, Pure Ratio2 10.0980 +Epoch [24/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0024, Loss2: 0.0030, Pure Ratio1: 9.9608, Pure Ratio2 9.9314 +Epoch [24/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0037, Loss2: 0.0040, Pure Ratio1: 9.9477, Pure Ratio2 9.9935 +Epoch [24/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0036, Loss2: 0.0043, Pure Ratio1: 9.8382, Pure Ratio2 9.9167 +Epoch [24/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 9.8431, Pure Ratio2 9.9255 +Epoch [24/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0035, Loss2: 0.0030, Pure Ratio1: 9.7549, Pure Ratio2 9.8333 +Epoch [24/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.9062, Loss1: 0.0027, Loss2: 0.0017, Pure Ratio1: 9.8011, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 61.9792 % Model2 62.7103 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0030, Loss2: 0.0021, Pure Ratio1: 10.6275, Pure Ratio2 10.5098 +Epoch [25/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0017, Pure Ratio1: 10.2549, Pure Ratio2 10.1667 +Epoch [25/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0023, Loss2: 0.0028, Pure Ratio1: 10.2353, Pure Ratio2 10.1634 +Epoch [25/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0029, Loss2: 0.0036, Pure Ratio1: 10.0294, Pure Ratio2 9.9608 +Epoch [25/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0044, Loss2: 0.0030, Pure Ratio1: 10.0118, Pure Ratio2 9.9765 +Epoch [25/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0042, Loss2: 0.0045, Pure Ratio1: 9.8693, Pure Ratio2 9.8301 +Epoch [25/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0024, Loss2: 0.0028, Pure Ratio1: 9.9104, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 61.3882 % Model2 63.7921 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0020, Loss2: 0.0022, Pure Ratio1: 10.3529, Pure Ratio2 10.1765 +Epoch [26/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 10.0490, Pure Ratio2 9.9412 +Epoch [26/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0034, Loss2: 0.0038, Pure Ratio1: 9.9673, Pure Ratio2 9.8954 +Epoch [26/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0023, Pure Ratio1: 10.1225, Pure Ratio2 10.0539 +Epoch [26/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0059, Loss2: 0.0052, Pure Ratio1: 9.9490, Pure Ratio2 9.9569 +Epoch [26/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0028, Loss2: 0.0026, Pure Ratio1: 9.9935, Pure Ratio2 10.0065 +Epoch [26/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0023, Loss2: 0.0027, Pure Ratio1: 9.9272, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 64.0525 % Model2 64.8137 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0014, Loss2: 0.0023, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [27/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0020, Loss2: 0.0022, Pure Ratio1: 9.8039, Pure Ratio2 9.8431 +Epoch [27/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0019, Loss2: 0.0014, Pure Ratio1: 9.7778, Pure Ratio2 9.7778 +Epoch [27/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0026, Pure Ratio1: 9.7843, Pure Ratio2 9.8039 +Epoch [27/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0026, Pure Ratio1: 9.8039, Pure Ratio2 9.7961 +Epoch [27/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0038, Loss2: 0.0042, Pure Ratio1: 9.8366, Pure Ratio2 9.8758 +Epoch [27/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0019, Pure Ratio1: 9.9608, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 61.5485 % Model2 61.6186 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0022, Loss2: 0.0015, Pure Ratio1: 9.1765, Pure Ratio2 9.1176 +Epoch [28/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0017, Loss2: 0.0023, Pure Ratio1: 9.6078, Pure Ratio2 9.6471 +Epoch [28/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0012, Pure Ratio1: 9.8627, Pure Ratio2 9.9216 +Epoch [28/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 81.2500, Loss1: 0.0029, Loss2: 0.0013, Pure Ratio1: 9.8186, Pure Ratio2 9.8922 +Epoch [28/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0019, Loss2: 0.0022, Pure Ratio1: 9.7725, Pure Ratio2 9.8118 +Epoch [28/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0020, Loss2: 0.0019, Pure Ratio1: 9.8431, Pure Ratio2 9.8464 +Epoch [28/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.0025, Loss2: 0.0023, Pure Ratio1: 9.8880, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 62.8706 % Model2 61.9591 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [29/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [29/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 10.2745, Pure Ratio2 10.2941 +Epoch [29/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0029, Loss2: 0.0022, Pure Ratio1: 10.0392, Pure Ratio2 10.0539 +Epoch [29/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.7812, Loss1: 0.0036, Loss2: 0.0025, Pure Ratio1: 9.8745, Pure Ratio2 9.8824 +Epoch [29/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0032, Loss2: 0.0026, Pure Ratio1: 9.8824, Pure Ratio2 9.8987 +Epoch [29/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0021, Loss2: 0.0024, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 61.1078 % Model2 59.4852 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0031, Loss2: 0.0020, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [30/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8235, Pure Ratio2 9.8922 +Epoch [30/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0014, Loss2: 0.0010, Pure Ratio1: 9.7974, Pure Ratio2 9.8431 +Epoch [30/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.7500, Pure Ratio2 9.7941 +Epoch [30/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.0023, Loss2: 0.0020, Pure Ratio1: 9.7529, Pure Ratio2 9.8157 +Epoch [30/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.8105, Pure Ratio2 9.8595 +Epoch [30/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0026, Pure Ratio1: 9.8347, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 59.2849 % Model2 60.3365 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.0000, Pure Ratio2 9.8431 +Epoch [31/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.0012, Loss2: 0.0019, Pure Ratio1: 9.9706, Pure Ratio2 9.7843 +Epoch [31/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0017, Loss2: 0.0014, Pure Ratio1: 9.9020, Pure Ratio2 9.7908 +Epoch [31/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0012, Loss2: 0.0015, Pure Ratio1: 10.0735, Pure Ratio2 9.9804 +Epoch [31/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0016, Loss2: 0.0010, Pure Ratio1: 10.2235, Pure Ratio2 10.1137 +Epoch [31/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 10.0261, Pure Ratio2 9.9837 +Epoch [31/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 79.6875, Loss1: 0.0027, Loss2: 0.0014, Pure Ratio1: 9.9412, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 61.5585 % Model2 60.5970 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0014, Loss2: 0.0010, Pure Ratio1: 10.0784, Pure Ratio2 9.8039 +Epoch [32/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0020, Loss2: 0.0020, Pure Ratio1: 9.8235, Pure Ratio2 9.6471 +Epoch [32/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.0012, Loss2: 0.0021, Pure Ratio1: 9.8301, Pure Ratio2 9.7320 +Epoch [32/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.9265, Pure Ratio2 9.8627 +Epoch [32/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 9.9412, Pure Ratio2 9.8980 +Epoch [32/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 77.3438, Loss1: 0.0042, Loss2: 0.0039, Pure Ratio1: 9.8660, Pure Ratio2 9.8758 +Epoch [32/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0020, Loss2: 0.0035, Pure Ratio1: 9.8964, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 60.4367 % Model2 60.8674 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0016, Loss2: 0.0016, Pure Ratio1: 9.5686, Pure Ratio2 9.6471 +Epoch [33/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 9.6373, Pure Ratio2 9.6471 +Epoch [33/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.6275, Pure Ratio2 9.6340 +Epoch [33/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [33/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.6667, Pure Ratio2 9.6824 +Epoch [33/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0018, Loss2: 0.0014, Pure Ratio1: 9.7026, Pure Ratio2 9.7059 +Epoch [33/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0017, Pure Ratio1: 9.7731, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 61.1979 % Model2 63.0509 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0028, Loss2: 0.0034, Pure Ratio1: 10.0980, Pure Ratio2 10.0588 +Epoch [34/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.9510, Pure Ratio2 9.9804 +Epoch [34/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0025, Loss2: 0.0016, Pure Ratio1: 10.0523, Pure Ratio2 10.1503 +Epoch [34/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.8775, Pure Ratio2 10.0000 +Epoch [34/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.8118, Pure Ratio2 9.9490 +Epoch [34/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 74.2188, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.8268, Pure Ratio2 9.9575 +Epoch [34/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8319, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 63.1410 % Model2 60.3966 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.6667, Pure Ratio2 9.5098 +Epoch [35/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 10.0000, Pure Ratio2 9.9118 +Epoch [35/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 10.0719, Pure Ratio2 10.0654 +Epoch [35/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 9.8529 +Epoch [35/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0018, Pure Ratio1: 9.7922, Pure Ratio2 9.7451 +Epoch [35/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 9.7941, Pure Ratio2 9.7680 +Epoch [35/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.8880, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 62.8706 % Model2 63.4115 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 8.9020, Pure Ratio2 8.8235 +Epoch [36/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.2157, Pure Ratio2 9.0980 +Epoch [36/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.6928, Pure Ratio2 9.6536 +Epoch [36/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0018, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [36/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 9.8941, Pure Ratio2 9.8627 +Epoch [36/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 86.7188, Loss1: 0.0017, Loss2: 0.0005, Pure Ratio1: 9.9150, Pure Ratio2 9.9052 +Epoch [36/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0018, Loss2: 0.0014, Pure Ratio1: 9.8543, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 61.2380 % Model2 63.0108 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.6275, Pure Ratio2 10.6078 +Epoch [37/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.2353, Pure Ratio2 10.1569 +Epoch [37/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8758, Pure Ratio2 9.8170 +Epoch [37/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.0686, Pure Ratio2 9.9412 +Epoch [37/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.8824, Pure Ratio2 9.8235 +Epoch [37/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.9118, Pure Ratio2 9.8660 +Epoch [37/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9048, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 60.4768 % Model2 61.7889 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.7059, Pure Ratio2 9.6471 +Epoch [38/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.8922, Pure Ratio2 9.7157 +Epoch [38/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.7908, Pure Ratio2 9.6797 +Epoch [38/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0013, Pure Ratio1: 9.7353, Pure Ratio2 9.6471 +Epoch [38/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.8471, Pure Ratio2 9.7608 +Epoch [38/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.7516, Pure Ratio2 9.6895 +Epoch [38/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.7479, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 61.0176 % Model2 63.3714 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.5490, Pure Ratio2 9.6667 +Epoch [39/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0588, Pure Ratio2 10.2451 +Epoch [39/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9150, Pure Ratio2 10.0261 +Epoch [39/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9510, Pure Ratio2 10.0637 +Epoch [39/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.8902, Pure Ratio2 10.0118 +Epoch [39/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.9346 +Epoch [39/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8543, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 60.3165 % Model2 60.4868 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 10.3922, Pure Ratio2 10.4706 +Epoch [40/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 10.2745, Pure Ratio2 10.1275 +Epoch [40/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0017, Pure Ratio1: 10.1046, Pure Ratio2 9.9412 +Epoch [40/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.9216, Pure Ratio2 9.7745 +Epoch [40/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.8314, Pure Ratio2 9.6784 +Epoch [40/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.8627, Pure Ratio2 9.7353 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8487, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 61.0276 % Model2 60.7472 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.7843, Pure Ratio2 9.7059 +Epoch [41/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.7941 +Epoch [41/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8105, Pure Ratio2 9.8235 +Epoch [41/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0016, Pure Ratio1: 9.8676, Pure Ratio2 9.8480 +Epoch [41/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8863, Pure Ratio2 9.8627 +Epoch [41/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8562, Pure Ratio2 9.8399 +Epoch [41/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.8992, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 61.7889 % Model2 59.6154 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.4314, Pure Ratio2 9.6471 +Epoch [42/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.6863, Pure Ratio2 9.6863 +Epoch [42/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0017, Pure Ratio1: 9.8170, Pure Ratio2 9.8824 +Epoch [42/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7549, Pure Ratio2 9.8627 +Epoch [42/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [42/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0014, Pure Ratio1: 9.8235, Pure Ratio2 9.8497 +Epoch [42/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8487, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 61.3281 % Model2 60.0260 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 10.0588 +Epoch [43/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0098, Pure Ratio2 10.1275 +Epoch [43/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8562, Pure Ratio2 9.9085 +Epoch [43/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9265, Pure Ratio2 10.0000 +Epoch [43/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0510, Pure Ratio2 10.1412 +Epoch [43/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.8627, Pure Ratio2 9.9346 +Epoch [43/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8796, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 58.0829 % Model2 59.7155 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.3922, Pure Ratio2 9.3529 +Epoch [44/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.6863 +Epoch [44/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.6732, Pure Ratio2 9.7320 +Epoch [44/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7892, Pure Ratio2 9.8039 +Epoch [44/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8902, Pure Ratio2 9.8980 +Epoch [44/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0008, Pure Ratio1: 9.9542, Pure Ratio2 9.9706 +Epoch [44/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9244, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 60.1162 % Model2 59.6855 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.4902 +Epoch [45/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 9.9020, Pure Ratio2 9.7647 +Epoch [45/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9346, Pure Ratio2 9.9150 +Epoch [45/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.7157, Pure Ratio2 9.7059 +Epoch [45/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.8667, Pure Ratio2 9.8510 +Epoch [45/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [45/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8599, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 59.8658 % Model2 61.1278 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9412, Pure Ratio2 9.8235 +Epoch [46/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8039 +Epoch [46/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.8562, Pure Ratio2 9.9281 +Epoch [46/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.8676, Pure Ratio2 9.9265 +Epoch [46/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 10.0078 +Epoch [46/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0020, Pure Ratio1: 9.8856, Pure Ratio2 9.9150 +Epoch [46/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8571, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 59.9459 % Model2 61.2981 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5098, Pure Ratio2 9.4902 +Epoch [47/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.7843 +Epoch [47/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9542, Pure Ratio2 9.9281 +Epoch [47/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7794, Pure Ratio2 9.7696 +Epoch [47/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.8863, Pure Ratio2 9.8824 +Epoch [47/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.8987, Pure Ratio2 9.9150 +Epoch [47/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 9.8852, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 60.7472 % Model2 62.5901 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.4118, Pure Ratio2 9.5686 +Epoch [48/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6275, Pure Ratio2 9.7059 +Epoch [48/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8562 +Epoch [48/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.8186, Pure Ratio2 9.8382 +Epoch [48/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.7804, Pure Ratio2 9.7569 +Epoch [48/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8007, Pure Ratio2 9.7712 +Epoch [48/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8515, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 61.3782 % Model2 60.7873 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.9804 +Epoch [49/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.1961, Pure Ratio2 10.2549 +Epoch [49/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.0019, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 10.1242 +Epoch [49/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0882, Pure Ratio2 10.2010 +Epoch [49/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0471, Pure Ratio2 10.1882 +Epoch [49/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.9150, Pure Ratio2 10.0131 +Epoch [49/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9132, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 61.2580 % Model2 59.9259 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 9.8627 +Epoch [50/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [50/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9869, Pure Ratio2 10.0458 +Epoch [50/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9167, Pure Ratio2 9.9804 +Epoch [50/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7922, Pure Ratio2 9.8510 +Epoch [50/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7680, Pure Ratio2 9.7974 +Epoch [50/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.8347, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 61.1979 % Model2 60.6370 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.2941 +Epoch [51/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0098 +Epoch [51/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.8758, Pure Ratio2 9.9281 +Epoch [51/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0020, Loss2: 0.0004, Pure Ratio1: 9.9755, Pure Ratio2 9.9804 +Epoch [51/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.9529, Pure Ratio2 9.9294 +Epoch [51/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9575, Pure Ratio2 9.9412 +Epoch [51/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9132, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 62.5200 % Model2 61.0677 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [52/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9902, Pure Ratio2 9.9510 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.8497 +Epoch [52/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9069, Pure Ratio2 9.8725 +Epoch [52/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9529, Pure Ratio2 9.9608 +Epoch [52/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9052, Pure Ratio2 9.9118 +Epoch [52/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9636, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 61.2780 % Model2 60.4167 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.8824, Pure Ratio2 10.1765 +Epoch [53/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.7941, Pure Ratio2 9.8824 +Epoch [53/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.7124, Pure Ratio2 9.7712 +Epoch [53/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.5882, Pure Ratio2 9.5882 +Epoch [53/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0013, Pure Ratio1: 9.6118, Pure Ratio2 9.5961 +Epoch [53/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7157, Pure Ratio2 9.6601 +Epoch [53/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.8011, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 59.8157 % Model2 60.5068 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0980 +Epoch [54/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9804, Pure Ratio2 9.8529 +Epoch [54/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 9.9935, Pure Ratio2 9.8366 +Epoch [54/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0016, Pure Ratio1: 9.9657, Pure Ratio2 9.8382 +Epoch [54/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.8980, Pure Ratio2 9.8471 +Epoch [54/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8072 +Epoch [54/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 60.6070 % Model2 62.5300 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.5686, Pure Ratio2 9.6667 +Epoch [55/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 9.5490, Pure Ratio2 9.6471 +Epoch [55/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5686, Pure Ratio2 9.6536 +Epoch [55/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8186, Pure Ratio2 9.9216 +Epoch [55/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.9098, Pure Ratio2 10.0000 +Epoch [55/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8791 +Epoch [55/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9244, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 59.2949 % Model2 61.7087 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7843, Pure Ratio2 9.6078 +Epoch [56/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.9314, Pure Ratio2 9.7647 +Epoch [56/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8758, Pure Ratio2 9.9020 +Epoch [56/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.9167 +Epoch [56/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.8706, Pure Ratio2 9.9216 +Epoch [56/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9510, Pure Ratio2 9.9967 +Epoch [56/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9524, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 62.0793 % Model2 60.1462 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.1373, Pure Ratio2 9.5490 +Epoch [57/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.6275, Pure Ratio2 9.9216 +Epoch [57/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6340, Pure Ratio2 9.8235 +Epoch [57/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0026, Pure Ratio1: 9.6471, Pure Ratio2 9.7843 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8353, Pure Ratio2 9.9373 +Epoch [57/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.7124, Pure Ratio2 9.8007 +Epoch [57/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 60.0661 % Model2 61.3381 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.1176, Pure Ratio2 10.0196 +Epoch [58/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.0882, Pure Ratio2 9.9706 +Epoch [58/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 9.9673 +Epoch [58/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 9.8824 +Epoch [58/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8863 +Epoch [58/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9281, Pure Ratio2 9.8627 +Epoch [58/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9076, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 60.8974 % Model2 61.0577 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.3725, Pure Ratio2 9.6863 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0882, Pure Ratio2 10.1961 +Epoch [59/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0065, Pure Ratio2 10.1111 +Epoch [59/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8480, Pure Ratio2 9.9265 +Epoch [59/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7608, Pure Ratio2 9.8392 +Epoch [59/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7614, Pure Ratio2 9.8660 +Epoch [59/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 62.1394 % Model2 61.6386 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.3725, Pure Ratio2 9.4510 +Epoch [60/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.4804, Pure Ratio2 9.7353 +Epoch [60/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7124, Pure Ratio2 9.9020 +Epoch [60/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.9314 +Epoch [60/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.9843 +Epoch [60/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8791, Pure Ratio2 9.9804 +Epoch [60/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8796, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 61.6687 % Model2 60.0361 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.9216 +Epoch [61/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 10.0882 +Epoch [61/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 10.1634, Pure Ratio2 10.1634 +Epoch [61/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [61/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8353, Pure Ratio2 9.8275 +Epoch [61/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8072, Pure Ratio2 9.7876 +Epoch [61/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0015, Loss2: 0.0006, Pure Ratio1: 9.8852, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 60.5869 % Model2 60.9776 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.6667, Pure Ratio2 9.7647 +Epoch [62/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5196, Pure Ratio2 9.5980 +Epoch [62/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7974, Pure Ratio2 9.9346 +Epoch [62/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0011, Pure Ratio1: 9.8725, Pure Ratio2 10.0098 +Epoch [62/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9176, Pure Ratio2 10.0549 +Epoch [62/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9085, Pure Ratio2 10.0490 +Epoch [62/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8515, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 61.4583 % Model2 61.1879 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.5490, Pure Ratio2 10.6078 +Epoch [63/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1176, Pure Ratio2 10.1667 +Epoch [63/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 9.9477 +Epoch [63/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.0735, Pure Ratio2 10.0735 +Epoch [63/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 10.0863, Pure Ratio2 10.0667 +Epoch [63/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0016, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [63/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9776, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 59.9159 % Model2 61.4083 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1961, Pure Ratio2 10.0784 +Epoch [64/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.6961 +Epoch [64/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.6536 +Epoch [64/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7990, Pure Ratio2 9.7647 +Epoch [64/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8314 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8856, Pure Ratio2 9.8856 +Epoch [64/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8683, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 60.9075 % Model2 62.2396 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7059, Pure Ratio2 9.7059 +Epoch [65/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7745, Pure Ratio2 9.7745 +Epoch [65/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8235, Pure Ratio2 9.7778 +Epoch [65/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.8529 +Epoch [65/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.9882, Pure Ratio2 9.9490 +Epoch [65/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9379, Pure Ratio2 9.9673 +Epoch [65/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9160, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 60.8974 % Model2 59.2748 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.5686 +Epoch [66/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.1471 +Epoch [66/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.2614 +Epoch [66/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9265, Pure Ratio2 9.9559 +Epoch [66/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0196, Pure Ratio2 10.0275 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.9771, Pure Ratio2 9.9935 +Epoch [66/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9692, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 60.6070 % Model2 60.5970 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.7647 +Epoch [67/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9706, Pure Ratio2 9.9804 +Epoch [67/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7386 +Epoch [67/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9363 +Epoch [67/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7882, Pure Ratio2 9.7882 +Epoch [67/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.7908, Pure Ratio2 9.7974 +Epoch [67/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8263, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 61.9591 % Model2 60.4768 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.1373, Pure Ratio2 9.1569 +Epoch [68/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.3333, Pure Ratio2 9.3333 +Epoch [68/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5425, Pure Ratio2 9.5033 +Epoch [68/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.6176 +Epoch [68/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8118, Pure Ratio2 9.7529 +Epoch [68/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8497, Pure Ratio2 9.7712 +Epoch [68/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.8291, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 59.1046 % Model2 60.5869 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3529, Pure Ratio2 10.1176 +Epoch [69/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.2941, Pure Ratio2 10.1471 +Epoch [69/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.1830, Pure Ratio2 10.0980 +Epoch [69/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0539, Pure Ratio2 10.0147 +Epoch [69/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9765, Pure Ratio2 9.9176 +Epoch [69/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8105, Pure Ratio2 9.7680 +Epoch [69/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.8908, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 60.0060 % Model2 59.8458 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.6275 +Epoch [70/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7353 +Epoch [70/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.5752, Pure Ratio2 9.5948 +Epoch [70/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 9.7696 +Epoch [70/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7922, Pure Ratio2 9.8118 +Epoch [70/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8333, Pure Ratio2 9.8235 +Epoch [70/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9524, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 61.9291 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.1176 +Epoch [71/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.3529, Pure Ratio2 10.2941 +Epoch [71/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0000 +Epoch [71/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 10.0490, Pure Ratio2 9.9118 +Epoch [71/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8863, Pure Ratio2 9.8000 +Epoch [71/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8399, Pure Ratio2 9.7549 +Epoch [71/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8291, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 60.8874 % Model2 61.0978 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.0196 +Epoch [72/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.1765 +Epoch [72/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.0719, Pure Ratio2 10.0850 +Epoch [72/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 9.9020 +Epoch [72/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8980, Pure Ratio2 9.8824 +Epoch [72/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8954, Pure Ratio2 9.8725 +Epoch [72/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8796, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 62.1094 % Model2 60.0761 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9804 +Epoch [73/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 9.9216 +Epoch [73/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.0719, Pure Ratio2 9.9477 +Epoch [73/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 9.9314 +Epoch [73/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0667, Pure Ratio2 9.9843 +Epoch [73/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 9.9575 +Epoch [73/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0532, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 61.6787 % Model2 59.4852 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.7451, Pure Ratio2 10.4314 +Epoch [74/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.3137, Pure Ratio2 10.2549 +Epoch [74/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0013, Loss2: 0.0003, Pure Ratio1: 10.2222, Pure Ratio2 10.1307 +Epoch [74/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.1324 +Epoch [74/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0549, Pure Ratio2 9.9647 +Epoch [74/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 10.0033, Pure Ratio2 9.9510 +Epoch [74/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9076, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 60.4667 % Model2 61.2280 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 10.1373, Pure Ratio2 9.8431 +Epoch [75/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.2745, Pure Ratio2 10.0882 +Epoch [75/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.0588 +Epoch [75/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0013, Loss2: 0.0002, Pure Ratio1: 10.2451, Pure Ratio2 10.1373 +Epoch [75/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1647, Pure Ratio2 10.0157 +Epoch [75/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0752, Pure Ratio2 9.9771 +Epoch [75/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9524, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 61.7388 % Model2 58.3033 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8824, Pure Ratio2 10.1176 +Epoch [76/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9118, Pure Ratio2 10.0980 +Epoch [76/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8758, Pure Ratio2 10.0523 +Epoch [76/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 10.0441 +Epoch [76/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8196, Pure Ratio2 9.8902 +Epoch [76/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7484, Pure Ratio2 9.7484 +Epoch [76/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.7479, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 61.9191 % Model2 59.1146 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.7647 +Epoch [77/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.8333 +Epoch [77/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7386, Pure Ratio2 9.6667 +Epoch [77/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.7990, Pure Ratio2 9.7500 +Epoch [77/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0471, Pure Ratio2 9.9569 +Epoch [77/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0523, Pure Ratio2 9.9967 +Epoch [77/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0280, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 60.4868 % Model2 60.2163 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.2941, Pure Ratio2 9.2745 +Epoch [78/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.7255 +Epoch [78/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9935, Pure Ratio2 9.8824 +Epoch [78/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.7843, Pure Ratio2 9.7206 +Epoch [78/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7608, Pure Ratio2 9.7529 +Epoch [78/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7876, Pure Ratio2 9.7941 +Epoch [78/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8711, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 61.5986 % Model2 61.0877 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.8824 +Epoch [79/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.5588 +Epoch [79/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.6667, Pure Ratio2 9.6601 +Epoch [79/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6961, Pure Ratio2 9.6275 +Epoch [79/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7098, Pure Ratio2 9.6196 +Epoch [79/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.7778, Pure Ratio2 9.7190 +Epoch [79/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8151, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 59.4351 % Model2 58.7340 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.0784 +Epoch [80/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.6961 +Epoch [80/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.7386, Pure Ratio2 9.7712 +Epoch [80/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7108, Pure Ratio2 9.7696 +Epoch [80/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.8471, Pure Ratio2 9.8941 +Epoch [80/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9444, Pure Ratio2 9.9771 +Epoch [80/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9440, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 60.0160 % Model2 57.9828 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8824 +Epoch [81/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0980, Pure Ratio2 10.1569 +Epoch [81/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1111, Pure Ratio2 10.1373 +Epoch [81/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7892, Pure Ratio2 9.8431 +Epoch [81/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8588, Pure Ratio2 9.9020 +Epoch [81/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8137, Pure Ratio2 9.8627 +Epoch [81/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8375, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 60.0160 % Model2 61.2380 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8235 +Epoch [82/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.7647 +Epoch [82/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6732, Pure Ratio2 9.5882 +Epoch [82/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.7108, Pure Ratio2 9.6127 +Epoch [82/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8549, Pure Ratio2 9.7843 +Epoch [82/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.8333 +Epoch [82/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8796, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 60.8073 % Model2 59.4651 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.2157, Pure Ratio2 10.2157 +Epoch [83/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.2255 +Epoch [83/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9935, Pure Ratio2 10.0850 +Epoch [83/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9657, Pure Ratio2 9.9804 +Epoch [83/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9843, Pure Ratio2 9.9765 +Epoch [83/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.9052 +Epoch [83/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0028, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 59.9760 % Model2 60.5469 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.1961, Pure Ratio2 9.3137 +Epoch [84/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.7647 +Epoch [84/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0036, Loss2: 0.0038, Pure Ratio1: 9.6536, Pure Ratio2 9.7059 +Epoch [84/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [84/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7569, Pure Ratio2 9.7647 +Epoch [84/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7876, Pure Ratio2 9.7484 +Epoch [84/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 60.0361 % Model2 61.9391 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2941, Pure Ratio2 10.2353 +Epoch [85/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.9118 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.8497 +Epoch [85/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8775, Pure Ratio2 9.7696 +Epoch [85/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9647, Pure Ratio2 9.9098 +Epoch [85/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9118, Pure Ratio2 9.8464 +Epoch [85/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9300, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 60.1162 % Model2 61.3281 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.9608 +Epoch [86/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8725, Pure Ratio2 9.9216 +Epoch [86/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7386, Pure Ratio2 9.7712 +Epoch [86/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9265, Pure Ratio2 9.9461 +Epoch [86/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9176 +Epoch [86/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7582, Pure Ratio2 9.7582 +Epoch [86/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8263, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 61.3682 % Model2 60.0060 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.1373, Pure Ratio2 9.2353 +Epoch [87/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.7255 +Epoch [87/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.7647 +Epoch [87/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [87/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.7137 +Epoch [87/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8007, Pure Ratio2 9.7843 +Epoch [87/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8515, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 61.0076 % Model2 59.2147 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.4118, Pure Ratio2 10.3333 +Epoch [88/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.8333 +Epoch [88/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 9.9869 +Epoch [88/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0049, Pure Ratio2 9.9902 +Epoch [88/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9569 +Epoch [88/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.8954 +Epoch [88/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8739, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 60.3265 % Model2 60.1763 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.4118 +Epoch [89/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.6765 +Epoch [89/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.1895 +Epoch [89/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0003, Pure Ratio1: 10.0539, Pure Ratio2 10.0735 +Epoch [89/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0018, Pure Ratio1: 10.0157, Pure Ratio2 10.0196 +Epoch [89/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9542 +Epoch [89/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9272, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 59.7957 % Model2 60.6971 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.5882 +Epoch [90/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.2059, Pure Ratio2 10.3235 +Epoch [90/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.1895, Pure Ratio2 10.3072 +Epoch [90/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0539 +Epoch [90/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9961, Pure Ratio2 10.0471 +Epoch [90/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9575, Pure Ratio2 10.0098 +Epoch [90/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8179, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 60.8474 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [91/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8725 +Epoch [91/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8301, Pure Ratio2 9.8824 +Epoch [91/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.9020 +Epoch [91/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7765, Pure Ratio2 9.7804 +Epoch [91/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7908, Pure Ratio2 9.7974 +Epoch [91/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 59.3750 % Model2 59.9259 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.5294 +Epoch [92/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.7157 +Epoch [92/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 9.9281 +Epoch [92/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9020 +Epoch [92/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7569 +Epoch [92/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.8170 +Epoch [92/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 61.6687 % Model2 60.5569 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3529, Pure Ratio2 10.2353 +Epoch [93/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0012, Pure Ratio1: 9.8627, Pure Ratio2 9.7059 +Epoch [93/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.8497 +Epoch [93/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.8186 +Epoch [93/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7961, Pure Ratio2 9.8510 +Epoch [93/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7386, Pure Ratio2 9.7974 +Epoch [93/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8291, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 60.5869 % Model2 59.5553 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.1765, Pure Ratio2 9.1765 +Epoch [94/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2647, Pure Ratio2 9.3137 +Epoch [94/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.4706, Pure Ratio2 9.3791 +Epoch [94/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7745, Pure Ratio2 9.7598 +Epoch [94/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.8078, Pure Ratio2 9.7882 +Epoch [94/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.7451 +Epoch [94/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 60.0461 % Model2 59.3550 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.7647 +Epoch [95/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.4706, Pure Ratio2 10.4020 +Epoch [95/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1830, Pure Ratio2 10.1307 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.1029 +Epoch [95/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0353, Pure Ratio2 9.9373 +Epoch [95/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0229, Pure Ratio2 9.8725 +Epoch [95/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0280, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 61.6286 % Model2 59.2448 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.5294 +Epoch [96/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8529 +Epoch [96/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8497 +Epoch [96/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9069 +Epoch [96/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [96/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0359, Pure Ratio2 9.9575 +Epoch [96/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0010, Pure Ratio1: 9.9972, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 59.7957 % Model2 61.2079 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 9.8824 +Epoch [97/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 9.7255 +Epoch [97/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.7843 +Epoch [97/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8186 +Epoch [97/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7686, Pure Ratio2 9.7569 +Epoch [97/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.8791 +Epoch [97/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8739, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 60.8474 % Model2 60.7672 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.4902, Pure Ratio2 9.2549 +Epoch [98/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.8725 +Epoch [98/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8562, Pure Ratio2 9.7582 +Epoch [98/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.7549 +Epoch [98/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7176, Pure Ratio2 9.7294 +Epoch [98/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8791 +Epoch [98/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8599, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 61.0777 % Model2 60.7071 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.2745 +Epoch [99/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0686, Pure Ratio2 10.0294 +Epoch [99/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9804 +Epoch [99/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8922, Pure Ratio2 9.9069 +Epoch [99/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9451, Pure Ratio2 10.0000 +Epoch [99/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.9020, Pure Ratio2 9.9706 +Epoch [99/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8123, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 61.3181 % Model2 62.0192 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.6863 +Epoch [100/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 9.9608 +Epoch [100/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8889, Pure Ratio2 9.8954 +Epoch [100/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6127, Pure Ratio2 9.7059 +Epoch [100/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7176, Pure Ratio2 9.8000 +Epoch [100/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.8235 +Epoch [100/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8459, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 60.2163 % Model2 60.2965 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.8824 +Epoch [101/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.7451 +Epoch [101/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.8431 +Epoch [101/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0735, Pure Ratio2 9.9951 +Epoch [101/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8745 +Epoch [101/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9673, Pure Ratio2 9.8725 +Epoch [101/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9272, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 60.5168 % Model2 59.8458 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 9.8627 +Epoch [102/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [102/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9150, Pure Ratio2 9.8693 +Epoch [102/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9461, Pure Ratio2 9.9020 +Epoch [102/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8941, Pure Ratio2 9.8314 +Epoch [102/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.7908 +Epoch [102/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9832, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 61.5986 % Model2 60.5669 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.3725 +Epoch [103/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.6569 +Epoch [103/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 9.8235 +Epoch [103/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 10.1814, Pure Ratio2 9.9755 +Epoch [103/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 9.8706 +Epoch [103/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0033, Pure Ratio2 9.8399 +Epoch [103/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0056, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 61.7188 % Model2 60.3766 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.1961 +Epoch [104/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.8431 +Epoch [104/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9281 +Epoch [104/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0017, Loss2: 0.0004, Pure Ratio1: 10.1029, Pure Ratio2 10.0539 +Epoch [104/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0667, Pure Ratio2 10.0196 +Epoch [104/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0000 +Epoch [104/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9188, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 60.5168 % Model2 59.1446 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0000 +Epoch [105/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8725, Pure Ratio2 9.9118 +Epoch [105/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8170 +Epoch [105/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6225, Pure Ratio2 9.6814 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8510, Pure Ratio2 9.8510 +Epoch [105/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8856, Pure Ratio2 9.8660 +Epoch [105/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 61.7588 % Model2 61.7288 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.7451 +Epoch [106/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7647 +Epoch [106/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7059 +Epoch [106/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.6863 +Epoch [106/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8275, Pure Ratio2 9.7843 +Epoch [106/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8660, Pure Ratio2 9.8268 +Epoch [106/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8543, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 60.2965 % Model2 60.2564 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.4510 +Epoch [107/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3235, Pure Ratio2 10.5392 +Epoch [107/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.2157 +Epoch [107/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.1275 +Epoch [107/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0745, Pure Ratio2 10.1020 +Epoch [107/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0163 +Epoch [107/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 61.3482 % Model2 62.7504 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.5490 +Epoch [108/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4608, Pure Ratio2 9.4412 +Epoch [108/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.3922, Pure Ratio2 9.3595 +Epoch [108/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4853, Pure Ratio2 9.4412 +Epoch [108/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6627, Pure Ratio2 9.6196 +Epoch [108/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.7876 +Epoch [108/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8403, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 60.7772 % Model2 60.9175 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.6667, Pure Ratio2 10.4706 +Epoch [109/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.0980 +Epoch [109/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2614, Pure Ratio2 10.2680 +Epoch [109/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2304, Pure Ratio2 10.1667 +Epoch [109/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0431, Pure Ratio2 10.0353 +Epoch [109/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0425 +Epoch [109/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9748, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 60.7672 % Model2 61.3682 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3137 +Epoch [110/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.3333, Pure Ratio2 10.4510 +Epoch [110/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.1438 +Epoch [110/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.1225 +Epoch [110/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0588 +Epoch [110/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 10.0065 +Epoch [110/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9552, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 59.2548 % Model2 61.1579 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7647 +Epoch [111/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 10.0196 +Epoch [111/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9869 +Epoch [111/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9804 +Epoch [111/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9373, Pure Ratio2 10.0157 +Epoch [111/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9837, Pure Ratio2 10.0425 +Epoch [111/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9776, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 61.2881 % Model2 61.3982 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1176, Pure Ratio2 9.9020 +Epoch [112/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8725 +Epoch [112/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 9.9869 +Epoch [112/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9510, Pure Ratio2 10.0049 +Epoch [112/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.9882 +Epoch [112/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8497, Pure Ratio2 9.9118 +Epoch [112/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8683, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 60.4667 % Model2 60.5569 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.4510, Pure Ratio2 10.2745 +Epoch [113/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.4020, Pure Ratio2 10.3235 +Epoch [113/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9346, Pure Ratio2 9.8693 +Epoch [113/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0539, Pure Ratio2 10.0343 +Epoch [113/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.8824 +Epoch [113/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9085, Pure Ratio2 9.8660 +Epoch [113/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 60.4868 % Model2 60.5369 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.6078 +Epoch [114/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 9.7843 +Epoch [114/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.8954 +Epoch [114/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0245, Pure Ratio2 9.9853 +Epoch [114/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1490, Pure Ratio2 10.0863 +Epoch [114/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 10.0098 +Epoch [114/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 60.5869 % Model2 61.2480 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8627 +Epoch [115/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8725 +Epoch [115/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.2222 +Epoch [115/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [115/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.9020 +Epoch [115/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8333 +Epoch [115/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 9.8655, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 61.2580 % Model2 60.6871 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 10.2941, Pure Ratio2 10.1373 +Epoch [116/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.2451 +Epoch [116/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.7974 +Epoch [116/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.8235 +Epoch [116/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.8235 +Epoch [116/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9052, Pure Ratio2 9.8660 +Epoch [116/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8796, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 60.8373 % Model2 60.0661 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1961 +Epoch [117/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0784 +Epoch [117/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 9.9739 +Epoch [117/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8333, Pure Ratio2 9.7696 +Epoch [117/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.7255 +Epoch [117/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7680 +Epoch [117/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8739, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 60.4267 % Model2 61.7688 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7451 +Epoch [118/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7745, Pure Ratio2 9.8922 +Epoch [118/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9608 +Epoch [118/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8775, Pure Ratio2 9.9118 +Epoch [118/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.9765 +Epoch [118/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 9.9281, Pure Ratio2 9.9477 +Epoch [118/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8992, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 61.0877 % Model2 59.9259 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.4314, Pure Ratio2 10.3725 +Epoch [119/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.4510, Pure Ratio2 10.4510 +Epoch [119/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1830 +Epoch [119/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.1520 +Epoch [119/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1333, Pure Ratio2 10.1333 +Epoch [119/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0327 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 60.6370 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [120/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7353 +Epoch [120/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9085 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9069 +Epoch [120/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9451, Pure Ratio2 9.8627 +Epoch [120/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.7680 +Epoch [120/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8852, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 60.3566 % Model2 59.4651 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.3333 +Epoch [121/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6765, Pure Ratio2 9.8725 +Epoch [121/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.4575, Pure Ratio2 9.6144 +Epoch [121/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8088 +Epoch [121/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6039, Pure Ratio2 9.7059 +Epoch [121/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.7974 +Epoch [121/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7675, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 60.7372 % Model2 60.6671 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [122/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0686 +Epoch [122/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 10.0588 +Epoch [122/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7598, Pure Ratio2 9.8971 +Epoch [122/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.8235 +Epoch [122/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.8170 +Epoch [122/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9468, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 60.4067 % Model2 60.6370 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [123/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9412 +Epoch [123/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.9477 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.0686 +Epoch [123/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8275, Pure Ratio2 9.8941 +Epoch [123/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8693 +Epoch [123/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8459, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 60.2464 % Model2 60.3365 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.8824 +Epoch [124/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [124/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1438 +Epoch [124/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 10.1422, Pure Ratio2 10.1569 +Epoch [124/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1647, Pure Ratio2 10.1373 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0556 +Epoch [124/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 59.5152 % Model2 61.0978 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [125/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8333 +Epoch [125/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.8431 +Epoch [125/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.7941 +Epoch [125/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Epoch [125/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8954 +Epoch [125/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 61.5184 % Model2 61.2380 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.4706 +Epoch [126/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3922, Pure Ratio2 9.2255 +Epoch [126/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4641, Pure Ratio2 9.3464 +Epoch [126/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7010, Pure Ratio2 9.5539 +Epoch [126/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7765, Pure Ratio2 9.7098 +Epoch [126/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7484 +Epoch [126/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8179, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 60.7372 % Model2 60.3666 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 9.7451 +Epoch [127/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 9.7647 +Epoch [127/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 9.9542 +Epoch [127/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8824 +Epoch [127/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8000, Pure Ratio2 9.7922 +Epoch [127/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7876 +Epoch [127/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 60.7672 % Model2 60.8474 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.5294 +Epoch [128/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8725 +Epoch [128/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8954 +Epoch [128/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0343 +Epoch [128/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0157, Pure Ratio2 9.9843 +Epoch [128/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0850, Pure Ratio2 10.0000 +Epoch [128/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.9720, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 60.2163 % Model2 59.5252 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.2745 +Epoch [129/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.9020 +Epoch [129/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.8889 +Epoch [129/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7402, Pure Ratio2 9.9069 +Epoch [129/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7216, Pure Ratio2 9.8588 +Epoch [129/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.8170 +Epoch [129/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8403, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 60.3265 % Model2 62.9107 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5294, Pure Ratio2 10.6863 +Epoch [130/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1863, Pure Ratio2 10.2451 +Epoch [130/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0654 +Epoch [130/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7010, Pure Ratio2 9.7451 +Epoch [130/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0050, Loss2: 0.0046, Pure Ratio1: 9.8941, Pure Ratio2 9.9451 +Epoch [130/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8856 +Epoch [130/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8711, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 61.1879 % Model2 61.3281 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6275 +Epoch [131/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.5196 +Epoch [131/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6797, Pure Ratio2 9.6993 +Epoch [131/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8578, Pure Ratio2 9.8039 +Epoch [131/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.8275, Pure Ratio2 9.7451 +Epoch [131/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7778 +Epoch [131/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7927, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 61.5685 % Model2 61.7788 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0784 +Epoch [132/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8235 +Epoch [132/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0016, Pure Ratio1: 9.8039, Pure Ratio2 9.6797 +Epoch [132/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.6324 +Epoch [132/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7020, Pure Ratio2 9.6275 +Epoch [132/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7222, Pure Ratio2 9.6536 +Epoch [132/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8123, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 61.1579 % Model2 61.4083 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9608 +Epoch [133/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.8824 +Epoch [133/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9216 +Epoch [133/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 9.9167 +Epoch [133/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0157 +Epoch [133/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8660, Pure Ratio2 9.7974 +Epoch [133/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 61.7288 % Model2 61.5785 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [134/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.3235, Pure Ratio2 10.4902 +Epoch [134/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0261 +Epoch [134/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7990, Pure Ratio2 9.8431 +Epoch [134/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.9216 +Epoch [134/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8693 +Epoch [134/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 62.2095 % Model2 60.6971 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0000 +Epoch [135/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.0588 +Epoch [135/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0588 +Epoch [135/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 9.9314 +Epoch [135/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8627 +Epoch [135/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 9.9085 +Epoch [135/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 94.5312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 62.1294 % Model2 61.4383 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.9804 +Epoch [136/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.9216 +Epoch [136/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4967, Pure Ratio2 9.6013 +Epoch [136/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5931, Pure Ratio2 9.6225 +Epoch [136/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6784, Pure Ratio2 9.6824 +Epoch [136/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7386 +Epoch [136/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8768, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 60.2564 % Model2 61.2780 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.2941 +Epoch [137/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.6471, Pure Ratio2 10.4804 +Epoch [137/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3464, Pure Ratio2 10.2418 +Epoch [137/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0588 +Epoch [137/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0824, Pure Ratio2 9.9451 +Epoch [137/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0014, Pure Ratio1: 10.0850, Pure Ratio2 9.9771 +Epoch [137/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9720, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 59.3249 % Model2 60.0561 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8235, Pure Ratio2 10.7843 +Epoch [138/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9804 +Epoch [138/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9020 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9951, Pure Ratio2 9.9902 +Epoch [138/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9529, Pure Ratio2 9.9490 +Epoch [138/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0011, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [138/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9132, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 61.4083 % Model2 60.2464 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [139/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7549 +Epoch [139/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7451 +Epoch [139/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7500 +Epoch [139/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.8667 +Epoch [139/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7745 +Epoch [139/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 60.0260 % Model2 60.8373 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8039 +Epoch [140/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9902 +Epoch [140/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.2222 +Epoch [140/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0392 +Epoch [140/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.1059 +Epoch [140/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9575, Pure Ratio2 9.9935 +Epoch [140/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9244, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 62.1595 % Model2 62.6903 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 9.9412 +Epoch [141/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.7451 +Epoch [141/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 9.8627 +Epoch [141/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7402 +Epoch [141/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.7176 +Epoch [141/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8072 +Epoch [141/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8936, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 61.8990 % Model2 61.9591 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.7255 +Epoch [142/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6961 +Epoch [142/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.9346 +Epoch [142/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9314 +Epoch [142/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6941, Pure Ratio2 9.8667 +Epoch [142/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9216 +Epoch [142/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 60.3766 % Model2 59.4351 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.6471 +Epoch [143/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5784, Pure Ratio2 9.8627 +Epoch [143/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.8562 +Epoch [143/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.9314 +Epoch [143/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9176, Pure Ratio2 10.1137 +Epoch [143/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7680, Pure Ratio2 9.9477 +Epoch [143/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6891, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 62.2196 % Model2 60.2965 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.3333 +Epoch [144/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9412 +Epoch [144/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7647 +Epoch [144/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9020 +Epoch [144/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.9059 +Epoch [144/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9346, Pure Ratio2 9.8791 +Epoch [144/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 62.0092 % Model2 62.1194 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.1961 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.3333 +Epoch [145/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.4575, Pure Ratio2 9.4575 +Epoch [145/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7010 +Epoch [145/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6902, Pure Ratio2 9.7490 +Epoch [145/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8039 +Epoch [145/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8123, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 62.0393 % Model2 62.6302 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0980 +Epoch [146/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1275 +Epoch [146/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9673 +Epoch [146/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.9461 +Epoch [146/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.0039 +Epoch [146/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.8889 +Epoch [146/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 60.9776 % Model2 61.8990 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0196 +Epoch [147/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 10.0784 +Epoch [147/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 10.0131 +Epoch [147/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.9755 +Epoch [147/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8078, Pure Ratio2 9.9176 +Epoch [147/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9641 +Epoch [147/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 61.7388 % Model2 60.2163 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.6471 +Epoch [148/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6275 +Epoch [148/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6209, Pure Ratio2 9.6340 +Epoch [148/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.7745 +Epoch [148/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8980 +Epoch [148/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8268 +Epoch [148/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8459, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 61.5585 % Model2 61.1679 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.1961, Pure Ratio2 9.1765 +Epoch [149/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.5686 +Epoch [149/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.6601 +Epoch [149/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.8775 +Epoch [149/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 10.0039 +Epoch [149/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9771 +Epoch [149/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8964, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 60.0962 % Model2 60.5869 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.4118 +Epoch [150/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3627 +Epoch [150/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 10.0458 +Epoch [150/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0735 +Epoch [150/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 10.0039 +Epoch [150/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9641 +Epoch [150/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 61.1378 % Model2 61.1579 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8824 +Epoch [151/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0014, Pure Ratio1: 9.5980, Pure Ratio2 9.6275 +Epoch [151/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9150 +Epoch [151/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8333 +Epoch [151/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7765 +Epoch [151/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8660 +Epoch [151/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 61.6887 % Model2 61.9291 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6078 +Epoch [152/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4510, Pure Ratio2 9.5098 +Epoch [152/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.5686 +Epoch [152/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5637, Pure Ratio2 9.5784 +Epoch [152/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6157, Pure Ratio2 9.6314 +Epoch [152/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6993 +Epoch [152/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 61.8690 % Model2 61.0276 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6667, Pure Ratio2 10.4314 +Epoch [153/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5784, Pure Ratio2 10.4314 +Epoch [153/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3660, Pure Ratio2 10.2549 +Epoch [153/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2598, Pure Ratio2 10.2255 +Epoch [153/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 9.9882 +Epoch [153/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 9.9542 +Epoch [153/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 60.6270 % Model2 62.6803 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.4118 +Epoch [154/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1863 +Epoch [154/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.2418 +Epoch [154/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0735 +Epoch [154/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 10.0275 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.8922 +Epoch [154/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8936, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 62.3798 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.6078 +Epoch [155/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7941 +Epoch [155/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.6863 +Epoch [155/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.7647 +Epoch [155/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 10.0039 +Epoch [155/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0686 +Epoch [155/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 60.5669 % Model2 61.9291 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.5686 +Epoch [156/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.7843 +Epoch [156/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Epoch [156/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9951, Pure Ratio2 10.0343 +Epoch [156/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.9804 +Epoch [156/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9281 +Epoch [156/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8515, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 60.5869 % Model2 61.3181 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.3137 +Epoch [157/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4020 +Epoch [157/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.6797 +Epoch [157/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 9.8775 +Epoch [157/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.8510 +Epoch [157/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7680 +Epoch [157/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 61.0577 % Model2 61.0477 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9020 +Epoch [158/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0490 +Epoch [158/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7712 +Epoch [158/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7647 +Epoch [158/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [158/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7288, Pure Ratio2 9.7059 +Epoch [158/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7899, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 60.6170 % Model2 61.1178 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.7451 +Epoch [159/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7059 +Epoch [159/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.6078 +Epoch [159/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7206 +Epoch [159/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7725 +Epoch [159/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.7941 +Epoch [159/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 59.9159 % Model2 60.3165 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4314, Pure Ratio2 10.5098 +Epoch [160/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9902 +Epoch [160/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0261 +Epoch [160/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0000 +Epoch [160/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9608 +Epoch [160/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8105 +Epoch [160/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7927, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 59.8057 % Model2 61.8189 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8627 +Epoch [161/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 9.9706 +Epoch [161/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [161/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.9069 +Epoch [161/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.9882 +Epoch [161/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9052 +Epoch [161/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8936, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 61.5385 % Model2 61.0176 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3529 +Epoch [162/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0098 +Epoch [162/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0131 +Epoch [162/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9314 +Epoch [162/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 9.9451 +Epoch [162/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.9020 +Epoch [162/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 61.7087 % Model2 60.6370 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5490 +Epoch [163/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6961, Pure Ratio2 9.5686 +Epoch [163/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7190 +Epoch [163/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.6961 +Epoch [163/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8353, Pure Ratio2 9.6902 +Epoch [163/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.8399 +Epoch [163/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 60.3065 % Model2 62.1294 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.2549 +Epoch [164/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.4706 +Epoch [164/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.5359 +Epoch [164/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8529, Pure Ratio2 9.7010 +Epoch [164/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.7922 +Epoch [164/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7680 +Epoch [164/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8263, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 61.1278 % Model2 61.9992 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [165/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7843 +Epoch [165/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.6863 +Epoch [165/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.7304 +Epoch [165/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9255, Pure Ratio2 9.7686 +Epoch [165/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.7778 +Epoch [165/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 60.8073 % Model2 61.8590 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.6667 +Epoch [166/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2059 +Epoch [166/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9935 +Epoch [166/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0686 +Epoch [166/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9490 +Epoch [166/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9216 +Epoch [166/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8543, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 61.1478 % Model2 62.1194 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8235 +Epoch [167/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.6765 +Epoch [167/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.6144 +Epoch [167/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [167/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.7412 +Epoch [167/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.8725 +Epoch [167/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9356, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 61.0477 % Model2 60.8173 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.1569, Pure Ratio2 10.0980 +Epoch [168/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.0392 +Epoch [168/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9281 +Epoch [168/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0392 +Epoch [168/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 9.9451 +Epoch [168/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 9.9804 +Epoch [168/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 61.1879 % Model2 60.7472 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 9.9804 +Epoch [169/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.1569 +Epoch [169/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.1830 +Epoch [169/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1127 +Epoch [169/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.9294 +Epoch [169/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.8889 +Epoch [169/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 62.1895 % Model2 61.6887 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.7059 +Epoch [170/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0098 +Epoch [170/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.8627 +Epoch [170/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9559 +Epoch [170/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9843 +Epoch [170/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8954 +Epoch [170/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 60.3065 % Model2 61.7087 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5490 +Epoch [171/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.4706 +Epoch [171/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.6797 +Epoch [171/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8480 +Epoch [171/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.7529 +Epoch [171/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.7288 +Epoch [171/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8739, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 61.5986 % Model2 61.3882 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9608 +Epoch [172/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.1765 +Epoch [172/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.1046 +Epoch [172/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1324, Pure Ratio2 10.0686 +Epoch [172/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2118, Pure Ratio2 10.1176 +Epoch [172/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 9.9248 +Epoch [172/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 61.3982 % Model2 61.5485 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9412 +Epoch [173/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [173/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0392 +Epoch [173/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0392 +Epoch [173/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.8275 +Epoch [173/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.8529 +Epoch [173/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 62.0393 % Model2 61.7989 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6078 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.0000 +Epoch [174/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 9.9739 +Epoch [174/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.0931 +Epoch [174/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.0471 +Epoch [174/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9216 +Epoch [174/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 61.8890 % Model2 60.5869 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [175/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.6078 +Epoch [175/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.8235 +Epoch [175/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0147 +Epoch [175/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9725 +Epoch [175/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9935 +Epoch [175/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9748, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 61.0978 % Model2 61.6186 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5882, Pure Ratio2 10.6275 +Epoch [176/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1275 +Epoch [176/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.0327 +Epoch [176/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0686 +Epoch [176/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9647, Pure Ratio2 9.9608 +Epoch [176/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8987 +Epoch [176/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8796, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 61.5785 % Model2 61.8089 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.2549 +Epoch [177/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.7745 +Epoch [177/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.7516 +Epoch [177/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8725 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7765 +Epoch [177/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.8497 +Epoch [177/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8655, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 61.0978 % Model2 61.0877 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8431 +Epoch [178/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8529 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.7974 +Epoch [178/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8529 +Epoch [178/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.9608 +Epoch [178/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7614, Pure Ratio2 9.8856 +Epoch [178/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7619, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 61.6286 % Model2 61.7688 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4314, Pure Ratio2 9.2549 +Epoch [179/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2451, Pure Ratio2 9.2353 +Epoch [179/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4052, Pure Ratio2 9.3268 +Epoch [179/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5735, Pure Ratio2 9.4951 +Epoch [179/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7569 +Epoch [179/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9314 +Epoch [179/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8151, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 61.0477 % Model2 61.4183 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.3137 +Epoch [180/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.1176 +Epoch [180/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 10.0261 +Epoch [180/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.9314 +Epoch [180/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8353, Pure Ratio2 9.9216 +Epoch [180/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9118 +Epoch [180/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8683, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 61.6086 % Model2 61.2179 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9804 +Epoch [181/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 10.0980 +Epoch [181/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 10.0654 +Epoch [181/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.9216 +Epoch [181/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9725 +Epoch [181/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9150 +Epoch [181/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8375, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 61.2981 % Model2 61.7087 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.7059 +Epoch [182/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.3235 +Epoch [182/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1830 +Epoch [182/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0539 +Epoch [182/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.9961 +Epoch [182/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9183 +Epoch [182/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 60.8373 % Model2 60.6871 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7843 +Epoch [183/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.6275 +Epoch [183/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7843 +Epoch [183/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 9.9951 +Epoch [183/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.8824 +Epoch [183/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [183/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 62.1895 % Model2 61.3381 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [184/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9706 +Epoch [184/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9608 +Epoch [184/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.9118 +Epoch [184/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.9529 +Epoch [184/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 10.0556 +Epoch [184/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 61.0677 % Model2 62.1795 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [185/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.6569 +Epoch [185/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 10.0000 +Epoch [185/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9265 +Epoch [185/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8667, Pure Ratio2 9.8706 +Epoch [185/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9739 +Epoch [185/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9552, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 61.1679 % Model2 60.7071 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5686 +Epoch [186/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4216, Pure Ratio2 9.5784 +Epoch [186/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8693 +Epoch [186/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8725 +Epoch [186/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.9843 +Epoch [186/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0033 +Epoch [186/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 60.9876 % Model2 61.8089 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.0392 +Epoch [187/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.2843 +Epoch [187/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9085 +Epoch [187/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9461 +Epoch [187/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9922 +Epoch [187/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8856 +Epoch [187/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 60.6170 % Model2 60.8373 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.2353 +Epoch [188/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 9.8725 +Epoch [188/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 9.8954 +Epoch [188/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.7549 +Epoch [188/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.7843 +Epoch [188/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.8987, Pure Ratio2 9.8366 +Epoch [188/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 61.0777 % Model2 61.7388 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1373 +Epoch [189/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8922 +Epoch [189/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8301 +Epoch [189/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.7255 +Epoch [189/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7137, Pure Ratio2 9.7529 +Epoch [189/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8693 +Epoch [189/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 61.6186 % Model2 61.3181 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.4706 +Epoch [190/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2647, Pure Ratio2 9.2353 +Epoch [190/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.7386 +Epoch [190/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7500, Pure Ratio2 9.8529 +Epoch [190/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8157, Pure Ratio2 9.8431 +Epoch [190/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8301 +Epoch [190/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 61.6887 % Model2 61.5986 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0588 +Epoch [191/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8725 +Epoch [191/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0007, Pure Ratio1: 10.0261, Pure Ratio2 10.1111 +Epoch [191/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0931 +Epoch [191/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9647 +Epoch [191/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0261 +Epoch [191/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 61.1779 % Model2 61.7788 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1961 +Epoch [192/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8627 +Epoch [192/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9020 +Epoch [192/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.8627 +Epoch [192/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7569 +Epoch [192/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.9542 +Epoch [192/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 61.0276 % Model2 61.7087 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.2549 +Epoch [193/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.1275 +Epoch [193/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1373 +Epoch [193/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8529 +Epoch [193/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.8745 +Epoch [193/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9052 +Epoch [193/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 61.3782 % Model2 61.3081 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0784 +Epoch [194/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.2353 +Epoch [194/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1569 +Epoch [194/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0049 +Epoch [194/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 9.9765 +Epoch [194/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9739 +Epoch [194/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 61.3682 % Model2 61.5885 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.4118 +Epoch [195/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2647, Pure Ratio2 9.3725 +Epoch [195/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5033, Pure Ratio2 9.6601 +Epoch [195/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8480 +Epoch [195/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6157, Pure Ratio2 9.7059 +Epoch [195/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7614 +Epoch [195/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 61.3281 % Model2 61.9391 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6471 +Epoch [196/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8922 +Epoch [196/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.8235 +Epoch [196/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.7402 +Epoch [196/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7529, Pure Ratio2 9.8157 +Epoch [196/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8072, Pure Ratio2 9.9052 +Epoch [196/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 61.4583 % Model2 61.4083 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8431 +Epoch [197/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.9020 +Epoch [197/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9935 +Epoch [197/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 10.0000 +Epoch [197/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9529 +Epoch [197/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8660, Pure Ratio2 9.8889 +Epoch [197/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8543, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 61.1078 % Model2 61.8890 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [198/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9412 +Epoch [198/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9216 +Epoch [198/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8235 +Epoch [198/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7608 +Epoch [198/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [198/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8179, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 61.4383 % Model2 61.5785 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.5490 +Epoch [199/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9706 +Epoch [199/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.7451 +Epoch [199/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.7059 +Epoch [199/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7020, Pure Ratio2 9.7255 +Epoch [199/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7712 +Epoch [199/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 61.6987 % Model2 61.2780 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3725 +Epoch [200/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3824 +Epoch [200/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2680, Pure Ratio2 10.2810 +Epoch [200/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1618 +Epoch [200/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.1176 +Epoch [200/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0654 +Epoch [200/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 61.5585 % Model2 61.0777 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9397 % diff --git a/other_methods/coteaching/coteaching_results/out_2_6.log b/other_methods/coteaching/coteaching_results/out_2_6.log new file mode 100644 index 0000000..64ee48f --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_2_6.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0165, Loss2: 0.0167, Pure Ratio1: 9.6960, Pure Ratio2 9.6480 +Epoch [2/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0168, Loss2: 0.0167, Pure Ratio1: 9.6080, Pure Ratio2 9.6320 +Epoch [2/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 25.7812, Loss1: 0.0171, Loss2: 0.0171, Pure Ratio1: 9.6053, Pure Ratio2 9.6107 +Epoch [2/200], Iter [200/390] Training Accuracy1: 21.0938, Training Accuracy2: 21.0938, Loss1: 0.0174, Loss2: 0.0175, Pure Ratio1: 9.7240, Pure Ratio2 9.7240 +Epoch [2/200], Iter [250/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0166, Loss2: 0.0169, Pure Ratio1: 9.7120, Pure Ratio2 9.7120 +Epoch [2/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 27.3438, Loss1: 0.0159, Loss2: 0.0161, Pure Ratio1: 9.7733, Pure Ratio2 9.7707 +Epoch [2/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.9062, Loss1: 0.0161, Loss2: 0.0159, Pure Ratio1: 9.7783, Pure Ratio2 9.7691 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 23.9984 % Model2 23.2372 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.8236 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0156, Loss2: 0.0155, Pure Ratio1: 9.3607, Pure Ratio2 9.3770 +Epoch [3/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 21.8750, Loss1: 0.0166, Loss2: 0.0164, Pure Ratio1: 9.4098, Pure Ratio2 9.4344 +Epoch [3/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0167, Pure Ratio1: 9.7978, Pure Ratio2 9.8415 +Epoch [3/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.1250, Loss1: 0.0157, Loss2: 0.0156, Pure Ratio1: 9.7008, Pure Ratio2 9.7131 +Epoch [3/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 28.1250, Loss1: 0.0156, Loss2: 0.0154, Pure Ratio1: 9.8131, Pure Ratio2 9.8098 +Epoch [3/200], Iter [300/390] Training Accuracy1: 16.4062, Training Accuracy2: 20.3125, Loss1: 0.0168, Loss2: 0.0169, Pure Ratio1: 9.8033, Pure Ratio2 9.8060 +Epoch [3/200], Iter [350/390] Training Accuracy1: 23.4375, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0161, Pure Ratio1: 9.7916, Pure Ratio2 9.7963 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 21.5345 % Model2 25.3706 %, Pure Ratio 1 9.8003 %, Pure Ratio 2 9.8108 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 28.9062, Loss1: 0.0159, Loss2: 0.0158, Pure Ratio1: 9.8319, Pure Ratio2 9.7983 +Epoch [4/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 28.9062, Loss1: 0.0172, Loss2: 0.0170, Pure Ratio1: 9.7059, Pure Ratio2 9.6639 +Epoch [4/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0160, Loss2: 0.0159, Pure Ratio1: 9.9328, Pure Ratio2 9.8599 +Epoch [4/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 21.0938, Loss1: 0.0176, Loss2: 0.0176, Pure Ratio1: 9.7353, Pure Ratio2 9.6555 +Epoch [4/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 23.4375, Loss1: 0.0163, Loss2: 0.0159, Pure Ratio1: 9.7143, Pure Ratio2 9.6571 +Epoch [4/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0152, Loss2: 0.0147, Pure Ratio1: 9.7731, Pure Ratio2 9.7283 +Epoch [4/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0143, Loss2: 0.0145, Pure Ratio1: 9.8079, Pure Ratio2 9.7839 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 23.4275 % Model2 24.0084 %, Pure Ratio 1 9.7802 %, Pure Ratio 2 9.7630 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0153, Loss2: 0.0159, Pure Ratio1: 9.4655, Pure Ratio2 9.3793 +Epoch [5/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0160, Loss2: 0.0159, Pure Ratio1: 9.7155, Pure Ratio2 9.6897 +Epoch [5/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0155, Loss2: 0.0154, Pure Ratio1: 9.9943, Pure Ratio2 9.9540 +Epoch [5/200], Iter [200/390] Training Accuracy1: 27.3438, Training Accuracy2: 27.3438, Loss1: 0.0158, Loss2: 0.0154, Pure Ratio1: 9.7284, Pure Ratio2 9.7112 +Epoch [5/200], Iter [250/390] Training Accuracy1: 25.0000, Training Accuracy2: 28.9062, Loss1: 0.0163, Loss2: 0.0164, Pure Ratio1: 9.7172, Pure Ratio2 9.6655 +Epoch [5/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0158, Loss2: 0.0165, Pure Ratio1: 9.7672, Pure Ratio2 9.7270 +Epoch [5/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 26.5625, Loss1: 0.0171, Loss2: 0.0173, Pure Ratio1: 9.7611, Pure Ratio2 9.7266 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 23.7280 % Model2 25.3005 %, Pure Ratio 1 9.7834 %, Pure Ratio 2 9.7569 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.0000, Loss1: 0.0173, Loss2: 0.0171, Pure Ratio1: 10.2301, Pure Ratio2 10.2124 +Epoch [6/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 24.2188, Loss1: 0.0163, Loss2: 0.0162, Pure Ratio1: 10.1239, Pure Ratio2 10.1239 +Epoch [6/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 38.2812, Loss1: 0.0138, Loss2: 0.0138, Pure Ratio1: 9.9115, Pure Ratio2 9.8938 +Epoch [6/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0156, Loss2: 0.0153, Pure Ratio1: 9.8540, Pure Ratio2 9.8938 +Epoch [6/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0160, Loss2: 0.0165, Pure Ratio1: 9.8195, Pure Ratio2 9.8336 +Epoch [6/200], Iter [300/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.0000, Loss1: 0.0159, Loss2: 0.0163, Pure Ratio1: 9.7345, Pure Ratio2 9.7434 +Epoch [6/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 28.1250, Loss1: 0.0149, Loss2: 0.0150, Pure Ratio1: 9.7219, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 28.8662 % Model2 29.0565 %, Pure Ratio 1 9.7617 %, Pure Ratio 2 9.7595 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0164, Loss2: 0.0162, Pure Ratio1: 9.4000, Pure Ratio2 9.4364 +Epoch [7/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.0000, Loss1: 0.0162, Loss2: 0.0163, Pure Ratio1: 9.5636, Pure Ratio2 9.5818 +Epoch [7/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 30.4688, Loss1: 0.0145, Loss2: 0.0147, Pure Ratio1: 9.6424, Pure Ratio2 9.6485 +Epoch [7/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0143, Loss2: 0.0141, Pure Ratio1: 9.6682, Pure Ratio2 9.6773 +Epoch [7/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 29.6875, Loss1: 0.0161, Loss2: 0.0156, Pure Ratio1: 9.7382, Pure Ratio2 9.7273 +Epoch [7/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 32.8125, Loss1: 0.0147, Loss2: 0.0147, Pure Ratio1: 9.7091, Pure Ratio2 9.6939 +Epoch [7/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0155, Loss2: 0.0160, Pure Ratio1: 9.7143, Pure Ratio2 9.7013 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 26.3121 % Model2 25.1102 %, Pure Ratio 1 9.7529 %, Pure Ratio 2 9.7226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0136, Loss2: 0.0140, Pure Ratio1: 9.5000, Pure Ratio2 9.4259 +Epoch [8/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0155, Loss2: 0.0149, Pure Ratio1: 9.2685, Pure Ratio2 9.2315 +Epoch [8/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0142, Loss2: 0.0141, Pure Ratio1: 9.4753, Pure Ratio2 9.4321 +Epoch [8/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0135, Loss2: 0.0137, Pure Ratio1: 9.4444, Pure Ratio2 9.4074 +Epoch [8/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0145, Loss2: 0.0143, Pure Ratio1: 9.6111, Pure Ratio2 9.5852 +Epoch [8/200], Iter [300/390] Training Accuracy1: 25.0000, Training Accuracy2: 29.6875, Loss1: 0.0157, Loss2: 0.0152, Pure Ratio1: 9.7377, Pure Ratio2 9.7284 +Epoch [8/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0146, Pure Ratio1: 9.6402, Pure Ratio2 9.6481 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 32.8926 % Model2 31.6206 %, Pure Ratio 1 9.7104 %, Pure Ratio 2 9.7270 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0140, Loss2: 0.0140, Pure Ratio1: 9.7524, Pure Ratio2 9.5238 +Epoch [9/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0143, Loss2: 0.0140, Pure Ratio1: 9.7333, Pure Ratio2 9.6286 +Epoch [9/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0134, Loss2: 0.0140, Pure Ratio1: 9.6000, Pure Ratio2 9.5238 +Epoch [9/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0153, Loss2: 0.0152, Pure Ratio1: 9.6381, Pure Ratio2 9.5762 +Epoch [9/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 25.7812, Loss1: 0.0150, Loss2: 0.0152, Pure Ratio1: 9.7600, Pure Ratio2 9.6990 +Epoch [9/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0126, Loss2: 0.0124, Pure Ratio1: 9.7683, Pure Ratio2 9.7079 +Epoch [9/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0148, Loss2: 0.0145, Pure Ratio1: 9.7524, Pure Ratio2 9.7007 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 25.7412 % Model2 24.2388 %, Pure Ratio 1 9.7582 %, Pure Ratio 2 9.7143 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0142, Loss2: 0.0140, Pure Ratio1: 9.6471, Pure Ratio2 9.6667 +Epoch [10/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0146, Loss2: 0.0148, Pure Ratio1: 9.3333, Pure Ratio2 9.3333 +Epoch [10/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0143, Pure Ratio1: 9.6078, Pure Ratio2 9.5948 +Epoch [10/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0150, Loss2: 0.0144, Pure Ratio1: 9.4755, Pure Ratio2 9.5000 +Epoch [10/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0146, Loss2: 0.0147, Pure Ratio1: 9.6157, Pure Ratio2 9.6196 +Epoch [10/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0145, Loss2: 0.0149, Pure Ratio1: 9.6471, Pure Ratio2 9.6373 +Epoch [10/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0146, Loss2: 0.0149, Pure Ratio1: 9.7171, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 30.9195 % Model2 32.0813 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.6858 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0131, Loss2: 0.0134, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [11/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 25.7812, Loss1: 0.0144, Loss2: 0.0148, Pure Ratio1: 9.8333, Pure Ratio2 9.9510 +Epoch [11/200], Iter [150/390] Training Accuracy1: 26.5625, Training Accuracy2: 24.2188, Loss1: 0.0160, Loss2: 0.0159, Pure Ratio1: 10.0196, Pure Ratio2 10.1176 +Epoch [11/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0138, Loss2: 0.0132, Pure Ratio1: 9.8186, Pure Ratio2 9.9216 +Epoch [11/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0140, Loss2: 0.0141, Pure Ratio1: 9.7961, Pure Ratio2 9.8902 +Epoch [11/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0152, Loss2: 0.0150, Pure Ratio1: 9.7745, Pure Ratio2 9.8660 +Epoch [11/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 29.6875, Loss1: 0.0134, Loss2: 0.0135, Pure Ratio1: 9.7367, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 30.0681 % Model2 30.7993 %, Pure Ratio 1 9.6983 %, Pure Ratio 2 9.7486 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 29.6875, Loss1: 0.0138, Loss2: 0.0132, Pure Ratio1: 9.7647, Pure Ratio2 9.7059 +Epoch [12/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0142, Loss2: 0.0142, Pure Ratio1: 10.1569, Pure Ratio2 10.0980 +Epoch [12/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0123, Loss2: 0.0124, Pure Ratio1: 9.9739, Pure Ratio2 9.9281 +Epoch [12/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.0312, Loss1: 0.0132, Loss2: 0.0141, Pure Ratio1: 9.6765, Pure Ratio2 9.6373 +Epoch [12/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0139, Loss2: 0.0138, Pure Ratio1: 9.6196, Pure Ratio2 9.5804 +Epoch [12/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0150, Loss2: 0.0155, Pure Ratio1: 9.7190, Pure Ratio2 9.6797 +Epoch [12/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0133, Loss2: 0.0135, Pure Ratio1: 9.7647, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 27.3738 % Model2 27.4038 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0157, Loss2: 0.0159, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [13/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0155, Loss2: 0.0147, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [13/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0131, Loss2: 0.0132, Pure Ratio1: 9.6405, Pure Ratio2 9.6013 +Epoch [13/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.8125, Loss1: 0.0153, Loss2: 0.0151, Pure Ratio1: 9.7745, Pure Ratio2 9.7451 +Epoch [13/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0143, Loss2: 0.0149, Pure Ratio1: 9.6431, Pure Ratio2 9.5922 +Epoch [13/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0145, Loss2: 0.0149, Pure Ratio1: 9.8235, Pure Ratio2 9.7876 +Epoch [13/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 31.2500, Loss1: 0.0129, Loss2: 0.0135, Pure Ratio1: 9.7731, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 29.3470 % Model2 29.8978 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.0312, Loss1: 0.0128, Loss2: 0.0132, Pure Ratio1: 9.2745, Pure Ratio2 9.1569 +Epoch [14/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 32.0312, Loss1: 0.0131, Loss2: 0.0135, Pure Ratio1: 9.6373, Pure Ratio2 9.5490 +Epoch [14/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0151, Loss2: 0.0154, Pure Ratio1: 9.8627, Pure Ratio2 9.8039 +Epoch [14/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 28.9062, Loss1: 0.0141, Loss2: 0.0141, Pure Ratio1: 9.9216, Pure Ratio2 9.8922 +Epoch [14/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0149, Loss2: 0.0149, Pure Ratio1: 9.8706, Pure Ratio2 9.8824 +Epoch [14/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0137, Loss2: 0.0130, Pure Ratio1: 9.6830, Pure Ratio2 9.6895 +Epoch [14/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 30.4688, Loss1: 0.0135, Loss2: 0.0137, Pure Ratio1: 9.7367, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 30.1282 % Model2 31.3802 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.7662 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0129, Loss2: 0.0134, Pure Ratio1: 9.4314, Pure Ratio2 9.4510 +Epoch [15/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0128, Loss2: 0.0123, Pure Ratio1: 9.5784, Pure Ratio2 9.5980 +Epoch [15/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0142, Loss2: 0.0139, Pure Ratio1: 9.7124, Pure Ratio2 9.7582 +Epoch [15/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 31.2500, Loss1: 0.0132, Loss2: 0.0140, Pure Ratio1: 9.7549, Pure Ratio2 9.8235 +Epoch [15/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 28.1250, Loss1: 0.0137, Loss2: 0.0140, Pure Ratio1: 9.6941, Pure Ratio2 9.7569 +Epoch [15/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0130, Loss2: 0.0131, Pure Ratio1: 9.6699, Pure Ratio2 9.7157 +Epoch [15/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0134, Loss2: 0.0134, Pure Ratio1: 9.6022, Pure Ratio2 9.6387 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 32.7825 % Model2 32.6923 %, Pure Ratio 1 9.6807 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0140, Loss2: 0.0144, Pure Ratio1: 9.7647, Pure Ratio2 9.6275 +Epoch [16/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0141, Loss2: 0.0133, Pure Ratio1: 9.5882, Pure Ratio2 9.5196 +Epoch [16/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0123, Loss2: 0.0123, Pure Ratio1: 9.6667, Pure Ratio2 9.6340 +Epoch [16/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0120, Loss2: 0.0125, Pure Ratio1: 9.6912, Pure Ratio2 9.7059 +Epoch [16/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0127, Loss2: 0.0129, Pure Ratio1: 9.6706, Pure Ratio2 9.6824 +Epoch [16/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0147, Pure Ratio1: 9.7353, Pure Ratio2 9.7484 +Epoch [16/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0140, Loss2: 0.0147, Pure Ratio1: 9.7339, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 29.2268 % Model2 28.1851 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0132, Loss2: 0.0131, Pure Ratio1: 9.6471, Pure Ratio2 9.5686 +Epoch [17/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 43.7500, Loss1: 0.0127, Loss2: 0.0124, Pure Ratio1: 9.6373, Pure Ratio2 9.6961 +Epoch [17/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0134, Loss2: 0.0128, Pure Ratio1: 9.7255, Pure Ratio2 9.8431 +Epoch [17/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0133, Loss2: 0.0135, Pure Ratio1: 9.7500, Pure Ratio2 9.8627 +Epoch [17/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0112, Loss2: 0.0113, Pure Ratio1: 9.7412, Pure Ratio2 9.8275 +Epoch [17/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0123, Loss2: 0.0127, Pure Ratio1: 9.7549, Pure Ratio2 9.8366 +Epoch [17/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0145, Loss2: 0.0139, Pure Ratio1: 9.7283, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 32.5521 % Model2 32.9527 %, Pure Ratio 1 9.7034 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0116, Loss2: 0.0111, Pure Ratio1: 8.9216, Pure Ratio2 8.9216 +Epoch [18/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0123, Loss2: 0.0124, Pure Ratio1: 9.3431, Pure Ratio2 9.4510 +Epoch [18/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0135, Loss2: 0.0126, Pure Ratio1: 9.4575, Pure Ratio2 9.5621 +Epoch [18/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0137, Loss2: 0.0136, Pure Ratio1: 9.3971, Pure Ratio2 9.4412 +Epoch [18/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0144, Loss2: 0.0145, Pure Ratio1: 9.4431, Pure Ratio2 9.4824 +Epoch [18/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0136, Pure Ratio1: 9.5621, Pure Ratio2 9.5915 +Epoch [18/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0146, Loss2: 0.0146, Pure Ratio1: 9.6835, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 30.1482 % Model2 30.1382 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0124, Loss2: 0.0124, Pure Ratio1: 9.2549, Pure Ratio2 9.1961 +Epoch [19/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0106, Loss2: 0.0107, Pure Ratio1: 9.5000, Pure Ratio2 9.4608 +Epoch [19/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0118, Loss2: 0.0119, Pure Ratio1: 9.6013, Pure Ratio2 9.4967 +Epoch [19/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0124, Loss2: 0.0126, Pure Ratio1: 9.7500, Pure Ratio2 9.6618 +Epoch [19/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0115, Loss2: 0.0120, Pure Ratio1: 9.7725, Pure Ratio2 9.7333 +Epoch [19/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.0625, Loss1: 0.0130, Loss2: 0.0121, Pure Ratio1: 9.7680, Pure Ratio2 9.7092 +Epoch [19/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 36.7188, Loss1: 0.0128, Loss2: 0.0129, Pure Ratio1: 9.8067, Pure Ratio2 9.7591 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 34.8958 % Model2 33.9042 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7511 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0106, Loss2: 0.0113, Pure Ratio1: 9.2745, Pure Ratio2 9.2941 +Epoch [20/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0108, Loss2: 0.0106, Pure Ratio1: 9.4314, Pure Ratio2 9.5588 +Epoch [20/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0118, Pure Ratio1: 9.7451, Pure Ratio2 9.8562 +Epoch [20/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0130, Loss2: 0.0125, Pure Ratio1: 9.7059, Pure Ratio2 9.8235 +Epoch [20/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0101, Loss2: 0.0101, Pure Ratio1: 9.6784, Pure Ratio2 9.7529 +Epoch [20/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0102, Loss2: 0.0110, Pure Ratio1: 9.6765, Pure Ratio2 9.7614 +Epoch [20/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0133, Loss2: 0.0121, Pure Ratio1: 9.6723, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 32.8225 % Model2 31.8209 %, Pure Ratio 1 9.7009 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0091, Loss2: 0.0091, Pure Ratio1: 10.1373, Pure Ratio2 10.0392 +Epoch [21/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0108, Loss2: 0.0114, Pure Ratio1: 9.7941, Pure Ratio2 9.6961 +Epoch [21/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0104, Loss2: 0.0104, Pure Ratio1: 9.6797, Pure Ratio2 9.5817 +Epoch [21/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0116, Loss2: 0.0109, Pure Ratio1: 9.8382, Pure Ratio2 9.7696 +Epoch [21/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0102, Loss2: 0.0104, Pure Ratio1: 9.8667, Pure Ratio2 9.8000 +Epoch [21/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0094, Loss2: 0.0101, Pure Ratio1: 9.7712, Pure Ratio2 9.6699 +Epoch [21/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0121, Loss2: 0.0122, Pure Ratio1: 9.8459, Pure Ratio2 9.7675 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 31.9712 % Model2 32.9026 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.6983 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0104, Loss2: 0.0107, Pure Ratio1: 9.0000, Pure Ratio2 8.9804 +Epoch [22/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0095, Loss2: 0.0095, Pure Ratio1: 9.2451, Pure Ratio2 9.3627 +Epoch [22/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0105, Loss2: 0.0114, Pure Ratio1: 9.4314, Pure Ratio2 9.4967 +Epoch [22/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0095, Loss2: 0.0100, Pure Ratio1: 9.5000, Pure Ratio2 9.5343 +Epoch [22/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0098, Loss2: 0.0097, Pure Ratio1: 9.5020, Pure Ratio2 9.5451 +Epoch [22/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0114, Loss2: 0.0116, Pure Ratio1: 9.5686, Pure Ratio2 9.5654 +Epoch [22/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0105, Loss2: 0.0107, Pure Ratio1: 9.6387, Pure Ratio2 9.6387 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 32.1114 % Model2 34.0044 %, Pure Ratio 1 9.6732 %, Pure Ratio 2 9.6682 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0102, Pure Ratio1: 9.8627, Pure Ratio2 9.9216 +Epoch [23/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0102, Loss2: 0.0098, Pure Ratio1: 9.4118, Pure Ratio2 9.4902 +Epoch [23/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0105, Loss2: 0.0117, Pure Ratio1: 9.5948, Pure Ratio2 9.6275 +Epoch [23/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 42.9688, Loss1: 0.0104, Loss2: 0.0111, Pure Ratio1: 9.5000, Pure Ratio2 9.4461 +Epoch [23/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0079, Loss2: 0.0083, Pure Ratio1: 9.5647, Pure Ratio2 9.5255 +Epoch [23/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0092, Loss2: 0.0100, Pure Ratio1: 9.7124, Pure Ratio2 9.6503 +Epoch [23/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0135, Pure Ratio1: 9.8403, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 31.9311 % Model2 31.1498 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7486 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0076, Loss2: 0.0084, Pure Ratio1: 10.0392, Pure Ratio2 9.8627 +Epoch [24/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0104, Loss2: 0.0114, Pure Ratio1: 9.7353, Pure Ratio2 9.7451 +Epoch [24/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0080, Loss2: 0.0093, Pure Ratio1: 9.7255, Pure Ratio2 9.7320 +Epoch [24/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0108, Loss2: 0.0108, Pure Ratio1: 9.7402, Pure Ratio2 9.7500 +Epoch [24/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 62.5000, Loss1: 0.0085, Loss2: 0.0079, Pure Ratio1: 9.6980, Pure Ratio2 9.7255 +Epoch [24/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0086, Loss2: 0.0085, Pure Ratio1: 9.5850, Pure Ratio2 9.6176 +Epoch [24/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.9688, Loss1: 0.0099, Loss2: 0.0100, Pure Ratio1: 9.6303, Pure Ratio2 9.6387 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 29.3169 % Model2 28.5056 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0076, Loss2: 0.0087, Pure Ratio1: 9.5490, Pure Ratio2 9.4706 +Epoch [25/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0060, Loss2: 0.0066, Pure Ratio1: 9.7549, Pure Ratio2 9.7549 +Epoch [25/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0105, Loss2: 0.0105, Pure Ratio1: 9.9412, Pure Ratio2 9.8562 +Epoch [25/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0080, Loss2: 0.0086, Pure Ratio1: 9.7990, Pure Ratio2 9.7304 +Epoch [25/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0075, Loss2: 0.0082, Pure Ratio1: 9.7412, Pure Ratio2 9.7059 +Epoch [25/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0088, Loss2: 0.0082, Pure Ratio1: 9.7026, Pure Ratio2 9.6863 +Epoch [25/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0089, Loss2: 0.0096, Pure Ratio1: 9.7563, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 30.3085 % Model2 29.7376 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7486 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0063, Loss2: 0.0071, Pure Ratio1: 9.5294, Pure Ratio2 9.4314 +Epoch [26/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0085, Loss2: 0.0076, Pure Ratio1: 9.6765, Pure Ratio2 9.6078 +Epoch [26/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0074, Loss2: 0.0081, Pure Ratio1: 9.5359, Pure Ratio2 9.5098 +Epoch [26/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0065, Loss2: 0.0068, Pure Ratio1: 9.8235, Pure Ratio2 9.8088 +Epoch [26/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0070, Loss2: 0.0081, Pure Ratio1: 9.7765, Pure Ratio2 9.7529 +Epoch [26/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0078, Loss2: 0.0077, Pure Ratio1: 9.8529, Pure Ratio2 9.7974 +Epoch [26/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0095, Loss2: 0.0106, Pure Ratio1: 9.7591, Pure Ratio2 9.7115 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 31.2600 % Model2 32.0112 %, Pure Ratio 1 9.7587 %, Pure Ratio 2 9.7109 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 64.0625, Loss1: 0.0051, Loss2: 0.0061, Pure Ratio1: 9.3529, Pure Ratio2 9.2353 +Epoch [27/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0059, Loss2: 0.0066, Pure Ratio1: 9.6569, Pure Ratio2 9.5784 +Epoch [27/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0058, Loss2: 0.0059, Pure Ratio1: 9.7124, Pure Ratio2 9.6536 +Epoch [27/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0072, Loss2: 0.0070, Pure Ratio1: 9.8186, Pure Ratio2 9.7598 +Epoch [27/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0060, Loss2: 0.0074, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [27/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0071, Loss2: 0.0075, Pure Ratio1: 9.7745, Pure Ratio2 9.7288 +Epoch [27/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0060, Loss2: 0.0059, Pure Ratio1: 9.7871, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 30.3786 % Model2 29.7576 %, Pure Ratio 1 9.7461 %, Pure Ratio 2 9.7235 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0054, Loss2: 0.0051, Pure Ratio1: 9.5294, Pure Ratio2 9.5686 +Epoch [28/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0043, Loss2: 0.0040, Pure Ratio1: 9.5000, Pure Ratio2 9.4020 +Epoch [28/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0043, Loss2: 0.0045, Pure Ratio1: 9.6144, Pure Ratio2 9.5229 +Epoch [28/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0068, Loss2: 0.0059, Pure Ratio1: 9.8873, Pure Ratio2 9.8382 +Epoch [28/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 62.5000, Loss1: 0.0071, Loss2: 0.0065, Pure Ratio1: 9.7137, Pure Ratio2 9.6745 +Epoch [28/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0054, Loss2: 0.0053, Pure Ratio1: 9.7974, Pure Ratio2 9.7778 +Epoch [28/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0069, Loss2: 0.0058, Pure Ratio1: 9.7871, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 29.6975 % Model2 30.3686 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0047, Loss2: 0.0043, Pure Ratio1: 9.6078, Pure Ratio2 9.6471 +Epoch [29/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0036, Loss2: 0.0042, Pure Ratio1: 9.8137, Pure Ratio2 9.8627 +Epoch [29/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0044, Loss2: 0.0042, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [29/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0052, Loss2: 0.0042, Pure Ratio1: 9.9853, Pure Ratio2 9.9951 +Epoch [29/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0052, Loss2: 0.0051, Pure Ratio1: 9.8353, Pure Ratio2 9.8510 +Epoch [29/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0052, Loss2: 0.0057, Pure Ratio1: 9.8137, Pure Ratio2 9.8137 +Epoch [29/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0065, Loss2: 0.0053, Pure Ratio1: 9.7787, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 30.3185 % Model2 30.7091 %, Pure Ratio 1 9.7386 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0034, Loss2: 0.0037, Pure Ratio1: 9.9412, Pure Ratio2 9.8235 +Epoch [30/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0045, Loss2: 0.0046, Pure Ratio1: 9.9706, Pure Ratio2 9.9804 +Epoch [30/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0039, Loss2: 0.0043, Pure Ratio1: 9.8170, Pure Ratio2 9.8170 +Epoch [30/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0040, Loss2: 0.0039, Pure Ratio1: 9.7794, Pure Ratio2 9.7304 +Epoch [30/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.8750, Loss1: 0.0058, Loss2: 0.0047, Pure Ratio1: 9.7647, Pure Ratio2 9.6980 +Epoch [30/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0042, Loss2: 0.0042, Pure Ratio1: 9.7941, Pure Ratio2 9.7353 +Epoch [30/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.0625, Loss1: 0.0046, Loss2: 0.0056, Pure Ratio1: 9.7171, Pure Ratio2 9.6723 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 29.6675 % Model2 30.2684 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7034 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0037, Loss2: 0.0043, Pure Ratio1: 9.1961, Pure Ratio2 9.0784 +Epoch [31/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.0040, Loss2: 0.0033, Pure Ratio1: 9.4314, Pure Ratio2 9.3431 +Epoch [31/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.0938, Loss1: 0.0055, Loss2: 0.0048, Pure Ratio1: 9.6144, Pure Ratio2 9.5817 +Epoch [31/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0048, Loss2: 0.0042, Pure Ratio1: 9.7745, Pure Ratio2 9.6667 +Epoch [31/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0032, Loss2: 0.0041, Pure Ratio1: 9.7333, Pure Ratio2 9.6157 +Epoch [31/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.9375, Loss1: 0.0059, Loss2: 0.0072, Pure Ratio1: 9.7320, Pure Ratio2 9.6176 +Epoch [31/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0046, Loss2: 0.0051, Pure Ratio1: 9.8179, Pure Ratio2 9.7227 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 30.7492 % Model2 30.4888 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.6631 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0033, Loss2: 0.0033, Pure Ratio1: 10.0000, Pure Ratio2 10.0588 +Epoch [32/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 74.2188, Loss1: 0.0045, Loss2: 0.0034, Pure Ratio1: 9.6078, Pure Ratio2 9.6275 +Epoch [32/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0033, Loss2: 0.0037, Pure Ratio1: 9.5752, Pure Ratio2 9.5294 +Epoch [32/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0030, Loss2: 0.0030, Pure Ratio1: 9.6569, Pure Ratio2 9.5784 +Epoch [32/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0025, Pure Ratio1: 9.6784, Pure Ratio2 9.6118 +Epoch [32/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 65.6250, Loss1: 0.0032, Loss2: 0.0048, Pure Ratio1: 9.6732, Pure Ratio2 9.6536 +Epoch [32/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0039, Loss2: 0.0053, Pure Ratio1: 9.7283, Pure Ratio2 9.7115 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 28.3854 % Model2 30.1482 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0025, Loss2: 0.0036, Pure Ratio1: 9.5686, Pure Ratio2 9.5686 +Epoch [33/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 77.3438, Loss1: 0.0016, Loss2: 0.0027, Pure Ratio1: 9.4804, Pure Ratio2 9.5882 +Epoch [33/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0058, Loss2: 0.0047, Pure Ratio1: 9.7190, Pure Ratio2 9.7908 +Epoch [33/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0045, Loss2: 0.0046, Pure Ratio1: 9.8284, Pure Ratio2 9.8676 +Epoch [33/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0027, Loss2: 0.0031, Pure Ratio1: 9.7333, Pure Ratio2 9.7569 +Epoch [33/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 76.5625, Loss1: 0.0027, Loss2: 0.0028, Pure Ratio1: 9.6797, Pure Ratio2 9.7092 +Epoch [33/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.0057, Loss2: 0.0027, Pure Ratio1: 9.7115, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 31.0497 % Model2 30.5188 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0024, Loss2: 0.0024, Pure Ratio1: 9.5882, Pure Ratio2 9.6275 +Epoch [34/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0028, Loss2: 0.0024, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [34/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0019, Pure Ratio1: 9.5752, Pure Ratio2 9.6340 +Epoch [34/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0023, Loss2: 0.0029, Pure Ratio1: 9.6127, Pure Ratio2 9.6471 +Epoch [34/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0033, Loss2: 0.0041, Pure Ratio1: 9.6471, Pure Ratio2 9.6824 +Epoch [34/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0035, Loss2: 0.0043, Pure Ratio1: 9.6895, Pure Ratio2 9.7157 +Epoch [34/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 71.0938, Loss1: 0.0022, Loss2: 0.0031, Pure Ratio1: 9.7283, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 29.9780 % Model2 30.4087 %, Pure Ratio 1 9.7159 %, Pure Ratio 2 9.7210 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0015, Loss2: 0.0017, Pure Ratio1: 9.6667, Pure Ratio2 9.7255 +Epoch [35/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0022, Loss2: 0.0015, Pure Ratio1: 9.8137, Pure Ratio2 9.8431 +Epoch [35/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0022, Loss2: 0.0025, Pure Ratio1: 9.8889, Pure Ratio2 9.9020 +Epoch [35/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0023, Loss2: 0.0019, Pure Ratio1: 9.8824, Pure Ratio2 9.8382 +Epoch [35/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0018, Pure Ratio1: 9.8745, Pure Ratio2 9.8078 +Epoch [35/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.0023, Loss2: 0.0030, Pure Ratio1: 9.8954, Pure Ratio2 9.8627 +Epoch [35/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0024, Pure Ratio1: 9.8179, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 29.5473 % Model2 29.8578 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0012, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [36/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.5784, Pure Ratio2 9.6078 +Epoch [36/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0017, Loss2: 0.0017, Pure Ratio1: 9.6209, Pure Ratio2 9.6405 +Epoch [36/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0024, Loss2: 0.0019, Pure Ratio1: 9.6275, Pure Ratio2 9.6716 +Epoch [36/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0016, Loss2: 0.0013, Pure Ratio1: 9.6784, Pure Ratio2 9.7176 +Epoch [36/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0025, Pure Ratio1: 9.5915, Pure Ratio2 9.6405 +Epoch [36/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0020, Loss2: 0.0025, Pure Ratio1: 9.6863, Pure Ratio2 9.7227 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 30.4788 % Model2 30.4387 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0020, Pure Ratio1: 9.7059, Pure Ratio2 9.6471 +Epoch [37/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 9.3529, Pure Ratio2 9.5000 +Epoch [37/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0016, Loss2: 0.0023, Pure Ratio1: 9.6078, Pure Ratio2 9.6209 +Epoch [37/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0020, Loss2: 0.0017, Pure Ratio1: 9.5784, Pure Ratio2 9.5686 +Epoch [37/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0021, Loss2: 0.0015, Pure Ratio1: 9.5529, Pure Ratio2 9.5412 +Epoch [37/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0021, Pure Ratio1: 9.5980, Pure Ratio2 9.6046 +Epoch [37/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0024, Pure Ratio1: 9.7451, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 29.3470 % Model2 28.7861 %, Pure Ratio 1 9.7486 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 9.1961, Pure Ratio2 9.3922 +Epoch [38/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0029, Pure Ratio1: 9.4412, Pure Ratio2 9.5588 +Epoch [38/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 9.6667, Pure Ratio2 9.7124 +Epoch [38/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.5784, Pure Ratio2 9.6422 +Epoch [38/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0023, Pure Ratio1: 9.7098, Pure Ratio2 9.7882 +Epoch [38/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0019, Loss2: 0.0019, Pure Ratio1: 9.6928, Pure Ratio2 9.7418 +Epoch [38/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0023, Loss2: 0.0026, Pure Ratio1: 9.6639, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 29.4772 % Model2 29.4071 %, Pure Ratio 1 9.6933 %, Pure Ratio 2 9.6883 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 9.2745, Pure Ratio2 9.2745 +Epoch [39/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.7941, Pure Ratio2 9.8627 +Epoch [39/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0016, Loss2: 0.0011, Pure Ratio1: 10.0327, Pure Ratio2 9.9673 +Epoch [39/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0014, Pure Ratio1: 9.9755, Pure Ratio2 9.9412 +Epoch [39/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0019, Pure Ratio1: 9.9529, Pure Ratio2 9.9176 +Epoch [39/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.8987, Pure Ratio2 9.8693 +Epoch [39/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0031, Loss2: 0.0023, Pure Ratio1: 9.7871, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 27.9447 % Model2 28.9463 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0024, Pure Ratio1: 9.8039, Pure Ratio2 9.5882 +Epoch [40/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0027, Pure Ratio1: 9.3431, Pure Ratio2 9.2647 +Epoch [40/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0013, Pure Ratio1: 9.4902, Pure Ratio2 9.4248 +Epoch [40/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.5049, Pure Ratio2 9.4608 +Epoch [40/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0018, Pure Ratio1: 9.4745, Pure Ratio2 9.4471 +Epoch [40/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.6144, Pure Ratio2 9.5784 +Epoch [40/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 86.7188, Loss1: 0.0017, Loss2: 0.0011, Pure Ratio1: 9.7619, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 29.3670 % Model2 30.6290 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7034 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.8235, Pure Ratio2 9.6078 +Epoch [41/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0017, Loss2: 0.0012, Pure Ratio1: 9.6765, Pure Ratio2 9.5784 +Epoch [41/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.6601, Pure Ratio2 9.5817 +Epoch [41/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0014, Loss2: 0.0016, Pure Ratio1: 9.6667, Pure Ratio2 9.6225 +Epoch [41/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.7451, Pure Ratio2 9.7098 +Epoch [41/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 9.8039, Pure Ratio2 9.7516 +Epoch [41/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.9062, Loss1: 0.0025, Loss2: 0.0011, Pure Ratio1: 9.7591, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 29.1567 % Model2 32.4619 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.5882, Pure Ratio2 9.6275 +Epoch [42/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0028, Loss2: 0.0014, Pure Ratio1: 9.7941, Pure Ratio2 9.7059 +Epoch [42/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0025, Pure Ratio1: 9.6275, Pure Ratio2 9.6405 +Epoch [42/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [42/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0028, Loss2: 0.0026, Pure Ratio1: 9.7451, Pure Ratio2 9.8000 +Epoch [42/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0019, Loss2: 0.0015, Pure Ratio1: 9.7092, Pure Ratio2 9.7288 +Epoch [42/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.7815, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 27.8045 % Model2 29.3369 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.3922, Pure Ratio2 9.4118 +Epoch [43/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.8725, Pure Ratio2 9.8725 +Epoch [43/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 77.3438, Loss1: 0.0007, Loss2: 0.0019, Pure Ratio1: 10.0588, Pure Ratio2 10.0523 +Epoch [43/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.9510, Pure Ratio2 9.9461 +Epoch [43/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0017, Pure Ratio1: 9.9059, Pure Ratio2 9.9059 +Epoch [43/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [43/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0020, Loss2: 0.0022, Pure Ratio1: 9.7675, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 27.9347 % Model2 29.6875 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0014, Pure Ratio1: 9.2745, Pure Ratio2 9.5490 +Epoch [44/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.5588, Pure Ratio2 9.6471 +Epoch [44/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.4771, Pure Ratio2 9.6340 +Epoch [44/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0013, Pure Ratio1: 9.4363, Pure Ratio2 9.5490 +Epoch [44/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.4471, Pure Ratio2 9.5373 +Epoch [44/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.5229, Pure Ratio2 9.5719 +Epoch [44/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0023, Pure Ratio1: 9.6331, Pure Ratio2 9.6583 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 29.4571 % Model2 28.4655 %, Pure Ratio 1 9.6807 %, Pure Ratio 2 9.7059 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0784, Pure Ratio2 10.4510 +Epoch [45/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 10.1667, Pure Ratio2 10.2353 +Epoch [45/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.9085, Pure Ratio2 9.9673 +Epoch [45/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0015, Pure Ratio1: 9.7157, Pure Ratio2 9.7108 +Epoch [45/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 9.7176, Pure Ratio2 9.7412 +Epoch [45/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.7157, Pure Ratio2 9.7288 +Epoch [45/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0017, Loss2: 0.0011, Pure Ratio1: 9.7899, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 29.6374 % Model2 30.1883 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0016, Pure Ratio1: 10.2745, Pure Ratio2 10.4902 +Epoch [46/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.7745, Pure Ratio2 9.7941 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.7255, Pure Ratio2 9.6275 +Epoch [46/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.8039, Pure Ratio2 9.6520 +Epoch [46/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8275, Pure Ratio2 9.6745 +Epoch [46/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 83.5938, Loss1: 0.0021, Loss2: 0.0014, Pure Ratio1: 9.8039, Pure Ratio2 9.7026 +Epoch [46/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 9.7703, Pure Ratio2 9.6667 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 28.6859 % Model2 28.3454 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.6958 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.5686, Pure Ratio2 9.3137 +Epoch [47/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.6078, Pure Ratio2 9.4902 +Epoch [47/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0010, Pure Ratio1: 9.9346, Pure Ratio2 9.8562 +Epoch [47/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.7108, Pure Ratio2 9.7108 +Epoch [47/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 9.7412, Pure Ratio2 9.6824 +Epoch [47/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 9.6732, Pure Ratio2 9.6732 +Epoch [47/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0016, Pure Ratio1: 9.7535, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 29.1166 % Model2 28.7159 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7084 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 9.4118, Pure Ratio2 9.1961 +Epoch [48/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0015, Pure Ratio1: 9.6373, Pure Ratio2 9.5098 +Epoch [48/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.6928, Pure Ratio2 9.6209 +Epoch [48/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.8480, Pure Ratio2 9.7451 +Epoch [48/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.8314, Pure Ratio2 9.7294 +Epoch [48/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 77.3438, Loss1: 0.0005, Loss2: 0.0016, Pure Ratio1: 9.7745, Pure Ratio2 9.6830 +Epoch [48/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.7563, Pure Ratio2 9.6807 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 30.1883 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [49/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9412, Pure Ratio2 9.7647 +Epoch [49/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8693, Pure Ratio2 9.7516 +Epoch [49/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0017, Pure Ratio1: 9.8676, Pure Ratio2 9.8039 +Epoch [49/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.7451, Pure Ratio2 9.6706 +Epoch [49/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0016, Loss2: 0.0007, Pure Ratio1: 9.6340, Pure Ratio2 9.5882 +Epoch [49/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 85.9375, Loss1: 0.0020, Loss2: 0.0006, Pure Ratio1: 9.7003, Pure Ratio2 9.6583 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 29.5873 % Model2 30.9896 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.6883 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 10.3725, Pure Ratio2 10.1961 +Epoch [50/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.9216, Pure Ratio2 9.7745 +Epoch [50/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.8954, Pure Ratio2 9.7778 +Epoch [50/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7843, Pure Ratio2 9.6814 +Epoch [50/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.6392, Pure Ratio2 9.5529 +Epoch [50/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 85.1562, Loss1: 0.0020, Loss2: 0.0010, Pure Ratio1: 9.6569, Pure Ratio2 9.6144 +Epoch [50/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.7759, Pure Ratio2 9.7227 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 28.9663 % Model2 30.5990 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0015, Loss2: 0.0019, Pure Ratio1: 9.9804, Pure Ratio2 9.8039 +Epoch [51/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.8824, Pure Ratio2 9.8235 +Epoch [51/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0017, Pure Ratio1: 10.0131, Pure Ratio2 9.9346 +Epoch [51/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.8529, Pure Ratio2 9.7647 +Epoch [51/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0014, Loss2: 0.0005, Pure Ratio1: 9.7725, Pure Ratio2 9.6863 +Epoch [51/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0017, Loss2: 0.0007, Pure Ratio1: 9.7026, Pure Ratio2 9.6569 +Epoch [51/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.6443, Pure Ratio2 9.6218 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 29.6575 % Model2 31.4403 %, Pure Ratio 1 9.7009 %, Pure Ratio 2 9.6807 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.2941, Pure Ratio2 9.2941 +Epoch [52/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 9.3922, Pure Ratio2 9.3039 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0012, Pure Ratio1: 9.4967, Pure Ratio2 9.3922 +Epoch [52/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0017, Loss2: 0.0005, Pure Ratio1: 9.5392, Pure Ratio2 9.3529 +Epoch [52/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0029, Pure Ratio1: 9.6941, Pure Ratio2 9.5412 +Epoch [52/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.7124, Pure Ratio2 9.5621 +Epoch [52/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.7675, Pure Ratio2 9.6246 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 29.5272 % Model2 31.1799 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.6958 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.2941, Pure Ratio2 9.5882 +Epoch [53/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.7549, Pure Ratio2 9.9216 +Epoch [53/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8301, Pure Ratio2 9.9412 +Epoch [53/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9118, Pure Ratio2 9.9412 +Epoch [53/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.8588, Pure Ratio2 9.8863 +Epoch [53/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0017, Pure Ratio1: 9.7157, Pure Ratio2 9.7353 +Epoch [53/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 9.6835, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 28.6158 % Model2 30.0781 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.7255, Pure Ratio2 9.5882 +Epoch [54/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.7451, Pure Ratio2 9.6471 +Epoch [54/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.6144, Pure Ratio2 9.5033 +Epoch [54/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0017, Loss2: 0.0012, Pure Ratio1: 9.5637, Pure Ratio2 9.4902 +Epoch [54/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.5255, Pure Ratio2 9.4510 +Epoch [54/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.6601, Pure Ratio2 9.6144 +Epoch [54/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0017, Pure Ratio1: 9.7311, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 29.7175 % Model2 29.9880 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.0980, Pure Ratio2 9.4902 +Epoch [55/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.0784, Pure Ratio2 9.2549 +Epoch [55/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.4183, Pure Ratio2 9.5752 +Epoch [55/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.6176, Pure Ratio2 9.7500 +Epoch [55/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0019, Pure Ratio1: 9.8510, Pure Ratio2 9.9686 +Epoch [55/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.7876, Pure Ratio2 9.8693 +Epoch [55/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.7675, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 28.4455 % Model2 29.3069 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.6078, Pure Ratio2 9.6275 +Epoch [56/200], Iter [100/390] Training Accuracy1: 94.5312, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8725, Pure Ratio2 9.7647 +Epoch [56/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 9.9346, Pure Ratio2 9.8758 +Epoch [56/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.7647, Pure Ratio2 9.7549 +Epoch [56/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8157, Pure Ratio2 9.8078 +Epoch [56/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 9.8366, Pure Ratio2 9.8203 +Epoch [56/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0024, Pure Ratio1: 9.8151, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 29.5473 % Model2 29.0565 %, Pure Ratio 1 9.7587 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 9.8235 +Epoch [57/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.7647, Pure Ratio2 9.7157 +Epoch [57/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.8954 +Epoch [57/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.8676, Pure Ratio2 9.8088 +Epoch [57/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8471, Pure Ratio2 9.7569 +Epoch [57/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7876, Pure Ratio2 9.7092 +Epoch [57/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0016, Loss2: 0.0009, Pure Ratio1: 9.7507, Pure Ratio2 9.6807 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 29.9780 % Model2 28.5958 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0014, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.3725 +Epoch [58/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0000, Pure Ratio2 9.8627 +Epoch [58/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [58/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8333, Pure Ratio2 9.7696 +Epoch [58/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.6863, Pure Ratio2 9.6392 +Epoch [58/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.6830, Pure Ratio2 9.6503 +Epoch [58/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0026, Pure Ratio1: 9.6723, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 29.5072 % Model2 29.5072 %, Pure Ratio 1 9.7009 %, Pure Ratio 2 9.7109 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [59/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [59/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9477, Pure Ratio2 9.9412 +Epoch [59/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.7696, Pure Ratio2 9.8088 +Epoch [59/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0017, Loss2: 0.0020, Pure Ratio1: 9.5882, Pure Ratio2 9.6431 +Epoch [59/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.7026, Pure Ratio2 9.7320 +Epoch [59/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0017, Pure Ratio1: 9.6555, Pure Ratio2 9.6723 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 28.0248 % Model2 28.2953 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [60/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.5980, Pure Ratio2 9.5000 +Epoch [60/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6732, Pure Ratio2 9.7255 +Epoch [60/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7402, Pure Ratio2 9.7549 +Epoch [60/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0023, Loss2: 0.0010, Pure Ratio1: 9.8431, Pure Ratio2 9.8196 +Epoch [60/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8758, Pure Ratio2 9.8072 +Epoch [60/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.8291, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 30.9295 % Model2 29.0966 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [61/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.8627, Pure Ratio2 10.0686 +Epoch [61/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.8497 +Epoch [61/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0019, Loss2: 0.0005, Pure Ratio1: 9.7010, Pure Ratio2 9.7941 +Epoch [61/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 9.7059, Pure Ratio2 9.7529 +Epoch [61/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.6765, Pure Ratio2 9.7092 +Epoch [61/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.6919, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 30.0180 % Model2 29.9279 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7255, Pure Ratio2 9.5686 +Epoch [62/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9804, Pure Ratio2 9.8137 +Epoch [62/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8497, Pure Ratio2 9.6993 +Epoch [62/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8775, Pure Ratio2 9.7745 +Epoch [62/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.8275, Pure Ratio2 9.6745 +Epoch [62/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8889, Pure Ratio2 9.7386 +Epoch [62/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8207, Pure Ratio2 9.6863 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 29.7676 % Model2 29.4772 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.7134 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.4314, Pure Ratio2 9.2745 +Epoch [63/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.6667, Pure Ratio2 9.6471 +Epoch [63/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.5033, Pure Ratio2 9.5621 +Epoch [63/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.6765, Pure Ratio2 9.6912 +Epoch [63/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.8235, Pure Ratio2 9.8078 +Epoch [63/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.6961, Pure Ratio2 9.6699 +Epoch [63/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7171, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 29.4772 % Model2 28.8261 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7285 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.3333 +Epoch [64/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.7451, Pure Ratio2 9.6667 +Epoch [64/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.6405, Pure Ratio2 9.6536 +Epoch [64/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.7010, Pure Ratio2 9.6618 +Epoch [64/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8000, Pure Ratio2 9.7373 +Epoch [64/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.8595, Pure Ratio2 9.7974 +Epoch [64/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.7787, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 29.2768 % Model2 30.2885 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.7059 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.0196, Pure Ratio2 9.1569 +Epoch [65/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.3235, Pure Ratio2 9.3824 +Epoch [65/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.3529, Pure Ratio2 9.3137 +Epoch [65/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.6225, Pure Ratio2 9.6373 +Epoch [65/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5961, Pure Ratio2 9.6000 +Epoch [65/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.6634, Pure Ratio2 9.7059 +Epoch [65/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7283, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 29.1266 % Model2 30.0982 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 10.5882, Pure Ratio2 10.6667 +Epoch [66/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [66/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0131, Pure Ratio2 10.0980 +Epoch [66/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.8137, Pure Ratio2 9.8627 +Epoch [66/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.7882, Pure Ratio2 9.8745 +Epoch [66/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.7190, Pure Ratio2 9.7843 +Epoch [66/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.7479, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 30.1883 % Model2 29.8778 %, Pure Ratio 1 9.7084 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [67/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [67/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.8170 +Epoch [67/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.8873, Pure Ratio2 9.9804 +Epoch [67/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.6706, Pure Ratio2 9.7412 +Epoch [67/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7876, Pure Ratio2 9.8268 +Epoch [67/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0018, Loss2: 0.0006, Pure Ratio1: 9.7563, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 28.5156 % Model2 30.3085 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0017, Pure Ratio1: 10.0980, Pure Ratio2 9.9608 +Epoch [68/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.6569, Pure Ratio2 9.4608 +Epoch [68/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.6928, Pure Ratio2 9.5752 +Epoch [68/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.7157, Pure Ratio2 9.5931 +Epoch [68/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.7176, Pure Ratio2 9.6471 +Epoch [68/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8105, Pure Ratio2 9.7157 +Epoch [68/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.7535, Pure Ratio2 9.6695 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 29.8778 % Model2 30.5889 %, Pure Ratio 1 9.7486 %, Pure Ratio 2 9.6757 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0016, Pure Ratio1: 9.3725, Pure Ratio2 9.6078 +Epoch [69/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 85.1562, Loss1: 0.0017, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.7059 +Epoch [69/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.7647, Pure Ratio2 9.8627 +Epoch [69/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8578, Pure Ratio2 9.9657 +Epoch [69/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.7451, Pure Ratio2 9.8196 +Epoch [69/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7157, Pure Ratio2 9.8039 +Epoch [69/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0019, Pure Ratio1: 9.6667, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 30.0781 % Model2 30.8594 %, Pure Ratio 1 9.7034 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0021, Loss2: 0.0010, Pure Ratio1: 10.1176, Pure Ratio2 10.2549 +Epoch [70/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [70/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.7843, Pure Ratio2 9.8235 +Epoch [70/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.9363, Pure Ratio2 9.9363 +Epoch [70/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8824, Pure Ratio2 9.9137 +Epoch [70/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.8203, Pure Ratio2 9.8105 +Epoch [70/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8067, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 29.0665 % Model2 30.0581 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.7059 +Epoch [71/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0098, Pure Ratio2 9.9706 +Epoch [71/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0458, Pure Ratio2 10.0131 +Epoch [71/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8578 +Epoch [71/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8706, Pure Ratio2 9.8745 +Epoch [71/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7418, Pure Ratio2 9.7484 +Epoch [71/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.6891, Pure Ratio2 9.6807 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 29.8077 % Model2 29.7376 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [72/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7745, Pure Ratio2 9.7353 +Epoch [72/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8758, Pure Ratio2 9.7843 +Epoch [72/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8873, Pure Ratio2 9.7647 +Epoch [72/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.0039, Pure Ratio2 9.8431 +Epoch [72/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8660, Pure Ratio2 9.6928 +Epoch [72/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.7703, Pure Ratio2 9.6162 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 29.2568 % Model2 29.8978 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.6606 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.2549, Pure Ratio2 9.3922 +Epoch [73/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.5490, Pure Ratio2 9.5000 +Epoch [73/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0037, Pure Ratio1: 9.7190, Pure Ratio2 9.6144 +Epoch [73/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0018, Loss2: 0.0008, Pure Ratio1: 9.8039, Pure Ratio2 9.7353 +Epoch [73/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 9.7373, Pure Ratio2 9.6863 +Epoch [73/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6961, Pure Ratio2 9.6601 +Epoch [73/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0023, Pure Ratio1: 9.8263, Pure Ratio2 9.7591 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 28.9463 % Model2 29.0365 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.2745, Pure Ratio2 9.3137 +Epoch [74/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.5196, Pure Ratio2 9.5490 +Epoch [74/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0018, Pure Ratio1: 9.5098, Pure Ratio2 9.6078 +Epoch [74/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 9.6520, Pure Ratio2 9.7206 +Epoch [74/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.7922 +Epoch [74/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.6242, Pure Ratio2 9.7059 +Epoch [74/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.6415, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 29.5072 % Model2 30.2784 %, Pure Ratio 1 9.7386 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.2941, Pure Ratio2 10.0392 +Epoch [75/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9118, Pure Ratio2 9.5490 +Epoch [75/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 10.0327, Pure Ratio2 9.7320 +Epoch [75/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 9.8480 +Epoch [75/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.9765, Pure Ratio2 9.8157 +Epoch [75/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.9673, Pure Ratio2 9.8333 +Epoch [75/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.8543, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 28.4255 % Model2 31.5705 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.6732 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.3333, Pure Ratio2 10.2941 +Epoch [76/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.8235, Pure Ratio2 9.8725 +Epoch [76/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8170, Pure Ratio2 9.8301 +Epoch [76/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.7794, Pure Ratio2 9.7990 +Epoch [76/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8000, Pure Ratio2 9.8157 +Epoch [76/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8072, Pure Ratio2 9.8072 +Epoch [76/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.7731, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 29.6174 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.9608 +Epoch [77/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 9.8235, Pure Ratio2 9.8922 +Epoch [77/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.5098, Pure Ratio2 9.5882 +Epoch [77/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.8480 +Epoch [77/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8471, Pure Ratio2 9.8902 +Epoch [77/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0016, Pure Ratio1: 9.8529, Pure Ratio2 9.8954 +Epoch [77/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8852, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 28.5757 % Model2 28.9964 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.1176, Pure Ratio2 9.1373 +Epoch [78/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [78/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8562, Pure Ratio2 9.9477 +Epoch [78/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7892, Pure Ratio2 9.8529 +Epoch [78/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 9.7098, Pure Ratio2 9.7765 +Epoch [78/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6536, Pure Ratio2 9.7549 +Epoch [78/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 9.6667, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 28.8462 % Model2 30.5589 %, Pure Ratio 1 9.6782 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.5294 +Epoch [79/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.4902, Pure Ratio2 9.4216 +Epoch [79/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6928, Pure Ratio2 9.7124 +Epoch [79/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7745, Pure Ratio2 9.7255 +Epoch [79/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.6902, Pure Ratio2 9.6353 +Epoch [79/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.6438, Pure Ratio2 9.5556 +Epoch [79/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0017, Loss2: 0.0009, Pure Ratio1: 9.7003, Pure Ratio2 9.6611 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 28.8562 % Model2 29.0865 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7109 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.6275 +Epoch [80/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.3627, Pure Ratio2 9.3529 +Epoch [80/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 9.3529, Pure Ratio2 9.3333 +Epoch [80/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8088, Pure Ratio2 9.6765 +Epoch [80/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7608, Pure Ratio2 9.5804 +Epoch [80/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7974, Pure Ratio2 9.6536 +Epoch [80/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.8263, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 28.7460 % Model2 28.4455 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.3529, Pure Ratio2 9.5882 +Epoch [81/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.6667 +Epoch [81/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 9.5882, Pure Ratio2 9.6340 +Epoch [81/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.3235, Pure Ratio2 9.4608 +Epoch [81/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.4549, Pure Ratio2 9.6157 +Epoch [81/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0001, Pure Ratio1: 9.3791, Pure Ratio2 9.5752 +Epoch [81/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.5350, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 29.4671 % Model2 28.8962 %, Pure Ratio 1 9.6531 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7059, Pure Ratio2 9.4314 +Epoch [82/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.2255, Pure Ratio2 9.0294 +Epoch [82/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.3137, Pure Ratio2 9.1373 +Epoch [82/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.4706, Pure Ratio2 9.3284 +Epoch [82/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.5922, Pure Ratio2 9.4863 +Epoch [82/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5196, Pure Ratio2 9.4869 +Epoch [82/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.6583, Pure Ratio2 9.6218 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 28.6859 % Model2 28.7360 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7034 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.3725, Pure Ratio2 9.7059 +Epoch [83/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.9118 +Epoch [83/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.5948, Pure Ratio2 9.7059 +Epoch [83/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.5294, Pure Ratio2 9.5637 +Epoch [83/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.6078, Pure Ratio2 9.6431 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.6340, Pure Ratio2 9.7288 +Epoch [83/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.6527, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 30.6490 % Model2 30.2684 %, Pure Ratio 1 9.6757 %, Pure Ratio 2 9.7763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.5882 +Epoch [84/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0980, Pure Ratio2 9.9608 +Epoch [84/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0327, Pure Ratio2 9.8824 +Epoch [84/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8088, Pure Ratio2 9.7010 +Epoch [84/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8784, Pure Ratio2 9.7647 +Epoch [84/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.6699 +Epoch [84/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8543, Pure Ratio2 9.7003 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 29.2368 % Model2 28.7961 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.7210 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [85/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9706, Pure Ratio2 9.8137 +Epoch [85/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9935, Pure Ratio2 9.9412 +Epoch [85/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.7794, Pure Ratio2 9.7549 +Epoch [85/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8275, Pure Ratio2 9.8000 +Epoch [85/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.8595, Pure Ratio2 9.8464 +Epoch [85/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.7647, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 30.3486 % Model2 30.7792 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.1765, Pure Ratio2 9.3137 +Epoch [86/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.3922, Pure Ratio2 9.3627 +Epoch [86/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.4183, Pure Ratio2 9.3660 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.4020, Pure Ratio2 9.4167 +Epoch [86/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.6078, Pure Ratio2 9.6353 +Epoch [86/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 9.5882, Pure Ratio2 9.6111 +Epoch [86/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6835, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 29.5172 % Model2 28.8161 %, Pure Ratio 1 9.7486 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 8.9608, Pure Ratio2 9.2353 +Epoch [87/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.4412, Pure Ratio2 9.5784 +Epoch [87/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5033, Pure Ratio2 9.6797 +Epoch [87/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.9688, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.5343, Pure Ratio2 9.6863 +Epoch [87/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6510, Pure Ratio2 9.7882 +Epoch [87/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5719, Pure Ratio2 9.6732 +Epoch [87/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0015, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 29.1166 % Model2 30.4988 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [88/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0015, Pure Ratio1: 9.5392, Pure Ratio2 9.4608 +Epoch [88/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.5817, Pure Ratio2 9.5359 +Epoch [88/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.5539, Pure Ratio2 9.5441 +Epoch [88/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.6118, Pure Ratio2 9.6471 +Epoch [88/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6307, Pure Ratio2 9.6307 +Epoch [88/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6835, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 28.6258 % Model2 28.8261 %, Pure Ratio 1 9.7185 %, Pure Ratio 2 9.7260 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9020 +Epoch [89/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.6373, Pure Ratio2 9.5588 +Epoch [89/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.7974 +Epoch [89/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9265, Pure Ratio2 9.8333 +Epoch [89/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8157, Pure Ratio2 9.7529 +Epoch [89/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6732, Pure Ratio2 9.6405 +Epoch [89/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.7143, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 29.4571 % Model2 30.2183 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3529, Pure Ratio2 10.3333 +Epoch [90/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.3431, Pure Ratio2 10.2157 +Epoch [90/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2810, Pure Ratio2 10.2092 +Epoch [90/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0784, Pure Ratio2 9.9461 +Epoch [90/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9843, Pure Ratio2 9.8863 +Epoch [90/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9902, Pure Ratio2 9.8954 +Epoch [90/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9328, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 29.6274 % Model2 28.7260 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [91/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [91/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 9.9346 +Epoch [91/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8676, Pure Ratio2 9.8137 +Epoch [91/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8353, Pure Ratio2 9.8745 +Epoch [91/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.7582 +Epoch [91/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7759, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 29.9079 % Model2 29.4671 %, Pure Ratio 1 9.7587 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2157, Pure Ratio2 10.2157 +Epoch [92/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 10.1765, Pure Ratio2 10.1667 +Epoch [92/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.9281 +Epoch [92/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.8137 +Epoch [92/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7843, Pure Ratio2 9.7765 +Epoch [92/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7320, Pure Ratio2 9.7451 +Epoch [92/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6891, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 30.8694 % Model2 30.6591 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.8627 +Epoch [93/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.7255 +Epoch [93/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7386, Pure Ratio2 9.7712 +Epoch [93/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.6765 +Epoch [93/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7333, Pure Ratio2 9.7490 +Epoch [93/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7222, Pure Ratio2 9.7320 +Epoch [93/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7675, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 28.2652 % Model2 29.7877 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0025, Pure Ratio1: 9.1961, Pure Ratio2 9.3725 +Epoch [94/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.3431, Pure Ratio2 9.5588 +Epoch [94/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.4444, Pure Ratio2 9.5294 +Epoch [94/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.3676, Pure Ratio2 9.4706 +Epoch [94/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.5294, Pure Ratio2 9.6118 +Epoch [94/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.5686, Pure Ratio2 9.6275 +Epoch [94/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.6387, Pure Ratio2 9.6499 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 28.7059 % Model2 29.8277 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.7059 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1373 +Epoch [95/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.2451 +Epoch [95/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 10.0654 +Epoch [95/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 10.0980, Pure Ratio2 10.1471 +Epoch [95/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.9059 +Epoch [95/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7418, Pure Ratio2 9.8203 +Epoch [95/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7591, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 28.3053 % Model2 28.3053 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.1961, Pure Ratio2 9.3137 +Epoch [96/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.6176, Pure Ratio2 9.6078 +Epoch [96/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6144, Pure Ratio2 9.5817 +Epoch [96/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5637, Pure Ratio2 9.5294 +Epoch [96/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4863, Pure Ratio2 9.4706 +Epoch [96/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.6405 +Epoch [96/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7759, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 28.9263 % Model2 28.4255 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [97/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.8137 +Epoch [97/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8301 +Epoch [97/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 9.8578 +Epoch [97/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8902, Pure Ratio2 9.8118 +Epoch [97/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7386, Pure Ratio2 9.6536 +Epoch [97/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7283, Pure Ratio2 9.6835 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 30.0982 % Model2 29.9279 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.7084 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.3725, Pure Ratio2 9.4902 +Epoch [98/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.6373 +Epoch [98/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [98/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7157, Pure Ratio2 9.6176 +Epoch [98/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5725, Pure Ratio2 9.5922 +Epoch [98/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.7190, Pure Ratio2 9.7026 +Epoch [98/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6807, Pure Ratio2 9.6499 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 28.6959 % Model2 29.1466 %, Pure Ratio 1 9.6757 %, Pure Ratio 2 9.6254 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 9.9412 +Epoch [99/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.1373 +Epoch [99/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0654, Pure Ratio2 10.0196 +Epoch [99/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.9657, Pure Ratio2 9.9314 +Epoch [99/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9882, Pure Ratio2 9.9451 +Epoch [99/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8987, Pure Ratio2 9.8987 +Epoch [99/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.8207, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 29.1667 % Model2 29.4571 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7084 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [100/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1961, Pure Ratio2 10.0490 +Epoch [100/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9281, Pure Ratio2 9.8235 +Epoch [100/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.8333, Pure Ratio2 9.6912 +Epoch [100/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.6784 +Epoch [100/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7778, Pure Ratio2 9.6863 +Epoch [100/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7899, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 29.0064 % Model2 29.6875 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7109 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [101/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0882 +Epoch [101/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [101/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 10.0098 +Epoch [101/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 90.6250, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8549 +Epoch [101/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.7418, Pure Ratio2 9.7647 +Epoch [101/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8151, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 29.0865 % Model2 28.1951 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.6078, Pure Ratio2 9.8039 +Epoch [102/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4314, Pure Ratio2 9.6078 +Epoch [102/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.4967, Pure Ratio2 9.6863 +Epoch [102/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.8039 +Epoch [102/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6196, Pure Ratio2 9.7137 +Epoch [102/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6340, Pure Ratio2 9.6601 +Epoch [102/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7367, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 29.1266 % Model2 30.5789 %, Pure Ratio 1 9.7059 %, Pure Ratio 2 9.7084 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.1765 +Epoch [103/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.3431 +Epoch [103/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8824, Pure Ratio2 9.5752 +Epoch [103/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8333, Pure Ratio2 9.5735 +Epoch [103/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.5137 +Epoch [103/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.8399, Pure Ratio2 9.6275 +Epoch [103/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.6555 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 29.1767 % Model2 30.0982 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.6556 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 8.8824, Pure Ratio2 8.8431 +Epoch [104/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.1078, Pure Ratio2 9.1667 +Epoch [104/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.2418, Pure Ratio2 9.2941 +Epoch [104/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.5196, Pure Ratio2 9.5539 +Epoch [104/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6980, Pure Ratio2 9.6941 +Epoch [104/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.6928, Pure Ratio2 9.6895 +Epoch [104/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.7535, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 28.8261 % Model2 29.4171 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8039 +Epoch [105/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4020, Pure Ratio2 9.3627 +Epoch [105/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.6013, Pure Ratio2 9.5621 +Epoch [105/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5833, Pure Ratio2 9.5098 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.7373, Pure Ratio2 9.6353 +Epoch [105/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6961, Pure Ratio2 9.6471 +Epoch [105/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7143, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 29.5773 % Model2 30.3886 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7134 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.2941, Pure Ratio2 9.8431 +Epoch [106/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 10.1078 +Epoch [106/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7255, Pure Ratio2 9.8758 +Epoch [106/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7990, Pure Ratio2 9.9167 +Epoch [106/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7725, Pure Ratio2 9.8941 +Epoch [106/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9608 +Epoch [106/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8543, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 28.8562 % Model2 29.5473 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1176 +Epoch [107/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [107/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6928, Pure Ratio2 9.6275 +Epoch [107/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7892, Pure Ratio2 9.7255 +Epoch [107/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7922, Pure Ratio2 9.7294 +Epoch [107/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7026, Pure Ratio2 9.6634 +Epoch [107/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7507, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 29.8377 % Model2 29.6274 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.3922, Pure Ratio2 9.6667 +Epoch [108/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8627 +Epoch [108/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6013, Pure Ratio2 9.7582 +Epoch [108/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5637, Pure Ratio2 9.7108 +Epoch [108/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6118, Pure Ratio2 9.7098 +Epoch [108/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6928, Pure Ratio2 9.7810 +Epoch [108/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7423, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 28.3854 % Model2 30.1883 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [109/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9706, Pure Ratio2 9.8333 +Epoch [109/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 9.9346 +Epoch [109/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9951, Pure Ratio2 9.9216 +Epoch [109/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 9.9451 +Epoch [109/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8595 +Epoch [109/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8319, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 29.3970 % Model2 29.2268 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0021, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.4314 +Epoch [110/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.8725 +Epoch [110/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9673 +Epoch [110/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 9.8971 +Epoch [110/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.8863 +Epoch [110/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [110/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8291, Pure Ratio2 9.7227 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 29.5072 % Model2 28.5457 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.1765, Pure Ratio2 9.0000 +Epoch [111/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5392, Pure Ratio2 9.5000 +Epoch [111/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7974, Pure Ratio2 9.7320 +Epoch [111/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8775, Pure Ratio2 9.8088 +Epoch [111/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8275, Pure Ratio2 9.7647 +Epoch [111/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8660, Pure Ratio2 9.8562 +Epoch [111/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 29.6074 % Model2 29.7175 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.1569 +Epoch [112/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 9.8039 +Epoch [112/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 9.8562 +Epoch [112/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9461, Pure Ratio2 9.7892 +Epoch [112/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9373, Pure Ratio2 9.7647 +Epoch [112/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8791, Pure Ratio2 9.7386 +Epoch [112/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8515, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 29.2768 % Model2 29.0665 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.6983 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7647 +Epoch [113/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8725, Pure Ratio2 9.9412 +Epoch [113/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.9020 +Epoch [113/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6814, Pure Ratio2 9.8186 +Epoch [113/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.7529 +Epoch [113/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7680, Pure Ratio2 9.8954 +Epoch [113/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7787, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 29.5873 % Model2 29.2869 %, Pure Ratio 1 9.6858 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.2745, Pure Ratio2 9.5490 +Epoch [114/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.1275, Pure Ratio2 9.0588 +Epoch [114/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.2745, Pure Ratio2 9.2288 +Epoch [114/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5245, Pure Ratio2 9.4167 +Epoch [114/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.5647 +Epoch [114/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0012, Pure Ratio1: 9.7092, Pure Ratio2 9.6405 +Epoch [114/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.7451, Pure Ratio2 9.6779 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 29.4671 % Model2 28.2252 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.6833 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7059 +Epoch [115/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7745 +Epoch [115/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 9.8366 +Epoch [115/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8088, Pure Ratio2 9.7990 +Epoch [115/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7412, Pure Ratio2 9.7451 +Epoch [115/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8105, Pure Ratio2 9.8399 +Epoch [115/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8403, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 29.3670 % Model2 28.6959 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.7662 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.0784, Pure Ratio2 9.1373 +Epoch [116/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8137, Pure Ratio2 9.8039 +Epoch [116/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.4837, Pure Ratio2 9.4641 +Epoch [116/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6029, Pure Ratio2 9.6275 +Epoch [116/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6196, Pure Ratio2 9.6353 +Epoch [116/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.6405 +Epoch [116/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6779, Pure Ratio2 9.6611 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 29.1266 % Model2 31.6306 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.7843 +Epoch [117/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.4804 +Epoch [117/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.6732 +Epoch [117/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.7157 +Epoch [117/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5569, Pure Ratio2 9.5882 +Epoch [117/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.6634 +Epoch [117/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7003, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 29.4872 % Model2 30.3085 %, Pure Ratio 1 9.7059 %, Pure Ratio 2 9.7235 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 8.9608, Pure Ratio2 8.8824 +Epoch [118/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5392, Pure Ratio2 9.4216 +Epoch [118/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7386, Pure Ratio2 9.6928 +Epoch [118/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.6912 +Epoch [118/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7529 +Epoch [118/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.7647 +Epoch [118/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7927, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 29.3069 % Model2 29.4671 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7134 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.9412 +Epoch [119/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0588 +Epoch [119/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.9935 +Epoch [119/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6961, Pure Ratio2 9.8529 +Epoch [119/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.8784 +Epoch [119/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7353, Pure Ratio2 9.8954 +Epoch [119/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7171, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 28.9964 % Model2 28.9062 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4510, Pure Ratio2 9.2745 +Epoch [120/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.5784 +Epoch [120/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7190, Pure Ratio2 9.5817 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7549, Pure Ratio2 9.6275 +Epoch [120/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7020, Pure Ratio2 9.5843 +Epoch [120/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6961, Pure Ratio2 9.6046 +Epoch [120/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6751, Pure Ratio2 9.6443 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 27.8345 % Model2 29.8678 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7109 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.2157, Pure Ratio2 9.1373 +Epoch [121/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.3039 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.3007, Pure Ratio2 9.2484 +Epoch [121/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.3873, Pure Ratio2 9.2598 +Epoch [121/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4824, Pure Ratio2 9.3569 +Epoch [121/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7124, Pure Ratio2 9.6536 +Epoch [121/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7871, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 28.9263 % Model2 30.2083 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.9020 +Epoch [122/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [122/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.9150 +Epoch [122/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.7157 +Epoch [122/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6431, Pure Ratio2 9.7451 +Epoch [122/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6895, Pure Ratio2 9.7614 +Epoch [122/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7815, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 29.9479 % Model2 28.3854 %, Pure Ratio 1 9.7386 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8235 +Epoch [123/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8627 +Epoch [123/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6340, Pure Ratio2 9.6209 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.6912 +Epoch [123/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7686, Pure Ratio2 9.7451 +Epoch [123/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.6863 +Epoch [123/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8095, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 28.7760 % Model2 28.5056 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.6471 +Epoch [124/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5490 +Epoch [124/200], Iter [150/390] Training Accuracy1: 96.0938, Training Accuracy2: 96.0938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6405, Pure Ratio2 9.6601 +Epoch [124/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7206, Pure Ratio2 9.7157 +Epoch [124/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6980, Pure Ratio2 9.6941 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.8072 +Epoch [124/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 29.3570 % Model2 29.5873 %, Pure Ratio 1 9.7059 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [125/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.6961 +Epoch [125/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.9739 +Epoch [125/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0245 +Epoch [125/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9686 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8497, Pure Ratio2 9.9314 +Epoch [125/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.6891, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 28.1050 % Model2 29.8578 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8431 +Epoch [126/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.3824 +Epoch [126/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3464, Pure Ratio2 9.2418 +Epoch [126/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5392, Pure Ratio2 9.3971 +Epoch [126/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5804, Pure Ratio2 9.5098 +Epoch [126/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6209, Pure Ratio2 9.6013 +Epoch [126/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6639, Pure Ratio2 9.6303 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 29.3470 % Model2 29.1967 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 9.9804 +Epoch [127/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8039 +Epoch [127/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9085, Pure Ratio2 9.7516 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1471, Pure Ratio2 10.0441 +Epoch [127/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8392 +Epoch [127/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8072, Pure Ratio2 9.7059 +Epoch [127/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.8011, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 28.7360 % Model2 30.0581 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 8.5686, Pure Ratio2 8.8627 +Epoch [128/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5490, Pure Ratio2 9.6569 +Epoch [128/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7778, Pure Ratio2 9.8824 +Epoch [128/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9559, Pure Ratio2 10.0000 +Epoch [128/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8196 +Epoch [128/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 9.8954 +Epoch [128/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 29.8177 % Model2 29.6875 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6667 +Epoch [129/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.4804, Pure Ratio2 9.3627 +Epoch [129/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.4118 +Epoch [129/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.6176 +Epoch [129/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.7922 +Epoch [129/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6993, Pure Ratio2 9.6536 +Epoch [129/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7507, Pure Ratio2 9.6975 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 29.1767 % Model2 29.3470 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.6833 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [130/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.7353 +Epoch [130/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7974 +Epoch [130/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.7010 +Epoch [130/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8196, Pure Ratio2 9.8667 +Epoch [130/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7810, Pure Ratio2 9.8366 +Epoch [130/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7115, Pure Ratio2 9.7367 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 28.5156 % Model2 29.9179 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 8.9412, Pure Ratio2 8.9412 +Epoch [131/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3431, Pure Ratio2 9.3333 +Epoch [131/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.5490 +Epoch [131/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6716, Pure Ratio2 9.5931 +Epoch [131/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6706, Pure Ratio2 9.5608 +Epoch [131/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8072, Pure Ratio2 9.6928 +Epoch [131/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 30.4988 % Model2 30.2885 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.0588, Pure Ratio2 9.3725 +Epoch [132/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.8137 +Epoch [132/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.8301 +Epoch [132/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9559 +Epoch [132/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9098 +Epoch [132/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6993, Pure Ratio2 9.7941 +Epoch [132/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7115, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 29.0164 % Model2 28.6458 %, Pure Ratio 1 9.6908 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1373 +Epoch [133/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7549 +Epoch [133/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6601 +Epoch [133/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.6275 +Epoch [133/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5843 +Epoch [133/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6078, Pure Ratio2 9.6699 +Epoch [133/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0023, Pure Ratio1: 9.6779, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 29.9279 % Model2 30.0881 %, Pure Ratio 1 9.6456 %, Pure Ratio 2 9.7285 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1961 +Epoch [134/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2059, Pure Ratio2 10.0882 +Epoch [134/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7712 +Epoch [134/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.7206 +Epoch [134/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8000, Pure Ratio2 9.7725 +Epoch [134/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7418, Pure Ratio2 9.7386 +Epoch [134/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7143, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 30.5288 % Model2 29.8077 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 8.9804, Pure Ratio2 8.8824 +Epoch [135/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.5000 +Epoch [135/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8954, Pure Ratio2 9.8889 +Epoch [135/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6765, Pure Ratio2 9.6078 +Epoch [135/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6235 +Epoch [135/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7026, Pure Ratio2 9.6013 +Epoch [135/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.6835, Pure Ratio2 9.5910 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 29.5773 % Model2 29.3770 %, Pure Ratio 1 9.7185 %, Pure Ratio 2 9.6405 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.1961 +Epoch [136/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 10.0588 +Epoch [136/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.8105 +Epoch [136/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5735, Pure Ratio2 9.7696 +Epoch [136/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5529, Pure Ratio2 9.6980 +Epoch [136/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5425, Pure Ratio2 9.6863 +Epoch [136/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6022, Pure Ratio2 9.7367 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 28.8462 % Model2 29.6374 %, Pure Ratio 1 9.6782 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.5294 +Epoch [137/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.0392 +Epoch [137/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.7908 +Epoch [137/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7255 +Epoch [137/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7882, Pure Ratio2 9.7686 +Epoch [137/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7745 +Epoch [137/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8263, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 28.0349 % Model2 29.4071 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.2353 +Epoch [138/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8333 +Epoch [138/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.8235 +Epoch [138/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6716, Pure Ratio2 9.7598 +Epoch [138/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.8745 +Epoch [138/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6438, Pure Ratio2 9.7288 +Epoch [138/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7563, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 29.7776 % Model2 29.3570 %, Pure Ratio 1 9.7109 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.2745 +Epoch [139/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7941 +Epoch [139/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.7647, Pure Ratio2 9.7320 +Epoch [139/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [139/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7412, Pure Ratio2 9.7804 +Epoch [139/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7451 +Epoch [139/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7283, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 29.0565 % Model2 29.0164 %, Pure Ratio 1 9.7360 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.6275 +Epoch [140/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8333 +Epoch [140/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7908, Pure Ratio2 9.8627 +Epoch [140/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6814, Pure Ratio2 9.8382 +Epoch [140/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8078 +Epoch [140/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.6373, Pure Ratio2 9.7647 +Epoch [140/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5966, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 29.3069 % Model2 29.3970 %, Pure Ratio 1 9.6833 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 8.9412, Pure Ratio2 8.9216 +Epoch [141/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.0294 +Epoch [141/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.5621, Pure Ratio2 9.4641 +Epoch [141/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5735, Pure Ratio2 9.5098 +Epoch [141/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7059 +Epoch [141/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6405, Pure Ratio2 9.6634 +Epoch [141/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7395, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 30.2484 % Model2 30.2584 %, Pure Ratio 1 9.7386 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9020 +Epoch [142/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9412 +Epoch [142/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [142/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8088, Pure Ratio2 9.8137 +Epoch [142/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7569, Pure Ratio2 9.7529 +Epoch [142/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.7680 +Epoch [142/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7311, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 30.0180 % Model2 30.2083 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.7647 +Epoch [143/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6176 +Epoch [143/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.5098 +Epoch [143/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.5833 +Epoch [143/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7804, Pure Ratio2 9.6902 +Epoch [143/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0022, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7157 +Epoch [143/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 29.8578 % Model2 30.8894 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.5686, Pure Ratio2 10.5490 +Epoch [144/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1471 +Epoch [144/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.8562 +Epoch [144/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7059 +Epoch [144/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6627, Pure Ratio2 9.6863 +Epoch [144/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.6340 +Epoch [144/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7199, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 30.3886 % Model2 30.0881 %, Pure Ratio 1 9.7059 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 10.0196 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8922 +Epoch [145/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.7974 +Epoch [145/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.7696 +Epoch [145/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7098, Pure Ratio2 9.7529 +Epoch [145/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6797, Pure Ratio2 9.6961 +Epoch [145/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6947, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 29.7977 % Model2 29.3770 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.5882 +Epoch [146/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8725 +Epoch [146/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.6209 +Epoch [146/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7402 +Epoch [146/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.8314 +Epoch [146/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.7941 +Epoch [146/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 29.0966 % Model2 30.3486 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 10.0588 +Epoch [147/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7843 +Epoch [147/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.7059 +Epoch [147/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6716, Pure Ratio2 9.7500 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6392, Pure Ratio2 9.6863 +Epoch [147/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7778 +Epoch [147/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7591, Pure Ratio2 9.7367 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 29.6875 % Model2 30.9295 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5686 +Epoch [148/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.6667 +Epoch [148/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7582 +Epoch [148/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.7010, Pure Ratio2 9.7353 +Epoch [148/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6235, Pure Ratio2 9.6706 +Epoch [148/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6307, Pure Ratio2 9.6340 +Epoch [148/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 29.4271 % Model2 30.2985 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4902, Pure Ratio2 9.5882 +Epoch [149/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5392, Pure Ratio2 9.6765 +Epoch [149/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 95.3125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.9150 +Epoch [149/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.6863, Pure Ratio2 9.8480 +Epoch [149/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6824, Pure Ratio2 9.7843 +Epoch [149/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.7157 +Epoch [149/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6499, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 29.5873 % Model2 30.5889 %, Pure Ratio 1 9.6481 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.7843 +Epoch [150/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.7353 +Epoch [150/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [150/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8382 +Epoch [150/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.7725 +Epoch [150/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7745, Pure Ratio2 9.7418 +Epoch [150/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7563, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 29.1667 % Model2 29.6274 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.9020 +Epoch [151/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3824, Pure Ratio2 9.3137 +Epoch [151/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5163, Pure Ratio2 9.5033 +Epoch [151/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5980, Pure Ratio2 9.5882 +Epoch [151/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5137, Pure Ratio2 9.5490 +Epoch [151/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6209, Pure Ratio2 9.5850 +Epoch [151/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6499, Pure Ratio2 9.6359 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 30.3085 % Model2 30.6891 %, Pure Ratio 1 9.6983 %, Pure Ratio 2 9.6657 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2157, Pure Ratio2 9.5686 +Epoch [152/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0017, Pure Ratio1: 9.5784, Pure Ratio2 9.7549 +Epoch [152/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.9281 +Epoch [152/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6814, Pure Ratio2 9.7353 +Epoch [152/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5451, Pure Ratio2 9.5922 +Epoch [152/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.6732 +Epoch [152/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 30.2885 % Model2 30.0581 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.3137 +Epoch [153/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.5098 +Epoch [153/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5621, Pure Ratio2 9.6405 +Epoch [153/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.8039 +Epoch [153/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6706, Pure Ratio2 9.6902 +Epoch [153/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.6993 +Epoch [153/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6779, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 28.8161 % Model2 29.2568 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7210 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 9.8039 +Epoch [154/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.5980 +Epoch [154/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6797, Pure Ratio2 9.5359 +Epoch [154/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.6667 +Epoch [154/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6902 +Epoch [154/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6634 +Epoch [154/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8347, Pure Ratio2 9.7591 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 29.0765 % Model2 29.9379 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [155/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.4314, Pure Ratio2 9.4216 +Epoch [155/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3007, Pure Ratio2 9.4052 +Epoch [155/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.3824, Pure Ratio2 9.4412 +Epoch [155/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6157, Pure Ratio2 9.6549 +Epoch [155/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7222 +Epoch [155/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7087, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 29.6274 % Model2 29.5272 %, Pure Ratio 1 9.7009 %, Pure Ratio 2 9.7260 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.4706 +Epoch [156/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7549 +Epoch [156/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8693 +Epoch [156/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.7500 +Epoch [156/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6196, Pure Ratio2 9.7255 +Epoch [156/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5817, Pure Ratio2 9.6732 +Epoch [156/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6162, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 30.4587 % Model2 29.5573 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [157/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8725 +Epoch [157/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7582, Pure Ratio2 9.8301 +Epoch [157/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7255 +Epoch [157/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7098, Pure Ratio2 9.6941 +Epoch [157/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6340, Pure Ratio2 9.6340 +Epoch [157/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7535, Pure Ratio2 9.7199 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 29.6274 % Model2 30.7292 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.9216, Pure Ratio2 9.0196 +Epoch [158/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6569 +Epoch [158/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3268, Pure Ratio2 9.3987 +Epoch [158/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3088, Pure Ratio2 9.2892 +Epoch [158/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5569, Pure Ratio2 9.4627 +Epoch [158/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.5000 +Epoch [158/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5378 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 29.9079 % Model2 29.6374 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.6807 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [159/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.8824 +Epoch [159/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.6797 +Epoch [159/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8775 +Epoch [159/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.7608 +Epoch [159/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.7778 +Epoch [159/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8123, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 29.1366 % Model2 29.4671 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.7059, Pure Ratio2 10.6078 +Epoch [160/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8922 +Epoch [160/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1307 +Epoch [160/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0392 +Epoch [160/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8314, Pure Ratio2 9.8000 +Epoch [160/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.7810 +Epoch [160/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6779 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 29.8978 % Model2 30.3986 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.7285 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.8039 +Epoch [161/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.9118 +Epoch [161/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6928, Pure Ratio2 9.8627 +Epoch [161/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.8382 +Epoch [161/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6706, Pure Ratio2 9.8549 +Epoch [161/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6830, Pure Ratio2 9.8137 +Epoch [161/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7115, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 30.0381 % Model2 30.6290 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3333 +Epoch [162/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [162/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0261 +Epoch [162/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8922 +Epoch [162/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 10.0039 +Epoch [162/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8529 +Epoch [162/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7283, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 30.2885 % Model2 29.7676 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0000, Pure Ratio2 8.8824 +Epoch [163/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2745, Pure Ratio2 9.2843 +Epoch [163/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5163, Pure Ratio2 9.5621 +Epoch [163/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4755, Pure Ratio2 9.5637 +Epoch [163/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5961, Pure Ratio2 9.6275 +Epoch [163/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6046, Pure Ratio2 9.6111 +Epoch [163/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6919, Pure Ratio2 9.7199 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 29.4171 % Model2 30.0881 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.7255 +Epoch [164/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.5000 +Epoch [164/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.6601 +Epoch [164/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5245 +Epoch [164/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6000, Pure Ratio2 9.5804 +Epoch [164/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6242 +Epoch [164/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6162 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 29.1567 % Model2 30.0280 %, Pure Ratio 1 9.6807 %, Pure Ratio 2 9.6858 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.8431 +Epoch [165/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7157 +Epoch [165/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.6144 +Epoch [165/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5833, Pure Ratio2 9.5931 +Epoch [165/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7529 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7026, Pure Ratio2 9.7386 +Epoch [165/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6695, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 30.1182 % Model2 29.9379 %, Pure Ratio 1 9.6933 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.4118 +Epoch [166/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8922 +Epoch [166/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7190 +Epoch [166/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7990 +Epoch [166/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8784, Pure Ratio2 9.7804 +Epoch [166/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.7320 +Epoch [166/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7731, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 29.8578 % Model2 30.0781 %, Pure Ratio 1 9.7360 %, Pure Ratio 2 9.7335 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.8627 +Epoch [167/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.8627 +Epoch [167/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8954 +Epoch [167/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8284 +Epoch [167/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7176 +Epoch [167/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7418 +Epoch [167/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7731, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 29.7576 % Model2 30.2684 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.5098 +Epoch [168/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7353 +Epoch [168/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.7647 +Epoch [168/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8676 +Epoch [168/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.9294 +Epoch [168/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9575 +Epoch [168/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 29.0966 % Model2 28.9764 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.3922 +Epoch [169/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0000 +Epoch [169/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8497 +Epoch [169/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7451 +Epoch [169/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6157 +Epoch [169/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6732 +Epoch [169/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7115, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 29.8177 % Model2 29.1967 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2745, Pure Ratio2 9.4314 +Epoch [170/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.7353 +Epoch [170/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5359, Pure Ratio2 9.6993 +Epoch [170/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6422, Pure Ratio2 9.7598 +Epoch [170/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.8941 +Epoch [170/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7026, Pure Ratio2 9.7974 +Epoch [170/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8571, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 29.4972 % Model2 30.0982 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7843 +Epoch [171/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9706 +Epoch [171/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8889 +Epoch [171/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.7696 +Epoch [171/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.7961 +Epoch [171/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7484 +Epoch [171/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7003, Pure Ratio2 9.6751 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 30.1282 % Model2 29.8377 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.6858 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.8235 +Epoch [172/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4902 +Epoch [172/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5359, Pure Ratio2 9.4641 +Epoch [172/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.4657 +Epoch [172/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.6235 +Epoch [172/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.6307 +Epoch [172/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 29.9880 % Model2 29.9780 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.7059 +Epoch [173/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.5392 +Epoch [173/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5229, Pure Ratio2 9.4575 +Epoch [173/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8725 +Epoch [173/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7765, Pure Ratio2 9.7961 +Epoch [173/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.7974 +Epoch [173/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7731, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 29.6474 % Model2 29.8077 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7647, Pure Ratio2 10.4902 +Epoch [174/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.8725 +Epoch [174/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.6471 +Epoch [174/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.7549 +Epoch [174/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7725, Pure Ratio2 9.7176 +Epoch [174/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7908, Pure Ratio2 9.7647 +Epoch [174/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7619, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 29.1767 % Model2 29.8377 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7260 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.4902 +Epoch [175/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7451 +Epoch [175/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.6471 +Epoch [175/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5588 +Epoch [175/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5569, Pure Ratio2 9.5804 +Epoch [175/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7876, Pure Ratio2 9.7647 +Epoch [175/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7535, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 30.0180 % Model2 30.2284 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.1961 +Epoch [176/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.0980 +Epoch [176/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3791, Pure Ratio2 9.3987 +Epoch [176/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6422, Pure Ratio2 9.6373 +Epoch [176/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6353, Pure Ratio2 9.5608 +Epoch [176/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6503, Pure Ratio2 9.6242 +Epoch [176/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6695 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 29.9279 % Model2 29.9279 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7210 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9608 +Epoch [177/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.1176 +Epoch [177/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9542 +Epoch [177/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.9804 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7647 +Epoch [177/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7810 +Epoch [177/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7563, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 29.2368 % Model2 29.5072 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6471 +Epoch [178/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3039 +Epoch [178/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0327 +Epoch [178/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8088 +Epoch [178/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6941, Pure Ratio2 9.7216 +Epoch [178/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.7582 +Epoch [178/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 29.3069 % Model2 30.0381 %, Pure Ratio 1 9.7185 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.5294 +Epoch [179/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [179/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8235 +Epoch [179/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7598 +Epoch [179/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8392 +Epoch [179/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8529 +Epoch [179/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8179, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 29.4972 % Model2 29.3369 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.6471, Pure Ratio2 8.7843 +Epoch [180/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.6667 +Epoch [180/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5817, Pure Ratio2 9.6536 +Epoch [180/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.6569 +Epoch [180/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.8039 +Epoch [180/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.8039 +Epoch [180/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7003, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 29.7276 % Model2 29.1967 %, Pure Ratio 1 9.6983 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.4314 +Epoch [181/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5196, Pure Ratio2 9.4902 +Epoch [181/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8105 +Epoch [181/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.6912 +Epoch [181/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9451 +Epoch [181/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.7124 +Epoch [181/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7535, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 29.7175 % Model2 30.1983 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.4510 +Epoch [182/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8137 +Epoch [182/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7778 +Epoch [182/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7451 +Epoch [182/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7922, Pure Ratio2 9.7961 +Epoch [182/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7320 +Epoch [182/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7143, Pure Ratio2 9.6975 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 28.5757 % Model2 30.3686 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2353, Pure Ratio2 9.2745 +Epoch [183/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7353 +Epoch [183/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.6732 +Epoch [183/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8676 +Epoch [183/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6941, Pure Ratio2 9.6745 +Epoch [183/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.6667 +Epoch [183/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7171, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 29.4872 % Model2 29.6875 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [184/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5588 +Epoch [184/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.6863 +Epoch [184/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.7206 +Epoch [184/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.7647 +Epoch [184/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8791 +Epoch [184/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 29.7676 % Model2 29.6675 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.6078 +Epoch [185/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0980 +Epoch [185/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0458 +Epoch [185/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8627 +Epoch [185/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.8196 +Epoch [185/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8889 +Epoch [185/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7703, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 30.3586 % Model2 30.0381 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [186/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7843 +Epoch [186/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [186/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6667 +Epoch [186/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8118, Pure Ratio2 9.7725 +Epoch [186/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.8105 +Epoch [186/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7339, Pure Ratio2 9.7199 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 29.6074 % Model2 29.7676 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7511 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5882 +Epoch [187/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.7059 +Epoch [187/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4183, Pure Ratio2 9.4641 +Epoch [187/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4167, Pure Ratio2 9.5000 +Epoch [187/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6157, Pure Ratio2 9.6667 +Epoch [187/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5719, Pure Ratio2 9.6078 +Epoch [187/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6359, Pure Ratio2 9.6695 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 29.3870 % Model2 29.3470 %, Pure Ratio 1 9.6833 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9804 +Epoch [188/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0490 +Epoch [188/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0000 +Epoch [188/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9559 +Epoch [188/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8196 +Epoch [188/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.8333 +Epoch [188/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 29.8077 % Model2 29.7676 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5882 +Epoch [189/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4804, Pure Ratio2 9.5882 +Epoch [189/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.7582 +Epoch [189/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.8725 +Epoch [189/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6392, Pure Ratio2 9.7490 +Epoch [189/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6438, Pure Ratio2 9.6895 +Epoch [189/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6751, Pure Ratio2 9.6863 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 29.7075 % Model2 29.4772 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.6471 +Epoch [190/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7745 +Epoch [190/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 9.9935 +Epoch [190/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0441, Pure Ratio2 9.9902 +Epoch [190/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9294 +Epoch [190/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8627 +Epoch [190/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 30.0180 % Model2 29.1867 %, Pure Ratio 1 9.7486 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.5294 +Epoch [191/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.6863 +Epoch [191/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8497 +Epoch [191/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.8578 +Epoch [191/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.8353 +Epoch [191/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.8007 +Epoch [191/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7339, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 30.7492 % Model2 29.7476 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.7159 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0588 +Epoch [192/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [192/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9150 +Epoch [192/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.7549 +Epoch [192/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.7216 +Epoch [192/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.8203 +Epoch [192/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 29.7877 % Model2 29.3870 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.0784 +Epoch [193/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0784 +Epoch [193/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1503 +Epoch [193/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8137 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8235 +Epoch [193/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7745 +Epoch [193/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7507, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 30.4487 % Model2 29.9279 %, Pure Ratio 1 9.7235 %, Pure Ratio 2 9.6782 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.5294 +Epoch [194/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.3431 +Epoch [194/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.2353 +Epoch [194/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.9804 +Epoch [194/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8784 +Epoch [194/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.9673 +Epoch [194/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 30.0080 % Model2 30.4087 %, Pure Ratio 1 9.7034 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9020 +Epoch [195/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4314 +Epoch [195/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.6601 +Epoch [195/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6814, Pure Ratio2 9.6863 +Epoch [195/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7373, Pure Ratio2 9.7608 +Epoch [195/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7124 +Epoch [195/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7619, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 30.0982 % Model2 29.6675 %, Pure Ratio 1 9.7386 %, Pure Ratio 2 9.7185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 9.9020 +Epoch [196/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.1176 +Epoch [196/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 9.8824 +Epoch [196/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7794, Pure Ratio2 9.5833 +Epoch [196/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.5647 +Epoch [196/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.6503 +Epoch [196/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 30.2384 % Model2 29.9579 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.7260 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [197/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0686 +Epoch [197/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0458 +Epoch [197/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.8431 +Epoch [197/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8353, Pure Ratio2 9.7882 +Epoch [197/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8268, Pure Ratio2 9.8562 +Epoch [197/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7731, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 30.4688 % Model2 29.9079 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.2941 +Epoch [198/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5882 +Epoch [198/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.6405 +Epoch [198/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.6618 +Epoch [198/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6549 +Epoch [198/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7974 +Epoch [198/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 29.9780 % Model2 29.0765 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.9020 +Epoch [199/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8529 +Epoch [199/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.7974 +Epoch [199/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.6520 +Epoch [199/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7412, Pure Ratio2 9.7020 +Epoch [199/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.7222 +Epoch [199/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6527, Pure Ratio2 9.6919 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 29.5673 % Model2 29.6575 %, Pure Ratio 1 9.7109 %, Pure Ratio 2 9.7562 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.1373 +Epoch [200/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.3137 +Epoch [200/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5621, Pure Ratio2 9.5490 +Epoch [200/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5931, Pure Ratio2 9.6176 +Epoch [200/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7098, Pure Ratio2 9.7176 +Epoch [200/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7614 +Epoch [200/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 29.9179 % Model2 30.2183 %, Pure Ratio 1 9.6858 %, Pure Ratio 2 9.7185 % diff --git a/other_methods/coteaching/coteaching_results/out_4_2.log b/other_methods/coteaching/coteaching_results/out_4_2.log new file mode 100644 index 0000000..a6cc8ba --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_4_2.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0159, Loss2: 0.0161, Pure Ratio1: 10.2240, Pure Ratio2 10.2080 +Epoch [2/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0141, Loss2: 0.0143, Pure Ratio1: 10.3360, Pure Ratio2 10.3200 +Epoch [2/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0133, Loss2: 0.0131, Pure Ratio1: 10.3253, Pure Ratio2 10.3040 +Epoch [2/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0126, Pure Ratio1: 10.3320, Pure Ratio2 10.3000 +Epoch [2/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0139, Pure Ratio1: 10.1856, Pure Ratio2 10.1504 +Epoch [2/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0123, Loss2: 0.0128, Pure Ratio1: 10.1307, Pure Ratio2 10.0880 +Epoch [2/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0117, Loss2: 0.0118, Pure Ratio1: 10.0571, Pure Ratio2 10.0160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 55.6290 % Model2 55.7091 %, Pure Ratio 1 10.0349 %, Pure Ratio 2 9.9959 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0105, Loss2: 0.0101, Pure Ratio1: 9.4426, Pure Ratio2 9.4754 +Epoch [3/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0117, Loss2: 0.0112, Pure Ratio1: 9.6393, Pure Ratio2 9.6721 +Epoch [3/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0122, Loss2: 0.0130, Pure Ratio1: 9.7596, Pure Ratio2 9.7869 +Epoch [3/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0117, Loss2: 0.0110, Pure Ratio1: 9.8074, Pure Ratio2 9.8115 +Epoch [3/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0118, Loss2: 0.0123, Pure Ratio1: 9.8984, Pure Ratio2 9.9016 +Epoch [3/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0116, Loss2: 0.0115, Pure Ratio1: 10.0164, Pure Ratio2 10.0219 +Epoch [3/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0108, Loss2: 0.0106, Pure Ratio1: 10.0375, Pure Ratio2 10.0468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 64.2829 % Model2 62.4900 %, Pure Ratio 1 9.9790 %, Pure Ratio 2 9.9916 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0098, Pure Ratio1: 9.9496, Pure Ratio2 9.7983 +Epoch [4/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0126, Loss2: 0.0125, Pure Ratio1: 9.9076, Pure Ratio2 9.8403 +Epoch [4/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0100, Loss2: 0.0100, Pure Ratio1: 9.9216, Pure Ratio2 9.8768 +Epoch [4/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0099, Loss2: 0.0095, Pure Ratio1: 9.8908, Pure Ratio2 9.8529 +Epoch [4/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0078, Loss2: 0.0089, Pure Ratio1: 9.9227, Pure Ratio2 9.9092 +Epoch [4/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0077, Loss2: 0.0077, Pure Ratio1: 9.9692, Pure Ratio2 9.9636 +Epoch [4/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0084, Loss2: 0.0080, Pure Ratio1: 9.9592, Pure Ratio2 9.9760 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 68.9203 % Model2 68.8702 %, Pure Ratio 1 9.9591 %, Pure Ratio 2 9.9763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0105, Loss2: 0.0107, Pure Ratio1: 9.9310, Pure Ratio2 9.8793 +Epoch [5/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0082, Loss2: 0.0088, Pure Ratio1: 10.0000, Pure Ratio2 9.9655 +Epoch [5/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0080, Loss2: 0.0081, Pure Ratio1: 10.2414, Pure Ratio2 10.2299 +Epoch [5/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0075, Loss2: 0.0080, Pure Ratio1: 10.2026, Pure Ratio2 10.1509 +Epoch [5/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0094, Loss2: 0.0096, Pure Ratio1: 10.2034, Pure Ratio2 10.1828 +Epoch [5/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0096, Loss2: 0.0091, Pure Ratio1: 10.0776, Pure Ratio2 10.0546 +Epoch [5/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0084, Loss2: 0.0084, Pure Ratio1: 10.1084, Pure Ratio2 10.0813 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 72.3157 % Model2 70.7732 %, Pure Ratio 1 10.0022 %, Pure Ratio 2 9.9867 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0087, Loss2: 0.0089, Pure Ratio1: 10.3717, Pure Ratio2 10.2301 +Epoch [6/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0087, Loss2: 0.0084, Pure Ratio1: 10.3097, Pure Ratio2 10.2566 +Epoch [6/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0062, Loss2: 0.0063, Pure Ratio1: 10.3186, Pure Ratio2 10.2832 +Epoch [6/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0066, Loss2: 0.0066, Pure Ratio1: 10.0664, Pure Ratio2 10.0841 +Epoch [6/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0092, Loss2: 0.0090, Pure Ratio1: 10.0779, Pure Ratio2 10.0885 +Epoch [6/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0092, Loss2: 0.0087, Pure Ratio1: 9.9086, Pure Ratio2 9.9440 +Epoch [6/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0050, Loss2: 0.0048, Pure Ratio1: 9.9368, Pure Ratio2 9.9722 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 75.6410 % Model2 74.9399 %, Pure Ratio 1 9.9637 %, Pure Ratio 2 9.9932 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0060, Loss2: 0.0059, Pure Ratio1: 10.3273, Pure Ratio2 10.2545 +Epoch [7/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0057, Loss2: 0.0050, Pure Ratio1: 10.1818, Pure Ratio2 10.1545 +Epoch [7/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0054, Loss2: 0.0058, Pure Ratio1: 10.4545, Pure Ratio2 10.4485 +Epoch [7/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0072, Loss2: 0.0075, Pure Ratio1: 10.1409, Pure Ratio2 10.1364 +Epoch [7/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0067, Loss2: 0.0058, Pure Ratio1: 10.1636, Pure Ratio2 10.1564 +Epoch [7/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0054, Loss2: 0.0058, Pure Ratio1: 10.0545, Pure Ratio2 10.0515 +Epoch [7/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0091, Loss2: 0.0091, Pure Ratio1: 9.9948, Pure Ratio2 9.9896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 76.0917 % Model2 76.1518 %, Pure Ratio 1 9.9767 %, Pure Ratio 2 9.9767 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0049, Loss2: 0.0049, Pure Ratio1: 10.5741, Pure Ratio2 10.5556 +Epoch [8/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0048, Loss2: 0.0048, Pure Ratio1: 10.1389, Pure Ratio2 10.0833 +Epoch [8/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0050, Loss2: 0.0049, Pure Ratio1: 9.9877, Pure Ratio2 9.9136 +Epoch [8/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0044, Loss2: 0.0046, Pure Ratio1: 9.8333, Pure Ratio2 9.8102 +Epoch [8/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0040, Loss2: 0.0046, Pure Ratio1: 9.9519, Pure Ratio2 9.9444 +Epoch [8/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0058, Loss2: 0.0066, Pure Ratio1: 9.9228, Pure Ratio2 9.9228 +Epoch [8/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0061, Loss2: 0.0050, Pure Ratio1: 9.9788, Pure Ratio2 9.9656 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 78.6558 % Model2 79.4571 %, Pure Ratio 1 10.0047 %, Pure Ratio 2 9.9810 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0038, Loss2: 0.0037, Pure Ratio1: 9.6381, Pure Ratio2 9.7143 +Epoch [9/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0035, Loss2: 0.0037, Pure Ratio1: 9.5619, Pure Ratio2 9.6000 +Epoch [9/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0033, Loss2: 0.0030, Pure Ratio1: 9.6317, Pure Ratio2 9.6698 +Epoch [9/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0049, Loss2: 0.0048, Pure Ratio1: 9.6048, Pure Ratio2 9.6048 +Epoch [9/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0051, Loss2: 0.0049, Pure Ratio1: 9.8971, Pure Ratio2 9.8743 +Epoch [9/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0062, Loss2: 0.0061, Pure Ratio1: 9.9429, Pure Ratio2 9.9365 +Epoch [9/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0047, Loss2: 0.0051, Pure Ratio1: 9.9102, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 78.6859 % Model2 78.6058 %, Pure Ratio 1 9.9634 %, Pure Ratio 2 9.9658 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0047, Loss2: 0.0059, Pure Ratio1: 10.7647, Pure Ratio2 10.5686 +Epoch [10/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0037, Loss2: 0.0038, Pure Ratio1: 10.4314, Pure Ratio2 10.3333 +Epoch [10/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0042, Loss2: 0.0040, Pure Ratio1: 10.0196, Pure Ratio2 9.9216 +Epoch [10/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0039, Loss2: 0.0046, Pure Ratio1: 9.9755, Pure Ratio2 9.8922 +Epoch [10/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0042, Loss2: 0.0039, Pure Ratio1: 10.0078, Pure Ratio2 9.8902 +Epoch [10/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0043, Loss2: 0.0042, Pure Ratio1: 10.0621, Pure Ratio2 9.9673 +Epoch [10/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0034, Loss2: 0.0036, Pure Ratio1: 10.1148, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 79.2768 % Model2 79.4571 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0023, Pure Ratio1: 9.6863, Pure Ratio2 9.7255 +Epoch [11/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.0033, Loss2: 0.0035, Pure Ratio1: 9.9020, Pure Ratio2 9.9510 +Epoch [11/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0040, Loss2: 0.0041, Pure Ratio1: 9.9020, Pure Ratio2 10.0261 +Epoch [11/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0057, Loss2: 0.0054, Pure Ratio1: 9.8676, Pure Ratio2 9.9314 +Epoch [11/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0042, Loss2: 0.0040, Pure Ratio1: 9.9608, Pure Ratio2 9.9765 +Epoch [11/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0042, Loss2: 0.0044, Pure Ratio1: 9.9869, Pure Ratio2 9.9967 +Epoch [11/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0037, Loss2: 0.0037, Pure Ratio1: 9.9916, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 81.1699 % Model2 80.9896 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0046, Loss2: 0.0038, Pure Ratio1: 9.4902, Pure Ratio2 9.4510 +Epoch [12/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0022, Pure Ratio1: 10.0196, Pure Ratio2 10.0000 +Epoch [12/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.0024, Loss2: 0.0016, Pure Ratio1: 10.1046, Pure Ratio2 10.1373 +Epoch [12/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0021, Pure Ratio1: 10.1127, Pure Ratio2 10.1078 +Epoch [12/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 67.9688, Loss1: 0.0031, Loss2: 0.0042, Pure Ratio1: 10.0431, Pure Ratio2 10.0392 +Epoch [12/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0031, Loss2: 0.0031, Pure Ratio1: 9.8725, Pure Ratio2 9.8693 +Epoch [12/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0040, Loss2: 0.0027, Pure Ratio1: 9.9804, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 80.9495 % Model2 80.2885 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0024, Loss2: 0.0020, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [13/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.0023, Loss2: 0.0026, Pure Ratio1: 10.3824, Pure Ratio2 10.3725 +Epoch [13/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0019, Pure Ratio1: 9.8039, Pure Ratio2 9.7974 +Epoch [13/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.0019, Loss2: 0.0016, Pure Ratio1: 9.7941, Pure Ratio2 9.7843 +Epoch [13/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0030, Loss2: 0.0034, Pure Ratio1: 9.8745, Pure Ratio2 9.8745 +Epoch [13/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0044, Loss2: 0.0034, Pure Ratio1: 9.9739, Pure Ratio2 9.9771 +Epoch [13/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0030, Loss2: 0.0028, Pure Ratio1: 10.0224, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 81.5505 % Model2 81.3201 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0018, Pure Ratio1: 10.2353, Pure Ratio2 10.1765 +Epoch [14/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0021, Loss2: 0.0022, Pure Ratio1: 9.9020, Pure Ratio2 9.9902 +Epoch [14/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.0026, Loss2: 0.0022, Pure Ratio1: 9.8301, Pure Ratio2 9.8954 +Epoch [14/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0021, Loss2: 0.0018, Pure Ratio1: 9.9069, Pure Ratio2 9.9265 +Epoch [14/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0022, Loss2: 0.0027, Pure Ratio1: 9.9255, Pure Ratio2 9.9255 +Epoch [14/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0022, Loss2: 0.0026, Pure Ratio1: 9.7876, Pure Ratio2 9.7778 +Epoch [14/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0017, Pure Ratio1: 9.9580, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 81.7308 % Model2 81.9311 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0021, Loss2: 0.0021, Pure Ratio1: 9.0392, Pure Ratio2 9.1176 +Epoch [15/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.5686, Pure Ratio2 9.6078 +Epoch [15/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0023, Loss2: 0.0017, Pure Ratio1: 9.6405, Pure Ratio2 9.7124 +Epoch [15/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0017, Pure Ratio1: 9.9314, Pure Ratio2 9.9510 +Epoch [15/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0028, Loss2: 0.0028, Pure Ratio1: 9.9608, Pure Ratio2 9.9961 +Epoch [15/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0028, Loss2: 0.0030, Pure Ratio1: 9.9771, Pure Ratio2 10.0229 +Epoch [15/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0026, Loss2: 0.0027, Pure Ratio1: 10.0364, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 82.4419 % Model2 82.0312 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0027, Loss2: 0.0030, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [16/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0026, Loss2: 0.0019, Pure Ratio1: 9.9902, Pure Ratio2 9.9412 +Epoch [16/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.0028, Loss2: 0.0020, Pure Ratio1: 10.1895, Pure Ratio2 10.1569 +Epoch [16/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.0000, Loss1: 0.0012, Loss2: 0.0019, Pure Ratio1: 10.1618, Pure Ratio2 10.1716 +Epoch [16/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0012, Pure Ratio1: 10.1569, Pure Ratio2 10.1882 +Epoch [16/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 10.1536, Pure Ratio2 10.1438 +Epoch [16/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.0616, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 81.5905 % Model2 82.1314 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0027, Loss2: 0.0017, Pure Ratio1: 10.0392, Pure Ratio2 9.9608 +Epoch [17/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0016, Pure Ratio1: 10.0588, Pure Ratio2 9.9412 +Epoch [17/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0022, Loss2: 0.0021, Pure Ratio1: 10.1961, Pure Ratio2 10.0915 +Epoch [17/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0014, Pure Ratio1: 10.1716, Pure Ratio2 10.1373 +Epoch [17/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 10.1412, Pure Ratio2 10.1137 +Epoch [17/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 10.0588, Pure Ratio2 10.0458 +Epoch [17/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 9.9692, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 81.3201 % Model2 82.4820 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [18/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0025, Loss2: 0.0023, Pure Ratio1: 10.0588, Pure Ratio2 10.0784 +Epoch [18/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0014, Loss2: 0.0009, Pure Ratio1: 10.0327, Pure Ratio2 10.0458 +Epoch [18/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0024, Pure Ratio1: 10.0245, Pure Ratio2 10.0441 +Epoch [18/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0012, Pure Ratio1: 10.0039, Pure Ratio2 10.0118 +Epoch [18/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0017, Pure Ratio1: 10.0621, Pure Ratio2 10.0752 +Epoch [18/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0037, Loss2: 0.0034, Pure Ratio1: 10.0616, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 81.0296 % Model2 79.3770 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0016, Loss2: 0.0017, Pure Ratio1: 9.5686, Pure Ratio2 9.5490 +Epoch [19/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0026, Loss2: 0.0018, Pure Ratio1: 9.5196, Pure Ratio2 9.5490 +Epoch [19/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.6993, Pure Ratio2 9.6928 +Epoch [19/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.7108 +Epoch [19/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.7373, Pure Ratio2 9.7373 +Epoch [19/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0015, Pure Ratio1: 9.8301, Pure Ratio2 9.8105 +Epoch [19/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.9440, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 81.9411 % Model2 80.4387 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.1765, Pure Ratio2 10.2157 +Epoch [20/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8431, Pure Ratio2 9.8922 +Epoch [20/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0012, Loss2: 0.0018, Pure Ratio1: 9.9804, Pure Ratio2 10.0980 +Epoch [20/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.9559, Pure Ratio2 10.0441 +Epoch [20/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.0017, Loss2: 0.0024, Pure Ratio1: 9.9451, Pure Ratio2 10.0039 +Epoch [20/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0009, Pure Ratio1: 10.0294, Pure Ratio2 10.1046 +Epoch [20/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0021, Loss2: 0.0015, Pure Ratio1: 10.0000, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 81.4103 % Model2 81.7208 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [21/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [21/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0017, Loss2: 0.0021, Pure Ratio1: 9.9542, Pure Ratio2 9.9804 +Epoch [21/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 10.1176, Pure Ratio2 10.0931 +Epoch [21/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.1137, Pure Ratio2 10.0980 +Epoch [21/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 10.0229, Pure Ratio2 10.0131 +Epoch [21/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 10.0056, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 82.1314 % Model2 82.1014 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0013, Pure Ratio1: 9.5098, Pure Ratio2 9.4706 +Epoch [22/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.7647, Pure Ratio2 9.7451 +Epoch [22/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7974, Pure Ratio2 9.7451 +Epoch [22/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8480, Pure Ratio2 9.7990 +Epoch [22/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9961, Pure Ratio2 10.0078 +Epoch [22/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9935, Pure Ratio2 10.0033 +Epoch [22/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 10.0616, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 81.7909 % Model2 83.4836 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0019, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [23/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.0008, Loss2: 0.0016, Pure Ratio1: 9.8039, Pure Ratio2 9.9118 +Epoch [23/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0019, Loss2: 0.0026, Pure Ratio1: 10.0131, Pure Ratio2 10.0850 +Epoch [23/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.0014, Loss2: 0.0019, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [23/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 9.9647, Pure Ratio2 9.9608 +Epoch [23/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 10.0392, Pure Ratio2 10.0261 +Epoch [23/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 10.0028, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 81.4804 % Model2 81.6707 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 9.8627 +Epoch [24/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.9510, Pure Ratio2 9.9412 +Epoch [24/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 10.0458 +Epoch [24/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0023, Loss2: 0.0022, Pure Ratio1: 10.1029, Pure Ratio2 10.1422 +Epoch [24/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [24/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.9771, Pure Ratio2 10.0327 +Epoch [24/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.9692, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 81.5805 % Model2 82.0312 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0013, Pure Ratio1: 10.4510, Pure Ratio2 10.7059 +Epoch [25/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0015, Loss2: 0.0005, Pure Ratio1: 10.2059, Pure Ratio2 10.2843 +Epoch [25/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0042, Loss2: 0.0051, Pure Ratio1: 10.0458, Pure Ratio2 10.0850 +Epoch [25/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.9363 +Epoch [25/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.0000, Pure Ratio2 9.9843 +Epoch [25/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.9706, Pure Ratio2 9.9542 +Epoch [25/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9664, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 81.4203 % Model2 80.8694 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.5490, Pure Ratio2 10.4706 +Epoch [26/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.6275, Pure Ratio2 10.6078 +Epoch [26/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.4052, Pure Ratio2 10.3856 +Epoch [26/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.3235, Pure Ratio2 10.2892 +Epoch [26/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.1333, Pure Ratio2 10.0745 +Epoch [26/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 10.1078, Pure Ratio2 10.0490 +Epoch [26/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 10.1008, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 81.9010 % Model2 82.4820 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.2941, Pure Ratio2 10.2353 +Epoch [27/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.1471, Pure Ratio2 10.0392 +Epoch [27/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.1242, Pure Ratio2 10.0654 +Epoch [27/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0147, Pure Ratio2 10.0049 +Epoch [27/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0353, Pure Ratio2 10.0353 +Epoch [27/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.0556, Pure Ratio2 10.0261 +Epoch [27/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 10.1261, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 81.1098 % Model2 81.0196 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.4314, Pure Ratio2 9.4706 +Epoch [28/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6176, Pure Ratio2 9.6176 +Epoch [28/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8039, Pure Ratio2 9.8301 +Epoch [28/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9069, Pure Ratio2 9.9118 +Epoch [28/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9176, Pure Ratio2 9.9373 +Epoch [28/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.9085, Pure Ratio2 9.9477 +Epoch [28/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9496, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 81.9511 % Model2 81.0397 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [29/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0020, Loss2: 0.0019, Pure Ratio1: 10.1176, Pure Ratio2 10.1961 +Epoch [29/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.1111, Pure Ratio2 10.0915 +Epoch [29/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 10.0098, Pure Ratio2 9.9804 +Epoch [29/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0431, Pure Ratio2 10.0275 +Epoch [29/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0015, Loss2: 0.0009, Pure Ratio1: 10.0131, Pure Ratio2 9.9902 +Epoch [29/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0013, Pure Ratio1: 9.9832, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 81.6506 % Model2 81.4603 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [30/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.7745, Pure Ratio2 9.8137 +Epoch [30/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9477, Pure Ratio2 10.0392 +Epoch [30/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8578, Pure Ratio2 9.9167 +Epoch [30/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0013, Pure Ratio1: 9.7804, Pure Ratio2 9.8000 +Epoch [30/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8660, Pure Ratio2 9.9085 +Epoch [30/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9496, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 82.1715 % Model2 81.3201 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.1031 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [31/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.9118, Pure Ratio2 9.8529 +Epoch [31/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0057, Loss2: 0.0047, Pure Ratio1: 9.9608, Pure Ratio2 9.9673 +Epoch [31/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.2353, Pure Ratio2 10.2353 +Epoch [31/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.2314, Pure Ratio2 10.2275 +Epoch [31/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1144, Pure Ratio2 10.1242 +Epoch [31/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0012, Loss2: 0.0014, Pure Ratio1: 10.0448, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 80.4187 % Model2 80.4688 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 10.5294, Pure Ratio2 10.7647 +Epoch [32/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0294, Pure Ratio2 10.2157 +Epoch [32/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.1307 +Epoch [32/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2108, Pure Ratio2 10.2549 +Epoch [32/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.1255, Pure Ratio2 10.2157 +Epoch [32/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0025, Loss2: 0.0021, Pure Ratio1: 10.0915, Pure Ratio2 10.1536 +Epoch [32/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.0018, Loss2: 0.0027, Pure Ratio1: 10.0364, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 81.4804 % Model2 81.6807 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0017, Pure Ratio1: 9.8235, Pure Ratio2 9.9608 +Epoch [33/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0007, Loss2: 0.0017, Pure Ratio1: 9.8529, Pure Ratio2 9.8431 +Epoch [33/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.0131, Pure Ratio2 10.0327 +Epoch [33/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0441, Pure Ratio2 10.0833 +Epoch [33/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9686, Pure Ratio2 10.0314 +Epoch [33/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.8856, Pure Ratio2 9.9379 +Epoch [33/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0008, Pure Ratio1: 9.8347, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 81.1599 % Model2 81.6206 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0980 +Epoch [34/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 10.1078, Pure Ratio2 10.1667 +Epoch [34/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.1307, Pure Ratio2 10.2418 +Epoch [34/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.1814 +Epoch [34/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0018, Loss2: 0.0010, Pure Ratio1: 10.0471, Pure Ratio2 10.0392 +Epoch [34/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9641, Pure Ratio2 9.9542 +Epoch [34/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9580, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 81.1999 % Model2 80.9295 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.2941, Pure Ratio2 9.2941 +Epoch [35/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 9.9706 +Epoch [35/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0458, Pure Ratio2 9.8693 +Epoch [35/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9657, Pure Ratio2 9.8382 +Epoch [35/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9059, Pure Ratio2 9.8000 +Epoch [35/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0261, Pure Ratio2 9.9281 +Epoch [35/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0015, Loss2: 0.0005, Pure Ratio1: 10.1008, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 81.1098 % Model2 80.8594 %, Pure Ratio 1 10.1006 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.2941, Pure Ratio2 9.4118 +Epoch [36/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.6471 +Epoch [36/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.9085, Pure Ratio2 9.9477 +Epoch [36/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 10.0343, Pure Ratio2 10.0441 +Epoch [36/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1137, Pure Ratio2 10.1059 +Epoch [36/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.1111 +Epoch [36/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.8438, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 79.7276 % Model2 81.5004 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 9.7843 +Epoch [37/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.8627 +Epoch [37/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9281, Pure Ratio2 9.8693 +Epoch [37/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.0686 +Epoch [37/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 10.0353, Pure Ratio2 9.9608 +Epoch [37/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0752, Pure Ratio2 9.9771 +Epoch [37/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0028, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 80.9796 % Model2 81.2300 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6078, Pure Ratio2 9.6471 +Epoch [38/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.2157 +Epoch [38/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 10.0719 +Epoch [38/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8529, Pure Ratio2 9.9118 +Epoch [38/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9294, Pure Ratio2 10.0000 +Epoch [38/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8497, Pure Ratio2 9.9314 +Epoch [38/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.8571, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 81.6506 % Model2 80.7392 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.4706, Pure Ratio2 9.3725 +Epoch [39/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.2549 +Epoch [39/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0588, Pure Ratio2 10.1307 +Epoch [39/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0441, Pure Ratio2 10.1471 +Epoch [39/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9412, Pure Ratio2 10.0824 +Epoch [39/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9542, Pure Ratio2 10.0359 +Epoch [39/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9860, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 81.4503 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.6078, Pure Ratio2 10.5490 +Epoch [40/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2647, Pure Ratio2 10.3333 +Epoch [40/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.1895, Pure Ratio2 10.3464 +Epoch [40/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1765 +Epoch [40/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9373, Pure Ratio2 10.0588 +Epoch [40/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.9673, Pure Ratio2 10.0098 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9748, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 81.2500 % Model2 81.7308 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 10.1765 +Epoch [41/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1471 +Epoch [41/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.1961 +Epoch [41/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.1176 +Epoch [41/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0588, Pure Ratio2 10.0941 +Epoch [41/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 10.0523 +Epoch [41/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0084, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 80.0681 % Model2 80.2284 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.3922, Pure Ratio2 10.3529 +Epoch [42/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1667, Pure Ratio2 10.1275 +Epoch [42/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0020, Pure Ratio1: 10.0523, Pure Ratio2 10.0784 +Epoch [42/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0098, Pure Ratio2 10.0245 +Epoch [42/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.0667, Pure Ratio2 10.0902 +Epoch [42/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0654, Pure Ratio2 10.0523 +Epoch [42/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9916, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 81.0597 % Model2 80.6390 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 10.1765 +Epoch [43/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0016, Pure Ratio1: 9.7941, Pure Ratio2 9.8922 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.8824, Pure Ratio2 9.9608 +Epoch [43/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.8137, Pure Ratio2 9.8873 +Epoch [43/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0118 +Epoch [43/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.9085, Pure Ratio2 9.9281 +Epoch [43/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9384, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 81.4403 % Model2 81.5104 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 8.9216, Pure Ratio2 9.0588 +Epoch [44/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.3627, Pure Ratio2 9.4118 +Epoch [44/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7516, Pure Ratio2 9.7320 +Epoch [44/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0011, Pure Ratio1: 9.8971, Pure Ratio2 9.8333 +Epoch [44/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.8706, Pure Ratio2 9.8235 +Epoch [44/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9248, Pure Ratio2 9.9150 +Epoch [44/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9888, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 80.7292 % Model2 81.4203 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.4510, Pure Ratio2 10.3137 +Epoch [45/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.3431, Pure Ratio2 10.2549 +Epoch [45/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.3007, Pure Ratio2 10.2092 +Epoch [45/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 9.9706 +Epoch [45/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.1882, Pure Ratio2 10.0745 +Epoch [45/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1144, Pure Ratio2 10.0229 +Epoch [45/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9972, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 81.0797 % Model2 80.9395 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [46/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.2745, Pure Ratio2 10.2941 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.2353 +Epoch [46/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0735, Pure Ratio2 10.1029 +Epoch [46/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.1647 +Epoch [46/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.0719, Pure Ratio2 10.1209 +Epoch [46/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0504, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 81.0096 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.5882 +Epoch [47/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7157, Pure Ratio2 9.6373 +Epoch [47/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8431, Pure Ratio2 9.8366 +Epoch [47/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8382, Pure Ratio2 9.8578 +Epoch [47/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0431 +Epoch [47/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 10.0654, Pure Ratio2 10.0817 +Epoch [47/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9972, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 80.7292 % Model2 81.1599 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.9020 +Epoch [48/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.7157, Pure Ratio2 9.9314 +Epoch [48/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8105, Pure Ratio2 10.0065 +Epoch [48/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8922, Pure Ratio2 10.0098 +Epoch [48/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9333, Pure Ratio2 10.0471 +Epoch [48/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9379, Pure Ratio2 10.0065 +Epoch [48/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9524, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 80.3185 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.3529 +Epoch [49/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.5392, Pure Ratio2 10.4804 +Epoch [49/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.3660, Pure Ratio2 10.3137 +Epoch [49/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2794, Pure Ratio2 10.2353 +Epoch [49/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.4157, Pure Ratio2 10.3529 +Epoch [49/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1634, Pure Ratio2 10.1078 +Epoch [49/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1317, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 79.3470 % Model2 81.0196 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.6078 +Epoch [50/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2059, Pure Ratio2 10.0686 +Epoch [50/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 10.1634, Pure Ratio2 10.0131 +Epoch [50/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1422, Pure Ratio2 10.0147 +Epoch [50/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9961, Pure Ratio2 9.8392 +Epoch [50/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9869, Pure Ratio2 9.8693 +Epoch [50/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0560, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 80.5990 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.7843, Pure Ratio2 10.8431 +Epoch [51/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.3922, Pure Ratio2 10.2549 +Epoch [51/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.1503, Pure Ratio2 10.0000 +Epoch [51/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0049, Pure Ratio2 9.8480 +Epoch [51/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.1412, Pure Ratio2 9.9765 +Epoch [51/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.1176, Pure Ratio2 9.9902 +Epoch [51/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0952, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 80.2484 % Model2 80.7692 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.6863 +Epoch [52/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.9020 +Epoch [52/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9167, Pure Ratio2 9.9167 +Epoch [52/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7961, Pure Ratio2 9.7843 +Epoch [52/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8595 +Epoch [52/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8768, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 80.5188 % Model2 80.8093 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.6471 +Epoch [53/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.1765, Pure Ratio2 10.3235 +Epoch [53/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 10.0850 +Epoch [53/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9559, Pure Ratio2 10.0343 +Epoch [53/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8353, Pure Ratio2 9.9176 +Epoch [53/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9967 +Epoch [53/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9328, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 80.3385 % Model2 81.5505 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 9.8235 +Epoch [54/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.0098 +Epoch [54/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.2810, Pure Ratio2 10.0327 +Epoch [54/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1716, Pure Ratio2 10.0000 +Epoch [54/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2039, Pure Ratio2 10.0824 +Epoch [54/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1013, Pure Ratio2 9.9967 +Epoch [54/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0644, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 79.1166 % Model2 80.2183 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0392 +Epoch [55/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 9.9902 +Epoch [55/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.9608, Pure Ratio2 9.9739 +Epoch [55/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0441 +Epoch [55/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.2510, Pure Ratio2 10.2157 +Epoch [55/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0261, Pure Ratio2 9.9804 +Epoch [55/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.0560, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 80.7492 % Model2 80.4487 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.0196 +Epoch [56/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0098 +Epoch [56/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8889, Pure Ratio2 9.9281 +Epoch [56/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8382, Pure Ratio2 9.9265 +Epoch [56/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.8863, Pure Ratio2 9.9725 +Epoch [56/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 10.1078 +Epoch [56/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0476, Pure Ratio2 10.1457 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 81.3201 % Model2 80.8594 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.5882 +Epoch [57/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.6765 +Epoch [57/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0014, Pure Ratio1: 9.9739, Pure Ratio2 9.9150 +Epoch [57/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0033, Loss2: 0.0029, Pure Ratio1: 9.8529, Pure Ratio2 9.8431 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8902, Pure Ratio2 9.8980 +Epoch [57/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8791, Pure Ratio2 9.8693 +Epoch [57/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9300, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 80.5589 % Model2 80.1983 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.4314, Pure Ratio2 9.7647 +Epoch [58/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0686 +Epoch [58/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.8170 +Epoch [58/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8088, Pure Ratio2 9.8186 +Epoch [58/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9137 +Epoch [58/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9248 +Epoch [58/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9664, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 81.0096 % Model2 80.6691 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.8235 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1275, Pure Ratio2 10.1569 +Epoch [59/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0016, Pure Ratio1: 10.1373, Pure Ratio2 10.2222 +Epoch [59/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0013, Loss2: 0.0002, Pure Ratio1: 10.1078, Pure Ratio2 10.1422 +Epoch [59/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0275, Pure Ratio2 10.0784 +Epoch [59/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0000, Pure Ratio2 10.0654 +Epoch [59/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9496, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 79.7676 % Model2 80.0881 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.3529, Pure Ratio2 9.2745 +Epoch [60/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6961, Pure Ratio2 9.6667 +Epoch [60/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9542, Pure Ratio2 9.9804 +Epoch [60/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0147, Pure Ratio2 10.0147 +Epoch [60/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.0471 +Epoch [60/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1275, Pure Ratio2 10.1307 +Epoch [60/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1008, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 79.4872 % Model2 79.1667 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0930 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3333, Pure Ratio2 10.3725 +Epoch [61/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3922, Pure Ratio2 10.5000 +Epoch [61/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.4575 +Epoch [61/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2402, Pure Ratio2 10.3186 +Epoch [61/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0745, Pure Ratio2 10.1176 +Epoch [61/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9739, Pure Ratio2 10.0327 +Epoch [61/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0112, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 80.6591 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.6471 +Epoch [62/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.7353, Pure Ratio2 9.6765 +Epoch [62/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8693 +Epoch [62/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9755, Pure Ratio2 9.9216 +Epoch [62/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 10.0392, Pure Ratio2 10.0510 +Epoch [62/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.0359 +Epoch [62/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0252, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 80.1382 % Model2 81.2800 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.7451, Pure Ratio2 10.5294 +Epoch [63/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2843, Pure Ratio2 10.1471 +Epoch [63/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0915, Pure Ratio2 10.0458 +Epoch [63/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1275, Pure Ratio2 10.0784 +Epoch [63/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2275, Pure Ratio2 10.2000 +Epoch [63/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1699, Pure Ratio2 10.1438 +Epoch [63/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1261, Pure Ratio2 10.1345 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 80.8093 % Model2 79.9579 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [64/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 9.9902 +Epoch [64/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9346 +Epoch [64/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8971, Pure Ratio2 9.9804 +Epoch [64/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9373, Pure Ratio2 9.9843 +Epoch [64/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9575, Pure Ratio2 10.0098 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 81.0797 % Model2 80.9095 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6078 +Epoch [65/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.8431 +Epoch [65/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.8170 +Epoch [65/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0245, Pure Ratio2 9.9510 +Epoch [65/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.1137, Pure Ratio2 10.0196 +Epoch [65/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0425, Pure Ratio2 9.9641 +Epoch [65/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0364, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 80.9595 % Model2 80.0681 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.7843, Pure Ratio2 10.8824 +Epoch [66/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.2059, Pure Ratio2 10.2255 +Epoch [66/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.2418 +Epoch [66/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0049, Pure Ratio2 9.9314 +Epoch [66/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1137, Pure Ratio2 10.0353 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 9.9869 +Epoch [66/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0728, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 80.3285 % Model2 81.1198 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1961, Pure Ratio2 9.9216 +Epoch [67/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0196 +Epoch [67/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.9346 +Epoch [67/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1324, Pure Ratio2 10.0686 +Epoch [67/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9922, Pure Ratio2 9.9373 +Epoch [67/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9837, Pure Ratio2 9.8791 +Epoch [67/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 80.2183 % Model2 80.6791 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.3529, Pure Ratio2 9.1569 +Epoch [68/200], Iter [100/390] Training Accuracy1: 94.5312, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 9.9608 +Epoch [68/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 9.9608 +Epoch [68/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.1373, Pure Ratio2 9.9510 +Epoch [68/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0471, Pure Ratio2 9.9294 +Epoch [68/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0425, Pure Ratio2 9.9804 +Epoch [68/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0448, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 79.7977 % Model2 80.7492 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4314 +Epoch [69/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.2255, Pure Ratio2 10.0980 +Epoch [69/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2092, Pure Ratio2 10.0523 +Epoch [69/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1961, Pure Ratio2 10.0784 +Epoch [69/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1529, Pure Ratio2 10.0824 +Epoch [69/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.0327 +Epoch [69/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0280, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 80.8393 % Model2 81.1398 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.2549 +Epoch [70/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.2843, Pure Ratio2 10.1765 +Epoch [70/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9412 +Epoch [70/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9951, Pure Ratio2 9.8824 +Epoch [70/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9137 +Epoch [70/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 9.9608 +Epoch [70/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 79.6074 % Model2 80.5088 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1373 +Epoch [71/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.4608 +Epoch [71/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [71/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8725, Pure Ratio2 9.9265 +Epoch [71/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.9804 +Epoch [71/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.9444 +Epoch [71/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9300, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 80.9495 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.8627, Pure Ratio2 10.6078 +Epoch [72/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.6176, Pure Ratio2 10.4608 +Epoch [72/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3987 +Epoch [72/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.1275 +Epoch [72/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0353, Pure Ratio2 10.0392 +Epoch [72/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 9.9967 +Epoch [72/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 80.6891 % Model2 81.0096 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 10.1373 +Epoch [73/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1078 +Epoch [73/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.1699 +Epoch [73/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0931, Pure Ratio2 10.1471 +Epoch [73/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0235, Pure Ratio2 10.0784 +Epoch [73/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0023, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 10.0229 +Epoch [73/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 79.8978 % Model2 80.5889 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.2941, Pure Ratio2 9.8039 +Epoch [74/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.2745, Pure Ratio2 9.9216 +Epoch [74/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2026, Pure Ratio2 9.9673 +Epoch [74/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2255, Pure Ratio2 10.0735 +Epoch [74/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1608, Pure Ratio2 10.0000 +Epoch [74/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 9.9444 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1064, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 81.5204 % Model2 80.7792 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.2157 +Epoch [75/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0588 +Epoch [75/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 9.9673 +Epoch [75/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0539, Pure Ratio2 10.0000 +Epoch [75/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0431, Pure Ratio2 10.0196 +Epoch [75/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0523, Pure Ratio2 10.0490 +Epoch [75/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9860, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 79.5373 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9412 +Epoch [76/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1275, Pure Ratio2 9.9804 +Epoch [76/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 9.9281 +Epoch [76/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9559, Pure Ratio2 9.9608 +Epoch [76/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9176, Pure Ratio2 9.9216 +Epoch [76/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8333 +Epoch [76/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8403, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 80.2784 % Model2 80.9195 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.2745 +Epoch [77/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.2059 +Epoch [77/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9346 +Epoch [77/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 10.0147, Pure Ratio2 9.9657 +Epoch [77/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1255, Pure Ratio2 10.0627 +Epoch [77/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1340, Pure Ratio2 10.0458 +Epoch [77/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1120, Pure Ratio2 10.0392 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 80.2885 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.7843 +Epoch [78/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7647, Pure Ratio2 9.7843 +Epoch [78/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8366 +Epoch [78/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.8137 +Epoch [78/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9686, Pure Ratio2 9.8196 +Epoch [78/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0425, Pure Ratio2 9.9575 +Epoch [78/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0140, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 79.2568 % Model2 79.9679 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.4706, Pure Ratio2 10.2353 +Epoch [79/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.0098 +Epoch [79/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 9.9739 +Epoch [79/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0637, Pure Ratio2 9.9314 +Epoch [79/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0745, Pure Ratio2 9.9686 +Epoch [79/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 9.9771 +Epoch [79/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0532, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 80.3185 % Model2 80.5589 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.2745 +Epoch [80/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1176 +Epoch [80/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8954 +Epoch [80/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8578, Pure Ratio2 9.8578 +Epoch [80/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9922, Pure Ratio2 9.9804 +Epoch [80/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0014, Loss2: 0.0002, Pure Ratio1: 10.0065, Pure Ratio2 9.9902 +Epoch [80/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0084, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 79.6875 % Model2 80.9896 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0784, Pure Ratio2 9.9412 +Epoch [81/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8529 +Epoch [81/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 9.9346 +Epoch [81/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.8775 +Epoch [81/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1098, Pure Ratio2 9.9804 +Epoch [81/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 9.9412 +Epoch [81/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9860, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 79.7276 % Model2 81.1298 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.2353 +Epoch [82/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [82/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9542 +Epoch [82/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9657, Pure Ratio2 10.0147 +Epoch [82/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0745, Pure Ratio2 10.1412 +Epoch [82/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.1503 +Epoch [82/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.1681 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 79.5272 % Model2 80.1883 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0955 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.1765, Pure Ratio2 10.3529 +Epoch [83/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8627 +Epoch [83/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8954 +Epoch [83/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.8676 +Epoch [83/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8941, Pure Ratio2 9.9686 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 10.0458 +Epoch [83/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0616, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 79.9079 % Model2 80.2083 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.8235 +Epoch [84/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8333, Pure Ratio2 9.7843 +Epoch [84/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 9.8366, Pure Ratio2 9.8039 +Epoch [84/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9755 +Epoch [84/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0235, Pure Ratio2 9.9725 +Epoch [84/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.9412 +Epoch [84/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9300, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 79.5473 % Model2 80.0481 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1961, Pure Ratio2 10.0588 +Epoch [85/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8529 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.8039 +Epoch [85/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8186 +Epoch [85/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9255, Pure Ratio2 9.9294 +Epoch [85/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8987, Pure Ratio2 9.8824 +Epoch [85/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 80.9796 % Model2 80.5288 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.7451 +Epoch [86/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 9.9608, Pure Ratio2 10.0588 +Epoch [86/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0245 +Epoch [86/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 10.0118 +Epoch [86/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 9.9575 +Epoch [86/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 81.4002 % Model2 80.4988 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8039 +Epoch [87/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9902 +Epoch [87/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9542 +Epoch [87/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9804 +Epoch [87/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0667, Pure Ratio2 9.9882 +Epoch [87/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 9.9837 +Epoch [87/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 79.8377 % Model2 78.1951 %, Pure Ratio 1 10.1408 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.1765 +Epoch [88/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.0882 +Epoch [88/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.2680, Pure Ratio2 10.2549 +Epoch [88/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.2402, Pure Ratio2 10.2794 +Epoch [88/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1451, Pure Ratio2 10.1451 +Epoch [88/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 10.1046 +Epoch [88/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 80.3385 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.3137, Pure Ratio2 10.2353 +Epoch [89/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9902 +Epoch [89/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2222, Pure Ratio2 10.1961 +Epoch [89/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.2304 +Epoch [89/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2275, Pure Ratio2 10.2235 +Epoch [89/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1732 +Epoch [89/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0168, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 80.0781 % Model2 79.9279 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.8627 +Epoch [90/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2647, Pure Ratio2 10.5588 +Epoch [90/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3333, Pure Ratio2 10.5425 +Epoch [90/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1716, Pure Ratio2 10.3480 +Epoch [90/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2118, Pure Ratio2 10.3412 +Epoch [90/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.2157 +Epoch [90/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 80.8494 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [91/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [91/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9150 +Epoch [91/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0735 +Epoch [91/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0039, Pure Ratio2 10.0392 +Epoch [91/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9935 +Epoch [91/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 80.1282 % Model2 80.3686 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8627 +Epoch [92/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.1961, Pure Ratio2 10.0980 +Epoch [92/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1111 +Epoch [92/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.1863, Pure Ratio2 10.0931 +Epoch [92/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1490, Pure Ratio2 10.0275 +Epoch [92/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 9.9869 +Epoch [92/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 81.1999 % Model2 81.3101 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.1176, Pure Ratio2 10.1569 +Epoch [93/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.9118, Pure Ratio2 9.9412 +Epoch [93/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0654, Pure Ratio2 10.0131 +Epoch [93/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1324 +Epoch [93/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9569, Pure Ratio2 9.9608 +Epoch [93/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8562, Pure Ratio2 9.8824 +Epoch [93/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 80.4087 % Model2 80.3285 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.6667 +Epoch [94/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.9020 +Epoch [94/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7908 +Epoch [94/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9167 +Epoch [94/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9059, Pure Ratio2 9.9176 +Epoch [94/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9216 +Epoch [94/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 80.4587 %, Pure Ratio 1 10.0855 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.0392 +Epoch [95/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.1863, Pure Ratio2 10.1176 +Epoch [95/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9869 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0000 +Epoch [95/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0745, Pure Ratio2 9.9373 +Epoch [95/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2059, Pure Ratio2 10.1111 +Epoch [95/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1485, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 80.6691 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4902 +Epoch [96/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.7451 +Epoch [96/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.6797 +Epoch [96/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0024, Pure Ratio1: 9.8235, Pure Ratio2 9.7059 +Epoch [96/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.7765 +Epoch [96/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.8660 +Epoch [96/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0056, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 80.4187 % Model2 80.8093 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 9.8431 +Epoch [97/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9608 +Epoch [97/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0000 +Epoch [97/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8775 +Epoch [97/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 9.9020 +Epoch [97/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 9.9248 +Epoch [97/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.0952, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 80.9195 % Model2 80.7392 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.4510, Pure Ratio2 9.3725 +Epoch [98/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9314 +Epoch [98/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8954, Pure Ratio2 9.9281 +Epoch [98/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8676, Pure Ratio2 9.8529 +Epoch [98/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9294, Pure Ratio2 9.9216 +Epoch [98/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0163, Pure Ratio2 9.9935 +Epoch [98/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0112, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 79.1466 % Model2 81.3702 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 10.2745 +Epoch [99/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.3235 +Epoch [99/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.1046 +Epoch [99/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 10.0343 +Epoch [99/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.1647 +Epoch [99/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9346, Pure Ratio2 10.0948 +Epoch [99/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8796, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 80.8894 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.5490 +Epoch [100/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7941 +Epoch [100/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8497 +Epoch [100/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7892, Pure Ratio2 9.7402 +Epoch [100/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8471, Pure Ratio2 9.8392 +Epoch [100/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0392 +Epoch [100/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.0336, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 79.6074 % Model2 81.2700 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [101/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.5392 +Epoch [101/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 9.9412 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0735, Pure Ratio2 9.9461 +Epoch [101/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0157, Pure Ratio2 9.9098 +Epoch [101/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.8693 +Epoch [101/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 80.1783 % Model2 79.8878 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6569, Pure Ratio2 9.7451 +Epoch [102/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0392 +Epoch [102/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0011, Pure Ratio1: 10.0588, Pure Ratio2 10.0637 +Epoch [102/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0471 +Epoch [102/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 9.9608 +Epoch [102/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1148, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 81.2901 %, Pure Ratio 1 10.1056 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [103/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.6275 +Epoch [103/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0131, Pure Ratio2 9.9412 +Epoch [103/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0735, Pure Ratio2 10.0441 +Epoch [103/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9843 +Epoch [103/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0654, Pure Ratio2 10.0719 +Epoch [103/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0336, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 79.5172 % Model2 80.6591 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2941 +Epoch [104/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6961, Pure Ratio2 9.8529 +Epoch [104/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0196 +Epoch [104/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9314, Pure Ratio2 10.0147 +Epoch [104/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 10.0588 +Epoch [104/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 10.0588 +Epoch [104/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 79.7476 % Model2 80.1282 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [105/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.1176 +Epoch [105/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [105/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8725 +Epoch [105/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9843 +Epoch [105/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0065 +Epoch [105/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 80.2484 % Model2 79.3369 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8627 +Epoch [106/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.9804 +Epoch [106/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9085, Pure Ratio2 10.0784 +Epoch [106/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 10.0588 +Epoch [106/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 10.0431 +Epoch [106/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.1275 +Epoch [106/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1541 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 80.5489 % Model2 80.6090 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3529, Pure Ratio2 10.3725 +Epoch [107/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3824, Pure Ratio2 10.4412 +Epoch [107/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0719 +Epoch [107/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.2549 +Epoch [107/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.2000 +Epoch [107/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0850, Pure Ratio2 10.1471 +Epoch [107/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0924, Pure Ratio2 10.1317 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 79.9179 % Model2 81.0096 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5882 +Epoch [108/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6078 +Epoch [108/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5621, Pure Ratio2 9.4837 +Epoch [108/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.4853 +Epoch [108/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.7569, Pure Ratio2 9.7216 +Epoch [108/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9052 +Epoch [108/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9440, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 80.7492 % Model2 80.6390 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.9020, Pure Ratio2 10.9020 +Epoch [109/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.3922, Pure Ratio2 10.3627 +Epoch [109/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3072, Pure Ratio2 10.3464 +Epoch [109/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 10.2451 +Epoch [109/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.1686 +Epoch [109/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1471 +Epoch [109/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1597, Pure Ratio2 10.1148 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 79.9079 % Model2 80.8594 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8431 +Epoch [110/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.1078 +Epoch [110/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.0392 +Epoch [110/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1912, Pure Ratio2 10.1716 +Epoch [110/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1686 +Epoch [110/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.1176 +Epoch [110/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 80.9896 % Model2 80.8794 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [111/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1373 +Epoch [111/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0588 +Epoch [111/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1225, Pure Ratio2 10.1471 +Epoch [111/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.1216 +Epoch [111/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0523, Pure Ratio2 10.0980 +Epoch [111/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0420, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 80.1282 % Model2 81.0196 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.7451 +Epoch [112/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9314 +Epoch [112/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.2288 +Epoch [112/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2108, Pure Ratio2 10.1961 +Epoch [112/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [112/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0621 +Epoch [112/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0252, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 81.1799 % Model2 81.0597 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.0588 +Epoch [113/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8922 +Epoch [113/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8758 +Epoch [113/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0882, Pure Ratio2 10.1029 +Epoch [113/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 9.9922 +Epoch [113/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0033 +Epoch [113/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1232, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 80.5589 % Model2 81.2600 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.0000 +Epoch [114/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [114/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1569 +Epoch [114/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0784 +Epoch [114/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2118, Pure Ratio2 10.1686 +Epoch [114/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1176 +Epoch [114/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0840, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 80.0881 % Model2 81.0296 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [115/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9314 +Epoch [115/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2680 +Epoch [115/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1422, Pure Ratio2 10.1520 +Epoch [115/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0784 +Epoch [115/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0359, Pure Ratio2 10.0490 +Epoch [115/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 80.9395 % Model2 81.1599 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0588 +Epoch [116/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9706 +Epoch [116/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.7908 +Epoch [116/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.8529 +Epoch [116/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.8784 +Epoch [116/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 9.9869 +Epoch [116/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 81.1799 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1569 +Epoch [117/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 9.9804 +Epoch [117/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.0000 +Epoch [117/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.8971 +Epoch [117/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8980, Pure Ratio2 9.8235 +Epoch [117/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.8235 +Epoch [117/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 80.1983 % Model2 80.3586 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 10.2745 +Epoch [118/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.1765 +Epoch [118/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.2418 +Epoch [118/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.1373 +Epoch [118/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.1765 +Epoch [118/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0229, Pure Ratio2 10.1928 +Epoch [118/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0056, Pure Ratio2 10.1653 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 80.7292 % Model2 80.5088 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0019, Loss2: 0.0020, Pure Ratio1: 10.0392, Pure Ratio2 10.0588 +Epoch [119/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3627, Pure Ratio2 10.4608 +Epoch [119/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.1242 +Epoch [119/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 10.0490 +Epoch [119/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8667, Pure Ratio2 10.0118 +Epoch [119/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0850 +Epoch [119/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 80.4287 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.5294 +Epoch [120/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.4314 +Epoch [120/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5490, Pure Ratio2 10.5033 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2794, Pure Ratio2 10.3137 +Epoch [120/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1765 +Epoch [120/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.1144 +Epoch [120/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1232, Pure Ratio2 10.1877 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 80.9495 % Model2 80.9996 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.1207 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.0980 +Epoch [121/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8922 +Epoch [121/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.8758, Pure Ratio2 9.9085 +Epoch [121/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1618 +Epoch [121/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [121/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.0131 +Epoch [121/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 80.5088 % Model2 80.2183 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [122/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.6275 +Epoch [122/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8627 +Epoch [122/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.7402 +Epoch [122/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8706 +Epoch [122/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9412 +Epoch [122/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 80.4087 % Model2 80.9395 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.5882 +Epoch [123/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.5000 +Epoch [123/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.3464 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3088, Pure Ratio2 10.2010 +Epoch [123/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1529, Pure Ratio2 10.1451 +Epoch [123/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0490 +Epoch [123/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 80.5288 % Model2 79.4071 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.3333 +Epoch [124/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9216 +Epoch [124/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [124/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0833 +Epoch [124/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.0784 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1144, Pure Ratio2 10.0882 +Epoch [124/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 81.0096 % Model2 79.8277 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1569 +Epoch [125/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.2157 +Epoch [125/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2288 +Epoch [125/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.2108 +Epoch [125/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.1490 +Epoch [125/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 10.1634 +Epoch [125/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 79.7175 % Model2 80.3986 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0729 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.5686 +Epoch [126/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.7059 +Epoch [126/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.6471 +Epoch [126/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9265 +Epoch [126/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8706 +Epoch [126/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [126/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 80.6490 % Model2 80.7993 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.6667 +Epoch [127/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.3039 +Epoch [127/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.2941 +Epoch [127/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2696 +Epoch [127/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0196 +Epoch [127/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9837, Pure Ratio2 10.0850 +Epoch [127/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9916, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 80.7392 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.1056 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [128/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [128/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9477, Pure Ratio2 9.9935 +Epoch [128/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1618 +Epoch [128/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0941, Pure Ratio2 10.1725 +Epoch [128/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.2647 +Epoch [128/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 80.6991 % Model2 80.4087 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2941 +Epoch [129/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.7451 +Epoch [129/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6340, Pure Ratio2 9.6993 +Epoch [129/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9461 +Epoch [129/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9961 +Epoch [129/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9771, Pure Ratio2 9.9477 +Epoch [129/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 80.5689 % Model2 81.0697 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6863, Pure Ratio2 10.7843 +Epoch [130/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2647 +Epoch [130/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0850 +Epoch [130/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7549, Pure Ratio2 9.7941 +Epoch [130/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9137 +Epoch [130/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9346 +Epoch [130/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 79.8678 % Model2 80.3085 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8431 +Epoch [131/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1863 +Epoch [131/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0065 +Epoch [131/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1176 +Epoch [131/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1647, Pure Ratio2 10.1412 +Epoch [131/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1797, Pure Ratio2 10.1340 +Epoch [131/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 80.6190 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9216 +Epoch [132/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 10.0490 +Epoch [132/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8105, Pure Ratio2 9.9608 +Epoch [132/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.8725 +Epoch [132/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8157, Pure Ratio2 9.9059 +Epoch [132/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0098 +Epoch [132/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 81.1398 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.3333, Pure Ratio2 10.2941 +Epoch [133/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2451 +Epoch [133/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9150 +Epoch [133/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9902 +Epoch [133/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0745, Pure Ratio2 10.0980 +Epoch [133/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9412 +Epoch [133/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9916, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 79.3870 % Model2 79.8377 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.4902, Pure Ratio2 10.3922 +Epoch [134/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4020 +Epoch [134/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0980 +Epoch [134/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0245 +Epoch [134/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [134/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0196 +Epoch [134/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 80.7692 % Model2 81.6807 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3529 +Epoch [135/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.3235 +Epoch [135/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1830, Pure Ratio2 10.2876 +Epoch [135/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 10.0882 +Epoch [135/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0431, Pure Ratio2 10.0392 +Epoch [135/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0523 +Epoch [135/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0364, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 80.7792 % Model2 80.5489 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5686, Pure Ratio2 10.4510 +Epoch [136/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [136/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.6863 +Epoch [136/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.7745 +Epoch [136/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 9.8510 +Epoch [136/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.9248 +Epoch [136/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 80.8093 % Model2 80.8894 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1765 +Epoch [137/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1569 +Epoch [137/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.1765 +Epoch [137/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.0392 +Epoch [137/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9647, Pure Ratio2 9.9725 +Epoch [137/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 10.0163 +Epoch [137/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 81.0196 % Model2 80.9395 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9020 +Epoch [138/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6667 +Epoch [138/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7451 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9069 +Epoch [138/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9569 +Epoch [138/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9052 +Epoch [138/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 81.1699 % Model2 80.3986 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.7059 +Epoch [139/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0980 +Epoch [139/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9477 +Epoch [139/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9510 +Epoch [139/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0549 +Epoch [139/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 9.9379 +Epoch [139/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 80.7893 % Model2 80.8994 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.1176 +Epoch [140/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9510 +Epoch [140/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0458 +Epoch [140/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7108, Pure Ratio2 9.7402 +Epoch [140/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8667 +Epoch [140/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9575 +Epoch [140/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 81.3201 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9608 +Epoch [141/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0392 +Epoch [141/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.1176 +Epoch [141/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0539 +Epoch [141/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0275 +Epoch [141/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.0490 +Epoch [141/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0280, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 79.9279 % Model2 80.1382 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0855 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9804 +Epoch [142/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.4608 +Epoch [142/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8301 +Epoch [142/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.8333 +Epoch [142/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.9098 +Epoch [142/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.9314 +Epoch [142/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 80.5789 % Model2 81.0296 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.0392 +Epoch [143/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9510 +Epoch [143/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.8889 +Epoch [143/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.0882 +Epoch [143/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2235, Pure Ratio2 10.1569 +Epoch [143/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0654 +Epoch [143/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 80.7592 % Model2 81.3001 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.6275, Pure Ratio2 10.5490 +Epoch [144/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1373 +Epoch [144/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9804 +Epoch [144/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0048, Loss2: 0.0045, Pure Ratio1: 10.1225, Pure Ratio2 10.1520 +Epoch [144/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.0706 +Epoch [144/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0327 +Epoch [144/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0644, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 80.8193 % Model2 80.8794 %, Pure Ratio 1 10.0754 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.1569 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4314, Pure Ratio2 9.3039 +Epoch [145/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [145/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9167, Pure Ratio2 9.9216 +Epoch [145/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [145/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0163, Pure Ratio2 9.9346 +Epoch [145/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9552, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 80.5489 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7059 +Epoch [146/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.1176 +Epoch [146/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9935 +Epoch [146/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0147 +Epoch [146/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0706, Pure Ratio2 10.1294 +Epoch [146/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0327 +Epoch [146/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 81.5004 % Model2 81.4303 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1765 +Epoch [147/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [147/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.1176 +Epoch [147/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0588 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 10.0039 +Epoch [147/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 10.0392 +Epoch [147/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 81.1799 % Model2 81.1198 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.0980 +Epoch [148/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.4902 +Epoch [148/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.6340 +Epoch [148/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7941 +Epoch [148/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 9.9333 +Epoch [148/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8758 +Epoch [148/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 80.6591 % Model2 81.5405 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.6471 +Epoch [149/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9804 +Epoch [149/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9739 +Epoch [149/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 10.0588 +Epoch [149/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9765 +Epoch [149/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9477, Pure Ratio2 9.9510 +Epoch [149/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 80.5188 % Model2 80.7993 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.4118 +Epoch [150/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5196, Pure Ratio2 10.4118 +Epoch [150/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.0523 +Epoch [150/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.1422 +Epoch [150/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 10.0980 +Epoch [150/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.1340 +Epoch [150/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 81.2901 % Model2 81.4904 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8431 +Epoch [151/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.0196 +Epoch [151/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.3529, Pure Ratio2 10.3072 +Epoch [151/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1373 +Epoch [151/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.0902 +Epoch [151/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1405, Pure Ratio2 10.0817 +Epoch [151/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 80.9395 % Model2 81.6506 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7843 +Epoch [152/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8333 +Epoch [152/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.9542 +Epoch [152/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8627 +Epoch [152/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7569, Pure Ratio2 9.8000 +Epoch [152/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8333 +Epoch [152/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 81.2099 % Model2 81.3301 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7647 +Epoch [153/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9706 +Epoch [153/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 10.0065 +Epoch [153/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9167 +Epoch [153/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 9.9451 +Epoch [153/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9902 +Epoch [153/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 80.1182 % Model2 80.9295 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4510, Pure Ratio2 10.3922 +Epoch [154/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.2941 +Epoch [154/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.2745 +Epoch [154/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.1029 +Epoch [154/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.9255, Pure Ratio2 10.0353 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0392 +Epoch [154/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 80.4888 % Model2 81.1498 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 10.0603 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0196 +Epoch [155/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.0490 +Epoch [155/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.8170 +Epoch [155/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 9.9853 +Epoch [155/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1098 +Epoch [155/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.2124 +Epoch [155/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1148, Pure Ratio2 10.1961 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 80.2584 % Model2 81.4002 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6078 +Epoch [156/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6471 +Epoch [156/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8693 +Epoch [156/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.0539 +Epoch [156/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.0902 +Epoch [156/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0719 +Epoch [156/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 80.9095 % Model2 81.1198 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7255 +Epoch [157/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8922 +Epoch [157/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8627 +Epoch [157/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0245 +Epoch [157/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 9.9882 +Epoch [157/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.9314 +Epoch [157/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 80.0581 % Model2 81.3201 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.2745 +Epoch [158/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.8137 +Epoch [158/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.8301 +Epoch [158/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.6814 +Epoch [158/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7255 +Epoch [158/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.7549 +Epoch [158/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 9.9580, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 80.8894 %, Pure Ratio 1 10.0679 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8431 +Epoch [159/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9118 +Epoch [159/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8235 +Epoch [159/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9069 +Epoch [159/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 9.9490 +Epoch [159/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0261 +Epoch [159/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 81.4603 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3922 +Epoch [160/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.3039 +Epoch [160/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3007, Pure Ratio2 10.3072 +Epoch [160/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.2255 +Epoch [160/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1725, Pure Ratio2 10.1961 +Epoch [160/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1013, Pure Ratio2 10.1144 +Epoch [160/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 81.1298 % Model2 81.2300 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.2745 +Epoch [161/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2451 +Epoch [161/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9935 +Epoch [161/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.0735 +Epoch [161/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2588, Pure Ratio2 10.1804 +Epoch [161/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.0850 +Epoch [161/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1064 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 80.2584 % Model2 80.9095 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.5294 +Epoch [162/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1373 +Epoch [162/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.1307 +Epoch [162/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9804 +Epoch [162/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9686 +Epoch [162/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9379 +Epoch [162/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 81.2800 % Model2 81.2600 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5686 +Epoch [163/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [163/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9412 +Epoch [163/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8627 +Epoch [163/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.8392 +Epoch [163/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9575 +Epoch [163/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 80.8794 % Model2 81.0096 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8039 +Epoch [164/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.6961 +Epoch [164/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.7712 +Epoch [164/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7108 +Epoch [164/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.8902 +Epoch [164/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.8987 +Epoch [164/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9916, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 80.6190 % Model2 81.3201 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6667 +Epoch [165/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.5686 +Epoch [165/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6732 +Epoch [165/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8137 +Epoch [165/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 9.9137 +Epoch [165/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8889 +Epoch [165/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 80.8393 % Model2 81.3301 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.5882 +Epoch [166/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2451 +Epoch [166/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0784 +Epoch [166/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1667 +Epoch [166/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0941 +Epoch [166/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [166/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 80.9796 % Model2 81.2300 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.5098 +Epoch [167/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0098 +Epoch [167/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8105 +Epoch [167/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8186 +Epoch [167/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9569 +Epoch [167/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9869 +Epoch [167/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 80.3486 % Model2 81.0697 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7451, Pure Ratio2 10.5490 +Epoch [168/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2255 +Epoch [168/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2680 +Epoch [168/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2794 +Epoch [168/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1882, Pure Ratio2 10.1490 +Epoch [168/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1895 +Epoch [168/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 80.5489 % Model2 81.3301 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Epoch [169/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9804 +Epoch [169/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0850 +Epoch [169/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9167 +Epoch [169/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0706, Pure Ratio2 10.0431 +Epoch [169/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 10.0556 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 81.3802 % Model2 80.6891 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.3725 +Epoch [170/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9118 +Epoch [170/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8693 +Epoch [170/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.0441 +Epoch [170/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9529 +Epoch [170/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8791 +Epoch [170/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 81.2500 % Model2 81.6206 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0196 +Epoch [171/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9314 +Epoch [171/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0196 +Epoch [171/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 9.9951 +Epoch [171/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9451 +Epoch [171/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9183 +Epoch [171/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 81.2400 % Model2 81.6306 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3137 +Epoch [172/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.4216 +Epoch [172/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2418, Pure Ratio2 10.2941 +Epoch [172/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2010 +Epoch [172/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2235, Pure Ratio2 10.1608 +Epoch [172/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0163 +Epoch [172/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 81.4503 % Model2 81.1599 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [173/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8627 +Epoch [173/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.7516 +Epoch [173/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9608 +Epoch [173/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.8588 +Epoch [173/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8333 +Epoch [173/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9664, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 81.4403 %, Pure Ratio 1 10.0603 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1373 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.1176 +Epoch [174/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1569 +Epoch [174/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1618 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1490 +Epoch [174/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0359 +Epoch [174/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9272, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 80.3486 % Model2 80.8494 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5294 +Epoch [175/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5000, Pure Ratio2 9.4902 +Epoch [175/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.5752 +Epoch [175/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 9.6618 +Epoch [175/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.7686 +Epoch [175/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.8595 +Epoch [175/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 81.0897 % Model2 81.4203 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.9020 +Epoch [176/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [176/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8562 +Epoch [176/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0490 +Epoch [176/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0000 +Epoch [176/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8725 +Epoch [176/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 80.8594 % Model2 81.4103 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.0000 +Epoch [177/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.4020 +Epoch [177/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.5098 +Epoch [177/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 9.9020 +Epoch [177/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [177/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9444 +Epoch [177/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9244, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 81.7208 % Model2 81.7007 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [178/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2451 +Epoch [178/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.1242 +Epoch [178/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.1765 +Epoch [178/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.0157 +Epoch [178/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 9.9510 +Epoch [178/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1429, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 81.2099 % Model2 81.1899 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.5882 +Epoch [179/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5588 +Epoch [179/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7516 +Epoch [179/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9020 +Epoch [179/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0941, Pure Ratio2 10.0196 +Epoch [179/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0784 +Epoch [179/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 81.9010 % Model2 81.3702 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1765 +Epoch [180/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.2059 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0850 +Epoch [180/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0343 +Epoch [180/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.0196 +Epoch [180/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.9314 +Epoch [180/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0056, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 81.1899 % Model2 80.9896 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3333 +Epoch [181/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3039, Pure Ratio2 10.2451 +Epoch [181/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.2484 +Epoch [181/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9755 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0784 +Epoch [181/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9346 +Epoch [181/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 80.6591 % Model2 81.3502 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.3137 +Epoch [182/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.1275 +Epoch [182/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.0392 +Epoch [182/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1569 +Epoch [182/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 10.0353 +Epoch [182/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9379 +Epoch [182/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 80.9696 % Model2 81.0697 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0196 +Epoch [183/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9314 +Epoch [183/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.1503 +Epoch [183/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1912 +Epoch [183/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0196 +Epoch [183/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9444 +Epoch [183/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 81.6707 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.1765 +Epoch [184/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9412 +Epoch [184/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9216 +Epoch [184/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8922 +Epoch [184/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0078 +Epoch [184/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0980 +Epoch [184/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0868, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 81.6406 % Model2 81.5104 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.5098 +Epoch [185/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.8431 +Epoch [185/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 9.9804 +Epoch [185/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1324 +Epoch [185/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1255, Pure Ratio2 10.0745 +Epoch [185/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0261 +Epoch [185/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1092, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 81.6406 % Model2 81.1699 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8039 +Epoch [186/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6961 +Epoch [186/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8562 +Epoch [186/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 9.9265 +Epoch [186/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.9961 +Epoch [186/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0556 +Epoch [186/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 81.6807 % Model2 81.7708 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2745 +Epoch [187/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.4412 +Epoch [187/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9935 +Epoch [187/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9363 +Epoch [187/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1216 +Epoch [187/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0882 +Epoch [187/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9636, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 81.3001 % Model2 81.7508 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.4314 +Epoch [188/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3137 +Epoch [188/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1961 +Epoch [188/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9608 +Epoch [188/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.9216 +Epoch [188/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0196 +Epoch [188/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 81.2901 % Model2 81.4303 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.2353 +Epoch [189/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [189/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7124 +Epoch [189/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.8088 +Epoch [189/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8627 +Epoch [189/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9216 +Epoch [189/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1204, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 81.2600 % Model2 81.0797 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.9216, Pure Ratio2 8.9412 +Epoch [190/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1667, Pure Ratio2 9.2353 +Epoch [190/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.8039 +Epoch [190/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8676 +Epoch [190/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.8157 +Epoch [190/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8137 +Epoch [190/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 81.5405 % Model2 81.7208 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7647 +Epoch [191/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8529 +Epoch [191/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0850 +Epoch [191/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.9755 +Epoch [191/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8588, Pure Ratio2 9.8980 +Epoch [191/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9412 +Epoch [191/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8992, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 81.3101 % Model2 81.8810 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.6275 +Epoch [192/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4412 +Epoch [192/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3595 +Epoch [192/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1912, Pure Ratio2 10.1520 +Epoch [192/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0392 +Epoch [192/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1013 +Epoch [192/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 81.1699 % Model2 81.7308 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7647, Pure Ratio2 10.9804 +Epoch [193/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3824, Pure Ratio2 10.6765 +Epoch [193/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3268, Pure Ratio2 10.5033 +Epoch [193/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.2304 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2667 +Epoch [193/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.1405 +Epoch [193/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 81.4403 % Model2 81.6206 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9216 +Epoch [194/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.0000 +Epoch [194/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2222, Pure Ratio2 10.1111 +Epoch [194/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0637 +Epoch [194/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 10.0431 +Epoch [194/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 10.0686 +Epoch [194/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 81.3101 % Model2 81.2600 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5490 +Epoch [195/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5588 +Epoch [195/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8105 +Epoch [195/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8725 +Epoch [195/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7412, Pure Ratio2 9.7294 +Epoch [195/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7745 +Epoch [195/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 81.5004 % Model2 81.2901 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.3333 +Epoch [196/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [196/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3007, Pure Ratio2 10.2092 +Epoch [196/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2500, Pure Ratio2 10.1912 +Epoch [196/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2196, Pure Ratio2 10.1569 +Epoch [196/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1209, Pure Ratio2 10.0392 +Epoch [196/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 81.4603 % Model2 81.4403 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0000 +Epoch [197/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1667 +Epoch [197/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.1373 +Epoch [197/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0441 +Epoch [197/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0392 +Epoch [197/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0556 +Epoch [197/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 81.5405 % Model2 81.3502 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1569 +Epoch [198/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3039 +Epoch [198/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1830 +Epoch [198/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.9069 +Epoch [198/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.8824 +Epoch [198/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8725 +Epoch [198/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 81.4704 % Model2 81.3802 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2745 +Epoch [199/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9118 +Epoch [199/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.7974 +Epoch [199/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.7304 +Epoch [199/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8706, Pure Ratio2 9.8392 +Epoch [199/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [199/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 81.6106 % Model2 81.4503 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5686, Pure Ratio2 10.4902 +Epoch [200/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.2843 +Epoch [200/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.2745 +Epoch [200/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.3088 +Epoch [200/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.1294 +Epoch [200/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0621, Pure Ratio2 10.0980 +Epoch [200/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 81.4403 % Model2 81.2901 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9397 % diff --git a/other_methods/coteaching/coteaching_results/out_4_4.log b/other_methods/coteaching/coteaching_results/out_4_4.log new file mode 100644 index 0000000..b4986a5 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_4_4.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0152, Pure Ratio1: 10.0480, Pure Ratio2 10.0800 +Epoch [2/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0142, Loss2: 0.0144, Pure Ratio1: 10.2240, Pure Ratio2 10.2560 +Epoch [2/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0156, Loss2: 0.0159, Pure Ratio1: 10.2080, Pure Ratio2 10.2240 +Epoch [2/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0150, Loss2: 0.0151, Pure Ratio1: 10.2200, Pure Ratio2 10.2440 +Epoch [2/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.0312, Loss1: 0.0153, Loss2: 0.0158, Pure Ratio1: 10.1984, Pure Ratio2 10.2272 +Epoch [2/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0132, Loss2: 0.0137, Pure Ratio1: 10.1867, Pure Ratio2 10.2187 +Epoch [2/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0133, Loss2: 0.0137, Pure Ratio1: 10.0023, Pure Ratio2 10.0389 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 44.5913 % Model2 42.5280 %, Pure Ratio 1 9.9897 %, Pure Ratio 2 10.0185 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0126, Pure Ratio1: 9.2131, Pure Ratio2 9.1475 +Epoch [3/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 35.9375, Loss1: 0.0140, Loss2: 0.0135, Pure Ratio1: 9.7049, Pure Ratio2 9.6967 +Epoch [3/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0126, Loss2: 0.0126, Pure Ratio1: 9.8689, Pure Ratio2 9.7923 +Epoch [3/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0126, Loss2: 0.0125, Pure Ratio1: 10.0328, Pure Ratio2 9.9836 +Epoch [3/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0133, Loss2: 0.0136, Pure Ratio1: 10.0623, Pure Ratio2 10.0164 +Epoch [3/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0127, Loss2: 0.0124, Pure Ratio1: 10.1202, Pure Ratio2 10.0710 +Epoch [3/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0128, Pure Ratio1: 10.0515, Pure Ratio2 10.0094 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 51.4824 % Model2 53.8161 %, Pure Ratio 1 10.0042 %, Pure Ratio 2 9.9643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0124, Loss2: 0.0126, Pure Ratio1: 9.8992, Pure Ratio2 9.9160 +Epoch [4/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0129, Loss2: 0.0130, Pure Ratio1: 9.7983, Pure Ratio2 9.7815 +Epoch [4/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0122, Loss2: 0.0120, Pure Ratio1: 9.8655, Pure Ratio2 9.9272 +Epoch [4/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0127, Loss2: 0.0125, Pure Ratio1: 9.8403, Pure Ratio2 9.8782 +Epoch [4/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0141, Loss2: 0.0136, Pure Ratio1: 9.7681, Pure Ratio2 9.7950 +Epoch [4/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0111, Loss2: 0.0119, Pure Ratio1: 9.8543, Pure Ratio2 9.8796 +Epoch [4/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0107, Loss2: 0.0114, Pure Ratio1: 9.9184, Pure Ratio2 9.9616 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 56.1098 % Model2 52.8846 %, Pure Ratio 1 9.9591 %, Pure Ratio 2 10.0043 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0129, Loss2: 0.0130, Pure Ratio1: 10.0000, Pure Ratio2 10.0862 +Epoch [5/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0100, Loss2: 0.0105, Pure Ratio1: 10.0776, Pure Ratio2 10.1379 +Epoch [5/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0124, Loss2: 0.0124, Pure Ratio1: 10.1034, Pure Ratio2 10.1379 +Epoch [5/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0123, Loss2: 0.0119, Pure Ratio1: 9.9871, Pure Ratio2 10.0086 +Epoch [5/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0105, Loss2: 0.0101, Pure Ratio1: 10.1172, Pure Ratio2 10.1379 +Epoch [5/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0108, Loss2: 0.0109, Pure Ratio1: 9.9885, Pure Ratio2 9.9914 +Epoch [5/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0122, Loss2: 0.0121, Pure Ratio1: 10.0074, Pure Ratio2 10.0025 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 57.5020 % Model2 55.7292 %, Pure Ratio 1 9.9624 %, Pure Ratio 2 9.9602 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0122, Pure Ratio1: 10.1062, Pure Ratio2 9.9823 +Epoch [6/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0129, Loss2: 0.0134, Pure Ratio1: 10.2478, Pure Ratio2 10.1858 +Epoch [6/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0104, Loss2: 0.0102, Pure Ratio1: 10.0649, Pure Ratio2 10.0295 +Epoch [6/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0107, Loss2: 0.0107, Pure Ratio1: 10.1018, Pure Ratio2 10.0619 +Epoch [6/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0101, Loss2: 0.0107, Pure Ratio1: 10.0814, Pure Ratio2 10.0743 +Epoch [6/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0113, Loss2: 0.0114, Pure Ratio1: 10.0088, Pure Ratio2 9.9971 +Epoch [6/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0096, Loss2: 0.0100, Pure Ratio1: 9.9393, Pure Ratio2 9.9393 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 57.8626 % Model2 54.5773 %, Pure Ratio 1 10.0068 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0109, Loss2: 0.0109, Pure Ratio1: 9.9273, Pure Ratio2 9.8000 +Epoch [7/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0107, Loss2: 0.0106, Pure Ratio1: 9.9909, Pure Ratio2 9.9182 +Epoch [7/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0118, Loss2: 0.0126, Pure Ratio1: 10.1636, Pure Ratio2 10.1333 +Epoch [7/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0098, Loss2: 0.0099, Pure Ratio1: 10.0500, Pure Ratio2 10.0636 +Epoch [7/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0111, Loss2: 0.0111, Pure Ratio1: 10.0291, Pure Ratio2 10.0400 +Epoch [7/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0094, Loss2: 0.0097, Pure Ratio1: 10.0152, Pure Ratio2 9.9788 +Epoch [7/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0107, Loss2: 0.0104, Pure Ratio1: 9.9688, Pure Ratio2 9.9299 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 67.5881 % Model2 67.8986 %, Pure Ratio 1 9.9347 %, Pure Ratio 2 9.8928 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0091, Loss2: 0.0085, Pure Ratio1: 10.0926, Pure Ratio2 10.2407 +Epoch [8/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0111, Loss2: 0.0107, Pure Ratio1: 9.7870, Pure Ratio2 9.8241 +Epoch [8/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0098, Loss2: 0.0099, Pure Ratio1: 9.9568, Pure Ratio2 9.9753 +Epoch [8/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0086, Loss2: 0.0090, Pure Ratio1: 9.8565, Pure Ratio2 9.8796 +Epoch [8/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0097, Loss2: 0.0106, Pure Ratio1: 9.9926, Pure Ratio2 10.0037 +Epoch [8/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0107, Loss2: 0.0095, Pure Ratio1: 9.9877, Pure Ratio2 10.0185 +Epoch [8/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.0000, Loss1: 0.0110, Loss2: 0.0105, Pure Ratio1: 9.9074, Pure Ratio2 9.9339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 64.9840 % Model2 67.0272 %, Pure Ratio 1 9.9406 %, Pure Ratio 2 9.9620 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0085, Loss2: 0.0088, Pure Ratio1: 9.7905, Pure Ratio2 9.8857 +Epoch [9/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0093, Loss2: 0.0089, Pure Ratio1: 9.5048, Pure Ratio2 9.5238 +Epoch [9/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0070, Loss2: 0.0071, Pure Ratio1: 9.6000, Pure Ratio2 9.6508 +Epoch [9/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0096, Loss2: 0.0100, Pure Ratio1: 9.6429, Pure Ratio2 9.6810 +Epoch [9/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0107, Loss2: 0.0108, Pure Ratio1: 9.7905, Pure Ratio2 9.8133 +Epoch [9/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0085, Loss2: 0.0082, Pure Ratio1: 9.9048, Pure Ratio2 9.9270 +Epoch [9/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0099, Loss2: 0.0098, Pure Ratio1: 9.8122, Pure Ratio2 9.8694 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 66.7268 % Model2 68.9303 %, Pure Ratio 1 9.8706 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0095, Loss2: 0.0096, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [10/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 39.8438, Loss1: 0.0102, Loss2: 0.0109, Pure Ratio1: 10.2059, Pure Ratio2 10.1275 +Epoch [10/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0108, Loss2: 0.0097, Pure Ratio1: 9.8431, Pure Ratio2 9.8301 +Epoch [10/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0100, Loss2: 0.0099, Pure Ratio1: 9.8873, Pure Ratio2 9.8725 +Epoch [10/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0085, Loss2: 0.0090, Pure Ratio1: 9.9020, Pure Ratio2 9.9176 +Epoch [10/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0102, Loss2: 0.0099, Pure Ratio1: 9.8856, Pure Ratio2 9.9150 +Epoch [10/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0104, Pure Ratio1: 9.9132, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 65.9555 % Model2 62.8806 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0080, Loss2: 0.0074, Pure Ratio1: 9.3922, Pure Ratio2 9.5686 +Epoch [11/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0082, Loss2: 0.0087, Pure Ratio1: 9.8431, Pure Ratio2 10.0196 +Epoch [11/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0085, Loss2: 0.0096, Pure Ratio1: 9.8562, Pure Ratio2 9.9739 +Epoch [11/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0084, Loss2: 0.0095, Pure Ratio1: 9.8922, Pure Ratio2 9.9755 +Epoch [11/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0087, Loss2: 0.0089, Pure Ratio1: 9.8078, Pure Ratio2 9.8941 +Epoch [11/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0098, Loss2: 0.0082, Pure Ratio1: 9.8497, Pure Ratio2 9.8824 +Epoch [11/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0081, Loss2: 0.0081, Pure Ratio1: 9.8179, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 63.9523 % Model2 66.2660 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0079, Loss2: 0.0079, Pure Ratio1: 9.2549, Pure Ratio2 9.2745 +Epoch [12/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0079, Loss2: 0.0079, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [12/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0071, Loss2: 0.0067, Pure Ratio1: 9.6863, Pure Ratio2 9.6993 +Epoch [12/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0065, Loss2: 0.0064, Pure Ratio1: 9.7304, Pure Ratio2 9.7598 +Epoch [12/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.0000, Loss1: 0.0077, Loss2: 0.0084, Pure Ratio1: 9.7333, Pure Ratio2 9.7216 +Epoch [12/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0105, Loss2: 0.0109, Pure Ratio1: 9.6863, Pure Ratio2 9.6993 +Epoch [12/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0067, Loss2: 0.0060, Pure Ratio1: 9.7675, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 64.3730 % Model2 65.8554 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0081, Loss2: 0.0080, Pure Ratio1: 9.2745, Pure Ratio2 9.5882 +Epoch [13/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0089, Loss2: 0.0074, Pure Ratio1: 9.6863, Pure Ratio2 9.7353 +Epoch [13/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0073, Loss2: 0.0069, Pure Ratio1: 9.4575, Pure Ratio2 9.4641 +Epoch [13/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0090, Loss2: 0.0087, Pure Ratio1: 9.5833, Pure Ratio2 9.5588 +Epoch [13/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 43.7500, Loss1: 0.0095, Loss2: 0.0106, Pure Ratio1: 9.7294, Pure Ratio2 9.7059 +Epoch [13/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0077, Loss2: 0.0078, Pure Ratio1: 9.8824, Pure Ratio2 9.8627 +Epoch [13/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0089, Loss2: 0.0090, Pure Ratio1: 9.9328, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 66.4463 % Model2 67.3978 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0065, Loss2: 0.0071, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [14/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0089, Loss2: 0.0097, Pure Ratio1: 9.7255, Pure Ratio2 9.6078 +Epoch [14/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0086, Loss2: 0.0085, Pure Ratio1: 9.7059, Pure Ratio2 9.6340 +Epoch [14/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0079, Loss2: 0.0074, Pure Ratio1: 9.8039, Pure Ratio2 9.7598 +Epoch [14/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0075, Loss2: 0.0078, Pure Ratio1: 9.7216, Pure Ratio2 9.6824 +Epoch [14/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0092, Loss2: 0.0090, Pure Ratio1: 9.7092, Pure Ratio2 9.6667 +Epoch [14/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0066, Loss2: 0.0067, Pure Ratio1: 9.8151, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 69.2808 % Model2 70.4828 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0080, Loss2: 0.0079, Pure Ratio1: 8.9412, Pure Ratio2 9.1176 +Epoch [15/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0063, Loss2: 0.0062, Pure Ratio1: 9.5980, Pure Ratio2 9.7353 +Epoch [15/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0083, Loss2: 0.0081, Pure Ratio1: 9.7516, Pure Ratio2 9.8627 +Epoch [15/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0080, Loss2: 0.0085, Pure Ratio1: 10.0098, Pure Ratio2 10.1422 +Epoch [15/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0085, Loss2: 0.0089, Pure Ratio1: 10.0824, Pure Ratio2 10.1529 +Epoch [15/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0069, Loss2: 0.0071, Pure Ratio1: 10.0458, Pure Ratio2 10.0980 +Epoch [15/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0072, Loss2: 0.0075, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 70.1923 % Model2 70.9435 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0095, Loss2: 0.0090, Pure Ratio1: 9.5098, Pure Ratio2 9.4118 +Epoch [16/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0090, Loss2: 0.0083, Pure Ratio1: 9.3039, Pure Ratio2 9.2941 +Epoch [16/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0077, Loss2: 0.0077, Pure Ratio1: 9.6797, Pure Ratio2 9.6667 +Epoch [16/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0066, Loss2: 0.0074, Pure Ratio1: 9.7402, Pure Ratio2 9.7157 +Epoch [16/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0069, Loss2: 0.0067, Pure Ratio1: 9.7922, Pure Ratio2 9.7804 +Epoch [16/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0114, Loss2: 0.0104, Pure Ratio1: 9.8399, Pure Ratio2 9.8431 +Epoch [16/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0074, Loss2: 0.0071, Pure Ratio1: 9.8179, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 68.6098 % Model2 65.6751 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0067, Loss2: 0.0073, Pure Ratio1: 9.4118, Pure Ratio2 9.3529 +Epoch [17/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0044, Loss2: 0.0045, Pure Ratio1: 9.5392, Pure Ratio2 9.4902 +Epoch [17/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0066, Loss2: 0.0054, Pure Ratio1: 9.7255, Pure Ratio2 9.7582 +Epoch [17/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0075, Loss2: 0.0078, Pure Ratio1: 9.8873, Pure Ratio2 9.8676 +Epoch [17/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0048, Loss2: 0.0053, Pure Ratio1: 9.9373, Pure Ratio2 9.8980 +Epoch [17/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0067, Loss2: 0.0071, Pure Ratio1: 9.9183, Pure Ratio2 9.8856 +Epoch [17/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0090, Loss2: 0.0102, Pure Ratio1: 9.8515, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 66.9071 % Model2 66.6266 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0049, Loss2: 0.0053, Pure Ratio1: 9.7647, Pure Ratio2 9.7843 +Epoch [18/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0067, Loss2: 0.0071, Pure Ratio1: 9.8431, Pure Ratio2 9.8529 +Epoch [18/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0086, Loss2: 0.0089, Pure Ratio1: 10.0392, Pure Ratio2 9.9935 +Epoch [18/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0050, Loss2: 0.0049, Pure Ratio1: 9.9608, Pure Ratio2 9.9461 +Epoch [18/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0096, Loss2: 0.0100, Pure Ratio1: 9.9294, Pure Ratio2 9.9176 +Epoch [18/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0059, Loss2: 0.0063, Pure Ratio1: 9.8497, Pure Ratio2 9.8529 +Epoch [18/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0071, Loss2: 0.0074, Pure Ratio1: 9.9048, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 66.1158 % Model2 66.6767 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0060, Loss2: 0.0057, Pure Ratio1: 10.0196, Pure Ratio2 9.9216 +Epoch [19/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0077, Loss2: 0.0078, Pure Ratio1: 9.6765, Pure Ratio2 9.5784 +Epoch [19/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0049, Loss2: 0.0044, Pure Ratio1: 9.6471, Pure Ratio2 9.6013 +Epoch [19/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0058, Loss2: 0.0056, Pure Ratio1: 9.6716, Pure Ratio2 9.6667 +Epoch [19/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0058, Loss2: 0.0058, Pure Ratio1: 9.7294, Pure Ratio2 9.6745 +Epoch [19/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0087, Loss2: 0.0077, Pure Ratio1: 9.8007, Pure Ratio2 9.7190 +Epoch [19/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0055, Loss2: 0.0055, Pure Ratio1: 9.8852, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 66.7768 % Model2 67.4980 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0042, Loss2: 0.0044, Pure Ratio1: 10.1569, Pure Ratio2 10.1961 +Epoch [20/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0043, Loss2: 0.0040, Pure Ratio1: 9.8431, Pure Ratio2 9.9216 +Epoch [20/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0054, Loss2: 0.0062, Pure Ratio1: 10.0131, Pure Ratio2 10.0065 +Epoch [20/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0065, Loss2: 0.0061, Pure Ratio1: 9.8922, Pure Ratio2 9.9461 +Epoch [20/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0070, Loss2: 0.0070, Pure Ratio1: 9.8471, Pure Ratio2 9.9059 +Epoch [20/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0049, Loss2: 0.0052, Pure Ratio1: 9.8954, Pure Ratio2 9.9706 +Epoch [20/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0049, Loss2: 0.0052, Pure Ratio1: 9.8123, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 64.4631 % Model2 65.4046 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0036, Loss2: 0.0038, Pure Ratio1: 10.0588, Pure Ratio2 10.1373 +Epoch [21/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0032, Loss2: 0.0035, Pure Ratio1: 9.6471, Pure Ratio2 9.6078 +Epoch [21/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0057, Loss2: 0.0057, Pure Ratio1: 9.8170, Pure Ratio2 9.7908 +Epoch [21/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0056, Loss2: 0.0050, Pure Ratio1: 9.9706, Pure Ratio2 9.9216 +Epoch [21/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0054, Loss2: 0.0052, Pure Ratio1: 10.0275, Pure Ratio2 9.9804 +Epoch [21/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0038, Loss2: 0.0043, Pure Ratio1: 9.9346, Pure Ratio2 9.9118 +Epoch [21/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0064, Loss2: 0.0067, Pure Ratio1: 9.8571, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 62.4900 % Model2 65.3546 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0047, Loss2: 0.0046, Pure Ratio1: 9.4314, Pure Ratio2 9.4118 +Epoch [22/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0051, Loss2: 0.0044, Pure Ratio1: 9.7451, Pure Ratio2 9.8627 +Epoch [22/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 68.7500, Loss1: 0.0028, Loss2: 0.0038, Pure Ratio1: 9.7059, Pure Ratio2 9.7124 +Epoch [22/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0051, Loss2: 0.0049, Pure Ratio1: 9.6961, Pure Ratio2 9.6618 +Epoch [22/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0057, Loss2: 0.0071, Pure Ratio1: 9.8157, Pure Ratio2 9.8314 +Epoch [22/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0054, Loss2: 0.0045, Pure Ratio1: 9.8105, Pure Ratio2 9.8235 +Epoch [22/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0039, Loss2: 0.0036, Pure Ratio1: 9.8936, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 65.0942 % Model2 64.9339 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 79.6875, Loss1: 0.0032, Loss2: 0.0026, Pure Ratio1: 9.9412, Pure Ratio2 9.8235 +Epoch [23/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0031, Loss2: 0.0040, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [23/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0050, Loss2: 0.0039, Pure Ratio1: 9.9085, Pure Ratio2 9.8824 +Epoch [23/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0068, Loss2: 0.0056, Pure Ratio1: 9.9069, Pure Ratio2 9.8578 +Epoch [23/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0026, Loss2: 0.0028, Pure Ratio1: 9.8980, Pure Ratio2 9.8863 +Epoch [23/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0032, Loss2: 0.0035, Pure Ratio1: 9.9935, Pure Ratio2 10.0000 +Epoch [23/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0049, Loss2: 0.0041, Pure Ratio1: 9.9860, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 62.9407 % Model2 63.9423 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0028, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [24/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.7812, Loss1: 0.0040, Loss2: 0.0027, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [24/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0036, Loss2: 0.0032, Pure Ratio1: 9.9477, Pure Ratio2 9.8693 +Epoch [24/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0029, Loss2: 0.0047, Pure Ratio1: 9.9167, Pure Ratio2 9.8627 +Epoch [24/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0030, Loss2: 0.0029, Pure Ratio1: 9.8588, Pure Ratio2 9.8196 +Epoch [24/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0023, Loss2: 0.0022, Pure Ratio1: 9.8203, Pure Ratio2 9.8007 +Epoch [24/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.0031, Loss2: 0.0026, Pure Ratio1: 9.8739, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 59.7155 % Model2 60.5068 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0025, Loss2: 0.0028, Pure Ratio1: 10.3529, Pure Ratio2 10.4314 +Epoch [25/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0026, Loss2: 0.0024, Pure Ratio1: 10.0686, Pure Ratio2 10.1275 +Epoch [25/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0037, Loss2: 0.0035, Pure Ratio1: 10.1961, Pure Ratio2 10.2549 +Epoch [25/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0034, Loss2: 0.0028, Pure Ratio1: 9.9265, Pure Ratio2 10.0049 +Epoch [25/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0042, Loss2: 0.0045, Pure Ratio1: 9.9373, Pure Ratio2 9.9922 +Epoch [25/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0045, Loss2: 0.0036, Pure Ratio1: 9.8203, Pure Ratio2 9.8693 +Epoch [25/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0024, Loss2: 0.0030, Pure Ratio1: 9.8908, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 62.0393 % Model2 62.1895 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0020, Loss2: 0.0029, Pure Ratio1: 10.0392, Pure Ratio2 10.1569 +Epoch [26/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.0023, Loss2: 0.0021, Pure Ratio1: 10.0098, Pure Ratio2 10.1373 +Epoch [26/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0029, Loss2: 0.0025, Pure Ratio1: 9.9608, Pure Ratio2 10.0719 +Epoch [26/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.0033, Loss2: 0.0022, Pure Ratio1: 10.0490, Pure Ratio2 10.1422 +Epoch [26/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0065, Loss2: 0.0046, Pure Ratio1: 9.9294, Pure Ratio2 10.0118 +Epoch [26/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0039, Loss2: 0.0035, Pure Ratio1: 9.9510, Pure Ratio2 10.0196 +Epoch [26/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0041, Loss2: 0.0037, Pure Ratio1: 9.9076, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 62.2296 % Model2 62.5601 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0028, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [27/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0033, Loss2: 0.0027, Pure Ratio1: 9.8333, Pure Ratio2 9.8922 +Epoch [27/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0022, Loss2: 0.0020, Pure Ratio1: 9.8235, Pure Ratio2 9.8562 +Epoch [27/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0028, Loss2: 0.0040, Pure Ratio1: 9.8725, Pure Ratio2 9.8480 +Epoch [27/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0023, Pure Ratio1: 9.8510, Pure Ratio2 9.8392 +Epoch [27/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.0026, Loss2: 0.0020, Pure Ratio1: 9.8399, Pure Ratio2 9.8529 +Epoch [27/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 80.4688, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 9.9356, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 62.6402 % Model2 60.9675 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 76.5625, Loss1: 0.0016, Loss2: 0.0021, Pure Ratio1: 9.1569, Pure Ratio2 9.1373 +Epoch [28/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0039, Loss2: 0.0035, Pure Ratio1: 9.5490, Pure Ratio2 9.5490 +Epoch [28/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0025, Loss2: 0.0017, Pure Ratio1: 9.8431, Pure Ratio2 9.8954 +Epoch [28/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.0021, Loss2: 0.0026, Pure Ratio1: 9.8333, Pure Ratio2 9.9461 +Epoch [28/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 9.6863, Pure Ratio2 9.8157 +Epoch [28/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0024, Loss2: 0.0021, Pure Ratio1: 9.7222, Pure Ratio2 9.8268 +Epoch [28/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0024, Pure Ratio1: 9.7703, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 61.3381 % Model2 61.9591 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 65.6250, Loss1: 0.0026, Loss2: 0.0043, Pure Ratio1: 9.5294, Pure Ratio2 9.5686 +Epoch [29/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0031, Loss2: 0.0034, Pure Ratio1: 9.7941, Pure Ratio2 9.9412 +Epoch [29/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.0015, Loss2: 0.0026, Pure Ratio1: 9.9739, Pure Ratio2 10.1111 +Epoch [29/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0019, Loss2: 0.0022, Pure Ratio1: 9.7843, Pure Ratio2 9.9118 +Epoch [29/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.0032, Loss2: 0.0028, Pure Ratio1: 9.6745, Pure Ratio2 9.7686 +Epoch [29/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.0016, Loss2: 0.0017, Pure Ratio1: 9.7288, Pure Ratio2 9.7843 +Epoch [29/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.0026, Loss2: 0.0030, Pure Ratio1: 9.7843, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 61.0677 % Model2 62.2296 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.0029, Loss2: 0.0016, Pure Ratio1: 9.9216, Pure Ratio2 10.0980 +Epoch [30/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0021, Pure Ratio1: 9.8137, Pure Ratio2 9.9608 +Epoch [30/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0015, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [30/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.7941, Pure Ratio2 9.7990 +Epoch [30/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0015, Pure Ratio1: 9.7490, Pure Ratio2 9.7647 +Epoch [30/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0027, Pure Ratio1: 9.7941, Pure Ratio2 9.8007 +Epoch [30/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0020, Loss2: 0.0030, Pure Ratio1: 9.8291, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 59.6655 % Model2 60.0962 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.5294, Pure Ratio2 9.3529 +Epoch [31/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 9.6275, Pure Ratio2 9.5588 +Epoch [31/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.8105, Pure Ratio2 9.7582 +Epoch [31/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [31/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0015, Pure Ratio1: 10.1765, Pure Ratio2 10.1922 +Epoch [31/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 10.0490, Pure Ratio2 10.0458 +Epoch [31/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.9748, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 61.5284 % Model2 63.0008 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.3137, Pure Ratio2 10.1569 +Epoch [32/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 73.4375, Loss1: 0.0013, Loss2: 0.0025, Pure Ratio1: 9.8824, Pure Ratio2 9.7647 +Epoch [32/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0025, Loss2: 0.0022, Pure Ratio1: 9.8758, Pure Ratio2 9.8105 +Epoch [32/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0020, Loss2: 0.0010, Pure Ratio1: 9.9804, Pure Ratio2 9.9216 +Epoch [32/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0010, Pure Ratio1: 10.0118, Pure Ratio2 9.9686 +Epoch [32/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0029, Loss2: 0.0031, Pure Ratio1: 10.0000, Pure Ratio2 9.9248 +Epoch [32/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0033, Loss2: 0.0031, Pure Ratio1: 10.0196, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 60.3666 % Model2 59.1947 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.0000, Pure Ratio2 9.1961 +Epoch [33/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.2255, Pure Ratio2 9.3333 +Epoch [33/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0015, Pure Ratio1: 9.4052, Pure Ratio2 9.5098 +Epoch [33/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.6863, Pure Ratio2 9.7549 +Epoch [33/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.7255, Pure Ratio2 9.7961 +Epoch [33/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.7812, Loss1: 0.0011, Loss2: 0.0021, Pure Ratio1: 9.7059, Pure Ratio2 9.7680 +Epoch [33/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0014, Pure Ratio1: 9.7339, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 60.2264 % Model2 60.4868 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0021, Loss2: 0.0011, Pure Ratio1: 9.7843, Pure Ratio2 9.9020 +Epoch [34/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [34/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 10.1046, Pure Ratio2 10.0523 +Epoch [34/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 10.0686, Pure Ratio2 10.0539 +Epoch [34/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0010, Pure Ratio1: 9.9765, Pure Ratio2 9.9333 +Epoch [34/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 9.9804, Pure Ratio2 9.9542 +Epoch [34/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0019, Pure Ratio1: 9.9328, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 60.0661 % Model2 59.5954 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0022, Loss2: 0.0022, Pure Ratio1: 9.9804, Pure Ratio2 9.7451 +Epoch [35/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.1078, Pure Ratio2 10.1569 +Epoch [35/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.0000, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 10.1176, Pure Ratio2 10.1634 +Epoch [35/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.8480, Pure Ratio2 9.9363 +Epoch [35/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0020, Pure Ratio1: 9.7725, Pure Ratio2 9.8745 +Epoch [35/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.8562, Pure Ratio2 9.9379 +Epoch [35/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0019, Pure Ratio1: 9.8319, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 58.9944 % Model2 61.3582 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 8.9608, Pure Ratio2 9.1569 +Epoch [36/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0023, Pure Ratio1: 9.4314, Pure Ratio2 9.4706 +Epoch [36/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.8562, Pure Ratio2 9.9542 +Epoch [36/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.8529, Pure Ratio2 9.9069 +Epoch [36/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0013, Pure Ratio1: 9.8745, Pure Ratio2 9.9529 +Epoch [36/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.9052, Pure Ratio2 9.9902 +Epoch [36/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.7955, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 58.8341 % Model2 61.8490 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.6863, Pure Ratio2 10.5882 +Epoch [37/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 10.1667, Pure Ratio2 10.0196 +Epoch [37/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.0850, Pure Ratio2 9.9608 +Epoch [37/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0026, Loss2: 0.0007, Pure Ratio1: 10.1275, Pure Ratio2 10.0931 +Epoch [37/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.9882, Pure Ratio2 9.9412 +Epoch [37/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0013, Pure Ratio1: 9.9575, Pure Ratio2 9.9379 +Epoch [37/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.9468, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 60.2364 % Model2 60.1162 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.7843, Pure Ratio2 9.6078 +Epoch [38/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.7157 +Epoch [38/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8497, Pure Ratio2 9.8105 +Epoch [38/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.7304, Pure Ratio2 9.7402 +Epoch [38/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.8980, Pure Ratio2 9.8745 +Epoch [38/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 9.8203, Pure Ratio2 9.8366 +Epoch [38/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7955, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 61.2380 % Model2 61.1679 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.5098, Pure Ratio2 9.8235 +Epoch [39/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 10.0000, Pure Ratio2 10.0490 +Epoch [39/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.9542, Pure Ratio2 9.8889 +Epoch [39/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0030, Loss2: 0.0014, Pure Ratio1: 9.8627, Pure Ratio2 9.8676 +Epoch [39/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.8588, Pure Ratio2 9.8941 +Epoch [39/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8922, Pure Ratio2 9.9575 +Epoch [39/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9216, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 59.7256 % Model2 59.2548 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 10.7059, Pure Ratio2 10.5882 +Epoch [40/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.2549, Pure Ratio2 10.1765 +Epoch [40/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0010, Pure Ratio1: 10.2353, Pure Ratio2 10.2092 +Epoch [40/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9265, Pure Ratio2 9.9559 +Epoch [40/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.7922, Pure Ratio2 9.8549 +Epoch [40/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8497, Pure Ratio2 9.8889 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8235, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 58.7540 % Model2 60.6170 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.0196 +Epoch [41/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0015, Loss2: 0.0006, Pure Ratio1: 9.8333, Pure Ratio2 9.8235 +Epoch [41/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8693, Pure Ratio2 9.7712 +Epoch [41/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8431, Pure Ratio2 9.7353 +Epoch [41/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 9.9686, Pure Ratio2 9.8196 +Epoch [41/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9641, Pure Ratio2 9.8170 +Epoch [41/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0013, Pure Ratio1: 9.9524, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 58.7240 % Model2 60.2163 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.6078, Pure Ratio2 9.6078 +Epoch [42/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.8333 +Epoch [42/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.9739, Pure Ratio2 9.9216 +Epoch [42/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0245, Pure Ratio2 9.9755 +Epoch [42/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.9843, Pure Ratio2 9.9490 +Epoch [42/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8497, Pure Ratio2 9.8431 +Epoch [42/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8571, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 59.6554 % Model2 60.9575 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.6667, Pure Ratio2 9.7843 +Epoch [43/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9608, Pure Ratio2 10.0098 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9020, Pure Ratio2 9.9346 +Epoch [43/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.9069, Pure Ratio2 9.8922 +Epoch [43/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 10.0745, Pure Ratio2 10.0549 +Epoch [43/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.8660, Pure Ratio2 9.8595 +Epoch [43/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0016, Loss2: 0.0007, Pure Ratio1: 9.9160, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 60.2965 % Model2 60.5268 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.3725, Pure Ratio2 9.3333 +Epoch [44/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.6078, Pure Ratio2 9.5980 +Epoch [44/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.7190 +Epoch [44/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.7059 +Epoch [44/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.9098, Pure Ratio2 9.8549 +Epoch [44/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9542, Pure Ratio2 9.8725 +Epoch [44/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 9.9048, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 61.5084 % Model2 59.6154 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.6667 +Epoch [45/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.6275, Pure Ratio2 9.6961 +Epoch [45/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.8039 +Epoch [45/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.7108 +Epoch [45/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8549, Pure Ratio2 9.8549 +Epoch [45/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7974, Pure Ratio2 9.8268 +Epoch [45/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8179, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 60.1462 % Model2 59.3450 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.7647, Pure Ratio2 9.8235 +Epoch [46/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.6863, Pure Ratio2 9.7451 +Epoch [46/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.7712, Pure Ratio2 9.8431 +Epoch [46/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.6912, Pure Ratio2 9.7549 +Epoch [46/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0015, Pure Ratio1: 9.8706, Pure Ratio2 9.9529 +Epoch [46/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0015, Loss2: 0.0029, Pure Ratio1: 9.8856, Pure Ratio2 9.9314 +Epoch [46/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.8655, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 59.2448 % Model2 61.3181 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.3922, Pure Ratio2 9.2745 +Epoch [47/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.6078, Pure Ratio2 9.5784 +Epoch [47/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8693, Pure Ratio2 9.9281 +Epoch [47/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.7990, Pure Ratio2 9.8284 +Epoch [47/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8902, Pure Ratio2 9.9294 +Epoch [47/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0013, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 9.9444 +Epoch [47/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0013, Pure Ratio1: 9.8459, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 60.6370 % Model2 60.7272 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.1961, Pure Ratio2 9.2745 +Epoch [48/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.6373 +Epoch [48/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.6340, Pure Ratio2 9.6275 +Epoch [48/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.7108, Pure Ratio2 9.7549 +Epoch [48/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7529, Pure Ratio2 9.7961 +Epoch [48/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7549, Pure Ratio2 9.7941 +Epoch [48/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8319, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 60.8273 % Model2 61.4283 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 10.0784 +Epoch [49/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.3824, Pure Ratio2 10.2647 +Epoch [49/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0016, Pure Ratio1: 10.0915, Pure Ratio2 10.0131 +Epoch [49/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 10.1422, Pure Ratio2 10.0490 +Epoch [49/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.1294, Pure Ratio2 10.0196 +Epoch [49/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0015, Pure Ratio1: 9.9837, Pure Ratio2 9.8595 +Epoch [49/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 59.7656 % Model2 58.6438 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.8039, Pure Ratio2 9.5882 +Epoch [50/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9314, Pure Ratio2 9.7549 +Epoch [50/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0261, Pure Ratio2 9.8693 +Epoch [50/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9951, Pure Ratio2 9.8824 +Epoch [50/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0014, Loss2: 0.0005, Pure Ratio1: 9.8667, Pure Ratio2 9.7608 +Epoch [50/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0015, Loss2: 0.0007, Pure Ratio1: 9.8431, Pure Ratio2 9.7157 +Epoch [50/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8796, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 58.3734 % Model2 61.5685 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.3137, Pure Ratio2 10.4314 +Epoch [51/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0784, Pure Ratio2 10.2059 +Epoch [51/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0458, Pure Ratio2 10.0850 +Epoch [51/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9069, Pure Ratio2 9.9951 +Epoch [51/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8784, Pure Ratio2 9.9804 +Epoch [51/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.9150, Pure Ratio2 10.0359 +Epoch [51/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8543, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 59.4151 % Model2 59.1346 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.5098, Pure Ratio2 9.3529 +Epoch [52/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.7941, Pure Ratio2 9.6765 +Epoch [52/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.7712, Pure Ratio2 9.7059 +Epoch [52/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8480 +Epoch [52/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0029, Loss2: 0.0026, Pure Ratio1: 9.9098, Pure Ratio2 9.8471 +Epoch [52/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.9052, Pure Ratio2 9.8268 +Epoch [52/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9048, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 59.3349 % Model2 60.4367 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [53/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [53/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7582, Pure Ratio2 9.8105 +Epoch [53/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.6373, Pure Ratio2 9.6863 +Epoch [53/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.6392, Pure Ratio2 9.6000 +Epoch [53/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.7190 +Epoch [53/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 59.6354 % Model2 60.0461 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.5882 +Epoch [54/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 9.9902, Pure Ratio2 9.6176 +Epoch [54/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 9.9346 +Epoch [54/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 10.0343, Pure Ratio2 9.9069 +Epoch [54/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8824, Pure Ratio2 9.7922 +Epoch [54/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9150, Pure Ratio2 9.8007 +Epoch [54/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8683, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 59.6154 % Model2 60.0060 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 10.1569 +Epoch [55/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [55/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.5294, Pure Ratio2 9.5686 +Epoch [55/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8333, Pure Ratio2 9.8922 +Epoch [55/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.8902, Pure Ratio2 9.9569 +Epoch [55/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7712, Pure Ratio2 9.8562 +Epoch [55/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8992, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 59.1947 % Model2 59.6955 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.6667, Pure Ratio2 9.6471 +Epoch [56/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8137, Pure Ratio2 9.9118 +Epoch [56/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7712, Pure Ratio2 9.8235 +Epoch [56/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.8529 +Epoch [56/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0005, Pure Ratio1: 9.8588, Pure Ratio2 9.9529 +Epoch [56/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.9771, Pure Ratio2 10.0523 +Epoch [56/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.9412, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 60.6370 % Model2 59.2849 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7843, Pure Ratio2 9.8431 +Epoch [57/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7157, Pure Ratio2 9.8137 +Epoch [57/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.6993, Pure Ratio2 9.7516 +Epoch [57/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6912, Pure Ratio2 9.7402 +Epoch [57/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8902 +Epoch [57/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.7484, Pure Ratio2 9.7876 +Epoch [57/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8375, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 59.9659 % Model2 62.6202 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.5882, Pure Ratio2 9.7843 +Epoch [58/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.1275, Pure Ratio2 10.1667 +Epoch [58/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [58/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8578, Pure Ratio2 9.9314 +Epoch [58/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.8627, Pure Ratio2 9.9255 +Epoch [58/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.8007, Pure Ratio2 9.8529 +Epoch [58/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.7759, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 59.9159 % Model2 59.9760 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0018, Loss2: 0.0003, Pure Ratio1: 9.3529, Pure Ratio2 9.2353 +Epoch [59/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1471 +Epoch [59/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.1895, Pure Ratio2 10.2614 +Epoch [59/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0294, Pure Ratio2 10.0833 +Epoch [59/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8980, Pure Ratio2 9.9451 +Epoch [59/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8366, Pure Ratio2 9.8758 +Epoch [59/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0014, Pure Ratio1: 9.8067, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 60.2364 % Model2 61.0076 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.6078, Pure Ratio2 9.3725 +Epoch [60/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 9.8039 +Epoch [60/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 9.9412 +Epoch [60/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.8725 +Epoch [60/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9922, Pure Ratio2 9.8549 +Epoch [60/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9641, Pure Ratio2 9.8464 +Epoch [60/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9916, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 60.4167 % Model2 61.7688 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.0980 +Epoch [61/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0784, Pure Ratio2 10.0392 +Epoch [61/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [61/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 10.0441 +Epoch [61/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8588, Pure Ratio2 9.8745 +Epoch [61/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8497, Pure Ratio2 9.8072 +Epoch [61/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0021, Pure Ratio1: 9.9748, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 59.4251 % Model2 59.7756 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Epoch [62/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.6275 +Epoch [62/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9281, Pure Ratio2 9.7778 +Epoch [62/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0392, Pure Ratio2 9.8725 +Epoch [62/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0706, Pure Ratio2 9.9373 +Epoch [62/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0654, Pure Ratio2 9.9085 +Epoch [62/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.0028, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 60.0160 % Model2 59.6354 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.9216, Pure Ratio2 11.0196 +Epoch [63/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.3824, Pure Ratio2 10.4706 +Epoch [63/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 10.0523 +Epoch [63/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0392, Pure Ratio2 10.1422 +Epoch [63/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.0706, Pure Ratio2 10.2000 +Epoch [63/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9739, Pure Ratio2 10.0588 +Epoch [63/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9832, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 60.0661 % Model2 59.7556 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.7059 +Epoch [64/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.6667, Pure Ratio2 9.5686 +Epoch [64/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.5948, Pure Ratio2 9.4837 +Epoch [64/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7990, Pure Ratio2 9.6765 +Epoch [64/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.7490 +Epoch [64/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8562, Pure Ratio2 9.7843 +Epoch [64/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9132, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 59.8958 % Model2 60.0661 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0014, Pure Ratio1: 9.7059, Pure Ratio2 9.6078 +Epoch [65/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 9.8235 +Epoch [65/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0007, Pure Ratio1: 9.9085, Pure Ratio2 9.9020 +Epoch [65/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8873 +Epoch [65/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8784 +Epoch [65/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.7974, Pure Ratio2 9.8954 +Epoch [65/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8768, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 59.9459 % Model2 59.4551 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.4314, Pure Ratio2 10.4706 +Epoch [66/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [66/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [66/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8775, Pure Ratio2 9.8039 +Epoch [66/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0039, Pure Ratio2 9.9608 +Epoch [66/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9216, Pure Ratio2 9.8693 +Epoch [66/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9272, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 59.2648 % Model2 59.9760 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.8431 +Epoch [67/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.8725 +Epoch [67/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6601, Pure Ratio2 9.8105 +Epoch [67/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 9.9461 +Epoch [67/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7765, Pure Ratio2 9.8471 +Epoch [67/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7778, Pure Ratio2 9.8497 +Epoch [67/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.8235, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 59.4752 % Model2 60.1663 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.2549, Pure Ratio2 9.3725 +Epoch [68/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5294 +Epoch [68/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7320, Pure Ratio2 9.7582 +Epoch [68/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.7157, Pure Ratio2 9.7402 +Epoch [68/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.8000, Pure Ratio2 9.7922 +Epoch [68/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9444 +Epoch [68/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8375, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 59.6454 % Model2 59.8958 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.1373 +Epoch [69/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.2843 +Epoch [69/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0015, Pure Ratio1: 10.0850, Pure Ratio2 10.2484 +Epoch [69/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0833, Pure Ratio2 10.1863 +Epoch [69/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 10.1216 +Epoch [69/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7974, Pure Ratio2 9.9673 +Epoch [69/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8655, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 60.6571 % Model2 61.6587 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.6078 +Epoch [70/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.7941, Pure Ratio2 9.8922 +Epoch [70/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.7647 +Epoch [70/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8676, Pure Ratio2 9.8284 +Epoch [70/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8863, Pure Ratio2 9.9098 +Epoch [70/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9183, Pure Ratio2 9.9281 +Epoch [70/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.9748, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 59.5753 % Model2 59.1546 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2745, Pure Ratio2 10.2157 +Epoch [71/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 10.6078, Pure Ratio2 10.5882 +Epoch [71/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.2614, Pure Ratio2 10.2157 +Epoch [71/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0882, Pure Ratio2 10.0833 +Epoch [71/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9765, Pure Ratio2 10.0157 +Epoch [71/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9641, Pure Ratio2 9.9706 +Epoch [71/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9216, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 59.4151 % Model2 59.0044 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1961 +Epoch [72/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.2941, Pure Ratio2 10.3431 +Epoch [72/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0392, Pure Ratio2 10.1961 +Epoch [72/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9118, Pure Ratio2 10.0539 +Epoch [72/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8157, Pure Ratio2 9.9608 +Epoch [72/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.7810, Pure Ratio2 9.9085 +Epoch [72/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8067, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 59.3950 % Model2 60.4267 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 10.2745 +Epoch [73/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [73/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1503 +Epoch [73/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0147, Pure Ratio2 10.0931 +Epoch [73/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0627, Pure Ratio2 10.0941 +Epoch [73/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9869, Pure Ratio2 10.0490 +Epoch [73/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9944, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 60.5869 % Model2 59.6554 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.0588 +Epoch [74/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.9412, Pure Ratio2 9.8431 +Epoch [74/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7516, Pure Ratio2 9.7451 +Epoch [74/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.9216, Pure Ratio2 9.9412 +Epoch [74/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8667, Pure Ratio2 9.8706 +Epoch [74/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0024, Loss2: 0.0041, Pure Ratio1: 9.9118, Pure Ratio2 9.9085 +Epoch [74/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8964, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 59.5753 % Model2 58.9543 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.8824 +Epoch [75/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0980, Pure Ratio2 10.3137 +Epoch [75/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 10.0458 +Epoch [75/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9363, Pure Ratio2 10.0392 +Epoch [75/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9765, Pure Ratio2 10.0235 +Epoch [75/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.0686, Pure Ratio2 10.0719 +Epoch [75/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9384, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 59.3550 % Model2 60.1863 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 10.1569, Pure Ratio2 9.9804 +Epoch [76/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8137 +Epoch [76/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0011, Pure Ratio1: 9.9150, Pure Ratio2 9.8693 +Epoch [76/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9363, Pure Ratio2 9.8873 +Epoch [76/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8902, Pure Ratio2 9.8000 +Epoch [76/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7582, Pure Ratio2 9.6569 +Epoch [76/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8403, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 60.2464 % Model2 60.3165 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.1373 +Epoch [77/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.9118 +Epoch [77/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6993, Pure Ratio2 9.6928 +Epoch [77/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8382, Pure Ratio2 9.8333 +Epoch [77/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9529, Pure Ratio2 9.9882 +Epoch [77/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.9902 +Epoch [77/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0364, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 59.8458 % Model2 59.4151 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.1765, Pure Ratio2 9.2157 +Epoch [78/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0019, Pure Ratio1: 9.7353, Pure Ratio2 9.7549 +Epoch [78/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 9.9020 +Epoch [78/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7549 +Epoch [78/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7725, Pure Ratio2 9.7647 +Epoch [78/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.8399, Pure Ratio2 9.8660 +Epoch [78/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.9384, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 59.3349 % Model2 60.8273 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.5490 +Epoch [79/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.5882, Pure Ratio2 9.6765 +Epoch [79/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.6536 +Epoch [79/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7108, Pure Ratio2 9.7500 +Epoch [79/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8196, Pure Ratio2 9.8824 +Epoch [79/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8595, Pure Ratio2 9.9020 +Epoch [79/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.9328, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 59.5553 % Model2 60.6671 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.3529, Pure Ratio2 10.1961 +Epoch [80/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.8725 +Epoch [80/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.7908 +Epoch [80/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.7059, Pure Ratio2 9.7549 +Epoch [80/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0016, Pure Ratio1: 9.8902, Pure Ratio2 9.9647 +Epoch [80/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9575, Pure Ratio2 9.9739 +Epoch [80/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8852, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 59.4151 % Model2 58.6939 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.5882, Pure Ratio2 9.7255 +Epoch [81/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 10.0686 +Epoch [81/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 10.0980 +Epoch [81/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.8137 +Epoch [81/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8706, Pure Ratio2 10.0039 +Epoch [81/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8366, Pure Ratio2 9.9706 +Epoch [81/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8599, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 60.1963 % Model2 61.2881 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.9020 +Epoch [82/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.8725 +Epoch [82/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7778 +Epoch [82/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6373, Pure Ratio2 9.6324 +Epoch [82/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8667, Pure Ratio2 9.8588 +Epoch [82/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.8301 +Epoch [82/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8543, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 61.2280 % Model2 61.1478 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0196, Pure Ratio2 10.1176 +Epoch [83/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.1471 +Epoch [83/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [83/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8627 +Epoch [83/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8118, Pure Ratio2 9.8667 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8203, Pure Ratio2 9.8627 +Epoch [83/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8487, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 58.5938 % Model2 59.6254 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [84/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 9.9118 +Epoch [84/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8758, Pure Ratio2 9.6928 +Epoch [84/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8775, Pure Ratio2 9.7010 +Epoch [84/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8706, Pure Ratio2 9.7490 +Epoch [84/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7549 +Epoch [84/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 58.6939 % Model2 58.9744 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 10.2157 +Epoch [85/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 10.0000 +Epoch [85/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 91.4062, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8889 +Epoch [85/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0005, Loss2: 0.0014, Pure Ratio1: 9.8824, Pure Ratio2 9.8824 +Epoch [85/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.8706, Pure Ratio2 9.9176 +Epoch [85/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7974, Pure Ratio2 9.8137 +Epoch [85/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7759, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 58.6839 % Model2 60.4467 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 10.1373 +Epoch [86/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 10.1078 +Epoch [86/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.7386 +Epoch [86/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7794, Pure Ratio2 9.9657 +Epoch [86/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8549, Pure Ratio2 9.9647 +Epoch [86/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.7647 +Epoch [86/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8067, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 59.5453 % Model2 61.0276 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 8.7059, Pure Ratio2 8.9412 +Epoch [87/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.5490, Pure Ratio2 9.6176 +Epoch [87/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.7516 +Epoch [87/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.6961, Pure Ratio2 9.7990 +Epoch [87/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.7569 +Epoch [87/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.7092, Pure Ratio2 9.7745 +Epoch [87/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8515, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 61.3982 % Model2 60.3466 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.5490, Pure Ratio2 10.3725 +Epoch [88/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1373, Pure Ratio2 10.0490 +Epoch [88/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1242, Pure Ratio2 10.0000 +Epoch [88/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 9.9706 +Epoch [88/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0016, Loss2: 0.0003, Pure Ratio1: 10.0118, Pure Ratio2 9.9020 +Epoch [88/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8856, Pure Ratio2 9.7810 +Epoch [88/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 58.7240 % Model2 59.4752 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.6471, Pure Ratio2 9.9020 +Epoch [89/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7353 +Epoch [89/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1176 +Epoch [89/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 10.0931 +Epoch [89/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [89/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8954, Pure Ratio2 9.9902 +Epoch [89/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8655, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 57.7825 % Model2 60.7272 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.3137, Pure Ratio2 10.3725 +Epoch [90/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2451, Pure Ratio2 10.3137 +Epoch [90/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.3529 +Epoch [90/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.1471 +Epoch [90/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0039, Pure Ratio2 10.1255 +Epoch [90/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.9444, Pure Ratio2 10.0654 +Epoch [90/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9328, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 61.6186 % Model2 60.1663 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 9.8431 +Epoch [91/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1667, Pure Ratio2 10.0294 +Epoch [91/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0458, Pure Ratio2 9.9412 +Epoch [91/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9461, Pure Ratio2 9.9461 +Epoch [91/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7804, Pure Ratio2 9.7608 +Epoch [91/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.8889, Pure Ratio2 9.8333 +Epoch [91/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 59.9960 % Model2 58.2532 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.5294 +Epoch [92/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 9.8431 +Epoch [92/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0980, Pure Ratio2 10.0065 +Epoch [92/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9657, Pure Ratio2 9.9608 +Epoch [92/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8353 +Epoch [92/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8725 +Epoch [92/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8403, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 59.2849 % Model2 58.6338 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.0196 +Epoch [93/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.7157 +Epoch [93/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7516, Pure Ratio2 9.6863 +Epoch [93/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7010, Pure Ratio2 9.7696 +Epoch [93/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7686, Pure Ratio2 9.8314 +Epoch [93/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7059, Pure Ratio2 9.7418 +Epoch [93/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7899, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 60.1462 % Model2 59.1847 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.4314 +Epoch [94/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5784, Pure Ratio2 9.5098 +Epoch [94/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5425, Pure Ratio2 9.5294 +Epoch [94/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7794, Pure Ratio2 9.7304 +Epoch [94/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7882, Pure Ratio2 9.7647 +Epoch [94/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7386, Pure Ratio2 9.7288 +Epoch [94/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8515, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 59.9960 % Model2 60.1562 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7843 +Epoch [95/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.1863 +Epoch [95/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.8824 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.0441 +Epoch [95/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8510, Pure Ratio2 9.8431 +Epoch [95/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9771, Pure Ratio2 9.9346 +Epoch [95/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9636, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 59.9159 % Model2 60.5669 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8627 +Epoch [96/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.9118 +Epoch [96/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.9739 +Epoch [96/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7745, Pure Ratio2 9.8873 +Epoch [96/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8196, Pure Ratio2 9.8118 +Epoch [96/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8856, Pure Ratio2 9.8660 +Epoch [96/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 60.1462 % Model2 59.8558 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1961 +Epoch [97/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 9.8725 +Epoch [97/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8366 +Epoch [97/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8578 +Epoch [97/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8510, Pure Ratio2 9.8157 +Epoch [97/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8856 +Epoch [97/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9524, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 59.9960 % Model2 58.8141 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.4706 +Epoch [98/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8529 +Epoch [98/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8366 +Epoch [98/200], Iter [200/390] Training Accuracy1: 95.3125, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9657, Pure Ratio2 9.8971 +Epoch [98/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.8157 +Epoch [98/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0033, Pure Ratio2 9.9412 +Epoch [98/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0672, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 60.0861 % Model2 59.6354 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.8824 +Epoch [99/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.0294 +Epoch [99/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1307, Pure Ratio2 10.1307 +Epoch [99/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0245 +Epoch [99/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2078, Pure Ratio2 10.1804 +Epoch [99/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [99/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9496, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 58.9543 % Model2 60.2464 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.4706 +Epoch [100/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.1078 +Epoch [100/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.0654 +Epoch [100/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8627 +Epoch [100/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8588 +Epoch [100/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.8203 +Epoch [100/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9468, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 58.9042 % Model2 60.6671 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.5686 +Epoch [101/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.7549 +Epoch [101/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 9.9804 +Epoch [101/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0245 +Epoch [101/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 9.9686 +Epoch [101/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 9.8889 +Epoch [101/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 59.2849 % Model2 60.8774 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8039 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.6471 +Epoch [102/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7516, Pure Ratio2 9.8301 +Epoch [102/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.9363 +Epoch [102/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8706, Pure Ratio2 9.9137 +Epoch [102/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8464 +Epoch [102/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9384, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 59.0745 % Model2 59.5453 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7255, Pure Ratio2 9.5686 +Epoch [103/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.7843 +Epoch [103/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0719, Pure Ratio2 9.9608 +Epoch [103/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.2059, Pure Ratio2 10.1029 +Epoch [103/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9725, Pure Ratio2 9.8745 +Epoch [103/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9052 +Epoch [103/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9832, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 60.4167 % Model2 58.2432 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [104/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.4510, Pure Ratio2 9.5098 +Epoch [104/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6928, Pure Ratio2 9.7190 +Epoch [104/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.9461 +Epoch [104/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9922 +Epoch [104/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9837, Pure Ratio2 9.9641 +Epoch [104/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9272, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 59.3349 % Model2 60.2163 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [105/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8922, Pure Ratio2 9.9020 +Epoch [105/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.7582 +Epoch [105/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7206 +Epoch [105/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7686, Pure Ratio2 9.8471 +Epoch [105/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8791, Pure Ratio2 9.9150 +Epoch [105/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 60.4667 % Model2 58.3433 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7059 +Epoch [106/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.7941 +Epoch [106/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.8366 +Epoch [106/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.7794 +Epoch [106/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8980, Pure Ratio2 9.8196 +Epoch [106/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0425, Pure Ratio2 9.9641 +Epoch [106/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.0280, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 59.6354 % Model2 59.5653 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.7647 +Epoch [107/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.7353 +Epoch [107/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.2484 +Epoch [107/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.2157 +Epoch [107/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0510, Pure Ratio2 10.1020 +Epoch [107/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9902, Pure Ratio2 10.0654 +Epoch [107/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 9.9468, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 58.4034 % Model2 60.3866 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.2157, Pure Ratio2 9.4118 +Epoch [108/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.3627, Pure Ratio2 9.3235 +Epoch [108/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.3987, Pure Ratio2 9.3922 +Epoch [108/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.4951 +Epoch [108/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7216, Pure Ratio2 9.6824 +Epoch [108/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8497 +Epoch [108/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9048, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 60.0160 % Model2 59.5753 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.8431, Pure Ratio2 10.6863 +Epoch [109/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.2255 +Epoch [109/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.3464, Pure Ratio2 10.3072 +Epoch [109/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1912, Pure Ratio2 10.1667 +Epoch [109/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0902, Pure Ratio2 10.1333 +Epoch [109/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.0654 +Epoch [109/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0224, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 61.3782 % Model2 61.4483 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.4902 +Epoch [110/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 10.1863 +Epoch [110/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [110/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 10.0147 +Epoch [110/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0314, Pure Ratio2 10.0353 +Epoch [110/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9935, Pure Ratio2 9.9935 +Epoch [110/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9188, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 60.8474 % Model2 59.9960 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.1569, Pure Ratio2 9.3529 +Epoch [111/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.8824 +Epoch [111/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.0196 +Epoch [111/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9559, Pure Ratio2 10.0343 +Epoch [111/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 10.0039 +Epoch [111/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9641, Pure Ratio2 10.0196 +Epoch [111/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9356, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 60.2865 % Model2 59.1246 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 10.1569 +Epoch [112/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 10.0294 +Epoch [112/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 10.1438 +Epoch [112/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0833, Pure Ratio2 10.1912 +Epoch [112/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 10.0902 +Epoch [112/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [112/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 59.3049 % Model2 59.5353 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.6078, Pure Ratio2 10.6667 +Epoch [113/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3431, Pure Ratio2 10.4020 +Epoch [113/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9673 +Epoch [113/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.0049, Pure Ratio2 10.0245 +Epoch [113/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Epoch [113/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9248, Pure Ratio2 9.9412 +Epoch [113/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 60.3766 % Model2 60.0361 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 10.0392 +Epoch [114/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8725, Pure Ratio2 9.9608 +Epoch [114/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0980 +Epoch [114/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0343, Pure Ratio2 10.1471 +Epoch [114/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1725, Pure Ratio2 10.2627 +Epoch [114/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.1078 +Epoch [114/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 60.0260 % Model2 61.3982 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8627 +Epoch [115/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.7745 +Epoch [115/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1307, Pure Ratio2 9.9804 +Epoch [115/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0245, Pure Ratio2 9.9608 +Epoch [115/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8941, Pure Ratio2 9.8824 +Epoch [115/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8595 +Epoch [115/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 60.6370 % Model2 59.2448 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 10.3137, Pure Ratio2 9.9216 +Epoch [116/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.0588 +Epoch [116/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9085 +Epoch [116/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 9.9069 +Epoch [116/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.9333 +Epoch [116/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 9.9444 +Epoch [116/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 59.7256 % Model2 61.7989 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 10.0784 +Epoch [117/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 9.9804 +Epoch [117/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9542 +Epoch [117/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8039 +Epoch [117/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7412, Pure Ratio2 9.7333 +Epoch [117/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.8007, Pure Ratio2 9.7745 +Epoch [117/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8571, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 60.4167 % Model2 59.5152 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.7647 +Epoch [118/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.7549 +Epoch [118/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 10.0719 +Epoch [118/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9265, Pure Ratio2 10.0049 +Epoch [118/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0235, Pure Ratio2 10.0706 +Epoch [118/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9869 +Epoch [118/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9440, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 60.8674 % Model2 59.8357 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.6275, Pure Ratio2 10.5686 +Epoch [119/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5000, Pure Ratio2 10.4216 +Epoch [119/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2418, Pure Ratio2 10.1895 +Epoch [119/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0294 +Epoch [119/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1137, Pure Ratio2 10.0706 +Epoch [119/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0654, Pure Ratio2 10.0359 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9944, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 60.7071 % Model2 60.8373 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8431 +Epoch [120/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1078 +Epoch [120/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 10.1046 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9755 +Epoch [120/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8980, Pure Ratio2 9.9412 +Epoch [120/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8725, Pure Ratio2 9.9052 +Epoch [120/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9160, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 59.6154 % Model2 59.3450 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1961 +Epoch [121/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 9.9706 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7647 +Epoch [121/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.8725 +Epoch [121/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7882, Pure Ratio2 9.7020 +Epoch [121/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.8497 +Epoch [121/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 58.5236 % Model2 59.9860 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 10.2157 +Epoch [122/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.0392 +Epoch [122/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0016, Loss2: 0.0001, Pure Ratio1: 9.6601, Pure Ratio2 9.8889 +Epoch [122/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.7745 +Epoch [122/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7098, Pure Ratio2 9.8039 +Epoch [122/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7974 +Epoch [122/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 60.1863 % Model2 60.2364 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2941, Pure Ratio2 10.4314 +Epoch [123/200], Iter [100/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1176 +Epoch [123/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.0261 +Epoch [123/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1029, Pure Ratio2 10.0196 +Epoch [123/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9725, Pure Ratio2 9.9216 +Epoch [123/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8399 +Epoch [123/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 60.5970 % Model2 61.0076 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.7843 +Epoch [124/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.7647 +Epoch [124/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.9281 +Epoch [124/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0980 +Epoch [124/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9373, Pure Ratio2 10.0549 +Epoch [124/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0621 +Epoch [124/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 58.7440 % Model2 59.6855 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4314, Pure Ratio2 9.7647 +Epoch [125/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.9412 +Epoch [125/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.1046 +Epoch [125/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9755 +Epoch [125/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8314, Pure Ratio2 9.9451 +Epoch [125/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.0392 +Epoch [125/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8739, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 59.1947 % Model2 60.4267 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [126/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6569, Pure Ratio2 9.7255 +Epoch [126/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6405, Pure Ratio2 9.6340 +Epoch [126/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8235 +Epoch [126/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8353, Pure Ratio2 9.8314 +Epoch [126/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.9183 +Epoch [126/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 59.5052 % Model2 60.2063 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.5686 +Epoch [127/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [127/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1373 +Epoch [127/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 10.0441, Pure Ratio2 9.9755 +Epoch [127/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8549, Pure Ratio2 9.8392 +Epoch [127/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8791 +Epoch [127/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 60.0861 % Model2 59.3349 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.4706 +Epoch [128/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [128/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [128/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.0588 +Epoch [128/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 9.9529 +Epoch [128/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0654, Pure Ratio2 10.0817 +Epoch [128/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9468, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 59.2849 % Model2 60.9976 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 9.9020 +Epoch [129/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.6863 +Epoch [129/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8824 +Epoch [129/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 10.0000 +Epoch [129/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8157, Pure Ratio2 9.9216 +Epoch [129/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.9052 +Epoch [129/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 59.9058 % Model2 60.3065 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 11.0392, Pure Ratio2 10.6471 +Epoch [130/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4216, Pure Ratio2 10.1176 +Epoch [130/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 9.9477 +Epoch [130/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.7206 +Epoch [130/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9882, Pure Ratio2 9.9569 +Epoch [130/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.9118 +Epoch [130/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 60.6370 % Model2 60.5369 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9804 +Epoch [131/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9216 +Epoch [131/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9804 +Epoch [131/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [131/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 10.0000 +Epoch [131/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9967 +Epoch [131/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9244, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 59.5252 % Model2 61.2981 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.4314 +Epoch [132/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.4216 +Epoch [132/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5163, Pure Ratio2 9.4118 +Epoch [132/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6029, Pure Ratio2 9.5294 +Epoch [132/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6784, Pure Ratio2 9.6157 +Epoch [132/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8105, Pure Ratio2 9.7353 +Epoch [132/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8880, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 60.5569 % Model2 60.4267 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.2941 +Epoch [133/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [133/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9542, Pure Ratio2 10.1438 +Epoch [133/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.1176 +Epoch [133/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0745, Pure Ratio2 10.1843 +Epoch [133/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.1013 +Epoch [133/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.1176 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 60.4768 % Model2 60.6270 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4706 +Epoch [134/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.5196 +Epoch [134/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1503 +Epoch [134/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8725 +Epoch [134/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9451, Pure Ratio2 9.9176 +Epoch [134/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9085 +Epoch [134/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 59.5553 % Model2 60.0761 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.7451 +Epoch [135/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8039 +Epoch [135/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9281 +Epoch [135/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0147 +Epoch [135/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 9.9294 +Epoch [135/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0131, Pure Ratio2 9.9608 +Epoch [135/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 60.9275 % Model2 59.3249 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6078 +Epoch [136/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.5980 +Epoch [136/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.5033 +Epoch [136/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.5833 +Epoch [136/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [136/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.6993 +Epoch [136/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8739, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 59.9960 % Model2 60.1963 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.3529, Pure Ratio2 10.1765 +Epoch [137/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2255 +Epoch [137/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 10.0261 +Epoch [137/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9902 +Epoch [137/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 10.0275 +Epoch [137/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1373 +Epoch [137/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9832, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 60.5268 % Model2 60.0861 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 11.0784, Pure Ratio2 10.7647 +Epoch [138/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.0588 +Epoch [138/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 9.9608 +Epoch [138/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1225 +Epoch [138/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0627, Pure Ratio2 10.0471 +Epoch [138/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 10.0033 +Epoch [138/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 60.8373 % Model2 61.1378 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1961 +Epoch [139/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.7255 +Epoch [139/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.6405 +Epoch [139/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6422 +Epoch [139/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8392 +Epoch [139/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8137, Pure Ratio2 9.7745 +Epoch [139/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 59.8357 % Model2 60.5970 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.8235 +Epoch [140/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9510 +Epoch [140/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 10.1046 +Epoch [140/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 10.0294 +Epoch [140/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0745, Pure Ratio2 10.1137 +Epoch [140/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0098 +Epoch [140/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 61.6286 % Model2 60.5669 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9216 +Epoch [141/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.9804 +Epoch [141/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.1176 +Epoch [141/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8137 +Epoch [141/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7176, Pure Ratio2 9.8000 +Epoch [141/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.9510 +Epoch [141/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8487, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 61.0276 % Model2 60.0060 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9412 +Epoch [142/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.6176 +Epoch [142/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9085 +Epoch [142/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8480 +Epoch [142/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7882 +Epoch [142/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.8170 +Epoch [142/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 60.2063 % Model2 60.4467 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 10.0392 +Epoch [143/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7745, Pure Ratio2 9.9706 +Epoch [143/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8889 +Epoch [143/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.8775 +Epoch [143/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9216 +Epoch [143/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.9150 +Epoch [143/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 59.0244 % Model2 59.6655 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 10.0784 +Epoch [144/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8725, Pure Ratio2 10.1863 +Epoch [144/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7778 +Epoch [144/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8480 +Epoch [144/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 10.0078 +Epoch [144/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9706 +Epoch [144/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 60.1663 % Model2 62.0292 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.9216, Pure Ratio2 8.9412 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.2451, Pure Ratio2 9.3333 +Epoch [145/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3791, Pure Ratio2 9.4575 +Epoch [145/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.7402 +Epoch [145/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7647 +Epoch [145/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.8203 +Epoch [145/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8123, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 59.9058 % Model2 59.9559 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [146/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [146/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1699 +Epoch [146/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0735, Pure Ratio2 10.1176 +Epoch [146/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0627, Pure Ratio2 10.0784 +Epoch [146/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 10.0131 +Epoch [146/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 59.9760 % Model2 59.5252 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [147/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 9.8922 +Epoch [147/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 9.9542 +Epoch [147/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8627 +Epoch [147/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 9.8314 +Epoch [147/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.9379 +Epoch [147/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9636, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 60.3065 % Model2 61.0076 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.4510 +Epoch [148/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8039 +Epoch [148/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9085, Pure Ratio2 9.8693 +Epoch [148/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 9.9510 +Epoch [148/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.0157 +Epoch [148/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [148/200], Iter [350/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 60.5669 % Model2 60.5769 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [149/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8529 +Epoch [149/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 9.9542 +Epoch [149/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9755 +Epoch [149/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 9.9882 +Epoch [149/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9739 +Epoch [149/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0252, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 60.2764 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.3333 +Epoch [150/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.5392, Pure Ratio2 10.2451 +Epoch [150/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 9.8235 +Epoch [150/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1716, Pure Ratio2 9.9608 +Epoch [150/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1333, Pure Ratio2 9.9294 +Epoch [150/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9379 +Epoch [150/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 59.3349 % Model2 59.7456 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6275 +Epoch [151/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.6275 +Epoch [151/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8693 +Epoch [151/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.9412 +Epoch [151/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9098, Pure Ratio2 9.8784 +Epoch [151/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8758 +Epoch [151/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 59.7556 % Model2 59.9659 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.4314 +Epoch [152/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.5392 +Epoch [152/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.6667 +Epoch [152/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.6373 +Epoch [152/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7412, Pure Ratio2 9.6667 +Epoch [152/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7288, Pure Ratio2 9.6601 +Epoch [152/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0017, Pure Ratio1: 9.8179, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 59.9259 % Model2 60.2063 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.2157 +Epoch [153/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.0490 +Epoch [153/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.0523 +Epoch [153/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0588 +Epoch [153/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9490 +Epoch [153/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.9248 +Epoch [153/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9636, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 60.5168 % Model2 59.9860 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.4706 +Epoch [154/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 10.0588 +Epoch [154/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1699 +Epoch [154/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8382, Pure Ratio2 9.9461 +Epoch [154/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 10.0078 +Epoch [154/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.9020 +Epoch [154/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8655, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 60.6070 % Model2 59.8357 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7059 +Epoch [155/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7745 +Epoch [155/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6209 +Epoch [155/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.7059 +Epoch [155/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8863 +Epoch [155/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9739 +Epoch [155/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 60.3966 % Model2 59.7556 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.7255 +Epoch [156/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.8235 +Epoch [156/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9281 +Epoch [156/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9755 +Epoch [156/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0235 +Epoch [156/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0621, Pure Ratio2 9.9118 +Epoch [156/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0756, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 60.4267 % Model2 60.2163 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.5098 +Epoch [157/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.7157 +Epoch [157/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9935 +Epoch [157/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1324 +Epoch [157/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0667 +Epoch [157/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.9706 +Epoch [157/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8683, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 60.7873 % Model2 60.8674 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [158/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.2843 +Epoch [158/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1503 +Epoch [158/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 10.0833 +Epoch [158/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.9647 +Epoch [158/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8203 +Epoch [158/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 60.4868 % Model2 59.8458 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8039 +Epoch [159/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.7843 +Epoch [159/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8105, Pure Ratio2 9.7059 +Epoch [159/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8676, Pure Ratio2 9.7696 +Epoch [159/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.7608 +Epoch [159/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0033, Pure Ratio2 9.8987 +Epoch [159/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 60.2664 % Model2 59.7756 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.3725 +Epoch [160/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.0784 +Epoch [160/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.0588 +Epoch [160/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1520, Pure Ratio2 10.0049 +Epoch [160/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2078, Pure Ratio2 10.0824 +Epoch [160/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 9.9444 +Epoch [160/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 60.0461 % Model2 59.7256 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.9412 +Epoch [161/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 10.0588 +Epoch [161/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5425, Pure Ratio2 9.6601 +Epoch [161/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6863 +Epoch [161/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8000, Pure Ratio2 9.7804 +Epoch [161/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.8105 +Epoch [161/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 60.6771 % Model2 60.3265 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.1961 +Epoch [162/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9804 +Epoch [162/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0023, Pure Ratio1: 10.0065, Pure Ratio2 9.9020 +Epoch [162/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8382 +Epoch [162/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0667, Pure Ratio2 10.0157 +Epoch [162/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9542 +Epoch [162/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 59.8257 % Model2 58.4936 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.8431 +Epoch [163/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.5686 +Epoch [163/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.8366 +Epoch [163/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7500, Pure Ratio2 9.8529 +Epoch [163/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8275, Pure Ratio2 9.9098 +Epoch [163/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 10.0131 +Epoch [163/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8487, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 60.4267 % Model2 60.4667 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6078 +Epoch [164/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.6765 +Epoch [164/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.8039 +Epoch [164/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8725 +Epoch [164/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.9451 +Epoch [164/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9183 +Epoch [164/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 60.6070 % Model2 61.0076 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7647 +Epoch [165/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.5588 +Epoch [165/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5556 +Epoch [165/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.7059 +Epoch [165/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.8902 +Epoch [165/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8922 +Epoch [165/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 60.7572 % Model2 61.5184 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [166/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.0294 +Epoch [166/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.8497 +Epoch [166/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.9363 +Epoch [166/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9961, Pure Ratio2 9.8863 +Epoch [166/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.8529 +Epoch [166/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 60.0661 % Model2 60.1663 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.6471 +Epoch [167/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [167/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8366 +Epoch [167/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8088 +Epoch [167/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9294 +Epoch [167/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9804 +Epoch [167/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0364, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 61.3181 % Model2 60.9275 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [168/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0882 +Epoch [168/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9673, Pure Ratio2 10.0131 +Epoch [168/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1078 +Epoch [168/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0902, Pure Ratio2 10.0941 +Epoch [168/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1765 +Epoch [168/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 60.2965 % Model2 61.3882 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [169/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2647 +Epoch [169/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.1438 +Epoch [169/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0441 +Epoch [169/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8980, Pure Ratio2 9.9608 +Epoch [169/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9575 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9552, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 60.3165 % Model2 61.2179 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6863 +Epoch [170/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9412 +Epoch [170/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.7059 +Epoch [170/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.8873 +Epoch [170/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.9216 +Epoch [170/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.8431 +Epoch [170/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 60.5569 % Model2 60.5268 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9804 +Epoch [171/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7647 +Epoch [171/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8039 +Epoch [171/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9314 +Epoch [171/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9686 +Epoch [171/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.9183 +Epoch [171/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 60.3666 % Model2 61.0978 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.4510 +Epoch [172/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1863 +Epoch [172/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1634, Pure Ratio2 10.1634 +Epoch [172/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0441, Pure Ratio2 9.9853 +Epoch [172/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1059 +Epoch [172/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0131 +Epoch [172/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 61.0777 % Model2 61.8990 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1373 +Epoch [173/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9216 +Epoch [173/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0458 +Epoch [173/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9412 +Epoch [173/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7529, Pure Ratio2 9.7725 +Epoch [173/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.8072 +Epoch [173/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8011, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 60.5369 % Model2 61.0877 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.0980 +Epoch [174/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0882 +Epoch [174/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [174/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0980 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.0549 +Epoch [174/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9346 +Epoch [174/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 61.9391 % Model2 60.7472 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.7843 +Epoch [175/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.8333 +Epoch [175/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.9150 +Epoch [175/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0196 +Epoch [175/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9686 +Epoch [175/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 10.0948 +Epoch [175/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 61.2981 % Model2 60.6771 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4902, Pure Ratio2 10.3529 +Epoch [176/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8039 +Epoch [176/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.7778 +Epoch [176/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.8431 +Epoch [176/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.7608 +Epoch [176/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.7320 +Epoch [176/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9244, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 60.9575 % Model2 61.3782 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.8627, Pure Ratio2 8.9608 +Epoch [177/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7157 +Epoch [177/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [177/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9412 +Epoch [177/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.9059 +Epoch [177/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9869 +Epoch [177/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 60.4667 % Model2 60.8474 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.4902 +Epoch [178/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1863 +Epoch [178/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1569 +Epoch [178/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.1422 +Epoch [178/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0745 +Epoch [178/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9281 +Epoch [178/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 60.6971 % Model2 60.8574 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7647 +Epoch [179/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7157 +Epoch [179/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7712 +Epoch [179/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.8039 +Epoch [179/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 9.9490 +Epoch [179/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.0784 +Epoch [179/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 61.7087 % Model2 60.9375 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.3725 +Epoch [180/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9510 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9346 +Epoch [180/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.8873 +Epoch [180/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.8902 +Epoch [180/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8431 +Epoch [180/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 60.4968 % Model2 60.7772 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [181/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.1078 +Epoch [181/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9346 +Epoch [181/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.8971 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0471 +Epoch [181/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.8758 +Epoch [181/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0224, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 61.2079 % Model2 60.0461 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4902 +Epoch [182/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.0882 +Epoch [182/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0980 +Epoch [182/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.1176 +Epoch [182/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0039 +Epoch [182/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8824 +Epoch [182/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 60.1963 % Model2 60.8574 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 10.0000 +Epoch [183/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8529 +Epoch [183/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.8627 +Epoch [183/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0490 +Epoch [183/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 10.0078 +Epoch [183/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0196 +Epoch [183/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 60.1763 % Model2 60.2364 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3529 +Epoch [184/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9020 +Epoch [184/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9150 +Epoch [184/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.8824 +Epoch [184/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.0392 +Epoch [184/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.0784 +Epoch [184/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 60.6671 % Model2 60.3666 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0042, Loss2: 0.0043, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [185/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7255 +Epoch [185/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 10.0261 +Epoch [185/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.1422 +Epoch [185/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [185/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.0752 +Epoch [185/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0560, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 61.0677 % Model2 60.9575 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.4510 +Epoch [186/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3824, Pure Ratio2 9.3824 +Epoch [186/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7190 +Epoch [186/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7500 +Epoch [186/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8667, Pure Ratio2 9.8039 +Epoch [186/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8954 +Epoch [186/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 60.2865 % Model2 61.0877 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0784 +Epoch [187/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2745 +Epoch [187/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.8235 +Epoch [187/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8971 +Epoch [187/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8667 +Epoch [187/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8333 +Epoch [187/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 61.6186 % Model2 59.9058 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8627, Pure Ratio2 10.5098 +Epoch [188/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.1569 +Epoch [188/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 9.8301 +Epoch [188/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.7206 +Epoch [188/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8118, Pure Ratio2 9.7373 +Epoch [188/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8301 +Epoch [188/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9272, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 61.2280 % Model2 61.5184 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7843, Pure Ratio2 10.8431 +Epoch [189/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2941 +Epoch [189/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0392 +Epoch [189/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.8922 +Epoch [189/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.8471 +Epoch [189/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.9314 +Epoch [189/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 61.4383 % Model2 61.3682 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9020 +Epoch [190/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7157 +Epoch [190/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9412 +Epoch [190/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 9.9853 +Epoch [190/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 9.8980 +Epoch [190/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0425, Pure Ratio2 9.9314 +Epoch [190/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 61.1679 % Model2 61.0176 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6275, Pure Ratio2 10.5686 +Epoch [191/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [191/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.1046 +Epoch [191/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.0490 +Epoch [191/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.8706 +Epoch [191/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9379 +Epoch [191/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 61.5084 % Model2 60.9375 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.8824 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8333 +Epoch [192/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7712 +Epoch [192/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7500, Pure Ratio2 9.7402 +Epoch [192/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6824, Pure Ratio2 9.6392 +Epoch [192/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7647 +Epoch [192/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 60.6170 % Model2 60.5669 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [193/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9608 +Epoch [193/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0458 +Epoch [193/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8284 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.9412 +Epoch [193/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9379 +Epoch [193/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 60.7472 % Model2 60.8474 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.0000 +Epoch [194/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5588, Pure Ratio2 10.1961 +Epoch [194/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3660, Pure Ratio2 10.0850 +Epoch [194/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3039, Pure Ratio2 10.0588 +Epoch [194/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2588, Pure Ratio2 10.0667 +Epoch [194/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2582, Pure Ratio2 10.0980 +Epoch [194/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1345, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 60.5769 % Model2 60.5168 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.3725 +Epoch [195/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.4314 +Epoch [195/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7059 +Epoch [195/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7990 +Epoch [195/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [195/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.7516 +Epoch [195/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 60.3766 % Model2 60.6270 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.0392 +Epoch [196/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1176 +Epoch [196/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 9.9673 +Epoch [196/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9755 +Epoch [196/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0706, Pure Ratio2 9.9882 +Epoch [196/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1797, Pure Ratio2 10.0719 +Epoch [196/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 60.3766 % Model2 60.7973 %, Pure Ratio 1 10.0553 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [197/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 9.9510 +Epoch [197/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 9.9346 +Epoch [197/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.8824 +Epoch [197/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 9.9451 +Epoch [197/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.8595 +Epoch [197/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 60.7572 % Model2 61.1278 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1176 +Epoch [198/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1471 +Epoch [198/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.1961 +Epoch [198/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0147 +Epoch [198/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.9294 +Epoch [198/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9869 +Epoch [198/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 60.7272 % Model2 60.5168 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8039, Pure Ratio2 10.8627 +Epoch [199/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2451 +Epoch [199/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [199/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8775 +Epoch [199/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.8667 +Epoch [199/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.8856 +Epoch [199/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 60.3666 % Model2 60.4868 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7255, Pure Ratio2 10.5490 +Epoch [200/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4216, Pure Ratio2 10.3235 +Epoch [200/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.2288 +Epoch [200/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.3627 +Epoch [200/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1843, Pure Ratio2 10.2549 +Epoch [200/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1144, Pure Ratio2 10.1569 +Epoch [200/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 60.7372 % Model2 60.9075 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9698 % diff --git a/other_methods/coteaching/coteaching_results/out_4_6.log b/other_methods/coteaching/coteaching_results/out_4_6.log new file mode 100644 index 0000000..d3084eb --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_4_6.log @@ -0,0 +1,2041 @@ +Files already downloaded and verified +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 22.6562, Loss1: 0.0167, Loss2: 0.0167, Pure Ratio1: 9.7440, Pure Ratio2 9.7760 +Epoch [2/200], Iter [100/390] Training Accuracy1: 22.6562, Training Accuracy2: 26.5625, Loss1: 0.0163, Loss2: 0.0164, Pure Ratio1: 9.7200, Pure Ratio2 9.7520 +Epoch [2/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.8750, Loss1: 0.0171, Loss2: 0.0171, Pure Ratio1: 9.9360, Pure Ratio2 9.9413 +Epoch [2/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 21.8750, Loss1: 0.0173, Loss2: 0.0175, Pure Ratio1: 9.8360, Pure Ratio2 9.8520 +Epoch [2/200], Iter [250/390] Training Accuracy1: 20.3125, Training Accuracy2: 17.9688, Loss1: 0.0167, Loss2: 0.0168, Pure Ratio1: 9.9104, Pure Ratio2 9.9264 +Epoch [2/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0157, Loss2: 0.0158, Pure Ratio1: 9.8720, Pure Ratio2 9.8800 +Epoch [2/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 28.1250, Loss1: 0.0147, Loss2: 0.0151, Pure Ratio1: 9.9291, Pure Ratio2 9.9451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 19.4411 % Model2 20.0621 %, Pure Ratio 1 9.9467 %, Pure Ratio 2 9.9631 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0149, Loss2: 0.0146, Pure Ratio1: 9.5574, Pure Ratio2 9.6557 +Epoch [3/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0148, Loss2: 0.0148, Pure Ratio1: 9.7541, Pure Ratio2 9.7951 +Epoch [3/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0149, Loss2: 0.0146, Pure Ratio1: 10.0328, Pure Ratio2 10.0765 +Epoch [3/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0142, Loss2: 0.0142, Pure Ratio1: 9.8730, Pure Ratio2 9.9057 +Epoch [3/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0146, Loss2: 0.0148, Pure Ratio1: 9.9016, Pure Ratio2 9.9016 +Epoch [3/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.0625, Loss1: 0.0147, Loss2: 0.0142, Pure Ratio1: 9.9071, Pure Ratio2 9.9126 +Epoch [3/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0154, Loss2: 0.0155, Pure Ratio1: 9.9321, Pure Ratio2 9.9391 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 17.7284 % Model2 18.2392 %, Pure Ratio 1 9.9517 %, Pure Ratio 2 9.9664 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0153, Loss2: 0.0153, Pure Ratio1: 9.6471, Pure Ratio2 9.6639 +Epoch [4/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.1250, Loss1: 0.0140, Loss2: 0.0145, Pure Ratio1: 9.8319, Pure Ratio2 9.7983 +Epoch [4/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.0312, Loss1: 0.0147, Loss2: 0.0142, Pure Ratio1: 9.9832, Pure Ratio2 9.9776 +Epoch [4/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.0000, Loss1: 0.0150, Loss2: 0.0151, Pure Ratio1: 9.8487, Pure Ratio2 9.8697 +Epoch [4/200], Iter [250/390] Training Accuracy1: 20.3125, Training Accuracy2: 22.6562, Loss1: 0.0158, Loss2: 0.0155, Pure Ratio1: 9.8857, Pure Ratio2 9.9025 +Epoch [4/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0133, Loss2: 0.0135, Pure Ratio1: 9.8627, Pure Ratio2 9.8543 +Epoch [4/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0134, Loss2: 0.0134, Pure Ratio1: 9.9184, Pure Ratio2 9.9040 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 20.7532 % Model2 21.2941 %, Pure Ratio 1 9.9440 %, Pure Ratio 2 9.9267 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 26.5625, Loss1: 0.0139, Loss2: 0.0140, Pure Ratio1: 10.0862, Pure Ratio2 10.1552 +Epoch [5/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 35.1562, Loss1: 0.0146, Loss2: 0.0141, Pure Ratio1: 9.8879, Pure Ratio2 9.9483 +Epoch [5/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0147, Loss2: 0.0149, Pure Ratio1: 9.9770, Pure Ratio2 10.0517 +Epoch [5/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0133, Loss2: 0.0133, Pure Ratio1: 9.9397, Pure Ratio2 9.9741 +Epoch [5/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.7812, Loss1: 0.0144, Loss2: 0.0144, Pure Ratio1: 9.9586, Pure Ratio2 9.9690 +Epoch [5/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0142, Loss2: 0.0141, Pure Ratio1: 9.9023, Pure Ratio2 9.9195 +Epoch [5/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0150, Loss2: 0.0150, Pure Ratio1: 9.9310, Pure Ratio2 9.9483 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 28.2953 % Model2 22.4659 %, Pure Ratio 1 9.9315 %, Pure Ratio 2 9.9602 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 28.9062, Loss1: 0.0145, Loss2: 0.0146, Pure Ratio1: 10.3009, Pure Ratio2 10.3186 +Epoch [6/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0133, Loss2: 0.0132, Pure Ratio1: 10.5575, Pure Ratio2 10.5221 +Epoch [6/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0124, Loss2: 0.0120, Pure Ratio1: 10.2183, Pure Ratio2 10.1829 +Epoch [6/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0120, Loss2: 0.0120, Pure Ratio1: 10.1150, Pure Ratio2 10.0841 +Epoch [6/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0130, Loss2: 0.0131, Pure Ratio1: 10.0035, Pure Ratio2 9.9823 +Epoch [6/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.9062, Loss1: 0.0134, Loss2: 0.0142, Pure Ratio1: 9.9823, Pure Ratio2 9.9705 +Epoch [6/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0139, Pure Ratio1: 9.9570, Pure Ratio2 9.9444 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 30.5489 % Model2 31.1398 %, Pure Ratio 1 9.9592 %, Pure Ratio 2 9.9206 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0144, Pure Ratio1: 9.0727, Pure Ratio2 9.0727 +Epoch [7/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0143, Loss2: 0.0143, Pure Ratio1: 9.7727, Pure Ratio2 9.7818 +Epoch [7/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0126, Loss2: 0.0128, Pure Ratio1: 10.2364, Pure Ratio2 10.2000 +Epoch [7/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0136, Loss2: 0.0139, Pure Ratio1: 9.9955, Pure Ratio2 9.9636 +Epoch [7/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0127, Loss2: 0.0129, Pure Ratio1: 9.9491, Pure Ratio2 9.9345 +Epoch [7/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0116, Loss2: 0.0116, Pure Ratio1: 9.9394, Pure Ratio2 9.9182 +Epoch [7/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0130, Loss2: 0.0133, Pure Ratio1: 9.9299, Pure Ratio2 9.9117 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 29.2969 % Model2 29.4171 %, Pure Ratio 1 9.9021 %, Pure Ratio 2 9.9068 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.9375, Loss1: 0.0128, Loss2: 0.0122, Pure Ratio1: 9.8704, Pure Ratio2 10.0185 +Epoch [8/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 34.3750, Loss1: 0.0145, Loss2: 0.0146, Pure Ratio1: 9.5833, Pure Ratio2 9.6759 +Epoch [8/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 34.3750, Loss1: 0.0115, Loss2: 0.0124, Pure Ratio1: 9.8704, Pure Ratio2 9.9198 +Epoch [8/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0122, Loss2: 0.0123, Pure Ratio1: 9.8380, Pure Ratio2 9.8843 +Epoch [8/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0115, Loss2: 0.0117, Pure Ratio1: 9.9963, Pure Ratio2 10.0519 +Epoch [8/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0127, Loss2: 0.0129, Pure Ratio1: 9.8302, Pure Ratio2 9.8704 +Epoch [8/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0126, Loss2: 0.0126, Pure Ratio1: 9.8254, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 29.7376 % Model2 29.8978 %, Pure Ratio 1 9.9074 %, Pure Ratio 2 9.9288 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0105, Loss2: 0.0109, Pure Ratio1: 9.3143, Pure Ratio2 9.2952 +Epoch [9/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0109, Loss2: 0.0117, Pure Ratio1: 9.6190, Pure Ratio2 9.5714 +Epoch [9/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0122, Loss2: 0.0120, Pure Ratio1: 9.9111, Pure Ratio2 9.8667 +Epoch [9/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0137, Loss2: 0.0135, Pure Ratio1: 10.0667, Pure Ratio2 10.0190 +Epoch [9/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0123, Loss2: 0.0123, Pure Ratio1: 10.0190, Pure Ratio2 10.0114 +Epoch [9/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 40.6250, Loss1: 0.0146, Loss2: 0.0137, Pure Ratio1: 9.9143, Pure Ratio2 9.9111 +Epoch [9/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0122, Loss2: 0.0120, Pure Ratio1: 9.9156, Pure Ratio2 9.8993 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 25.6711 % Model2 26.2420 %, Pure Ratio 1 9.9536 %, Pure Ratio 2 9.9463 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0122, Loss2: 0.0114, Pure Ratio1: 10.2745, Pure Ratio2 10.3333 +Epoch [10/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0119, Loss2: 0.0118, Pure Ratio1: 9.9608, Pure Ratio2 9.9902 +Epoch [10/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0134, Loss2: 0.0135, Pure Ratio1: 9.9216, Pure Ratio2 9.9281 +Epoch [10/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0121, Loss2: 0.0125, Pure Ratio1: 9.8922, Pure Ratio2 9.8676 +Epoch [10/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0102, Loss2: 0.0101, Pure Ratio1: 9.9255, Pure Ratio2 9.9216 +Epoch [10/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 33.5938, Loss1: 0.0134, Loss2: 0.0133, Pure Ratio1: 9.8268, Pure Ratio2 9.8268 +Epoch [10/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0121, Loss2: 0.0115, Pure Ratio1: 9.8459, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 30.1482 % Model2 29.6474 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0103, Loss2: 0.0104, Pure Ratio1: 9.9412, Pure Ratio2 10.0980 +Epoch [11/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0112, Loss2: 0.0115, Pure Ratio1: 10.2157, Pure Ratio2 10.3529 +Epoch [11/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 42.1875, Loss1: 0.0121, Loss2: 0.0115, Pure Ratio1: 10.1111, Pure Ratio2 10.2092 +Epoch [11/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0124, Loss2: 0.0125, Pure Ratio1: 9.9363, Pure Ratio2 10.0000 +Epoch [11/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0111, Loss2: 0.0105, Pure Ratio1: 10.0627, Pure Ratio2 10.1137 +Epoch [11/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0111, Loss2: 0.0109, Pure Ratio1: 9.9379, Pure Ratio2 9.9837 +Epoch [11/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0106, Loss2: 0.0112, Pure Ratio1: 9.8487, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 26.6927 % Model2 29.7576 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0123, Loss2: 0.0120, Pure Ratio1: 9.7059, Pure Ratio2 9.8039 +Epoch [12/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0116, Loss2: 0.0119, Pure Ratio1: 9.6471, Pure Ratio2 9.7353 +Epoch [12/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0098, Loss2: 0.0090, Pure Ratio1: 9.7712, Pure Ratio2 9.7908 +Epoch [12/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0099, Loss2: 0.0099, Pure Ratio1: 9.8775, Pure Ratio2 9.9363 +Epoch [12/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0116, Loss2: 0.0111, Pure Ratio1: 9.8471, Pure Ratio2 9.8980 +Epoch [12/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0124, Loss2: 0.0121, Pure Ratio1: 9.9183, Pure Ratio2 9.9575 +Epoch [12/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0113, Loss2: 0.0111, Pure Ratio1: 9.9524, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 29.7175 % Model2 30.1783 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0106, Loss2: 0.0106, Pure Ratio1: 10.1176, Pure Ratio2 10.1176 +Epoch [13/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 35.9375, Loss1: 0.0122, Loss2: 0.0118, Pure Ratio1: 9.9804, Pure Ratio2 10.0098 +Epoch [13/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0107, Loss2: 0.0106, Pure Ratio1: 9.8431, Pure Ratio2 9.9150 +Epoch [13/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0121, Loss2: 0.0115, Pure Ratio1: 9.8186, Pure Ratio2 9.9363 +Epoch [13/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0113, Loss2: 0.0119, Pure Ratio1: 9.8706, Pure Ratio2 9.9882 +Epoch [13/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0108, Loss2: 0.0109, Pure Ratio1: 9.9150, Pure Ratio2 10.0163 +Epoch [13/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0100, Loss2: 0.0093, Pure Ratio1: 9.9104, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 28.3353 % Model2 28.4054 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0116, Loss2: 0.0114, Pure Ratio1: 9.4314, Pure Ratio2 9.1765 +Epoch [14/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 46.0938, Loss1: 0.0112, Loss2: 0.0107, Pure Ratio1: 9.4804, Pure Ratio2 9.4216 +Epoch [14/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 39.0625, Loss1: 0.0118, Loss2: 0.0122, Pure Ratio1: 9.5948, Pure Ratio2 9.5425 +Epoch [14/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0115, Loss2: 0.0118, Pure Ratio1: 9.7794, Pure Ratio2 9.7598 +Epoch [14/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 40.6250, Loss1: 0.0113, Loss2: 0.0112, Pure Ratio1: 9.8902, Pure Ratio2 9.8784 +Epoch [14/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 37.5000, Loss1: 0.0118, Loss2: 0.0116, Pure Ratio1: 9.8235, Pure Ratio2 9.8301 +Epoch [14/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0097, Loss2: 0.0102, Pure Ratio1: 9.9160, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 29.7776 % Model2 30.4187 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0104, Loss2: 0.0103, Pure Ratio1: 9.9804, Pure Ratio2 9.9804 +Epoch [15/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0097, Loss2: 0.0096, Pure Ratio1: 10.0490, Pure Ratio2 10.0490 +Epoch [15/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0102, Loss2: 0.0108, Pure Ratio1: 10.1176, Pure Ratio2 10.1634 +Epoch [15/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0113, Loss2: 0.0120, Pure Ratio1: 10.2255, Pure Ratio2 10.2402 +Epoch [15/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 30.4688, Loss1: 0.0133, Loss2: 0.0132, Pure Ratio1: 10.0039, Pure Ratio2 10.0157 +Epoch [15/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0108, Loss2: 0.0101, Pure Ratio1: 9.9641, Pure Ratio2 9.9837 +Epoch [15/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0113, Loss2: 0.0106, Pure Ratio1: 9.8992, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 30.9896 % Model2 30.7492 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0107, Loss2: 0.0107, Pure Ratio1: 10.6275, Pure Ratio2 10.3529 +Epoch [16/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0113, Loss2: 0.0110, Pure Ratio1: 10.4020, Pure Ratio2 10.2549 +Epoch [16/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0116, Loss2: 0.0110, Pure Ratio1: 10.4706, Pure Ratio2 10.3464 +Epoch [16/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0101, Loss2: 0.0100, Pure Ratio1: 10.3186, Pure Ratio2 10.2157 +Epoch [16/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0110, Loss2: 0.0105, Pure Ratio1: 10.1569, Pure Ratio2 10.0745 +Epoch [16/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0125, Loss2: 0.0123, Pure Ratio1: 10.0556, Pure Ratio2 9.9771 +Epoch [16/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0129, Loss2: 0.0119, Pure Ratio1: 10.0112, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 25.9816 % Model2 27.0032 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0099, Loss2: 0.0107, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [17/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0096, Loss2: 0.0093, Pure Ratio1: 9.6667, Pure Ratio2 9.7353 +Epoch [17/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 45.3125, Loss1: 0.0100, Loss2: 0.0108, Pure Ratio1: 9.7843, Pure Ratio2 9.7582 +Epoch [17/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0103, Loss2: 0.0097, Pure Ratio1: 9.6716, Pure Ratio2 9.6373 +Epoch [17/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0084, Loss2: 0.0079, Pure Ratio1: 9.9059, Pure Ratio2 9.8784 +Epoch [17/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0092, Loss2: 0.0100, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [17/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 49.2188, Loss1: 0.0100, Loss2: 0.0098, Pure Ratio1: 9.8683, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 30.0280 % Model2 27.6943 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0091, Loss2: 0.0097, Pure Ratio1: 9.5098, Pure Ratio2 9.4510 +Epoch [18/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0092, Loss2: 0.0094, Pure Ratio1: 10.0000, Pure Ratio2 9.9216 +Epoch [18/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0101, Loss2: 0.0099, Pure Ratio1: 9.9281, Pure Ratio2 9.8758 +Epoch [18/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0095, Loss2: 0.0097, Pure Ratio1: 9.9853, Pure Ratio2 9.9216 +Epoch [18/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0105, Loss2: 0.0101, Pure Ratio1: 9.9843, Pure Ratio2 9.9255 +Epoch [18/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0088, Loss2: 0.0093, Pure Ratio1: 10.0033, Pure Ratio2 9.9542 +Epoch [18/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0095, Loss2: 0.0093, Pure Ratio1: 9.9916, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 25.7011 % Model2 27.7143 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0084, Loss2: 0.0082, Pure Ratio1: 8.9804, Pure Ratio2 8.8824 +Epoch [19/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0083, Loss2: 0.0087, Pure Ratio1: 9.6176, Pure Ratio2 9.6373 +Epoch [19/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0077, Loss2: 0.0075, Pure Ratio1: 9.6928, Pure Ratio2 9.6536 +Epoch [19/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0090, Loss2: 0.0096, Pure Ratio1: 9.7500, Pure Ratio2 9.7157 +Epoch [19/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0081, Loss2: 0.0086, Pure Ratio1: 9.9686, Pure Ratio2 9.8902 +Epoch [19/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0081, Loss2: 0.0083, Pure Ratio1: 9.9281, Pure Ratio2 9.8627 +Epoch [19/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0077, Loss2: 0.0086, Pure Ratio1: 10.0028, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 29.8478 % Model2 29.7776 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0071, Loss2: 0.0072, Pure Ratio1: 10.4118, Pure Ratio2 10.2745 +Epoch [20/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0067, Loss2: 0.0068, Pure Ratio1: 10.0294, Pure Ratio2 9.9216 +Epoch [20/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0087, Loss2: 0.0084, Pure Ratio1: 10.1307, Pure Ratio2 10.0327 +Epoch [20/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0093, Loss2: 0.0103, Pure Ratio1: 10.1324, Pure Ratio2 10.0245 +Epoch [20/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0078, Loss2: 0.0086, Pure Ratio1: 9.9373, Pure Ratio2 9.8157 +Epoch [20/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0090, Loss2: 0.0092, Pure Ratio1: 9.9837, Pure Ratio2 9.8889 +Epoch [20/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0099, Loss2: 0.0093, Pure Ratio1: 9.9664, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 28.9263 % Model2 28.6258 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0087, Loss2: 0.0087, Pure Ratio1: 10.1373, Pure Ratio2 10.0392 +Epoch [21/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0071, Loss2: 0.0078, Pure Ratio1: 10.1176, Pure Ratio2 10.0098 +Epoch [21/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0077, Loss2: 0.0074, Pure Ratio1: 10.1569, Pure Ratio2 10.0523 +Epoch [21/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0074, Loss2: 0.0077, Pure Ratio1: 10.1225, Pure Ratio2 10.0539 +Epoch [21/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0095, Loss2: 0.0086, Pure Ratio1: 10.0980, Pure Ratio2 10.0353 +Epoch [21/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0087, Loss2: 0.0090, Pure Ratio1: 10.0065, Pure Ratio2 9.9804 +Epoch [21/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0101, Loss2: 0.0089, Pure Ratio1: 10.0168, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 27.7244 % Model2 28.8562 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 57.8125, Loss1: 0.0057, Loss2: 0.0066, Pure Ratio1: 9.7451, Pure Ratio2 9.6863 +Epoch [22/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0057, Loss2: 0.0059, Pure Ratio1: 9.8235, Pure Ratio2 9.7353 +Epoch [22/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0086, Loss2: 0.0086, Pure Ratio1: 9.6732, Pure Ratio2 9.5621 +Epoch [22/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0088, Loss2: 0.0087, Pure Ratio1: 9.4020, Pure Ratio2 9.3725 +Epoch [22/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0079, Loss2: 0.0080, Pure Ratio1: 9.4941, Pure Ratio2 9.4235 +Epoch [22/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0093, Loss2: 0.0080, Pure Ratio1: 9.6993, Pure Ratio2 9.6438 +Epoch [22/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0068, Loss2: 0.0061, Pure Ratio1: 9.8263, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 28.7861 % Model2 29.0966 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0060, Loss2: 0.0070, Pure Ratio1: 9.1373, Pure Ratio2 9.0980 +Epoch [23/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0065, Loss2: 0.0079, Pure Ratio1: 9.4412, Pure Ratio2 9.3235 +Epoch [23/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0056, Loss2: 0.0054, Pure Ratio1: 9.6209, Pure Ratio2 9.5490 +Epoch [23/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 66.4062, Loss1: 0.0073, Loss2: 0.0071, Pure Ratio1: 9.7745, Pure Ratio2 9.7206 +Epoch [23/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0052, Loss2: 0.0058, Pure Ratio1: 9.8235, Pure Ratio2 9.7765 +Epoch [23/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0063, Loss2: 0.0057, Pure Ratio1: 9.8399, Pure Ratio2 9.7647 +Epoch [23/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0080, Loss2: 0.0079, Pure Ratio1: 9.9664, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 29.0865 % Model2 29.0064 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0058, Loss2: 0.0055, Pure Ratio1: 10.0196, Pure Ratio2 9.8235 +Epoch [24/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0080, Loss2: 0.0071, Pure Ratio1: 9.9118, Pure Ratio2 9.7255 +Epoch [24/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0064, Loss2: 0.0058, Pure Ratio1: 9.9150, Pure Ratio2 9.8039 +Epoch [24/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0069, Loss2: 0.0085, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [24/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0043, Loss2: 0.0057, Pure Ratio1: 10.0235, Pure Ratio2 9.9412 +Epoch [24/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0068, Loss2: 0.0065, Pure Ratio1: 9.9510, Pure Ratio2 9.8791 +Epoch [24/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0060, Loss2: 0.0068, Pure Ratio1: 9.9216, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 26.6326 % Model2 27.4639 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0050, Loss2: 0.0046, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [25/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.0042, Loss2: 0.0037, Pure Ratio1: 9.9608, Pure Ratio2 10.0098 +Epoch [25/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0055, Loss2: 0.0057, Pure Ratio1: 10.1765, Pure Ratio2 10.1765 +Epoch [25/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0049, Loss2: 0.0057, Pure Ratio1: 9.9216, Pure Ratio2 9.9755 +Epoch [25/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0057, Loss2: 0.0066, Pure Ratio1: 9.8863, Pure Ratio2 9.9412 +Epoch [25/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0069, Loss2: 0.0066, Pure Ratio1: 9.8105, Pure Ratio2 9.8268 +Epoch [25/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0056, Loss2: 0.0051, Pure Ratio1: 9.9132, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 27.4840 % Model2 29.3770 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0032, Loss2: 0.0038, Pure Ratio1: 9.9608, Pure Ratio2 10.1569 +Epoch [26/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.0023, Loss2: 0.0031, Pure Ratio1: 10.0588, Pure Ratio2 10.1863 +Epoch [26/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0042, Loss2: 0.0034, Pure Ratio1: 10.0065, Pure Ratio2 10.0850 +Epoch [26/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0045, Loss2: 0.0043, Pure Ratio1: 10.0000, Pure Ratio2 10.0833 +Epoch [26/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0050, Loss2: 0.0055, Pure Ratio1: 9.9294, Pure Ratio2 10.0078 +Epoch [26/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0052, Loss2: 0.0047, Pure Ratio1: 9.8856, Pure Ratio2 9.9379 +Epoch [26/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0053, Loss2: 0.0058, Pure Ratio1: 9.8375, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 28.9463 % Model2 28.8261 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0036, Loss2: 0.0028, Pure Ratio1: 10.1176, Pure Ratio2 10.0588 +Epoch [27/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0068, Loss2: 0.0065, Pure Ratio1: 10.0490, Pure Ratio2 10.1275 +Epoch [27/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0040, Loss2: 0.0038, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [27/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0041, Loss2: 0.0041, Pure Ratio1: 9.9706, Pure Ratio2 10.0343 +Epoch [27/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0042, Loss2: 0.0054, Pure Ratio1: 9.9451, Pure Ratio2 9.9529 +Epoch [27/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0054, Loss2: 0.0053, Pure Ratio1: 9.9020, Pure Ratio2 9.9248 +Epoch [27/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0047, Loss2: 0.0053, Pure Ratio1: 9.8880, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 26.7929 % Model2 27.1534 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0026, Loss2: 0.0028, Pure Ratio1: 9.5294, Pure Ratio2 9.5490 +Epoch [28/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0026, Loss2: 0.0024, Pure Ratio1: 9.5490, Pure Ratio2 9.5490 +Epoch [28/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0031, Loss2: 0.0032, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [28/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 73.4375, Loss1: 0.0023, Loss2: 0.0039, Pure Ratio1: 9.8137, Pure Ratio2 9.8676 +Epoch [28/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0044, Loss2: 0.0044, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [28/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 9.6863, Pure Ratio2 9.7908 +Epoch [28/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0043, Loss2: 0.0033, Pure Ratio1: 9.9020, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 27.5841 % Model2 25.6010 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0033, Loss2: 0.0028, Pure Ratio1: 9.1765, Pure Ratio2 9.1569 +Epoch [29/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.7745, Pure Ratio2 9.6765 +Epoch [29/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0028, Loss2: 0.0030, Pure Ratio1: 9.7059, Pure Ratio2 9.6797 +Epoch [29/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 9.7794, Pure Ratio2 9.7647 +Epoch [29/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0034, Loss2: 0.0031, Pure Ratio1: 9.7529, Pure Ratio2 9.7765 +Epoch [29/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 67.1875, Loss1: 0.0023, Loss2: 0.0041, Pure Ratio1: 9.8137, Pure Ratio2 9.8105 +Epoch [29/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0032, Loss2: 0.0039, Pure Ratio1: 9.8936, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 27.7845 % Model2 28.3554 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0022, Pure Ratio1: 10.5490, Pure Ratio2 10.4118 +Epoch [30/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.9062, Loss1: 0.0025, Loss2: 0.0020, Pure Ratio1: 10.4412, Pure Ratio2 10.3333 +Epoch [30/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0024, Pure Ratio1: 10.3399, Pure Ratio2 10.2484 +Epoch [30/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0029, Loss2: 0.0033, Pure Ratio1: 10.1373, Pure Ratio2 10.0735 +Epoch [30/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0023, Loss2: 0.0033, Pure Ratio1: 10.0314, Pure Ratio2 9.9608 +Epoch [30/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0040, Pure Ratio1: 10.0359, Pure Ratio2 9.9575 +Epoch [30/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.0024, Loss2: 0.0022, Pure Ratio1: 9.9832, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 29.5974 % Model2 27.9147 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.0025, Loss2: 0.0019, Pure Ratio1: 10.0588, Pure Ratio2 10.2353 +Epoch [31/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0028, Loss2: 0.0019, Pure Ratio1: 9.8824, Pure Ratio2 9.9314 +Epoch [31/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 10.1634, Pure Ratio2 10.1765 +Epoch [31/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0024, Loss2: 0.0023, Pure Ratio1: 10.3480, Pure Ratio2 10.3382 +Epoch [31/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0020, Pure Ratio1: 10.0980, Pure Ratio2 10.1020 +Epoch [31/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0026, Loss2: 0.0023, Pure Ratio1: 9.9869, Pure Ratio2 9.9771 +Epoch [31/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0032, Loss2: 0.0036, Pure Ratio1: 10.0140, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 27.5441 % Model2 28.3253 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0022, Pure Ratio1: 11.1373, Pure Ratio2 11.0784 +Epoch [32/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0022, Loss2: 0.0027, Pure Ratio1: 10.1765, Pure Ratio2 10.1667 +Epoch [32/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0033, Loss2: 0.0023, Pure Ratio1: 10.2222, Pure Ratio2 10.2484 +Epoch [32/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.0016, Loss2: 0.0024, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [32/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0018, Pure Ratio1: 10.0000, Pure Ratio2 9.9882 +Epoch [32/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0033, Loss2: 0.0032, Pure Ratio1: 9.9150, Pure Ratio2 9.9248 +Epoch [32/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 77.3438, Loss1: 0.0019, Loss2: 0.0024, Pure Ratio1: 9.9748, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 26.2320 % Model2 26.6426 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0015, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [33/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 82.8125, Loss1: 0.0024, Loss2: 0.0011, Pure Ratio1: 9.7647, Pure Ratio2 9.7549 +Epoch [33/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0023, Loss2: 0.0022, Pure Ratio1: 9.9150, Pure Ratio2 9.8954 +Epoch [33/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0025, Loss2: 0.0033, Pure Ratio1: 10.0441, Pure Ratio2 10.0441 +Epoch [33/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0014, Pure Ratio1: 10.0235, Pure Ratio2 9.9765 +Epoch [33/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 78.9062, Loss1: 0.0014, Loss2: 0.0019, Pure Ratio1: 9.9608, Pure Ratio2 9.9085 +Epoch [33/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.9062, Loss1: 0.0021, Loss2: 0.0014, Pure Ratio1: 9.8908, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 27.4740 % Model2 26.9932 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.5098, Pure Ratio2 9.5294 +Epoch [34/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0030, Loss2: 0.0038, Pure Ratio1: 9.7157, Pure Ratio2 9.6176 +Epoch [34/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0014, Loss2: 0.0017, Pure Ratio1: 9.7124, Pure Ratio2 9.6536 +Epoch [34/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 80.4688, Loss1: 0.0028, Loss2: 0.0013, Pure Ratio1: 9.8431, Pure Ratio2 9.8333 +Epoch [34/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 76.5625, Loss1: 0.0028, Loss2: 0.0021, Pure Ratio1: 9.9137, Pure Ratio2 9.9255 +Epoch [34/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.0027, Loss2: 0.0018, Pure Ratio1: 9.9444, Pure Ratio2 10.0000 +Epoch [34/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.7812, Loss1: 0.0024, Loss2: 0.0031, Pure Ratio1: 9.9328, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 26.8129 % Model2 28.1550 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0023, Pure Ratio1: 9.6275, Pure Ratio2 9.8824 +Epoch [35/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0021, Pure Ratio1: 10.0686, Pure Ratio2 10.3627 +Epoch [35/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0017, Loss2: 0.0011, Pure Ratio1: 9.9085, Pure Ratio2 10.0915 +Epoch [35/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.8627, Pure Ratio2 10.0000 +Epoch [35/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0017, Pure Ratio1: 9.9569, Pure Ratio2 10.0902 +Epoch [35/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.0028, Loss2: 0.0015, Pure Ratio1: 10.0196, Pure Ratio2 10.1471 +Epoch [35/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0023, Loss2: 0.0021, Pure Ratio1: 9.9748, Pure Ratio2 10.0980 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 26.2921 % Model2 27.6042 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [36/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0014, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [36/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0019, Loss2: 0.0012, Pure Ratio1: 9.9739, Pure Ratio2 9.9673 +Epoch [36/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0014, Pure Ratio1: 10.0245, Pure Ratio2 9.9608 +Epoch [36/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0019, Pure Ratio1: 10.0275, Pure Ratio2 9.9412 +Epoch [36/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0020, Loss2: 0.0013, Pure Ratio1: 10.0163, Pure Ratio2 9.9510 +Epoch [36/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0021, Loss2: 0.0012, Pure Ratio1: 10.0336, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 26.2019 % Model2 25.9415 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.7255, Pure Ratio2 9.7059 +Epoch [37/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0009, Pure Ratio1: 9.6569, Pure Ratio2 9.7647 +Epoch [37/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.8105, Pure Ratio2 9.8954 +Epoch [37/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.9265, Pure Ratio2 9.9510 +Epoch [37/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0020, Loss2: 0.0013, Pure Ratio1: 10.0314, Pure Ratio2 10.0588 +Epoch [37/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0022, Pure Ratio1: 9.9346, Pure Ratio2 9.9739 +Epoch [37/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0024, Pure Ratio1: 9.9132, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 26.5224 % Model2 27.1134 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.1569, Pure Ratio2 9.2353 +Epoch [38/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.6667, Pure Ratio2 9.6863 +Epoch [38/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.7320, Pure Ratio2 9.7582 +Epoch [38/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0015, Pure Ratio1: 9.7843, Pure Ratio2 9.7794 +Epoch [38/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0019, Loss2: 0.0010, Pure Ratio1: 9.9490, Pure Ratio2 9.9333 +Epoch [38/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.0024, Loss2: 0.0022, Pure Ratio1: 9.7974, Pure Ratio2 9.7778 +Epoch [38/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0018, Pure Ratio1: 9.8179, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 27.4038 % Model2 27.5641 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0021, Loss2: 0.0009, Pure Ratio1: 9.6667, Pure Ratio2 9.5686 +Epoch [39/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 10.1863, Pure Ratio2 10.0686 +Epoch [39/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 10.1176, Pure Ratio2 10.0458 +Epoch [39/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0020, Loss2: 0.0012, Pure Ratio1: 10.0294, Pure Ratio2 9.9951 +Epoch [39/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 10.0431, Pure Ratio2 10.0118 +Epoch [39/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0016, Loss2: 0.0021, Pure Ratio1: 9.9314, Pure Ratio2 9.8824 +Epoch [39/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0015, Pure Ratio1: 9.9692, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 26.9030 % Model2 27.0533 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [40/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0010, Pure Ratio1: 10.1078, Pure Ratio2 10.1667 +Epoch [40/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 10.1830, Pure Ratio2 10.2092 +Epoch [40/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0015, Pure Ratio1: 9.9706, Pure Ratio2 9.9118 +Epoch [40/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0008, Loss2: 0.0014, Pure Ratio1: 9.8980, Pure Ratio2 9.8275 +Epoch [40/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.9379, Pure Ratio2 9.8824 +Epoch [40/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0014, Pure Ratio1: 10.0252, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 28.0549 % Model2 26.7228 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 10.2353, Pure Ratio2 10.1569 +Epoch [41/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 10.2353, Pure Ratio2 10.0392 +Epoch [41/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 9.9869, Pure Ratio2 9.7908 +Epoch [41/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0020, Pure Ratio1: 9.9167, Pure Ratio2 9.7500 +Epoch [41/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 10.0353, Pure Ratio2 9.8667 +Epoch [41/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0015, Pure Ratio1: 9.9869, Pure Ratio2 9.8529 +Epoch [41/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 10.0700, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 27.2035 % Model2 28.9163 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.7451 +Epoch [42/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [42/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 9.7386, Pure Ratio2 9.7582 +Epoch [42/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 9.9314, Pure Ratio2 9.9559 +Epoch [42/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0020, Pure Ratio1: 9.9373, Pure Ratio2 9.9451 +Epoch [42/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.9542, Pure Ratio2 9.9542 +Epoch [42/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 9.9720, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 28.1150 % Model2 29.1867 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.8235 +Epoch [43/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.1471, Pure Ratio2 10.2647 +Epoch [43/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.9739, Pure Ratio2 10.1438 +Epoch [43/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.1275, Pure Ratio2 10.2451 +Epoch [43/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 91.4062, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 10.1569, Pure Ratio2 10.2510 +Epoch [43/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.9510, Pure Ratio2 10.0196 +Epoch [43/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0017, Loss2: 0.0005, Pure Ratio1: 9.8683, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 28.0950 % Model2 27.8446 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.2745, Pure Ratio2 9.2157 +Epoch [44/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.6765, Pure Ratio2 9.6176 +Epoch [44/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.6863, Pure Ratio2 9.5752 +Epoch [44/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 9.6520, Pure Ratio2 9.5637 +Epoch [44/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.7020, Pure Ratio2 9.6863 +Epoch [44/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.7124, Pure Ratio2 9.6928 +Epoch [44/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0021, Pure Ratio1: 9.8739, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 26.9131 % Model2 28.1450 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [45/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.0098, Pure Ratio2 10.0000 +Epoch [45/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.6797, Pure Ratio2 9.6471 +Epoch [45/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.6912, Pure Ratio2 9.5833 +Epoch [45/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.0011, Loss2: 0.0022, Pure Ratio1: 9.8157, Pure Ratio2 9.7843 +Epoch [45/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.8431, Pure Ratio2 9.8725 +Epoch [45/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8655, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 27.6943 % Model2 27.9547 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [46/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9118, Pure Ratio2 9.8922 +Epoch [46/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0021, Pure Ratio1: 9.8366, Pure Ratio2 9.8627 +Epoch [46/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 9.8873, Pure Ratio2 9.8529 +Epoch [46/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8431, Pure Ratio2 9.8392 +Epoch [46/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0013, Pure Ratio1: 9.8922, Pure Ratio2 9.8529 +Epoch [46/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.8403, Pure Ratio2 9.7871 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 26.8229 % Model2 26.2320 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.3922, Pure Ratio2 9.1569 +Epoch [47/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 9.7255, Pure Ratio2 9.5686 +Epoch [47/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0051, Loss2: 0.0051, Pure Ratio1: 9.9608, Pure Ratio2 9.8170 +Epoch [47/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.8529, Pure Ratio2 9.7402 +Epoch [47/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0014, Pure Ratio1: 9.8196, Pure Ratio2 9.7451 +Epoch [47/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.8987, Pure Ratio2 9.8366 +Epoch [47/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0018, Pure Ratio1: 9.9664, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 26.2019 % Model2 27.6943 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.6863, Pure Ratio2 9.3529 +Epoch [48/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.7647, Pure Ratio2 9.5882 +Epoch [48/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9020, Pure Ratio2 9.8170 +Epoch [48/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 9.8088 +Epoch [48/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8627, Pure Ratio2 9.8549 +Epoch [48/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 9.9739, Pure Ratio2 9.9542 +Epoch [48/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9104, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 27.0833 % Model2 27.3638 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.3529, Pure Ratio2 10.5294 +Epoch [49/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0015, Pure Ratio1: 10.5196, Pure Ratio2 10.5980 +Epoch [49/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 10.3268, Pure Ratio2 10.4052 +Epoch [49/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0017, Pure Ratio1: 10.1127, Pure Ratio2 10.2059 +Epoch [49/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.0941, Pure Ratio2 10.1176 +Epoch [49/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 10.0392, Pure Ratio2 10.0294 +Epoch [49/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0018, Pure Ratio1: 9.9832, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 27.4038 % Model2 27.9247 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 10.4902, Pure Ratio2 10.5686 +Epoch [50/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.8137, Pure Ratio2 9.9510 +Epoch [50/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.9346, Pure Ratio2 10.0392 +Epoch [50/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.7843, Pure Ratio2 9.8333 +Epoch [50/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.7059, Pure Ratio2 9.7333 +Epoch [50/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7222, Pure Ratio2 9.7484 +Epoch [50/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8319, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 27.5341 % Model2 26.7628 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0023, Pure Ratio1: 10.3137, Pure Ratio2 10.2157 +Epoch [51/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 10.1471, Pure Ratio2 10.0588 +Epoch [51/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.9346 +Epoch [51/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.9657, Pure Ratio2 9.9951 +Epoch [51/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.9529, Pure Ratio2 9.9725 +Epoch [51/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.8595, Pure Ratio2 9.8693 +Epoch [51/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.8908, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 27.2135 % Model2 26.2921 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [52/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0013, Pure Ratio1: 9.5882, Pure Ratio2 9.5980 +Epoch [52/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.5556, Pure Ratio2 9.5229 +Epoch [52/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.6176, Pure Ratio2 9.6029 +Epoch [52/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.6980, Pure Ratio2 9.6549 +Epoch [52/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.8693, Pure Ratio2 9.8105 +Epoch [52/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.8179, Pure Ratio2 9.7675 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 27.4639 % Model2 27.1635 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.1765, Pure Ratio2 10.1569 +Epoch [53/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9314, Pure Ratio2 9.9314 +Epoch [53/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 9.9869 +Epoch [53/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 9.9069 +Epoch [53/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.8314, Pure Ratio2 9.7098 +Epoch [53/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.8758, Pure Ratio2 9.7908 +Epoch [53/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8992, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 28.1751 % Model2 25.9215 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.2941, Pure Ratio2 8.9608 +Epoch [54/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.4020, Pure Ratio2 9.1373 +Epoch [54/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.7059, Pure Ratio2 9.4314 +Epoch [54/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8382, Pure Ratio2 9.6618 +Epoch [54/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.8588, Pure Ratio2 9.7412 +Epoch [54/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9118, Pure Ratio2 9.8137 +Epoch [54/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0025, Pure Ratio1: 9.9132, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 28.0950 % Model2 26.8029 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.3529, Pure Ratio2 10.2353 +Epoch [55/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.6275, Pure Ratio2 9.5392 +Epoch [55/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.6013, Pure Ratio2 9.5163 +Epoch [55/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0014, Pure Ratio1: 9.6716, Pure Ratio2 9.6961 +Epoch [55/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 9.7529, Pure Ratio2 9.7412 +Epoch [55/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.7124, Pure Ratio2 9.7157 +Epoch [55/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.8151, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 26.3822 % Model2 26.4623 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [56/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9608, Pure Ratio2 9.9216 +Epoch [56/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0014, Loss2: 0.0004, Pure Ratio1: 10.0458, Pure Ratio2 9.9608 +Epoch [56/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.0539, Pure Ratio2 10.0343 +Epoch [56/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.1176, Pure Ratio2 10.1137 +Epoch [56/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 10.0425 +Epoch [56/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 9.9468, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 27.3337 % Model2 27.9347 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 8.8235, Pure Ratio2 9.0392 +Epoch [57/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.6765, Pure Ratio2 9.9412 +Epoch [57/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 9.5817, Pure Ratio2 9.7582 +Epoch [57/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0013, Loss2: 0.0007, Pure Ratio1: 9.7010, Pure Ratio2 9.8725 +Epoch [57/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0019, Pure Ratio1: 9.8157, Pure Ratio2 10.0078 +Epoch [57/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8268, Pure Ratio2 10.0000 +Epoch [57/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.8403, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 28.2652 % Model2 26.8029 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.5490, Pure Ratio2 9.5098 +Epoch [58/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.9608, Pure Ratio2 10.0098 +Epoch [58/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8693, Pure Ratio2 9.9020 +Epoch [58/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.8775, Pure Ratio2 9.8873 +Epoch [58/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.8118, Pure Ratio2 9.8314 +Epoch [58/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 9.7941, Pure Ratio2 9.8529 +Epoch [58/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8964, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 26.7027 % Model2 27.9547 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0016, Pure Ratio1: 10.4902, Pure Ratio2 10.3333 +Epoch [59/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.5196, Pure Ratio2 10.3824 +Epoch [59/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0014, Loss2: 0.0008, Pure Ratio1: 10.1503, Pure Ratio2 9.9673 +Epoch [59/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.9706, Pure Ratio2 9.8873 +Epoch [59/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9255, Pure Ratio2 9.9137 +Epoch [59/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 9.9706, Pure Ratio2 9.9085 +Epoch [59/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.9244, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 26.7428 % Model2 26.5325 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Epoch [60/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [60/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9542, Pure Ratio2 9.9281 +Epoch [60/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.8824, Pure Ratio2 9.8676 +Epoch [60/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 82.8125, Loss1: 0.0019, Loss2: 0.0004, Pure Ratio1: 10.0471, Pure Ratio2 10.0510 +Epoch [60/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9967 +Epoch [60/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 9.9104, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 26.1719 % Model2 27.7945 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.4314, Pure Ratio2 10.1373 +Epoch [61/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 10.3235, Pure Ratio2 10.2745 +Epoch [61/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0523, Pure Ratio2 9.9608 +Epoch [61/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.9902, Pure Ratio2 10.0098 +Epoch [61/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8784, Pure Ratio2 9.8980 +Epoch [61/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.9444, Pure Ratio2 9.9706 +Epoch [61/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.9160, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 28.1350 % Model2 28.5857 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 10.0392 +Epoch [62/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9902, Pure Ratio2 10.1765 +Epoch [62/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9542, Pure Ratio2 10.0458 +Epoch [62/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8382, Pure Ratio2 9.9216 +Epoch [62/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 90.6250, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.9137, Pure Ratio2 10.0118 +Epoch [62/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 9.9052, Pure Ratio2 10.0294 +Epoch [62/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0016, Pure Ratio1: 9.8599, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 26.6627 % Model2 28.6959 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0000, Pure Ratio2 9.9216 +Epoch [63/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.0784, Pure Ratio2 10.0490 +Epoch [63/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.7320, Pure Ratio2 9.7320 +Epoch [63/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8186, Pure Ratio2 9.7990 +Epoch [63/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0013, Pure Ratio1: 10.0588, Pure Ratio2 10.0039 +Epoch [63/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.9314, Pure Ratio2 9.8889 +Epoch [63/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.9496, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 26.2620 % Model2 27.1034 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.3529, Pure Ratio2 9.1373 +Epoch [64/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.0686, Pure Ratio2 9.0000 +Epoch [64/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.5752, Pure Ratio2 9.4837 +Epoch [64/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8039, Pure Ratio2 9.7500 +Epoch [64/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9059, Pure Ratio2 9.8510 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.0163, Pure Ratio2 9.8954 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.9664, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 26.0917 % Model2 27.9447 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9412, Pure Ratio2 10.1176 +Epoch [65/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9706 +Epoch [65/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.7582, Pure Ratio2 9.7647 +Epoch [65/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8137, Pure Ratio2 9.8725 +Epoch [65/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8980, Pure Ratio2 9.9216 +Epoch [65/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0229, Pure Ratio2 10.0327 +Epoch [65/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0015, Pure Ratio1: 9.9300, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 25.9716 % Model2 27.7043 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 10.3137, Pure Ratio2 10.2157 +Epoch [66/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 87.5000, Loss1: 0.0013, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.2647 +Epoch [66/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.4183, Pure Ratio2 10.4771 +Epoch [66/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.2451, Pure Ratio2 10.2647 +Epoch [66/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1529, Pure Ratio2 10.1804 +Epoch [66/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.1471, Pure Ratio2 10.1797 +Epoch [66/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.9804, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 27.1835 % Model2 25.9315 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 10.0196 +Epoch [67/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.1667 +Epoch [67/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9346, Pure Ratio2 10.1569 +Epoch [67/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.1765 +Epoch [67/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1765 +Epoch [67/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.9902, Pure Ratio2 10.1536 +Epoch [67/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9748, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 27.3938 % Model2 26.8029 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.5686, Pure Ratio2 9.7059 +Epoch [68/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.5392, Pure Ratio2 9.5882 +Epoch [68/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7908, Pure Ratio2 9.7908 +Epoch [68/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8578, Pure Ratio2 9.8676 +Epoch [68/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.7961 +Epoch [68/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9379 +Epoch [68/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9076, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 26.6927 % Model2 27.5140 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.5490 +Epoch [69/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.7647 +Epoch [69/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8366, Pure Ratio2 9.9020 +Epoch [69/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0196, Pure Ratio2 10.0343 +Epoch [69/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8941, Pure Ratio2 9.8863 +Epoch [69/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9118, Pure Ratio2 9.9183 +Epoch [69/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.9440, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 27.3838 % Model2 27.3938 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.1765, Pure Ratio2 10.1373 +Epoch [70/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 10.1078, Pure Ratio2 10.1275 +Epoch [70/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.9281, Pure Ratio2 9.9869 +Epoch [70/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.8627, Pure Ratio2 9.9755 +Epoch [70/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0012, Pure Ratio1: 9.8627, Pure Ratio2 9.9922 +Epoch [70/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0359 +Epoch [70/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.9860, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 28.1550 % Model2 27.9347 %, Pure Ratio 1 9.8517 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0588, Pure Ratio2 9.8431 +Epoch [71/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9902, Pure Ratio2 9.9706 +Epoch [71/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1111, Pure Ratio2 10.0523 +Epoch [71/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.0245, Pure Ratio2 10.0000 +Epoch [71/200], Iter [250/390] Training Accuracy1: 95.3125, Training Accuracy2: 92.9688, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1451, Pure Ratio2 10.1647 +Epoch [71/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0229, Pure Ratio2 9.9902 +Epoch [71/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0280, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 28.0950 % Model2 28.2853 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.1176 +Epoch [72/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9608, Pure Ratio2 10.0980 +Epoch [72/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [72/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7745, Pure Ratio2 9.8088 +Epoch [72/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8510 +Epoch [72/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8595, Pure Ratio2 9.8954 +Epoch [72/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.9104, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 27.6943 % Model2 27.6843 %, Pure Ratio 1 9.8844 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0014, Loss2: 0.0002, Pure Ratio1: 9.4510, Pure Ratio2 9.3725 +Epoch [73/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 9.9412, Pure Ratio2 10.0294 +Epoch [73/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0065, Pure Ratio2 10.0131 +Epoch [73/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.9706 +Epoch [73/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0020, Loss2: 0.0005, Pure Ratio1: 9.9490, Pure Ratio2 9.9725 +Epoch [73/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.8889 +Epoch [73/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 9.8768, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 27.2837 % Model2 26.8830 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0017, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [74/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7255, Pure Ratio2 9.5882 +Epoch [74/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6471 +Epoch [74/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8333 +Epoch [74/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8745 +Epoch [74/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.7810, Pure Ratio2 9.7778 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8992, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 26.7829 % Model2 26.4724 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 10.1961 +Epoch [75/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9902, Pure Ratio2 10.1275 +Epoch [75/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0523, Pure Ratio2 10.1242 +Epoch [75/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.1471, Pure Ratio2 10.1912 +Epoch [75/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0941, Pure Ratio2 10.1373 +Epoch [75/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0523, Pure Ratio2 10.0784 +Epoch [75/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0018, Loss2: 0.0005, Pure Ratio1: 9.9440, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 27.8446 % Model2 27.1334 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.4314, Pure Ratio2 10.6667 +Epoch [76/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.3333, Pure Ratio2 10.2059 +Epoch [76/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [76/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.1569 +Epoch [76/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0745, Pure Ratio2 9.9725 +Epoch [76/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9575, Pure Ratio2 9.9183 +Epoch [76/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8599, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 27.6643 % Model2 25.7512 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1373 +Epoch [77/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 10.0196 +Epoch [77/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.9673 +Epoch [77/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 10.0049, Pure Ratio2 10.0882 +Epoch [77/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9843, Pure Ratio2 10.0863 +Epoch [77/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9379, Pure Ratio2 10.0294 +Epoch [77/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9328, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 28.4054 % Model2 27.5341 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Epoch [78/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.6078, Pure Ratio2 9.5784 +Epoch [78/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.7712 +Epoch [78/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 9.7255, Pure Ratio2 9.7794 +Epoch [78/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.7608, Pure Ratio2 9.7843 +Epoch [78/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [78/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.8936, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 27.4940 % Model2 27.8145 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0392, Pure Ratio2 10.1961 +Epoch [79/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7059 +Epoch [79/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6405, Pure Ratio2 9.7516 +Epoch [79/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.7892 +Epoch [79/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.8824 +Epoch [79/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.7222, Pure Ratio2 9.8497 +Epoch [79/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 27.0333 % Model2 27.1835 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.7059, Pure Ratio2 9.8627 +Epoch [80/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.5098, Pure Ratio2 9.7059 +Epoch [80/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.6340, Pure Ratio2 9.7843 +Epoch [80/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.7892 +Epoch [80/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.8941, Pure Ratio2 9.9686 +Epoch [80/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.9967, Pure Ratio2 10.0523 +Epoch [80/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 27.8345 % Model2 26.8129 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.3529, Pure Ratio2 9.4706 +Epoch [81/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.8333 +Epoch [81/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.7778, Pure Ratio2 9.8170 +Epoch [81/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7745, Pure Ratio2 9.8235 +Epoch [81/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8314, Pure Ratio2 9.8353 +Epoch [81/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9150, Pure Ratio2 9.8758 +Epoch [81/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9244, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 27.4239 % Model2 27.4139 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [82/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.6863, Pure Ratio2 9.7647 +Epoch [82/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.6601, Pure Ratio2 9.6928 +Epoch [82/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7010, Pure Ratio2 9.7010 +Epoch [82/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7412, Pure Ratio2 9.7961 +Epoch [82/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7712, Pure Ratio2 9.8039 +Epoch [82/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8543, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 27.8546 % Model2 27.2536 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0019, Pure Ratio1: 9.7059, Pure Ratio2 9.3922 +Epoch [83/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.6961 +Epoch [83/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.6863 +Epoch [83/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9363, Pure Ratio2 9.8529 +Epoch [83/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.9059, Pure Ratio2 9.8902 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.8889, Pure Ratio2 9.8693 +Epoch [83/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 26.8630 % Model2 26.8329 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.8431 +Epoch [84/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.9902 +Epoch [84/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.6536, Pure Ratio2 9.6667 +Epoch [84/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.6765, Pure Ratio2 9.6471 +Epoch [84/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6549, Pure Ratio2 9.6235 +Epoch [84/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7124 +Epoch [84/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7367, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 27.0433 % Model2 26.5425 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 9.5098, Pure Ratio2 9.6667 +Epoch [85/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6176, Pure Ratio2 9.6471 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.8366, Pure Ratio2 9.8431 +Epoch [85/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8431, Pure Ratio2 9.7402 +Epoch [85/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9294, Pure Ratio2 9.8627 +Epoch [85/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9281, Pure Ratio2 9.8725 +Epoch [85/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8571, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 27.0933 % Model2 27.9147 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.7843 +Epoch [86/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 9.9412 +Epoch [86/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7582 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9069, Pure Ratio2 9.8382 +Epoch [86/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9529, Pure Ratio2 9.9569 +Epoch [86/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8954, Pure Ratio2 9.8725 +Epoch [86/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 27.7444 % Model2 27.5841 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.7451 +Epoch [87/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0980, Pure Ratio2 9.8922 +Epoch [87/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 10.1438, Pure Ratio2 9.9216 +Epoch [87/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.2059, Pure Ratio2 10.0735 +Epoch [87/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.0941, Pure Ratio2 9.9569 +Epoch [87/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0490, Pure Ratio2 9.9052 +Epoch [87/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9972, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 27.9447 % Model2 26.0317 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.5098 +Epoch [88/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6569, Pure Ratio2 9.5882 +Epoch [88/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7190, Pure Ratio2 9.6928 +Epoch [88/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8088, Pure Ratio2 9.7696 +Epoch [88/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9255, Pure Ratio2 9.9451 +Epoch [88/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.8497, Pure Ratio2 9.9608 +Epoch [88/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8964, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 27.3638 % Model2 26.7428 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.6275, Pure Ratio2 9.5882 +Epoch [89/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.4608, Pure Ratio2 9.4216 +Epoch [89/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7647, Pure Ratio2 9.7974 +Epoch [89/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9461, Pure Ratio2 9.8971 +Epoch [89/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9255, Pure Ratio2 9.8549 +Epoch [89/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8431 +Epoch [89/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9048, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 27.1635 % Model2 26.6827 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.5686, Pure Ratio2 10.6863 +Epoch [90/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0012, Pure Ratio1: 10.3922, Pure Ratio2 10.4020 +Epoch [90/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 10.3529, Pure Ratio2 10.3333 +Epoch [90/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0686 +Epoch [90/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1059, Pure Ratio2 10.0784 +Epoch [90/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0686, Pure Ratio2 10.0719 +Epoch [90/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0021, Loss2: 0.0017, Pure Ratio1: 9.9748, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 26.6627 % Model2 26.0617 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6863, Pure Ratio2 9.4902 +Epoch [91/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.7941 +Epoch [91/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 9.8889 +Epoch [91/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0049 +Epoch [91/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1059, Pure Ratio2 9.9882 +Epoch [91/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 9.9379 +Epoch [91/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0812, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 27.0533 % Model2 27.0733 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.8517 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7451, Pure Ratio2 9.4706 +Epoch [92/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6176, Pure Ratio2 9.4902 +Epoch [92/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8105, Pure Ratio2 9.7320 +Epoch [92/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8137, Pure Ratio2 9.7598 +Epoch [92/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8078, Pure Ratio2 9.7765 +Epoch [92/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8464 +Epoch [92/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 26.6126 % Model2 28.5156 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [93/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.6863, Pure Ratio2 9.7451 +Epoch [93/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9739 +Epoch [93/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9314, Pure Ratio2 9.9461 +Epoch [93/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9294 +Epoch [93/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9314 +Epoch [93/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8908, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 28.9263 % Model2 27.2837 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8039, Pure Ratio2 9.9020 +Epoch [94/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6569, Pure Ratio2 9.7059 +Epoch [94/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8627, Pure Ratio2 9.9150 +Epoch [94/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8039, Pure Ratio2 9.8480 +Epoch [94/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8392, Pure Ratio2 9.8000 +Epoch [94/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0015, Loss2: 0.0006, Pure Ratio1: 9.8497, Pure Ratio2 9.7908 +Epoch [94/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.8627, Pure Ratio2 9.7983 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 26.8029 % Model2 25.1903 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.7451 +Epoch [95/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 9.9510 +Epoch [95/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 10.1307, Pure Ratio2 10.1438 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1667, Pure Ratio2 10.1765 +Epoch [95/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0471, Pure Ratio2 10.0471 +Epoch [95/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0012, Pure Ratio1: 9.9739, Pure Ratio2 9.9902 +Epoch [95/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 27.2135 % Model2 26.3522 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.3725, Pure Ratio2 9.6275 +Epoch [96/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.8039 +Epoch [96/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6340, Pure Ratio2 9.8366 +Epoch [96/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.5980, Pure Ratio2 9.7647 +Epoch [96/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7686, Pure Ratio2 9.8941 +Epoch [96/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8333, Pure Ratio2 9.8824 +Epoch [96/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8375, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 27.3738 % Model2 26.1318 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.1961 +Epoch [97/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 10.0784 +Epoch [97/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 9.9477 +Epoch [97/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9167, Pure Ratio2 9.9314 +Epoch [97/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8588, Pure Ratio2 9.8353 +Epoch [97/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8791, Pure Ratio2 9.8529 +Epoch [97/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8067, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 28.8662 % Model2 26.6126 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.2549, Pure Ratio2 9.3725 +Epoch [98/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [98/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9412, Pure Ratio2 10.0065 +Epoch [98/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8529 +Epoch [98/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.8549, Pure Ratio2 9.7804 +Epoch [98/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.8824 +Epoch [98/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9468, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 28.2352 % Model2 27.9647 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.0980 +Epoch [99/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.2157 +Epoch [99/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2680, Pure Ratio2 10.2941 +Epoch [99/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.1814 +Epoch [99/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2000, Pure Ratio2 10.1569 +Epoch [99/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0621, Pure Ratio2 10.0425 +Epoch [99/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0308, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 28.5557 % Model2 27.5541 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.1373 +Epoch [100/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2353, Pure Ratio2 10.2059 +Epoch [100/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 9.9477 +Epoch [100/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9069, Pure Ratio2 9.7745 +Epoch [100/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7765 +Epoch [100/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.7549 +Epoch [100/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0000, Pure Ratio1: 9.9272, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 28.5857 % Model2 27.7344 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [101/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.1176 +Epoch [101/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 10.0131 +Epoch [101/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9461, Pure Ratio2 10.0441 +Epoch [101/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8902, Pure Ratio2 9.9804 +Epoch [101/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8758, Pure Ratio2 9.9150 +Epoch [101/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8711, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 27.5040 % Model2 27.8245 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 9.9608 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7353 +Epoch [102/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0013, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [102/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8824 +Epoch [102/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0235, Pure Ratio2 9.9686 +Epoch [102/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 9.9150 +Epoch [102/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 26.9131 % Model2 26.1118 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.1176 +Epoch [103/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.1961 +Epoch [103/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1830, Pure Ratio2 10.2484 +Epoch [103/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0931, Pure Ratio2 10.1127 +Epoch [103/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 9.9569 +Epoch [103/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.9379, Pure Ratio2 9.9477 +Epoch [103/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9048, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 27.4339 % Model2 28.0849 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0012, Pure Ratio1: 9.7059, Pure Ratio2 9.7059 +Epoch [104/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.5196, Pure Ratio2 9.6373 +Epoch [104/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7778, Pure Ratio2 9.8431 +Epoch [104/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9265, Pure Ratio2 9.9265 +Epoch [104/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8980, Pure Ratio2 9.9098 +Epoch [104/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [104/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 27.6743 % Model2 27.2236 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.4118 +Epoch [105/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.7059, Pure Ratio2 9.6471 +Epoch [105/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.6405 +Epoch [105/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6618, Pure Ratio2 9.6471 +Epoch [105/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8078, Pure Ratio2 9.7804 +Epoch [105/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7974 +Epoch [105/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8880, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 26.9832 % Model2 26.9631 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [106/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [106/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.9346, Pure Ratio2 9.9542 +Epoch [106/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9167 +Epoch [106/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9176, Pure Ratio2 9.8784 +Epoch [106/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8922, Pure Ratio2 9.8889 +Epoch [106/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.9104, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 27.0533 % Model2 27.6042 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 10.0392 +Epoch [107/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1569 +Epoch [107/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0051, Loss2: 0.0048, Pure Ratio1: 10.1111, Pure Ratio2 10.0784 +Epoch [107/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1324 +Epoch [107/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1137, Pure Ratio2 10.0275 +Epoch [107/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0012, Loss2: 0.0001, Pure Ratio1: 10.0556, Pure Ratio2 9.9085 +Epoch [107/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9748, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 27.0833 % Model2 28.3153 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7843 +Epoch [108/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.6569 +Epoch [108/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5621, Pure Ratio2 9.6536 +Epoch [108/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7059 +Epoch [108/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6118, Pure Ratio2 9.5725 +Epoch [108/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.8758, Pure Ratio2 9.8007 +Epoch [108/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9720, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 26.6326 % Model2 26.8530 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 11.0588, Pure Ratio2 10.9608 +Epoch [109/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.2549, Pure Ratio2 10.3137 +Epoch [109/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0327, Pure Ratio2 10.0196 +Epoch [109/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0049 +Epoch [109/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9882 +Epoch [109/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 9.9739 +Epoch [109/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 27.9347 % Model2 27.8846 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 9.9804 +Epoch [110/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.1765 +Epoch [110/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.8431 +Epoch [110/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0441, Pure Ratio2 9.9853 +Epoch [110/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9765, Pure Ratio2 9.9529 +Epoch [110/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9608, Pure Ratio2 9.9804 +Epoch [110/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9160, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 28.2552 % Model2 28.5457 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8627 +Epoch [111/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [111/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1438, Pure Ratio2 10.1503 +Epoch [111/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1275, Pure Ratio2 10.1078 +Epoch [111/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0980, Pure Ratio2 10.1098 +Epoch [111/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0980 +Epoch [111/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9552, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 27.8045 % Model2 27.9247 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5686, Pure Ratio2 10.4510 +Epoch [112/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2745 +Epoch [112/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3203, Pure Ratio2 10.3922 +Epoch [112/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.2353 +Epoch [112/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.1882 +Epoch [112/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9641, Pure Ratio2 10.0098 +Epoch [112/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9468, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 27.8546 % Model2 27.0733 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.9412 +Epoch [113/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 9.9902 +Epoch [113/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.5948 +Epoch [113/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.7402, Pure Ratio2 9.6765 +Epoch [113/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.8824 +Epoch [113/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9673 +Epoch [113/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0112, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 27.6042 % Model2 27.3037 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9020 +Epoch [114/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1078, Pure Ratio2 10.1078 +Epoch [114/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.0523 +Epoch [114/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9951, Pure Ratio2 9.9657 +Epoch [114/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 9.9412 +Epoch [114/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 9.9510 +Epoch [114/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9328, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 28.4155 % Model2 27.4840 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.5294, Pure Ratio2 10.4118 +Epoch [115/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.0882 +Epoch [115/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 10.0784 +Epoch [115/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0539 +Epoch [115/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1608, Pure Ratio2 10.1647 +Epoch [115/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0294 +Epoch [115/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0616, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 27.8546 % Model2 28.1150 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [116/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8529 +Epoch [116/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.8366 +Epoch [116/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9167, Pure Ratio2 9.9216 +Epoch [116/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8784, Pure Ratio2 9.8941 +Epoch [116/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8595, Pure Ratio2 9.8856 +Epoch [116/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8403, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 28.2552 % Model2 27.3838 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.3529 +Epoch [117/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0014, Pure Ratio1: 9.6078, Pure Ratio2 9.4804 +Epoch [117/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.6863 +Epoch [117/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8186, Pure Ratio2 9.7059 +Epoch [117/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [117/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.8203 +Epoch [117/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 27.9447 % Model2 27.5641 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 9.6667 +Epoch [118/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 9.8529 +Epoch [118/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.8039 +Epoch [118/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8873 +Epoch [118/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9373, Pure Ratio2 9.9098 +Epoch [118/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7320, Pure Ratio2 9.7386 +Epoch [118/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7759, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 27.1034 % Model2 27.8345 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.4510, Pure Ratio2 10.5686 +Epoch [119/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3431, Pure Ratio2 10.3431 +Epoch [119/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3595, Pure Ratio2 10.2614 +Epoch [119/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2402, Pure Ratio2 10.2206 +Epoch [119/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1882, Pure Ratio2 10.2549 +Epoch [119/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 10.1078 +Epoch [119/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 27.3538 % Model2 28.4255 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9804, Pure Ratio2 9.9020 +Epoch [120/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9804 +Epoch [120/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1111 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2598, Pure Ratio2 10.1275 +Epoch [120/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0549, Pure Ratio2 9.9216 +Epoch [120/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0229, Pure Ratio2 9.8824 +Epoch [120/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 28.1250 % Model2 26.9030 %, Pure Ratio 1 10.0302 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.5882 +Epoch [121/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.8627 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7320 +Epoch [121/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.6863 +Epoch [121/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.7490 +Epoch [121/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8660, Pure Ratio2 9.8235 +Epoch [121/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9692, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 26.2019 % Model2 27.1735 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2157 +Epoch [122/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9804 +Epoch [122/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1895, Pure Ratio2 10.2222 +Epoch [122/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0784 +Epoch [122/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8902 +Epoch [122/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.8889 +Epoch [122/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9160, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 27.6743 % Model2 27.6743 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 9.8627 +Epoch [123/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.6471 +Epoch [123/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [123/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9265, Pure Ratio2 9.7794 +Epoch [123/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0353, Pure Ratio2 9.8941 +Epoch [123/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8431 +Epoch [123/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 28.4756 % Model2 28.3353 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.6078 +Epoch [124/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.5196 +Epoch [124/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 9.8039 +Epoch [124/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1716, Pure Ratio2 9.9608 +Epoch [124/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 9.9098 +Epoch [124/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.0261, Pure Ratio2 9.8889 +Epoch [124/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 27.9147 % Model2 26.9331 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.6078 +Epoch [125/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7353, Pure Ratio2 9.6863 +Epoch [125/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Epoch [125/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0147 +Epoch [125/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.8667 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9183 +Epoch [125/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8936, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 27.7043 % Model2 27.5541 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8627 +Epoch [126/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.7157 +Epoch [126/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.5163, Pure Ratio2 9.6340 +Epoch [126/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4804, Pure Ratio2 9.5882 +Epoch [126/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7176, Pure Ratio2 9.7922 +Epoch [126/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8464 +Epoch [126/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8515, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 28.3654 % Model2 28.3654 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.3137 +Epoch [127/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1176 +Epoch [127/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9935, Pure Ratio2 10.0784 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9755 +Epoch [127/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7686, Pure Ratio2 9.7882 +Epoch [127/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7516 +Epoch [127/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8487, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 26.2620 % Model2 26.5425 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Epoch [128/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1667, Pure Ratio2 10.0490 +Epoch [128/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1307, Pure Ratio2 10.0065 +Epoch [128/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0294 +Epoch [128/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.8510 +Epoch [128/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0019, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 9.9575 +Epoch [128/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0168, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 27.8946 % Model2 28.5357 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.2549 +Epoch [129/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.6961, Pure Ratio2 9.7941 +Epoch [129/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6993, Pure Ratio2 9.6993 +Epoch [129/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7843 +Epoch [129/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8392, Pure Ratio2 9.8863 +Epoch [129/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9052 +Epoch [129/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9636, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 28.1751 % Model2 27.1034 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5294, Pure Ratio2 10.5294 +Epoch [130/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3235, Pure Ratio2 10.3529 +Epoch [130/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.2680 +Epoch [130/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8971, Pure Ratio2 9.8627 +Epoch [130/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9686, Pure Ratio2 9.9098 +Epoch [130/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 9.9118 +Epoch [130/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9944, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 27.1835 % Model2 28.5657 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9608 +Epoch [131/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 10.0294 +Epoch [131/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0458, Pure Ratio2 10.1569 +Epoch [131/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 10.1520 +Epoch [131/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [131/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0000 +Epoch [131/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 27.7444 % Model2 27.7544 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.8824, Pure Ratio2 10.0588 +Epoch [132/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0490 +Epoch [132/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.0458 +Epoch [132/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.0882 +Epoch [132/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9686, Pure Ratio2 9.9451 +Epoch [132/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.9085 +Epoch [132/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 27.5942 % Model2 27.4339 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [133/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2843, Pure Ratio2 10.2745 +Epoch [133/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 9.9935 +Epoch [133/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9069, Pure Ratio2 9.8039 +Epoch [133/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8510 +Epoch [133/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9118, Pure Ratio2 9.8072 +Epoch [133/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0004, Pure Ratio1: 9.8880, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 26.7228 % Model2 25.9916 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [134/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9314, Pure Ratio2 9.8235 +Epoch [134/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.5359, Pure Ratio2 9.5556 +Epoch [134/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.7892 +Epoch [134/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8941 +Epoch [134/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9314, Pure Ratio2 9.9118 +Epoch [134/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0504, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 27.7244 % Model2 28.2452 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 8.9216, Pure Ratio2 8.9412 +Epoch [135/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.4216 +Epoch [135/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.8497 +Epoch [135/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7696 +Epoch [135/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.7412 +Epoch [135/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7974 +Epoch [135/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8375, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 28.5757 % Model2 28.4756 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [136/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.0392 +Epoch [136/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0065 +Epoch [136/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0343, Pure Ratio2 10.1176 +Epoch [136/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9333, Pure Ratio2 9.9765 +Epoch [136/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.9183 +Epoch [136/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8347, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 28.8762 % Model2 27.8345 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.6078 +Epoch [137/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0000 +Epoch [137/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9281 +Epoch [137/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8333 +Epoch [137/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9098, Pure Ratio2 9.9098 +Epoch [137/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 9.9575 +Epoch [137/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8711, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 26.6627 % Model2 28.5657 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.8039, Pure Ratio2 10.6471 +Epoch [138/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2745 +Epoch [138/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0719, Pure Ratio2 10.1765 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1176 +Epoch [138/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.9961 +Epoch [138/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9542, Pure Ratio2 9.9771 +Epoch [138/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9580, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 28.0549 % Model2 27.6042 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8431 +Epoch [139/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9804 +Epoch [139/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6471, Pure Ratio2 9.6471 +Epoch [139/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7892 +Epoch [139/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9569, Pure Ratio2 10.0510 +Epoch [139/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.9575 +Epoch [139/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9552, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 27.3438 % Model2 27.0433 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.4902 +Epoch [140/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.0196 +Epoch [140/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.2222, Pure Ratio2 10.0784 +Epoch [140/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9461, Pure Ratio2 9.8284 +Epoch [140/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.8157 +Epoch [140/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9477, Pure Ratio2 9.7908 +Epoch [140/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9440, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 27.1635 % Model2 27.1935 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3922 +Epoch [141/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1765 +Epoch [141/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.2418 +Epoch [141/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0931 +Epoch [141/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9843, Pure Ratio2 10.0980 +Epoch [141/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 10.0098 +Epoch [141/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9104, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 27.7143 % Model2 26.5425 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.4510, Pure Ratio2 9.6667 +Epoch [142/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7157 +Epoch [142/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7124 +Epoch [142/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7941 +Epoch [142/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8588, Pure Ratio2 9.7569 +Epoch [142/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9444, Pure Ratio2 9.8595 +Epoch [142/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 27.5741 % Model2 28.6759 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 9.9196 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5882, Pure Ratio2 9.7451 +Epoch [143/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5882, Pure Ratio2 9.5294 +Epoch [143/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6405 +Epoch [143/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6029, Pure Ratio2 9.6078 +Epoch [143/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7608, Pure Ratio2 9.7569 +Epoch [143/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7614, Pure Ratio2 9.7549 +Epoch [143/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8571, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 27.5541 % Model2 28.7961 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 93.7500, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 9.7451 +Epoch [144/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.6569 +Epoch [144/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.6209 +Epoch [144/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7500, Pure Ratio2 9.6716 +Epoch [144/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7686 +Epoch [144/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 9.8268 +Epoch [144/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9468, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 27.3538 % Model2 27.1234 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.8819 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 9.9216 +Epoch [145/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8431 +Epoch [145/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.8497 +Epoch [145/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8775 +Epoch [145/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.8980 +Epoch [145/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9314 +Epoch [145/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 27.4339 % Model2 27.6943 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5882 +Epoch [146/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7059 +Epoch [146/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.6275 +Epoch [146/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7745 +Epoch [146/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.9020 +Epoch [146/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8693 +Epoch [146/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 28.2853 % Model2 27.2336 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.6471 +Epoch [147/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7451 +Epoch [147/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0011, Pure Ratio1: 9.8301, Pure Ratio2 9.8039 +Epoch [147/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.8775 +Epoch [147/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.8667 +Epoch [147/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8987 +Epoch [147/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0280, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 27.1034 % Model2 27.4539 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.5098 +Epoch [148/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.5392 +Epoch [148/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7451 +Epoch [148/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.7745 +Epoch [148/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.7686 +Epoch [148/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8333, Pure Ratio2 9.7810 +Epoch [148/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8711, Pure Ratio2 9.8627 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 27.8045 % Model2 26.7829 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.1765 +Epoch [149/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.4706, Pure Ratio2 9.5000 +Epoch [149/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8301 +Epoch [149/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0049 +Epoch [149/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0078, Pure Ratio2 9.9686 +Epoch [149/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.8431 +Epoch [149/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 9.9160, Pure Ratio2 9.8739 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 27.6242 % Model2 27.9447 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.8844 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.6863 +Epoch [150/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.2255 +Epoch [150/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 10.0850 +Epoch [150/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0245 +Epoch [150/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9608 +Epoch [150/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.9837 +Epoch [150/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 28.4255 % Model2 28.2552 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.3333 +Epoch [151/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2647, Pure Ratio2 9.3922 +Epoch [151/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.8039 +Epoch [151/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8186 +Epoch [151/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7020, Pure Ratio2 9.7451 +Epoch [151/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7712, Pure Ratio2 9.7941 +Epoch [151/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8095, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 27.6242 % Model2 27.3137 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1961 +Epoch [152/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0588 +Epoch [152/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.1438 +Epoch [152/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 9.9461 +Epoch [152/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.8863 +Epoch [152/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.8007 +Epoch [152/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 27.8546 % Model2 28.5056 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.9608 +Epoch [153/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.7647 +Epoch [153/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6993, Pure Ratio2 9.7320 +Epoch [153/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8039 +Epoch [153/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8118, Pure Ratio2 9.8627 +Epoch [153/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.8595 +Epoch [153/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8571, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 26.8730 % Model2 27.7143 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 10.1765 +Epoch [154/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1863 +Epoch [154/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0784 +Epoch [154/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.9216 +Epoch [154/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.9176 +Epoch [154/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9444 +Epoch [154/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8936, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 26.6226 % Model2 27.2837 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.3725 +Epoch [155/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.6471, Pure Ratio2 9.5098 +Epoch [155/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5752 +Epoch [155/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8186, Pure Ratio2 9.7255 +Epoch [155/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.8549 +Epoch [155/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 9.9118 +Epoch [155/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 27.1134 % Model2 27.4840 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.4902 +Epoch [156/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.7745 +Epoch [156/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7516 +Epoch [156/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8088 +Epoch [156/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 9.9765 +Epoch [156/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.8954 +Epoch [156/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 27.4740 % Model2 27.8446 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7451 +Epoch [157/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9118 +Epoch [157/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7386 +Epoch [157/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 9.8873 +Epoch [157/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 9.9529 +Epoch [157/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9346 +Epoch [157/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 28.7861 % Model2 28.0148 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8431 +Epoch [158/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8137, Pure Ratio2 9.8627 +Epoch [158/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.6797 +Epoch [158/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.7402 +Epoch [158/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0314, Pure Ratio2 9.8431 +Epoch [158/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 9.8497 +Epoch [158/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 27.8846 % Model2 28.0849 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [159/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [159/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.8170 +Epoch [159/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 9.8873 +Epoch [159/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9255, Pure Ratio2 9.8196 +Epoch [159/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 9.9118 +Epoch [159/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 27.8946 % Model2 27.7244 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2549 +Epoch [160/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.8529 +Epoch [160/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.9346 +Epoch [160/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8971, Pure Ratio2 9.9853 +Epoch [160/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8471, Pure Ratio2 9.8824 +Epoch [160/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9085 +Epoch [160/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 27.6843 % Model2 28.1751 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8718 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.8431 +Epoch [161/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 10.0686 +Epoch [161/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0065 +Epoch [161/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 10.0343 +Epoch [161/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9451 +Epoch [161/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8497 +Epoch [161/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8459, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 28.1150 % Model2 27.8946 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.6863 +Epoch [162/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.6765 +Epoch [162/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9150 +Epoch [162/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.8627 +Epoch [162/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8471 +Epoch [162/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8529 +Epoch [162/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 27.9247 % Model2 27.2336 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2941 +Epoch [163/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.1078 +Epoch [163/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.9477 +Epoch [163/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8431 +Epoch [163/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8392, Pure Ratio2 9.8235 +Epoch [163/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8595, Pure Ratio2 9.8431 +Epoch [163/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8543, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 27.5942 % Model2 27.2636 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.2745 +Epoch [164/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7647 +Epoch [164/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7712 +Epoch [164/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8627 +Epoch [164/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8667 +Epoch [164/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9020 +Epoch [164/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 27.6142 % Model2 26.8329 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6078 +Epoch [165/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2549, Pure Ratio2 9.2353 +Epoch [165/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.5425 +Epoch [165/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6324 +Epoch [165/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7216, Pure Ratio2 9.6706 +Epoch [165/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8203 +Epoch [165/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9272, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 28.0649 % Model2 27.2736 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.4902 +Epoch [166/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1569 +Epoch [166/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7778 +Epoch [166/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9265 +Epoch [166/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 9.9373 +Epoch [166/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8399 +Epoch [166/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 27.5841 % Model2 27.2035 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8627 +Epoch [167/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6373 +Epoch [167/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.6209 +Epoch [167/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.7745 +Epoch [167/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.8549 +Epoch [167/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0131 +Epoch [167/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 27.1534 % Model2 27.5741 %, Pure Ratio 1 9.8718 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0000 +Epoch [168/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.5196, Pure Ratio2 10.4902 +Epoch [168/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0719 +Epoch [168/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [168/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0471 +Epoch [168/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.0065 +Epoch [168/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 27.9647 % Model2 26.9832 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5490 +Epoch [169/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.7745 +Epoch [169/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.7843 +Epoch [169/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.7745 +Epoch [169/200], Iter [250/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8745 +Epoch [169/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8529 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 27.8746 % Model2 28.1150 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1569 +Epoch [170/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9118 +Epoch [170/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7908, Pure Ratio2 9.7386 +Epoch [170/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.7990 +Epoch [170/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.8392 +Epoch [170/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8791 +Epoch [170/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8936, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 27.7744 % Model2 27.4539 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2941 +Epoch [171/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0392 +Epoch [171/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8497 +Epoch [171/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0490 +Epoch [171/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 10.0196 +Epoch [171/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9935 +Epoch [171/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 26.8730 % Model2 27.7143 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [172/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.5294 +Epoch [172/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.7582 +Epoch [172/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7696, Pure Ratio2 9.7304 +Epoch [172/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8510, Pure Ratio2 9.8078 +Epoch [172/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.8627 +Epoch [172/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9916, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 28.7360 % Model2 28.6759 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1569 +Epoch [173/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0980 +Epoch [173/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8170 +Epoch [173/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.8578 +Epoch [173/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8667, Pure Ratio2 9.7765 +Epoch [173/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8105 +Epoch [173/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 27.9647 % Model2 27.2536 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 9.9608 +Epoch [174/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9118 +Epoch [174/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8497 +Epoch [174/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9363 +Epoch [174/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8706, Pure Ratio2 9.8824 +Epoch [174/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9412 +Epoch [174/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 27.4740 % Model2 27.4439 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9608 +Epoch [175/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8235 +Epoch [175/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0784 +Epoch [175/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8480 +Epoch [175/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 9.8980 +Epoch [175/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.9216 +Epoch [175/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 27.3738 % Model2 27.3538 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.1961 +Epoch [176/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.3333 +Epoch [176/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5359 +Epoch [176/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.7647 +Epoch [176/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.7843 +Epoch [176/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8595 +Epoch [176/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 27.2837 % Model2 28.5256 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7451 +Epoch [177/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1176 +Epoch [177/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.2810 +Epoch [177/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.2059 +Epoch [177/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 10.0549 +Epoch [177/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8889 +Epoch [177/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 27.9347 % Model2 28.0349 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0980 +Epoch [178/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3039, Pure Ratio2 10.2451 +Epoch [178/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.2549 +Epoch [178/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0392 +Epoch [178/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9961, Pure Ratio2 9.9333 +Epoch [178/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.8268 +Epoch [178/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 27.2636 % Model2 27.7043 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.2745 +Epoch [179/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7157 +Epoch [179/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8627 +Epoch [179/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7794, Pure Ratio2 9.7598 +Epoch [179/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0235, Pure Ratio2 9.9725 +Epoch [179/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9575, Pure Ratio2 9.9379 +Epoch [179/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 27.0433 % Model2 28.0248 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1176, Pure Ratio2 9.0588 +Epoch [180/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.6176 +Epoch [180/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6667 +Epoch [180/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8578 +Epoch [180/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7137 +Epoch [180/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.8072 +Epoch [180/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 27.7744 % Model2 27.1735 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [181/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7549 +Epoch [181/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8170 +Epoch [181/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.7892 +Epoch [181/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.8588 +Epoch [181/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.7516 +Epoch [181/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8571, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 28.1951 % Model2 27.3738 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.2745 +Epoch [182/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.2353 +Epoch [182/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1895 +Epoch [182/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0000 +Epoch [182/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 9.9725 +Epoch [182/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.9020 +Epoch [182/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 28.5457 % Model2 27.5040 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.4902 +Epoch [183/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.5784 +Epoch [183/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6928 +Epoch [183/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8578, Pure Ratio2 9.7059 +Epoch [183/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8118, Pure Ratio2 9.6588 +Epoch [183/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7418 +Epoch [183/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 27.6142 % Model2 27.2536 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.9020 +Epoch [184/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.5098 +Epoch [184/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4967, Pure Ratio2 9.6405 +Epoch [184/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4951, Pure Ratio2 9.4951 +Epoch [184/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7529, Pure Ratio2 9.7176 +Epoch [184/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.8758 +Epoch [184/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 27.5040 % Model2 27.2636 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.5686 +Epoch [185/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.5980 +Epoch [185/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9085 +Epoch [185/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0147, Pure Ratio2 9.9853 +Epoch [185/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9294 +Epoch [185/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9771, Pure Ratio2 9.9706 +Epoch [185/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 27.6843 % Model2 27.6743 %, Pure Ratio 1 9.8994 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6863 +Epoch [186/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.9510 +Epoch [186/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7386 +Epoch [186/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8137 +Epoch [186/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6235, Pure Ratio2 9.7686 +Epoch [186/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8824 +Epoch [186/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7983, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 27.3938 % Model2 26.8129 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0392 +Epoch [187/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0490 +Epoch [187/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.9477 +Epoch [187/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9608 +Epoch [187/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9255, Pure Ratio2 10.1176 +Epoch [187/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.9771 +Epoch [187/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 27.7644 % Model2 26.8930 %, Pure Ratio 1 9.8567 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8627 +Epoch [188/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0098 +Epoch [188/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0000 +Epoch [188/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.8627 +Epoch [188/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 9.9686 +Epoch [188/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.0327 +Epoch [188/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0700, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 27.8646 % Model2 28.0449 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 9.9608 +Epoch [189/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8039 +Epoch [189/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8758 +Epoch [189/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.7549 +Epoch [189/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.7490 +Epoch [189/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8660, Pure Ratio2 9.8627 +Epoch [189/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 28.2752 % Model2 27.9547 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.4118 +Epoch [190/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0588 +Epoch [190/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9542 +Epoch [190/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9853 +Epoch [190/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 9.9608 +Epoch [190/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8301 +Epoch [190/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 27.2837 % Model2 27.7043 %, Pure Ratio 1 9.9120 %, Pure Ratio 2 9.8944 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 11.0392, Pure Ratio2 10.7647 +Epoch [191/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.1471 +Epoch [191/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2222, Pure Ratio2 10.0523 +Epoch [191/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 9.9755 +Epoch [191/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.0000 +Epoch [191/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9641 +Epoch [191/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 27.8145 % Model2 26.8730 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.5490 +Epoch [192/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.3725 +Epoch [192/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0028, Loss2: 0.0031, Pure Ratio1: 9.6993, Pure Ratio2 9.6993 +Epoch [192/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.7010 +Epoch [192/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7529, Pure Ratio2 9.7451 +Epoch [192/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9020 +Epoch [192/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8796, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 27.3538 % Model2 27.5541 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.8869 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.3137 +Epoch [193/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7451 +Epoch [193/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8301 +Epoch [193/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7157 +Epoch [193/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.8431 +Epoch [193/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8203 +Epoch [193/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 28.2151 % Model2 27.2135 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.2157 +Epoch [194/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0294 +Epoch [194/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0523 +Epoch [194/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9412 +Epoch [194/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 9.9843 +Epoch [194/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1209, Pure Ratio2 10.0784 +Epoch [194/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1064, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 27.1835 % Model2 27.4239 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6078 +Epoch [195/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.6667 +Epoch [195/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7386 +Epoch [195/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 9.7941 +Epoch [195/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.9176 +Epoch [195/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9412 +Epoch [195/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9244, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 27.4740 % Model2 27.5441 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.2157 +Epoch [196/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9804 +Epoch [196/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0327 +Epoch [196/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0098 +Epoch [196/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0627, Pure Ratio2 10.0039 +Epoch [196/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.9248 +Epoch [196/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9972, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 27.6743 % Model2 27.1334 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 10.0784 +Epoch [197/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.9020 +Epoch [197/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.9542 +Epoch [197/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.9412 +Epoch [197/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9765 +Epoch [197/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8954 +Epoch [197/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 27.4539 % Model2 27.4840 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9412 +Epoch [198/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.6569 +Epoch [198/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.9020 +Epoch [198/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.8382 +Epoch [198/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7569, Pure Ratio2 9.8667 +Epoch [198/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9477 +Epoch [198/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 27.8946 % Model2 27.6643 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [199/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0588 +Epoch [199/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9869 +Epoch [199/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.7892 +Epoch [199/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.8784 +Epoch [199/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9967 +Epoch [199/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 27.4339 % Model2 27.8446 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.8039 +Epoch [200/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.4118 +Epoch [200/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0915 +Epoch [200/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.0245 +Epoch [200/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9882 +Epoch [200/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.0327 +Epoch [200/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 27.3838 % Model2 27.6943 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9296 % diff --git a/other_methods/coteaching/coteaching_results/out_6_2.log b/other_methods/coteaching/coteaching_results/out_6_2.log new file mode 100644 index 0000000..5e384d6 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_6_2.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0150, Loss2: 0.0149, Pure Ratio1: 10.2080, Pure Ratio2 10.1440 +Epoch [2/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0134, Loss2: 0.0134, Pure Ratio1: 10.2960, Pure Ratio2 10.2640 +Epoch [2/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0130, Loss2: 0.0132, Pure Ratio1: 10.2453, Pure Ratio2 10.1973 +Epoch [2/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0128, Loss2: 0.0128, Pure Ratio1: 10.2720, Pure Ratio2 10.2520 +Epoch [2/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0130, Loss2: 0.0134, Pure Ratio1: 10.1824, Pure Ratio2 10.1568 +Epoch [2/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0113, Loss2: 0.0111, Pure Ratio1: 10.1573, Pure Ratio2 10.1227 +Epoch [2/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0113, Loss2: 0.0113, Pure Ratio1: 10.1189, Pure Ratio2 10.1006 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 54.2969 % Model2 53.9163 %, Pure Ratio 1 10.0841 %, Pure Ratio 2 10.0574 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0100, Loss2: 0.0096, Pure Ratio1: 9.4918, Pure Ratio2 9.5246 +Epoch [3/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0116, Loss2: 0.0115, Pure Ratio1: 9.8607, Pure Ratio2 9.8770 +Epoch [3/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0116, Loss2: 0.0113, Pure Ratio1: 9.8962, Pure Ratio2 9.9180 +Epoch [3/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0103, Loss2: 0.0103, Pure Ratio1: 9.9836, Pure Ratio2 9.9959 +Epoch [3/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0107, Loss2: 0.0109, Pure Ratio1: 10.0230, Pure Ratio2 10.0230 +Epoch [3/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0110, Loss2: 0.0110, Pure Ratio1: 10.0956, Pure Ratio2 10.0984 +Epoch [3/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0101, Loss2: 0.0098, Pure Ratio1: 10.1054, Pure Ratio2 10.1101 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 65.5749 % Model2 65.6851 %, Pure Ratio 1 10.0189 %, Pure Ratio 2 10.0210 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0088, Loss2: 0.0090, Pure Ratio1: 9.7143, Pure Ratio2 9.6975 +Epoch [4/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0111, Loss2: 0.0108, Pure Ratio1: 10.0924, Pure Ratio2 10.0252 +Epoch [4/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0085, Loss2: 0.0085, Pure Ratio1: 10.0560, Pure Ratio2 10.0112 +Epoch [4/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0082, Loss2: 0.0082, Pure Ratio1: 9.9076, Pure Ratio2 9.8529 +Epoch [4/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0078, Loss2: 0.0084, Pure Ratio1: 9.8588, Pure Ratio2 9.8218 +Epoch [4/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0082, Loss2: 0.0079, Pure Ratio1: 9.9496, Pure Ratio2 9.9132 +Epoch [4/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0072, Loss2: 0.0076, Pure Ratio1: 9.9712, Pure Ratio2 9.9568 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 68.2192 % Model2 68.2192 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0085, Loss2: 0.0085, Pure Ratio1: 9.8793, Pure Ratio2 9.9138 +Epoch [5/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0069, Loss2: 0.0076, Pure Ratio1: 9.8621, Pure Ratio2 9.8793 +Epoch [5/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0076, Loss2: 0.0082, Pure Ratio1: 10.1264, Pure Ratio2 10.1322 +Epoch [5/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0071, Loss2: 0.0076, Pure Ratio1: 10.0862, Pure Ratio2 10.0991 +Epoch [5/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0082, Loss2: 0.0087, Pure Ratio1: 10.1310, Pure Ratio2 10.1448 +Epoch [5/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0083, Loss2: 0.0080, Pure Ratio1: 9.9483, Pure Ratio2 9.9655 +Epoch [5/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0069, Loss2: 0.0073, Pure Ratio1: 10.0320, Pure Ratio2 10.0493 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 70.6931 % Model2 71.5946 %, Pure Ratio 1 9.9668 %, Pure Ratio 2 9.9823 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0078, Loss2: 0.0089, Pure Ratio1: 10.2655, Pure Ratio2 10.3009 +Epoch [6/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0066, Loss2: 0.0074, Pure Ratio1: 10.2832, Pure Ratio2 10.3009 +Epoch [6/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0055, Loss2: 0.0055, Pure Ratio1: 10.2478, Pure Ratio2 10.2360 +Epoch [6/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0058, Loss2: 0.0065, Pure Ratio1: 10.0796, Pure Ratio2 10.1195 +Epoch [6/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0079, Loss2: 0.0079, Pure Ratio1: 10.0814, Pure Ratio2 10.0885 +Epoch [6/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0079, Loss2: 0.0074, Pure Ratio1: 9.9764, Pure Ratio2 9.9882 +Epoch [6/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.0050, Loss2: 0.0055, Pure Ratio1: 9.9621, Pure Ratio2 9.9646 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 73.6178 % Model2 71.0136 %, Pure Ratio 1 9.9523 %, Pure Ratio 2 9.9750 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0059, Loss2: 0.0060, Pure Ratio1: 10.2545, Pure Ratio2 10.2909 +Epoch [7/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.1875, Loss1: 0.0046, Loss2: 0.0049, Pure Ratio1: 10.1000, Pure Ratio2 10.1364 +Epoch [7/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0049, Loss2: 0.0042, Pure Ratio1: 10.3879, Pure Ratio2 10.3576 +Epoch [7/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0064, Loss2: 0.0058, Pure Ratio1: 10.1227, Pure Ratio2 10.0682 +Epoch [7/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0056, Loss2: 0.0061, Pure Ratio1: 10.1273, Pure Ratio2 10.1091 +Epoch [7/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.9375, Loss1: 0.0048, Loss2: 0.0063, Pure Ratio1: 10.0303, Pure Ratio2 10.0152 +Epoch [7/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0075, Loss2: 0.0072, Pure Ratio1: 9.9792, Pure Ratio2 9.9740 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 67.4179 % Model2 76.8329 %, Pure Ratio 1 9.9580 %, Pure Ratio 2 9.9277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0042, Loss2: 0.0043, Pure Ratio1: 10.6481, Pure Ratio2 10.4815 +Epoch [8/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.0047, Loss2: 0.0052, Pure Ratio1: 10.1019, Pure Ratio2 10.1204 +Epoch [8/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0036, Loss2: 0.0042, Pure Ratio1: 9.9506, Pure Ratio2 9.9753 +Epoch [8/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0044, Loss2: 0.0039, Pure Ratio1: 9.8009, Pure Ratio2 9.8472 +Epoch [8/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.8750, Loss1: 0.0044, Loss2: 0.0041, Pure Ratio1: 9.9704, Pure Ratio2 9.9926 +Epoch [8/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0057, Loss2: 0.0060, Pure Ratio1: 9.9290, Pure Ratio2 9.9630 +Epoch [8/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0044, Loss2: 0.0046, Pure Ratio1: 9.9709, Pure Ratio2 9.9815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 78.5056 % Model2 75.8413 %, Pure Ratio 1 9.9905 %, Pure Ratio 2 9.9881 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0034, Loss2: 0.0033, Pure Ratio1: 9.6190, Pure Ratio2 9.6762 +Epoch [9/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0039, Loss2: 0.0037, Pure Ratio1: 9.6095, Pure Ratio2 9.6381 +Epoch [9/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0024, Loss2: 0.0024, Pure Ratio1: 9.7270, Pure Ratio2 9.7778 +Epoch [9/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0056, Loss2: 0.0055, Pure Ratio1: 9.6952, Pure Ratio2 9.7810 +Epoch [9/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0044, Loss2: 0.0042, Pure Ratio1: 9.9048, Pure Ratio2 9.9543 +Epoch [9/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0046, Loss2: 0.0042, Pure Ratio1: 9.9397, Pure Ratio2 9.9968 +Epoch [9/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0041, Loss2: 0.0045, Pure Ratio1: 9.8966, Pure Ratio2 9.9510 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 79.5873 % Model2 77.9748 %, Pure Ratio 1 9.9292 %, Pure Ratio 2 9.9805 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0043, Loss2: 0.0039, Pure Ratio1: 10.6078, Pure Ratio2 10.6863 +Epoch [10/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0029, Loss2: 0.0028, Pure Ratio1: 10.1863, Pure Ratio2 10.1863 +Epoch [10/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0048, Loss2: 0.0039, Pure Ratio1: 9.8301, Pure Ratio2 9.8039 +Epoch [10/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0042, Loss2: 0.0043, Pure Ratio1: 9.8824, Pure Ratio2 9.8725 +Epoch [10/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.0032, Loss2: 0.0031, Pure Ratio1: 9.8588, Pure Ratio2 9.8588 +Epoch [10/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0030, Loss2: 0.0035, Pure Ratio1: 9.8824, Pure Ratio2 9.9183 +Epoch [10/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0034, Loss2: 0.0035, Pure Ratio1: 9.9356, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 76.9231 % Model2 77.5240 %, Pure Ratio 1 9.8693 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 74.2188, Loss1: 0.0016, Loss2: 0.0029, Pure Ratio1: 9.6275, Pure Ratio2 9.5294 +Epoch [11/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.0027, Loss2: 0.0023, Pure Ratio1: 9.8824, Pure Ratio2 9.8333 +Epoch [11/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0036, Loss2: 0.0029, Pure Ratio1: 9.8562, Pure Ratio2 9.8497 +Epoch [11/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0040, Loss2: 0.0043, Pure Ratio1: 9.8382, Pure Ratio2 9.8333 +Epoch [11/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0040, Loss2: 0.0046, Pure Ratio1: 9.8784, Pure Ratio2 9.8784 +Epoch [11/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0040, Loss2: 0.0055, Pure Ratio1: 9.8889, Pure Ratio2 9.9118 +Epoch [11/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0039, Loss2: 0.0036, Pure Ratio1: 9.8796, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 79.3169 % Model2 78.9563 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0028, Loss2: 0.0036, Pure Ratio1: 9.1765, Pure Ratio2 9.1569 +Epoch [12/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0024, Pure Ratio1: 9.8922, Pure Ratio2 9.8824 +Epoch [12/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0020, Loss2: 0.0015, Pure Ratio1: 10.0458, Pure Ratio2 10.0392 +Epoch [12/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0019, Loss2: 0.0021, Pure Ratio1: 10.0490, Pure Ratio2 10.0196 +Epoch [12/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0026, Loss2: 0.0028, Pure Ratio1: 9.9725, Pure Ratio2 9.9804 +Epoch [12/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0036, Loss2: 0.0033, Pure Ratio1: 9.8333, Pure Ratio2 9.8333 +Epoch [12/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.0023, Loss2: 0.0027, Pure Ratio1: 9.8992, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 78.3654 % Model2 80.4087 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0024, Loss2: 0.0025, Pure Ratio1: 10.0196, Pure Ratio2 9.9216 +Epoch [13/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0021, Loss2: 0.0022, Pure Ratio1: 10.3529, Pure Ratio2 10.3137 +Epoch [13/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.7843, Pure Ratio2 9.7712 +Epoch [13/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 9.7500, Pure Ratio2 9.7304 +Epoch [13/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0032, Loss2: 0.0031, Pure Ratio1: 9.8157, Pure Ratio2 9.8000 +Epoch [13/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0034, Loss2: 0.0034, Pure Ratio1: 9.8889, Pure Ratio2 9.8922 +Epoch [13/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0033, Loss2: 0.0028, Pure Ratio1: 9.9188, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 81.1198 % Model2 80.2885 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0015, Loss2: 0.0016, Pure Ratio1: 10.2745, Pure Ratio2 10.4314 +Epoch [14/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0026, Loss2: 0.0024, Pure Ratio1: 9.8431, Pure Ratio2 10.0196 +Epoch [14/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0040, Loss2: 0.0041, Pure Ratio1: 9.8105, Pure Ratio2 9.8889 +Epoch [14/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.0024, Loss2: 0.0022, Pure Ratio1: 9.8873, Pure Ratio2 9.9314 +Epoch [14/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0029, Loss2: 0.0025, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Epoch [14/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0021, Loss2: 0.0025, Pure Ratio1: 9.7778, Pure Ratio2 9.7778 +Epoch [14/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0017, Loss2: 0.0017, Pure Ratio1: 9.9468, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 81.0397 % Model2 81.6106 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0022, Loss2: 0.0023, Pure Ratio1: 9.0000, Pure Ratio2 9.2941 +Epoch [15/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0010, Pure Ratio1: 9.5588, Pure Ratio2 9.7647 +Epoch [15/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.0017, Loss2: 0.0019, Pure Ratio1: 9.6471, Pure Ratio2 9.7255 +Epoch [15/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0026, Loss2: 0.0026, Pure Ratio1: 9.9608, Pure Ratio2 9.9657 +Epoch [15/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0024, Pure Ratio1: 9.9686, Pure Ratio2 10.0000 +Epoch [15/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0019, Loss2: 0.0027, Pure Ratio1: 9.9641, Pure Ratio2 10.0000 +Epoch [15/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 10.0252, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 81.8710 % Model2 81.9111 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0025, Loss2: 0.0021, Pure Ratio1: 10.2941, Pure Ratio2 10.1373 +Epoch [16/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0031, Loss2: 0.0026, Pure Ratio1: 9.9216, Pure Ratio2 9.8137 +Epoch [16/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0025, Pure Ratio1: 10.0523, Pure Ratio2 10.0000 +Epoch [16/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 75.0000, Loss1: 0.0009, Loss2: 0.0022, Pure Ratio1: 10.0343, Pure Ratio2 9.9853 +Epoch [16/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0016, Loss2: 0.0021, Pure Ratio1: 10.0667, Pure Ratio2 10.0157 +Epoch [16/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0033, Loss2: 0.0027, Pure Ratio1: 10.0392, Pure Ratio2 9.9706 +Epoch [16/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0019, Loss2: 0.0012, Pure Ratio1: 9.9692, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 81.0096 % Model2 80.9395 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0022, Loss2: 0.0027, Pure Ratio1: 10.0784, Pure Ratio2 10.0588 +Epoch [17/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0023, Loss2: 0.0027, Pure Ratio1: 9.9020, Pure Ratio2 9.8922 +Epoch [17/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.9935, Pure Ratio2 10.0784 +Epoch [17/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.9608, Pure Ratio2 10.0637 +Epoch [17/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0016, Loss2: 0.0011, Pure Ratio1: 9.9490, Pure Ratio2 10.0392 +Epoch [17/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 72.6562, Loss1: 0.0018, Loss2: 0.0022, Pure Ratio1: 9.8758, Pure Ratio2 9.9869 +Epoch [17/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0027, Loss2: 0.0021, Pure Ratio1: 9.8291, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 81.1098 % Model2 81.0697 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [18/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0030, Loss2: 0.0030, Pure Ratio1: 9.8922, Pure Ratio2 9.9608 +Epoch [18/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0012, Loss2: 0.0018, Pure Ratio1: 9.8693, Pure Ratio2 9.8758 +Epoch [18/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0030, Loss2: 0.0024, Pure Ratio1: 9.8824, Pure Ratio2 9.8725 +Epoch [18/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0018, Loss2: 0.0014, Pure Ratio1: 9.9059, Pure Ratio2 9.8745 +Epoch [18/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0015, Loss2: 0.0015, Pure Ratio1: 9.9902, Pure Ratio2 9.9706 +Epoch [18/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0043, Loss2: 0.0041, Pure Ratio1: 9.9944, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 80.8694 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0017, Loss2: 0.0015, Pure Ratio1: 9.5294, Pure Ratio2 9.4706 +Epoch [19/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 78.1250, Loss1: 0.0019, Loss2: 0.0018, Pure Ratio1: 9.5000, Pure Ratio2 9.4216 +Epoch [19/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0022, Loss2: 0.0021, Pure Ratio1: 9.5948, Pure Ratio2 9.4641 +Epoch [19/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.6667, Pure Ratio2 9.5539 +Epoch [19/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0010, Loss2: 0.0015, Pure Ratio1: 9.7098, Pure Ratio2 9.6196 +Epoch [19/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.7451, Pure Ratio2 9.6863 +Epoch [19/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0018, Loss2: 0.0010, Pure Ratio1: 9.9048, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 81.6106 % Model2 81.1699 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.8919 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.0980, Pure Ratio2 10.2157 +Epoch [20/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.8333, Pure Ratio2 9.8725 +Epoch [20/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0012, Pure Ratio1: 9.9935, Pure Ratio2 9.9935 +Epoch [20/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.9755, Pure Ratio2 10.0147 +Epoch [20/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0025, Loss2: 0.0024, Pure Ratio1: 9.9294, Pure Ratio2 9.9412 +Epoch [20/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 10.0098, Pure Ratio2 10.0425 +Epoch [20/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.9580, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 80.8894 % Model2 80.8494 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [21/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6078, Pure Ratio2 9.5882 +Epoch [21/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0029, Loss2: 0.0016, Pure Ratio1: 9.8693, Pure Ratio2 9.8170 +Epoch [21/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 10.0049, Pure Ratio2 9.9608 +Epoch [21/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 10.0431, Pure Ratio2 9.9686 +Epoch [21/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0018, Pure Ratio1: 9.9346, Pure Ratio2 9.8824 +Epoch [21/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 80.3686 % Model2 80.3686 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 9.6078, Pure Ratio2 9.6863 +Epoch [22/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.8333, Pure Ratio2 9.8725 +Epoch [22/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.7320, Pure Ratio2 9.8235 +Epoch [22/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.8382, Pure Ratio2 9.8627 +Epoch [22/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0011, Pure Ratio1: 9.9804, Pure Ratio2 9.9765 +Epoch [22/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0013, Loss2: 0.0009, Pure Ratio1: 9.9967, Pure Ratio2 9.9608 +Epoch [22/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0014, Pure Ratio1: 10.0112, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 81.4002 % Model2 82.3417 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.9216, Pure Ratio2 10.0196 +Epoch [23/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.9020, Pure Ratio2 9.9412 +Epoch [23/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0021, Pure Ratio1: 10.0654, Pure Ratio2 10.1176 +Epoch [23/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0016, Pure Ratio1: 9.9951, Pure Ratio2 10.0441 +Epoch [23/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0014, Loss2: 0.0015, Pure Ratio1: 9.9608, Pure Ratio2 10.0314 +Epoch [23/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.0458, Pure Ratio2 10.0654 +Epoch [23/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.0084, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 81.5805 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 10.0784 +Epoch [24/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.0012, Loss2: 0.0017, Pure Ratio1: 9.8333, Pure Ratio2 9.8922 +Epoch [24/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.9281, Pure Ratio2 9.9804 +Epoch [24/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0015, Loss2: 0.0026, Pure Ratio1: 10.0196, Pure Ratio2 10.0784 +Epoch [24/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 9.9569, Pure Ratio2 10.0667 +Epoch [24/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0022, Pure Ratio1: 9.8987, Pure Ratio2 9.9771 +Epoch [24/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.9132, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 80.7993 % Model2 79.5773 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.4118, Pure Ratio2 10.4902 +Epoch [25/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.1275, Pure Ratio2 10.1961 +Epoch [25/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0020, Loss2: 0.0023, Pure Ratio1: 10.0719, Pure Ratio2 10.1373 +Epoch [25/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.9804 +Epoch [25/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0019, Pure Ratio1: 9.9569, Pure Ratio2 10.0157 +Epoch [25/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8987, Pure Ratio2 9.9477 +Epoch [25/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.9160, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 80.0781 % Model2 80.2885 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.7255, Pure Ratio2 10.7255 +Epoch [26/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.6471, Pure Ratio2 10.7059 +Epoch [26/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.4706, Pure Ratio2 10.5229 +Epoch [26/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.3137, Pure Ratio2 10.3382 +Epoch [26/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 10.1412, Pure Ratio2 10.1216 +Epoch [26/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 10.1242, Pure Ratio2 10.1209 +Epoch [26/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0012, Loss2: 0.0009, Pure Ratio1: 10.1036, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 81.1599 % Model2 79.7175 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0012, Pure Ratio1: 10.3725, Pure Ratio2 10.3137 +Epoch [27/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 10.1667, Pure Ratio2 10.1961 +Epoch [27/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0012, Loss2: 0.0004, Pure Ratio1: 10.1242, Pure Ratio2 10.1111 +Epoch [27/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9853, Pure Ratio2 10.0049 +Epoch [27/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9922, Pure Ratio2 10.0314 +Epoch [27/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.0000, Pure Ratio2 10.0229 +Epoch [27/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 10.0476, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 79.6575 % Model2 79.4972 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.3137, Pure Ratio2 9.3137 +Epoch [28/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0015, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.5490 +Epoch [28/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.8170, Pure Ratio2 9.7255 +Epoch [28/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.8922, Pure Ratio2 9.8235 +Epoch [28/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9176, Pure Ratio2 9.8745 +Epoch [28/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9346, Pure Ratio2 9.8791 +Epoch [28/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0017, Pure Ratio1: 9.9496, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 81.1699 % Model2 81.0296 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.5686, Pure Ratio2 9.6078 +Epoch [29/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0018, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [29/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.8889, Pure Ratio2 9.9150 +Epoch [29/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0022, Loss2: 0.0031, Pure Ratio1: 9.8382, Pure Ratio2 9.8627 +Epoch [29/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.9137, Pure Ratio2 9.9529 +Epoch [29/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9281, Pure Ratio2 9.9379 +Epoch [29/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.0013, Loss2: 0.0012, Pure Ratio1: 9.8992, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 79.8778 % Model2 81.7107 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.5490, Pure Ratio2 9.6667 +Epoch [30/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [30/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8758, Pure Ratio2 9.8693 +Epoch [30/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.8284, Pure Ratio2 9.8039 +Epoch [30/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7373, Pure Ratio2 9.7333 +Epoch [30/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0006, Loss2: 0.0024, Pure Ratio1: 9.8399, Pure Ratio2 9.8268 +Epoch [30/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9496, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 79.6575 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.1569 +Epoch [31/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9706, Pure Ratio2 9.9216 +Epoch [31/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.0915, Pure Ratio2 10.0196 +Epoch [31/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0017, Pure Ratio1: 10.3578, Pure Ratio2 10.2451 +Epoch [31/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.2824, Pure Ratio2 10.2078 +Epoch [31/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 76.5625, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 10.1144, Pure Ratio2 10.0784 +Epoch [31/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.0017, Loss2: 0.0012, Pure Ratio1: 10.0168, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 78.6558 % Model2 79.8377 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.7451, Pure Ratio2 10.7059 +Epoch [32/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.1863, Pure Ratio2 10.1275 +Epoch [32/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0850, Pure Ratio2 10.0327 +Epoch [32/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.2451, Pure Ratio2 10.2108 +Epoch [32/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.1882, Pure Ratio2 10.1529 +Epoch [32/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 10.0980, Pure Ratio2 10.0850 +Epoch [32/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 10.0588, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 80.0581 % Model2 79.2668 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.8039 +Epoch [33/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 9.7353, Pure Ratio2 9.7843 +Epoch [33/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.8431 +Epoch [33/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0006, Pure Ratio1: 9.9314, Pure Ratio2 9.9314 +Epoch [33/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9569, Pure Ratio2 9.9373 +Epoch [33/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 9.9020, Pure Ratio2 9.8497 +Epoch [33/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8515, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 79.7676 % Model2 79.6274 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0196, Pure Ratio2 9.9804 +Epoch [34/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0000, Pure Ratio2 9.9706 +Epoch [34/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 10.0654, Pure Ratio2 10.0458 +Epoch [34/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0931, Pure Ratio2 10.0784 +Epoch [34/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0039, Pure Ratio2 9.9765 +Epoch [34/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9150, Pure Ratio2 9.9248 +Epoch [34/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0008, Pure Ratio1: 9.9384, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 81.0697 % Model2 80.9295 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.4510, Pure Ratio2 9.4314 +Epoch [35/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9706 +Epoch [35/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9673, Pure Ratio2 9.8039 +Epoch [35/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.7843 +Epoch [35/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9059, Pure Ratio2 9.8039 +Epoch [35/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0012, Pure Ratio1: 9.9869, Pure Ratio2 9.8987 +Epoch [35/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0700, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 79.8978 % Model2 81.1599 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.2549, Pure Ratio2 9.1373 +Epoch [36/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.5294, Pure Ratio2 9.4608 +Epoch [36/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8105, Pure Ratio2 9.8170 +Epoch [36/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8971, Pure Ratio2 9.8971 +Epoch [36/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0157, Pure Ratio2 10.0588 +Epoch [36/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0523, Pure Ratio2 10.1242 +Epoch [36/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9664, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 79.2268 % Model2 80.4788 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.6863, Pure Ratio2 9.6667 +Epoch [37/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.7157, Pure Ratio2 9.7255 +Epoch [37/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8693, Pure Ratio2 9.9020 +Epoch [37/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.0588, Pure Ratio2 10.1127 +Epoch [37/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9137, Pure Ratio2 9.9529 +Epoch [37/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9869, Pure Ratio2 10.0556 +Epoch [37/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8964, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 80.4087 % Model2 81.0296 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7059, Pure Ratio2 9.7255 +Epoch [38/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.1667, Pure Ratio2 10.1078 +Epoch [38/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9608, Pure Ratio2 9.8824 +Epoch [38/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8382, Pure Ratio2 9.7794 +Epoch [38/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9529, Pure Ratio2 9.9176 +Epoch [38/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8987, Pure Ratio2 9.8562 +Epoch [38/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9244, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 81.4503 % Model2 81.1198 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.2941 +Epoch [39/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.0980 +Epoch [39/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0065, Pure Ratio2 9.9804 +Epoch [39/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.0003, Loss2: 0.0014, Pure Ratio1: 10.0735, Pure Ratio2 10.0245 +Epoch [39/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0275, Pure Ratio2 9.9843 +Epoch [39/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0131, Pure Ratio2 9.9575 +Epoch [39/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 10.0224, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 79.3770 % Model2 78.5256 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.3922, Pure Ratio2 10.4706 +Epoch [40/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.2059, Pure Ratio2 10.2255 +Epoch [40/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.2353, Pure Ratio2 10.2745 +Epoch [40/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.0490, Pure Ratio2 10.0686 +Epoch [40/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0078, Pure Ratio2 9.9804 +Epoch [40/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9967, Pure Ratio2 9.9771 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0014, Pure Ratio1: 10.0504, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 79.7276 % Model2 79.8778 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.1961 +Epoch [41/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.1863 +Epoch [41/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0719, Pure Ratio2 10.1438 +Epoch [41/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9608, Pure Ratio2 10.0098 +Epoch [41/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0017, Pure Ratio1: 9.9765, Pure Ratio2 10.0275 +Epoch [41/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.9542 +Epoch [41/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9384, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 79.3970 % Model2 80.1082 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.1961, Pure Ratio2 10.4706 +Epoch [42/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.0686, Pure Ratio2 10.1961 +Epoch [42/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1046, Pure Ratio2 10.1961 +Epoch [42/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.1373 +Epoch [42/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 10.0471, Pure Ratio2 10.1412 +Epoch [42/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0523, Pure Ratio2 10.1503 +Epoch [42/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9664, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 79.3470 % Model2 78.9563 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1961, Pure Ratio2 9.8627 +Epoch [43/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0014, Pure Ratio1: 10.0000, Pure Ratio2 9.8039 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [43/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.8186, Pure Ratio2 9.7696 +Epoch [43/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9725, Pure Ratio2 9.9098 +Epoch [43/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 9.8725, Pure Ratio2 9.8203 +Epoch [43/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 79.7676 % Model2 80.0581 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.0980, Pure Ratio2 9.2353 +Epoch [44/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6275 +Epoch [44/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7516, Pure Ratio2 9.9020 +Epoch [44/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8676, Pure Ratio2 9.9559 +Epoch [44/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8745, Pure Ratio2 9.9294 +Epoch [44/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.9477, Pure Ratio2 10.0033 +Epoch [44/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9860, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 80.2985 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.3333, Pure Ratio2 10.4706 +Epoch [45/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.1471, Pure Ratio2 10.3137 +Epoch [45/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.1961, Pure Ratio2 10.2745 +Epoch [45/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 10.0784 +Epoch [45/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1098, Pure Ratio2 10.1647 +Epoch [45/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 10.0784 +Epoch [45/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9580, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 79.9980 % Model2 80.0280 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.0196 +Epoch [46/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.1176 +Epoch [46/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1634, Pure Ratio2 10.1307 +Epoch [46/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9510, Pure Ratio2 9.9363 +Epoch [46/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.0275 +Epoch [46/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9706, Pure Ratio2 10.0000 +Epoch [46/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.9636, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 79.6775 % Model2 80.3185 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [47/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7157, Pure Ratio2 9.7941 +Epoch [47/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.8889 +Epoch [47/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.8186, Pure Ratio2 9.8676 +Epoch [47/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.9176 +Epoch [47/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9771, Pure Ratio2 9.9444 +Epoch [47/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9048, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 80.4487 % Model2 78.4455 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6667, Pure Ratio2 9.7255 +Epoch [48/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [48/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8758, Pure Ratio2 9.8627 +Epoch [48/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0012, Loss2: 0.0002, Pure Ratio1: 9.8775, Pure Ratio2 9.9363 +Epoch [48/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9098, Pure Ratio2 10.0000 +Epoch [48/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9314, Pure Ratio2 10.0065 +Epoch [48/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 80.7893 % Model2 80.5489 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.1765 +Epoch [49/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.4510 +Epoch [49/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.2680, Pure Ratio2 10.2549 +Epoch [49/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.1225 +Epoch [49/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.3098, Pure Ratio2 10.2471 +Epoch [49/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0784, Pure Ratio2 10.0359 +Epoch [49/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0644, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 80.8393 % Model2 80.6390 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.8039 +Epoch [50/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1765 +Epoch [50/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0850, Pure Ratio2 10.1176 +Epoch [50/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0686, Pure Ratio2 10.1127 +Epoch [50/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9059, Pure Ratio2 9.9843 +Epoch [50/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9379, Pure Ratio2 9.9771 +Epoch [50/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9692, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 79.9079 % Model2 79.1567 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 10.4118, Pure Ratio2 10.5686 +Epoch [51/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.3137, Pure Ratio2 10.3137 +Epoch [51/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 10.0196, Pure Ratio2 10.0392 +Epoch [51/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.9167, Pure Ratio2 9.9020 +Epoch [51/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0471, Pure Ratio2 10.0000 +Epoch [51/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0425, Pure Ratio2 9.9804 +Epoch [51/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0308, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 79.8377 % Model2 78.7460 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.8824 +Epoch [52/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0294 +Epoch [52/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8954 +Epoch [52/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 9.9755 +Epoch [52/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.8510 +Epoch [52/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9052, Pure Ratio2 9.9020 +Epoch [52/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9580, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 79.5873 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.2745 +Epoch [53/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.2745 +Epoch [53/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [53/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8284, Pure Ratio2 9.9314 +Epoch [53/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8157, Pure Ratio2 9.8588 +Epoch [53/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8758, Pure Ratio2 9.9608 +Epoch [53/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 80.6090 % Model2 79.9379 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [54/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0000, Pure Ratio2 10.0098 +Epoch [54/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 10.1503 +Epoch [54/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9853, Pure Ratio2 9.9902 +Epoch [54/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0353, Pure Ratio2 10.0706 +Epoch [54/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9869, Pure Ratio2 10.0000 +Epoch [54/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9860, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 80.0681 % Model2 79.5072 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.1961, Pure Ratio2 10.3137 +Epoch [55/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.2255, Pure Ratio2 10.1961 +Epoch [55/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0850, Pure Ratio2 10.0654 +Epoch [55/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0882, Pure Ratio2 10.1127 +Epoch [55/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2235, Pure Ratio2 10.2431 +Epoch [55/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0327, Pure Ratio2 10.0163 +Epoch [55/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.0672, Pure Ratio2 10.0504 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 79.9379 % Model2 78.2252 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7843, Pure Ratio2 9.8039 +Epoch [56/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.7941 +Epoch [56/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7712, Pure Ratio2 9.7516 +Epoch [56/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7353 +Epoch [56/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.8667 +Epoch [56/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0131, Pure Ratio2 9.9771 +Epoch [56/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0700, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.5589 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [57/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9510, Pure Ratio2 9.9118 +Epoch [57/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9935, Pure Ratio2 9.9150 +Epoch [57/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8873, Pure Ratio2 9.7745 +Epoch [57/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9922, Pure Ratio2 9.9098 +Epoch [57/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 9.8758 +Epoch [57/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0224, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 80.4988 % Model2 79.7376 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.7647 +Epoch [58/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9216, Pure Ratio2 10.0784 +Epoch [58/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8105, Pure Ratio2 9.8824 +Epoch [58/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8873 +Epoch [58/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9765, Pure Ratio2 10.0000 +Epoch [58/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9477, Pure Ratio2 9.9281 +Epoch [58/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9580, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 78.7059 % Model2 79.8778 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6275, Pure Ratio2 9.7255 +Epoch [59/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2451 +Epoch [59/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 10.1830, Pure Ratio2 10.2941 +Epoch [59/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9853, Pure Ratio2 10.1275 +Epoch [59/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.9490, Pure Ratio2 10.0706 +Epoch [59/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9542, Pure Ratio2 10.0523 +Epoch [59/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9244, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 79.0966 % Model2 80.8193 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.5098, Pure Ratio2 9.3137 +Epoch [60/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.7941, Pure Ratio2 9.8039 +Epoch [60/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9150, Pure Ratio2 10.0523 +Epoch [60/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9363, Pure Ratio2 10.0833 +Epoch [60/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9059, Pure Ratio2 10.0431 +Epoch [60/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9902, Pure Ratio2 10.1438 +Epoch [60/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 10.0028, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 79.3570 % Model2 79.9479 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.4118, Pure Ratio2 10.3725 +Epoch [61/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.4510, Pure Ratio2 10.4314 +Epoch [61/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.4314, Pure Ratio2 10.3203 +Epoch [61/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3137, Pure Ratio2 10.2598 +Epoch [61/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0013, Pure Ratio1: 10.1686, Pure Ratio2 10.0824 +Epoch [61/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0915, Pure Ratio2 10.0229 +Epoch [61/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1485, Pure Ratio2 10.0840 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 80.1382 % Model2 79.4571 %, Pure Ratio 1 10.1156 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.6667 +Epoch [62/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7353 +Epoch [62/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8366, Pure Ratio2 9.7974 +Epoch [62/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9118, Pure Ratio2 9.8676 +Epoch [62/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 10.0431 +Epoch [62/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0392 +Epoch [62/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0007, Pure Ratio1: 9.9692, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 79.4471 % Model2 79.2969 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.7255, Pure Ratio2 10.6078 +Epoch [63/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2157 +Epoch [63/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0458, Pure Ratio2 10.0261 +Epoch [63/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1373, Pure Ratio2 10.1225 +Epoch [63/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.2431, Pure Ratio2 10.1725 +Epoch [63/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1830, Pure Ratio2 10.0621 +Epoch [63/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1821, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 78.7059 % Model2 79.6474 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.1373 +Epoch [64/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8725 +Epoch [64/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.8105 +Epoch [64/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8676, Pure Ratio2 9.8480 +Epoch [64/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9686, Pure Ratio2 9.9608 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9869, Pure Ratio2 9.9739 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0112, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 80.5789 % Model2 78.3053 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.6078 +Epoch [65/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7549, Pure Ratio2 9.7647 +Epoch [65/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8170, Pure Ratio2 9.7582 +Epoch [65/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8627, Pure Ratio2 9.7941 +Epoch [65/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.8980 +Epoch [65/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.9118, Pure Ratio2 9.8987 +Epoch [65/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.9468, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 79.7576 % Model2 79.9279 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.8627, Pure Ratio2 10.7647 +Epoch [66/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.3137, Pure Ratio2 10.3137 +Epoch [66/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3399, Pure Ratio2 10.2745 +Epoch [66/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9657, Pure Ratio2 9.9216 +Epoch [66/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1059, Pure Ratio2 10.0510 +Epoch [66/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 9.9706 +Epoch [66/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 79.6875 % Model2 80.3886 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.6863 +Epoch [67/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 9.9118 +Epoch [67/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9085 +Epoch [67/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.1520, Pure Ratio2 10.1373 +Epoch [67/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0510, Pure Ratio2 10.0118 +Epoch [67/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0229 +Epoch [67/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0504, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 79.7576 % Model2 80.3886 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.2941, Pure Ratio2 9.2549 +Epoch [68/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.1176 +Epoch [68/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 10.1307 +Epoch [68/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [68/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0078, Pure Ratio2 10.0627 +Epoch [68/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0359, Pure Ratio2 10.1013 +Epoch [68/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0308, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 80.1382 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.5098 +Epoch [69/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0980 +Epoch [69/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0458, Pure Ratio2 10.0719 +Epoch [69/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0441, Pure Ratio2 10.0637 +Epoch [69/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0627 +Epoch [69/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.9641, Pure Ratio2 9.9869 +Epoch [69/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9132, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 79.9179 % Model2 80.9195 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.0784 +Epoch [70/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 9.9902 +Epoch [70/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9085 +Epoch [70/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.8627 +Epoch [70/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9961, Pure Ratio2 9.9569 +Epoch [70/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0621, Pure Ratio2 10.0000 +Epoch [70/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0728, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 80.4287 % Model2 79.3570 %, Pure Ratio 1 10.0578 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 9.9020 +Epoch [71/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 10.3235, Pure Ratio2 10.2157 +Epoch [71/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8170 +Epoch [71/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8137 +Epoch [71/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.9333, Pure Ratio2 9.8353 +Epoch [71/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.8203 +Epoch [71/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9244, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 80.3085 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.5686, Pure Ratio2 10.7843 +Epoch [72/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.4706, Pure Ratio2 10.6471 +Epoch [72/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2614, Pure Ratio2 10.4248 +Epoch [72/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1814 +Epoch [72/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1412 +Epoch [72/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0882 +Epoch [72/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8936, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 77.6843 % Model2 80.3486 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0980, Pure Ratio2 10.1765 +Epoch [73/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1275 +Epoch [73/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1895 +Epoch [73/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.1225 +Epoch [73/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0471, Pure Ratio2 10.0510 +Epoch [73/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0098 +Epoch [73/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0168, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 79.8478 % Model2 79.0465 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5882, Pure Ratio2 10.2549 +Epoch [74/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2647, Pure Ratio2 9.9804 +Epoch [74/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 9.9608 +Epoch [74/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1471, Pure Ratio2 10.0196 +Epoch [74/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1020, Pure Ratio2 9.9922 +Epoch [74/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0817, Pure Ratio2 9.9542 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0728, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 79.3570 % Model2 79.2067 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4314, Pure Ratio2 10.3529 +Epoch [75/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1863, Pure Ratio2 10.1569 +Epoch [75/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1242, Pure Ratio2 10.0392 +Epoch [75/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 9.9853 +Epoch [75/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1490, Pure Ratio2 10.0627 +Epoch [75/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.0294 +Epoch [75/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0168, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 78.9663 % Model2 79.8377 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.1569 +Epoch [76/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2353, Pure Ratio2 10.0980 +Epoch [76/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1438, Pure Ratio2 10.0392 +Epoch [76/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 10.0637, Pure Ratio2 9.9951 +Epoch [76/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0510, Pure Ratio2 9.9608 +Epoch [76/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.8301 +Epoch [76/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 79.1066 % Model2 80.1382 %, Pure Ratio 1 10.0628 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.6275 +Epoch [77/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.7353 +Epoch [77/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7190 +Epoch [77/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.7745 +Epoch [77/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1529, Pure Ratio2 10.0000 +Epoch [77/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1209, Pure Ratio2 9.9314 +Epoch [77/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 90.6250, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.0756, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 80.1583 % Model2 79.6875 %, Pure Ratio 1 10.0352 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5098, Pure Ratio2 9.8627 +Epoch [78/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0020, Loss2: 0.0005, Pure Ratio1: 9.5294, Pure Ratio2 9.6765 +Epoch [78/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.8627 +Epoch [78/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7696, Pure Ratio2 9.8088 +Epoch [78/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8353, Pure Ratio2 9.8118 +Epoch [78/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9869, Pure Ratio2 9.9641 +Epoch [78/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 79.4271 % Model2 80.5889 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.3922 +Epoch [79/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0784 +Epoch [79/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.0588 +Epoch [79/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0147 +Epoch [79/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9137, Pure Ratio2 10.0078 +Epoch [79/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0425 +Epoch [79/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9580, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 79.4671 % Model2 79.1466 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.1006 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3725 +Epoch [80/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1863 +Epoch [80/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9739, Pure Ratio2 10.0196 +Epoch [80/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [80/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0353, Pure Ratio2 10.1843 +Epoch [80/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 10.1895 +Epoch [80/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0056, Pure Ratio2 10.1821 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 81.0998 % Model2 79.9980 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.1282 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [81/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [81/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 9.9673 +Epoch [81/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9265 +Epoch [81/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0863, Pure Ratio2 10.0980 +Epoch [81/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0163 +Epoch [81/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9916, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 79.9379 % Model2 79.4772 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1569 +Epoch [82/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.8529 +Epoch [82/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7908, Pure Ratio2 9.8105 +Epoch [82/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8775, Pure Ratio2 9.8578 +Epoch [82/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9333 +Epoch [82/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0490, Pure Ratio2 10.0131 +Epoch [82/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0700, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 78.0649 % Model2 79.1266 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0028, Loss2: 0.0036, Pure Ratio1: 10.2549, Pure Ratio2 10.4118 +Epoch [83/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [83/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 9.9281 +Epoch [83/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.8676 +Epoch [83/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9725 +Epoch [83/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0784 +Epoch [83/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0700, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 79.8578 % Model2 79.8377 %, Pure Ratio 1 10.0251 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.7647 +Epoch [84/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [84/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.8693 +Epoch [84/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [84/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9569 +Epoch [84/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8660, Pure Ratio2 9.8889 +Epoch [84/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9160, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 80.1783 % Model2 78.7861 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9608 +Epoch [85/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [85/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8105, Pure Ratio2 9.7647 +Epoch [85/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8529, Pure Ratio2 9.8137 +Epoch [85/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9059, Pure Ratio2 9.8431 +Epoch [85/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8660, Pure Ratio2 9.8203 +Epoch [85/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9104, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 78.1651 % Model2 78.0349 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.7059 +Epoch [86/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 9.8922 +Epoch [86/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.8824 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0539, Pure Ratio2 9.9314 +Epoch [86/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0745, Pure Ratio2 9.9490 +Epoch [86/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 9.8562 +Epoch [86/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0140, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 79.0765 % Model2 79.5773 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7451, Pure Ratio2 9.7255 +Epoch [87/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9706, Pure Ratio2 10.0490 +Epoch [87/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9935, Pure Ratio2 9.9085 +Epoch [87/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0147, Pure Ratio2 9.9461 +Epoch [87/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0078, Pure Ratio2 9.9765 +Epoch [87/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 9.9967 +Epoch [87/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0616, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 79.1366 % Model2 79.1366 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1765 +Epoch [88/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0196 +Epoch [88/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 10.2222 +Epoch [88/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2206 +Epoch [88/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0471, Pure Ratio2 10.0784 +Epoch [88/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9673, Pure Ratio2 9.9739 +Epoch [88/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 79.0565 % Model2 80.1783 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0980, Pure Ratio2 10.5294 +Epoch [89/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8137, Pure Ratio2 10.0980 +Epoch [89/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1111, Pure Ratio2 10.2418 +Epoch [89/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1324, Pure Ratio2 10.2451 +Epoch [89/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.2000 +Epoch [89/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0915, Pure Ratio2 10.1144 +Epoch [89/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 79.9279 % Model2 79.7977 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.6667, Pure Ratio2 10.6078 +Epoch [90/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.4412, Pure Ratio2 10.3725 +Epoch [90/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5033, Pure Ratio2 10.4183 +Epoch [90/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3039, Pure Ratio2 10.2647 +Epoch [90/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.3020, Pure Ratio2 10.2549 +Epoch [90/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.1601 +Epoch [90/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1373, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 78.7961 % Model2 80.8193 %, Pure Ratio 1 10.0704 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8627 +Epoch [91/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0014, Pure Ratio1: 10.0294, Pure Ratio2 10.0686 +Epoch [91/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9935, Pure Ratio2 10.1373 +Epoch [91/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0931, Pure Ratio2 10.1667 +Epoch [91/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 10.0706 +Epoch [91/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 10.0588 +Epoch [91/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1008, Pure Ratio2 10.1597 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.0581 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8431 +Epoch [92/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1569 +Epoch [92/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2810, Pure Ratio2 10.1895 +Epoch [92/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1863 +Epoch [92/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2039, Pure Ratio2 10.1176 +Epoch [92/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.0196 +Epoch [92/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 78.8061 % Model2 79.1567 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [93/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.9608 +Epoch [93/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9542 +Epoch [93/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [93/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8784 +Epoch [93/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8595, Pure Ratio2 9.8529 +Epoch [93/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8992, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 79.7576 % Model2 79.5573 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.4118, Pure Ratio2 9.6863 +Epoch [94/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0025, Loss2: 0.0011, Pure Ratio1: 9.6765, Pure Ratio2 9.8235 +Epoch [94/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [94/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8480, Pure Ratio2 9.9265 +Epoch [94/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8706, Pure Ratio2 9.9137 +Epoch [94/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.8987 +Epoch [94/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9748, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 79.7977 % Model2 80.3586 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [95/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1863, Pure Ratio2 10.2059 +Epoch [95/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9935 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9755, Pure Ratio2 10.0196 +Epoch [95/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.9529 +Epoch [95/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1046, Pure Ratio2 10.1176 +Epoch [95/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 80.0381 % Model2 79.1166 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.2353 +Epoch [96/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7157, Pure Ratio2 9.6667 +Epoch [96/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7451 +Epoch [96/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.7353 +Epoch [96/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8784 +Epoch [96/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9314 +Epoch [96/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 79.2969 % Model2 80.0280 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0000 +Epoch [97/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9706 +Epoch [97/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1111, Pure Ratio2 10.0915 +Epoch [97/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.9902 +Epoch [97/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0549 +Epoch [97/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1046 +Epoch [97/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0756, Pure Ratio2 10.1092 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 78.6558 % Model2 79.1867 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0478 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [98/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.0392 +Epoch [98/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.0065 +Epoch [98/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 9.9363 +Epoch [98/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 9.9451 +Epoch [98/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.0033 +Epoch [98/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 79.5473 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7843 +Epoch [99/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 10.0490 +Epoch [99/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9935 +Epoch [99/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9951, Pure Ratio2 10.0000 +Epoch [99/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1294, Pure Ratio2 10.1333 +Epoch [99/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0018, Pure Ratio1: 10.0621, Pure Ratio2 10.0850 +Epoch [99/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 79.6875 % Model2 79.1767 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.6275 +Epoch [100/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9510 +Epoch [100/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8301, Pure Ratio2 9.8431 +Epoch [100/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.7059 +Epoch [100/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.8235 +Epoch [100/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9052, Pure Ratio2 9.9706 +Epoch [100/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9384, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 78.5056 % Model2 78.9363 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [101/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7745 +Epoch [101/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.2026, Pure Ratio2 10.2157 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1275 +Epoch [101/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0392 +Epoch [101/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 9.9575 +Epoch [101/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0728, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 78.8862 % Model2 79.3269 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [102/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5980, Pure Ratio2 9.6569 +Epoch [102/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.8954 +Epoch [102/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9363, Pure Ratio2 9.9657 +Epoch [102/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9294, Pure Ratio2 9.9373 +Epoch [102/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9150 +Epoch [102/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0308, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 80.3986 % Model2 78.6158 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.7059 +Epoch [103/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7647 +Epoch [103/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0028, Loss2: 0.0025, Pure Ratio1: 10.1242, Pure Ratio2 10.0327 +Epoch [103/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.0588 +Epoch [103/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 9.9255 +Epoch [103/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 9.9869 +Epoch [103/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 79.8177 % Model2 79.7276 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.0980 +Epoch [104/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7745 +Epoch [104/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.8889 +Epoch [104/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9706 +Epoch [104/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 10.0392 +Epoch [104/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [104/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 79.8077 % Model2 79.6274 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 10.0588 +Epoch [105/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.2059, Pure Ratio2 10.1667 +Epoch [105/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9216 +Epoch [105/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.7941 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 9.9216 +Epoch [105/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 9.9673 +Epoch [105/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9328, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 80.0681 % Model2 79.0665 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6275 +Epoch [106/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9118 +Epoch [106/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.0065 +Epoch [106/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9706, Pure Ratio2 9.9461 +Epoch [106/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.8784 +Epoch [106/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0065, Pure Ratio2 9.9706 +Epoch [106/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0532, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 78.7861 % Model2 80.2584 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.3922 +Epoch [107/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.3137 +Epoch [107/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.9020 +Epoch [107/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 10.0735 +Epoch [107/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0667, Pure Ratio2 10.1255 +Epoch [107/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0458, Pure Ratio2 10.0784 +Epoch [107/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 78.4054 % Model2 79.4471 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.8039 +Epoch [108/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6863 +Epoch [108/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6536, Pure Ratio2 9.5556 +Epoch [108/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6569 +Epoch [108/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8706 +Epoch [108/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0065 +Epoch [108/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9944, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 79.5072 % Model2 79.6575 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.9412, Pure Ratio2 11.1373 +Epoch [109/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4118, Pure Ratio2 10.6373 +Epoch [109/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.4118 +Epoch [109/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2206, Pure Ratio2 10.3284 +Epoch [109/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1294, Pure Ratio2 10.2196 +Epoch [109/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0948, Pure Ratio2 10.1601 +Epoch [109/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0896, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 79.4671 % Model2 79.6775 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.4314 +Epoch [110/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2941 +Epoch [110/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.1699 +Epoch [110/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.2010 +Epoch [110/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1216 +Epoch [110/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.0490 +Epoch [110/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 79.5974 % Model2 79.4271 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [111/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9902 +Epoch [111/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7908 +Epoch [111/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9167, Pure Ratio2 9.9412 +Epoch [111/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 9.9765 +Epoch [111/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9771, Pure Ratio2 9.9052 +Epoch [111/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0280, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 79.9880 % Model2 80.0381 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8824 +Epoch [112/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9902 +Epoch [112/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.3399 +Epoch [112/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.3186 +Epoch [112/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1490, Pure Ratio2 10.2196 +Epoch [112/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.1242 +Epoch [112/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 10.0644 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 79.3870 % Model2 80.1482 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1373, Pure Ratio2 10.2157 +Epoch [113/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0588 +Epoch [113/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9281 +Epoch [113/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 10.1520, Pure Ratio2 10.1275 +Epoch [113/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0667, Pure Ratio2 10.0549 +Epoch [113/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0098 +Epoch [113/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1120, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 79.9279 % Model2 79.8978 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1569 +Epoch [114/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.0686 +Epoch [114/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.3595 +Epoch [114/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0637, Pure Ratio2 10.2304 +Epoch [114/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0941, Pure Ratio2 10.2667 +Epoch [114/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1797 +Epoch [114/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 79.9579 % Model2 79.4671 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.2549 +Epoch [115/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.2647 +Epoch [115/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3856, Pure Ratio2 10.4706 +Epoch [115/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.2745 +Epoch [115/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1922, Pure Ratio2 10.1922 +Epoch [115/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.1340 +Epoch [115/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0644, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 80.5889 % Model2 79.1466 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0804 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0392 +Epoch [116/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0784 +Epoch [116/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.8235 +Epoch [116/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8382, Pure Ratio2 9.9363 +Epoch [116/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0706 +Epoch [116/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.1405 +Epoch [116/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 79.1166 % Model2 79.8778 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1176, Pure Ratio2 10.0784 +Epoch [117/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1176 +Epoch [117/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.1895 +Epoch [117/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 10.0588, Pure Ratio2 10.0686 +Epoch [117/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9490, Pure Ratio2 9.9294 +Epoch [117/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [117/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 79.9679 % Model2 79.4271 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.0980 +Epoch [118/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1373 +Epoch [118/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.0850 +Epoch [118/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1029, Pure Ratio2 9.9265 +Epoch [118/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1294, Pure Ratio2 10.0510 +Epoch [118/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1601, Pure Ratio2 10.1078 +Epoch [118/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0924, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 79.8978 % Model2 79.4071 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0009, Pure Ratio1: 10.3137, Pure Ratio2 10.1373 +Epoch [119/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.3627 +Epoch [119/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1765 +Epoch [119/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1618 +Epoch [119/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0078, Pure Ratio2 10.0314 +Epoch [119/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 10.0196 +Epoch [119/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 79.6775 % Model2 80.2484 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.0000 +Epoch [120/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.2353 +Epoch [120/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.3856, Pure Ratio2 10.2745 +Epoch [120/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2206, Pure Ratio2 10.0882 +Epoch [120/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1137, Pure Ratio2 9.9490 +Epoch [120/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9085 +Epoch [120/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0812, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 79.1567 % Model2 79.6775 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0392 +Epoch [121/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9118 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8889 +Epoch [121/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.0980 +Epoch [121/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9333 +Epoch [121/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0065 +Epoch [121/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0224, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 80.0280 % Model2 79.5072 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.0588 +Epoch [122/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [122/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8627 +Epoch [122/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7794 +Epoch [122/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9529, Pure Ratio2 9.8745 +Epoch [122/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0261, Pure Ratio2 9.9379 +Epoch [122/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 79.3970 % Model2 79.5873 %, Pure Ratio 1 10.1483 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9804 +Epoch [123/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2255 +Epoch [123/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.2288 +Epoch [123/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.1765 +Epoch [123/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0902 +Epoch [123/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8660, Pure Ratio2 10.0229 +Epoch [123/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9076, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 79.4671 % Model2 79.0465 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4118, Pure Ratio2 10.5686 +Epoch [124/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 10.1078 +Epoch [124/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.1111 +Epoch [124/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.1863 +Epoch [124/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1686 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.1503 +Epoch [124/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0532, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 79.7877 % Model2 79.7476 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3725 +Epoch [125/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2451, Pure Ratio2 10.2843 +Epoch [125/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.2484 +Epoch [125/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1814 +Epoch [125/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0510, Pure Ratio2 10.1373 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1013, Pure Ratio2 10.1111 +Epoch [125/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 80.0080 % Model2 79.8878 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 10.0588 +Epoch [126/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8137 +Epoch [126/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.6928 +Epoch [126/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9069 +Epoch [126/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7608, Pure Ratio2 9.7765 +Epoch [126/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8791 +Epoch [126/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 79.9679 % Model2 79.1767 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4902 +Epoch [127/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2255 +Epoch [127/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.3464 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2108, Pure Ratio2 10.2794 +Epoch [127/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9216, Pure Ratio2 10.0314 +Epoch [127/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.1046 +Epoch [127/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 80.6891 % Model2 80.1282 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0392 +Epoch [128/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0686 +Epoch [128/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0784 +Epoch [128/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1961 +Epoch [128/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1333, Pure Ratio2 10.1686 +Epoch [128/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2320 +Epoch [128/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1036 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 79.1767 % Model2 79.5373 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7059 +Epoch [129/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.5588 +Epoch [129/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7843 +Epoch [129/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0245 +Epoch [129/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0667 +Epoch [129/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.0327 +Epoch [129/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0420, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 79.4171 % Model2 79.6174 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2745, Pure Ratio2 10.3137 +Epoch [130/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0980 +Epoch [130/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1242 +Epoch [130/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7549, Pure Ratio2 9.7745 +Epoch [130/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9412 +Epoch [130/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9314 +Epoch [130/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 79.3970 % Model2 78.8061 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.9216 +Epoch [131/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.2451 +Epoch [131/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 10.1765 +Epoch [131/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0245, Pure Ratio2 10.1863 +Epoch [131/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.2000 +Epoch [131/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.1536 +Epoch [131/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 78.8562 % Model2 78.6358 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6667 +Epoch [132/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8039 +Epoch [132/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7908 +Epoch [132/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7402, Pure Ratio2 9.8235 +Epoch [132/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7569, Pure Ratio2 9.8510 +Epoch [132/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9477 +Epoch [132/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8852, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 79.4471 % Model2 80.4287 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.5490 +Epoch [133/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.4216 +Epoch [133/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.0000 +Epoch [133/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0637 +Epoch [133/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1490, Pure Ratio2 10.1373 +Epoch [133/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 9.9869 +Epoch [133/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0840, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 80.4387 % Model2 79.9980 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1373 +Epoch [134/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3039, Pure Ratio2 10.2647 +Epoch [134/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9542 +Epoch [134/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.9216 +Epoch [134/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9569 +Epoch [134/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9183 +Epoch [134/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 79.3770 % Model2 80.1382 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3725 +Epoch [135/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.2941 +Epoch [135/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.2353 +Epoch [135/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9951 +Epoch [135/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9451 +Epoch [135/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9575, Pure Ratio2 10.0065 +Epoch [135/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 80.1683 % Model2 80.3285 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.9020 +Epoch [136/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.2157 +Epoch [136/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.8301 +Epoch [136/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9020 +Epoch [136/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9137 +Epoch [136/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9771 +Epoch [136/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0840, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 79.9479 % Model2 79.1266 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.2353 +Epoch [137/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.1765 +Epoch [137/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2614, Pure Ratio2 10.0980 +Epoch [137/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.0294 +Epoch [137/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0314, Pure Ratio2 10.0039 +Epoch [137/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0294 +Epoch [137/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0560, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 80.4587 % Model2 79.4972 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1961 +Epoch [138/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8039 +Epoch [138/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.9020 +Epoch [138/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0441 +Epoch [138/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.9686 +Epoch [138/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9804 +Epoch [138/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9944, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 79.8778 % Model2 79.9079 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.4118 +Epoch [139/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.9314 +Epoch [139/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8235 +Epoch [139/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9412 +Epoch [139/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0157, Pure Ratio2 10.0706 +Epoch [139/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9771 +Epoch [139/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 9.9748, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 79.1166 % Model2 79.7376 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 9.9804 +Epoch [140/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9412 +Epoch [140/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 10.0000 +Epoch [140/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7549 +Epoch [140/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.8745 +Epoch [140/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0012, Loss2: 0.0000, Pure Ratio1: 10.0621, Pure Ratio2 10.0131 +Epoch [140/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 79.8377 % Model2 79.1767 %, Pure Ratio 1 10.0528 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8235 +Epoch [141/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.9020 +Epoch [141/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9673 +Epoch [141/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.8529 +Epoch [141/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8745 +Epoch [141/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.8922 +Epoch [141/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9244, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 79.4071 % Model2 80.1182 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9608 +Epoch [142/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.6863 +Epoch [142/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.9281 +Epoch [142/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8088, Pure Ratio2 9.8922 +Epoch [142/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9412 +Epoch [142/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.9118 +Epoch [142/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8655, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 79.8277 % Model2 79.9980 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0784 +Epoch [143/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9902 +Epoch [143/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9608 +Epoch [143/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0833 +Epoch [143/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1608 +Epoch [143/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.0882 +Epoch [143/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 79.3069 % Model2 79.4872 %, Pure Ratio 1 10.0779 %, Pure Ratio 2 10.0452 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3922, Pure Ratio2 10.4706 +Epoch [144/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1863, Pure Ratio2 10.1275 +Epoch [144/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0065 +Epoch [144/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0060, Loss2: 0.0072, Pure Ratio1: 10.0882, Pure Ratio2 10.0980 +Epoch [144/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0510 +Epoch [144/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0359 +Epoch [144/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 79.8978 % Model2 80.3285 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.1961 +Epoch [145/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3824, Pure Ratio2 9.2353 +Epoch [145/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.6078 +Epoch [145/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.7990 +Epoch [145/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.8275 +Epoch [145/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8431 +Epoch [145/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 79.3870 % Model2 79.3870 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 10.0196 +Epoch [146/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.2059 +Epoch [146/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0654 +Epoch [146/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 10.0539 +Epoch [146/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0157, Pure Ratio2 10.1294 +Epoch [146/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 10.0654 +Epoch [146/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 79.9679 % Model2 80.4187 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.0980 +Epoch [147/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0196 +Epoch [147/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0523 +Epoch [147/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0118, Pure Ratio2 9.9647 +Epoch [147/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0098 +Epoch [147/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0896, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 79.6875 % Model2 79.5773 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.3922 +Epoch [148/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5000 +Epoch [148/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [148/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7794, Pure Ratio2 9.7451 +Epoch [148/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8784 +Epoch [148/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8268, Pure Ratio2 9.7549 +Epoch [148/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8992, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 80.0280 % Model2 79.5974 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4510, Pure Ratio2 9.5686 +Epoch [149/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8039 +Epoch [149/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.9869 +Epoch [149/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0637 +Epoch [149/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9922 +Epoch [149/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 9.9935 +Epoch [149/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 80.1883 % Model2 80.2985 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1373 +Epoch [150/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.3039 +Epoch [150/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9542 +Epoch [150/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 10.0147 +Epoch [150/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0039 +Epoch [150/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0654 +Epoch [150/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9748, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 79.3069 % Model2 79.3870 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [151/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8725 +Epoch [151/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.2026 +Epoch [151/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0147, Pure Ratio2 10.0588 +Epoch [151/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.9412, Pure Ratio2 9.9882 +Epoch [151/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [151/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 79.8778 % Model2 79.8377 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 10.0603 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.1373 +Epoch [152/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 10.0392 +Epoch [152/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0980 +Epoch [152/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8971 +Epoch [152/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8275 +Epoch [152/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8529 +Epoch [152/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 79.6675 % Model2 79.9679 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9804 +Epoch [153/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1961 +Epoch [153/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.1830 +Epoch [153/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0024, Loss2: 0.0026, Pure Ratio1: 10.0196, Pure Ratio2 10.1324 +Epoch [153/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 10.1333 +Epoch [153/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0163, Pure Ratio2 10.1340 +Epoch [153/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 79.6675 % Model2 80.2484 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.0980 +Epoch [154/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0882 +Epoch [154/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.1242 +Epoch [154/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.8873 +Epoch [154/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.8431 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.8856 +Epoch [154/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 79.7676 % Model2 79.9980 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0784 +Epoch [155/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0588 +Epoch [155/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9542, Pure Ratio2 9.8627 +Epoch [155/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0000 +Epoch [155/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0015, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0392 +Epoch [155/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1242, Pure Ratio2 10.1405 +Epoch [155/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 79.9780 % Model2 80.3185 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9412 +Epoch [156/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7647 +Epoch [156/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.0588 +Epoch [156/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2990, Pure Ratio2 10.2108 +Epoch [156/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1451 +Epoch [156/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.0784 +Epoch [156/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 10.0224 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 80.0982 % Model2 79.3670 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [157/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9510 +Epoch [157/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8366 +Epoch [157/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9167 +Epoch [157/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.9569 +Epoch [157/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9510 +Epoch [157/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0056, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 80.0982 % Model2 79.5974 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7255 +Epoch [158/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0686 +Epoch [158/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9216 +Epoch [158/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.7941 +Epoch [158/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.7804 +Epoch [158/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.8497 +Epoch [158/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8375, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 79.2268 % Model2 79.2869 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0980 +Epoch [159/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9412 +Epoch [159/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8497 +Epoch [159/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9461, Pure Ratio2 9.9510 +Epoch [159/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9843 +Epoch [159/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0229 +Epoch [159/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 79.8978 % Model2 80.6190 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4314, Pure Ratio2 10.6078 +Epoch [160/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3725 +Epoch [160/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3072, Pure Ratio2 10.3922 +Epoch [160/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.3137 +Epoch [160/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2824 +Epoch [160/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 10.1993 +Epoch [160/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 79.4071 % Model2 79.4671 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.0196 +Epoch [161/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2843, Pure Ratio2 10.1667 +Epoch [161/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9739 +Epoch [161/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1127, Pure Ratio2 10.0490 +Epoch [161/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2431, Pure Ratio2 10.2157 +Epoch [161/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0948 +Epoch [161/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0952 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 79.8177 % Model2 79.7877 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.3922 +Epoch [162/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0882 +Epoch [162/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0850 +Epoch [162/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9951 +Epoch [162/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 9.9961 +Epoch [162/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.9281 +Epoch [162/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 80.3486 % Model2 80.7392 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0392 +Epoch [163/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.1471 +Epoch [163/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.1111 +Epoch [163/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0000 +Epoch [163/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 10.0039 +Epoch [163/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0915, Pure Ratio2 10.1373 +Epoch [163/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 80.3285 % Model2 80.1883 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1176 +Epoch [164/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0392 +Epoch [164/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 10.0458 +Epoch [164/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9069 +Epoch [164/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0118 +Epoch [164/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0490 +Epoch [164/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 79.9679 % Model2 80.3586 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 10.0603 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.6667 +Epoch [165/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6275 +Epoch [165/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6667 +Epoch [165/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.8382 +Epoch [165/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.9137 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8987 +Epoch [165/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 79.6374 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0025 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3529 +Epoch [166/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.1471 +Epoch [166/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0196 +Epoch [166/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0931, Pure Ratio2 10.1716 +Epoch [166/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 10.0667 +Epoch [166/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0588 +Epoch [166/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0112, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 79.7175 % Model2 79.6975 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.2941 +Epoch [167/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0098 +Epoch [167/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.7190 +Epoch [167/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8186, Pure Ratio2 9.7255 +Epoch [167/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0039, Pure Ratio2 9.9451 +Epoch [167/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0033, Pure Ratio2 9.9673 +Epoch [167/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9860, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 79.2568 % Model2 79.8778 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6471, Pure Ratio2 10.7843 +Epoch [168/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3235 +Epoch [168/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2484, Pure Ratio2 10.3072 +Epoch [168/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.3529 +Epoch [168/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2627 +Epoch [168/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.3203 +Epoch [168/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.1541 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 79.5072 % Model2 80.3185 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.3333 +Epoch [169/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0882, Pure Ratio2 10.2353 +Epoch [169/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.2353 +Epoch [169/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.1029 +Epoch [169/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 10.0784 +Epoch [169/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 10.0752 +Epoch [169/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 79.4171 % Model2 80.2784 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.5490 +Epoch [170/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8627 +Epoch [170/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.8627 +Epoch [170/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 10.0049 +Epoch [170/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8157, Pure Ratio2 9.9255 +Epoch [170/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9150 +Epoch [170/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 10.0168 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 79.8778 % Model2 79.7075 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [171/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7059 +Epoch [171/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9281, Pure Ratio2 9.8693 +Epoch [171/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.8873 +Epoch [171/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9725 +Epoch [171/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9314 +Epoch [171/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 79.9579 % Model2 80.1182 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1373 +Epoch [172/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.4804 +Epoch [172/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3595, Pure Ratio2 10.3333 +Epoch [172/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2402, Pure Ratio2 10.2598 +Epoch [172/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2275 +Epoch [172/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.0719 +Epoch [172/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0532, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 79.7676 % Model2 80.0180 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.5882 +Epoch [173/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8039 +Epoch [173/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.7582 +Epoch [173/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 10.0392 +Epoch [173/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.9725 +Epoch [173/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 10.0033 +Epoch [173/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8571, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 80.0481 % Model2 80.3986 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 10.0880 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1569 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2647 +Epoch [174/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.3007 +Epoch [174/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2794, Pure Ratio2 10.3627 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1529, Pure Ratio2 10.2353 +Epoch [174/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0980 +Epoch [174/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 79.9880 % Model2 79.4371 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 97.6562, Training Accuracy2: 97.6562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6863 +Epoch [175/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [175/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.6405 +Epoch [175/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7990 +Epoch [175/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.9059 +Epoch [175/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0196 +Epoch [175/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0336, Pure Ratio2 10.0392 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 79.8277 % Model2 79.4972 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6275 +Epoch [176/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0686 +Epoch [176/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0065 +Epoch [176/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.0784 +Epoch [176/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0471 +Epoch [176/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.8954 +Epoch [176/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 79.5673 % Model2 79.5573 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.3333 +Epoch [177/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.5000 +Epoch [177/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4967, Pure Ratio2 9.5948 +Epoch [177/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.9314 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8667, Pure Ratio2 9.9059 +Epoch [177/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 10.0392 +Epoch [177/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 80.0982 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1765 +Epoch [178/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2157 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 10.0850 +Epoch [178/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.1176 +Epoch [178/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 10.0039 +Epoch [178/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9183 +Epoch [178/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 10.0392 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 80.1783 % Model2 80.4888 %, Pure Ratio 1 10.0377 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8039 +Epoch [179/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.8039 +Epoch [179/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.9281 +Epoch [179/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9902 +Epoch [179/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.1373 +Epoch [179/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.2222 +Epoch [179/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 80.4187 % Model2 80.0481 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9216 +Epoch [180/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1569 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.1046 +Epoch [180/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9853 +Epoch [180/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.9647 +Epoch [180/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8660, Pure Ratio2 9.9379 +Epoch [180/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 80.6090 % Model2 80.1983 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.5882 +Epoch [181/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5000, Pure Ratio2 10.5196 +Epoch [181/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3856, Pure Ratio2 10.3529 +Epoch [181/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0245 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 10.1176 +Epoch [181/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9608 +Epoch [181/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 80.0381 % Model2 79.9980 %, Pure Ratio 1 10.0880 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.4314 +Epoch [182/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.1863 +Epoch [182/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1373 +Epoch [182/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1814 +Epoch [182/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1490, Pure Ratio2 10.1451 +Epoch [182/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 10.0033 +Epoch [182/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0448, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 79.6174 % Model2 80.2284 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.0392 +Epoch [183/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.6765 +Epoch [183/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [183/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.1912 +Epoch [183/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.1373 +Epoch [183/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0425 +Epoch [183/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 80.4587 % Model2 80.1282 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4510, Pure Ratio2 10.3333 +Epoch [184/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.9020 +Epoch [184/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 9.9673 +Epoch [184/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9461 +Epoch [184/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1098, Pure Ratio2 10.0314 +Epoch [184/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.1144 +Epoch [184/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1821, Pure Ratio2 10.0924 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 80.0881 % Model2 79.9479 %, Pure Ratio 1 10.0955 %, Pure Ratio 2 10.0176 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8039 +Epoch [185/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.8922 +Epoch [185/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0458 +Epoch [185/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.1275 +Epoch [185/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.0235 +Epoch [185/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 9.9804 +Epoch [185/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1120, Pure Ratio2 10.0308 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 79.9379 % Model2 79.9279 %, Pure Ratio 1 10.0830 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.9020 +Epoch [186/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.7451 +Epoch [186/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 10.0000 +Epoch [186/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9069, Pure Ratio2 9.9853 +Epoch [186/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.0824 +Epoch [186/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0654, Pure Ratio2 10.1503 +Epoch [186/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0756, Pure Ratio2 10.1569 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 80.0180 % Model2 79.9980 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0905 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.4902 +Epoch [187/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.5490 +Epoch [187/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9869 +Epoch [187/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.9314 +Epoch [187/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.1294 +Epoch [187/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9771, Pure Ratio2 10.0719 +Epoch [187/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 79.9179 % Model2 80.2584 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0503 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.5882 +Epoch [188/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.4020 +Epoch [188/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.1961 +Epoch [188/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8971, Pure Ratio2 9.9314 +Epoch [188/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 10.0078 +Epoch [188/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0425, Pure Ratio2 10.0817 +Epoch [188/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 80.1082 % Model2 80.7993 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.1156 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0011, Loss2: 0.0013, Pure Ratio1: 10.2549, Pure Ratio2 10.3725 +Epoch [189/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.1471 +Epoch [189/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8301 +Epoch [189/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8676 +Epoch [189/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8510, Pure Ratio2 9.8824 +Epoch [189/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 10.0098 +Epoch [189/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0952, Pure Ratio2 10.1373 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 79.8277 % Model2 80.4587 %, Pure Ratio 1 10.0503 %, Pure Ratio 2 10.0830 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.0784 +Epoch [190/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.1961, Pure Ratio2 9.2843 +Epoch [190/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.7451 +Epoch [190/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8725 +Epoch [190/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9333 +Epoch [190/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9412 +Epoch [190/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0840, Pure Ratio2 10.0784 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 80.2183 % Model2 80.3085 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [191/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.9020 +Epoch [191/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1111, Pure Ratio2 10.1046 +Epoch [191/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0147 +Epoch [191/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9725, Pure Ratio2 9.9765 +Epoch [191/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0392 +Epoch [191/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0224, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 80.6190 % Model2 80.4187 %, Pure Ratio 1 10.0729 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.5490 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.3235 +Epoch [192/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2222, Pure Ratio2 10.2614 +Epoch [192/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1814, Pure Ratio2 10.1618 +Epoch [192/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0471 +Epoch [192/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1275 +Epoch [192/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 79.5473 % Model2 80.6390 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0377 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7451, Pure Ratio2 10.8235 +Epoch [193/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4412, Pure Ratio2 10.6275 +Epoch [193/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.4641 +Epoch [193/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1618, Pure Ratio2 10.1569 +Epoch [193/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.2353 +Epoch [193/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1405, Pure Ratio2 10.1634 +Epoch [193/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 80.2985 % Model2 80.3385 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 10.0277 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.8431 +Epoch [194/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.1275 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.2549 +Epoch [194/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9167, Pure Ratio2 10.1275 +Epoch [194/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0627 +Epoch [194/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 10.0850 +Epoch [194/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 80.4087 % Model2 80.3586 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0980, Pure Ratio2 9.2941 +Epoch [195/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3137, Pure Ratio2 9.4216 +Epoch [195/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6144, Pure Ratio2 9.7516 +Epoch [195/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7304, Pure Ratio2 9.8235 +Epoch [195/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6157, Pure Ratio2 9.7412 +Epoch [195/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.8333 +Epoch [195/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8711, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 80.4788 % Model2 80.4788 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.1131 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3137 +Epoch [196/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [196/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.1438 +Epoch [196/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0637, Pure Ratio2 10.0784 +Epoch [196/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1255, Pure Ratio2 10.1333 +Epoch [196/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0359 +Epoch [196/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 80.1482 % Model2 80.2784 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [197/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.2549 +Epoch [197/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.2026 +Epoch [197/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 10.0833 +Epoch [197/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.1098 +Epoch [197/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0229, Pure Ratio2 10.0556 +Epoch [197/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 80.2684 % Model2 81.0096 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1569 +Epoch [198/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3529, Pure Ratio2 10.3039 +Epoch [198/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.2353 +Epoch [198/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 9.9853 +Epoch [198/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.8863 +Epoch [198/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8987 +Epoch [198/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 80.3586 % Model2 80.6490 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.7843, Pure Ratio2 10.5882 +Epoch [199/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4412, Pure Ratio2 10.3431 +Epoch [199/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1634 +Epoch [199/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0294 +Epoch [199/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0078, Pure Ratio2 10.0392 +Epoch [199/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 9.9412 +Epoch [199/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 10.0364 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 80.4688 % Model2 80.6090 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4706, Pure Ratio2 10.7059 +Epoch [200/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3235, Pure Ratio2 10.4804 +Epoch [200/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2222, Pure Ratio2 10.3922 +Epoch [200/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1912, Pure Ratio2 10.2941 +Epoch [200/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1020, Pure Ratio2 10.1529 +Epoch [200/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0850, Pure Ratio2 10.1667 +Epoch [200/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.1485 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 80.6390 % Model2 80.2784 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 10.0804 % diff --git a/other_methods/coteaching/coteaching_results/out_6_4.log b/other_methods/coteaching/coteaching_results/out_6_4.log new file mode 100644 index 0000000..aabcbea --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_6_4.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 28.1250, Training Accuracy2: 35.1562, Loss1: 0.0156, Loss2: 0.0155, Pure Ratio1: 9.6480, Pure Ratio2 9.6640 +Epoch [2/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0140, Loss2: 0.0142, Pure Ratio1: 9.9440, Pure Ratio2 9.9440 +Epoch [2/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 29.6875, Loss1: 0.0148, Loss2: 0.0155, Pure Ratio1: 10.0213, Pure Ratio2 10.0267 +Epoch [2/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0146, Loss2: 0.0146, Pure Ratio1: 10.1160, Pure Ratio2 10.1400 +Epoch [2/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0143, Loss2: 0.0138, Pure Ratio1: 10.1152, Pure Ratio2 10.1280 +Epoch [2/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0129, Loss2: 0.0124, Pure Ratio1: 10.0773, Pure Ratio2 10.0773 +Epoch [2/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0134, Loss2: 0.0132, Pure Ratio1: 9.9726, Pure Ratio2 9.9589 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 41.1058 % Model2 39.2628 %, Pure Ratio 1 9.9795 %, Pure Ratio 2 9.9569 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0126, Loss2: 0.0118, Pure Ratio1: 9.1639, Pure Ratio2 9.1311 +Epoch [3/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0130, Loss2: 0.0123, Pure Ratio1: 9.5000, Pure Ratio2 9.4754 +Epoch [3/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0123, Loss2: 0.0123, Pure Ratio1: 9.7322, Pure Ratio2 9.7486 +Epoch [3/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 50.7812, Loss1: 0.0116, Loss2: 0.0114, Pure Ratio1: 9.9016, Pure Ratio2 9.9303 +Epoch [3/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0131, Loss2: 0.0127, Pure Ratio1: 9.9869, Pure Ratio2 10.0230 +Epoch [3/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0109, Loss2: 0.0110, Pure Ratio1: 10.0137, Pure Ratio2 10.0464 +Epoch [3/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.1875, Loss1: 0.0118, Loss2: 0.0122, Pure Ratio1: 9.9578, Pure Ratio2 9.9719 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 45.0921 % Model2 44.9419 %, Pure Ratio 1 9.9369 %, Pure Ratio 2 9.9601 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0116, Loss2: 0.0114, Pure Ratio1: 10.1008, Pure Ratio2 10.0672 +Epoch [4/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0123, Loss2: 0.0124, Pure Ratio1: 9.7647, Pure Ratio2 9.7731 +Epoch [4/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0106, Loss2: 0.0110, Pure Ratio1: 9.8375, Pure Ratio2 9.8431 +Epoch [4/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0112, Loss2: 0.0107, Pure Ratio1: 9.7269, Pure Ratio2 9.7353 +Epoch [4/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0113, Loss2: 0.0119, Pure Ratio1: 9.7479, Pure Ratio2 9.7815 +Epoch [4/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0106, Loss2: 0.0110, Pure Ratio1: 9.8431, Pure Ratio2 9.8515 +Epoch [4/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0106, Loss2: 0.0106, Pure Ratio1: 9.9376, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 51.5925 % Model2 51.9131 %, Pure Ratio 1 9.9569 %, Pure Ratio 2 9.9504 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0116, Loss2: 0.0115, Pure Ratio1: 10.3793, Pure Ratio2 10.4828 +Epoch [5/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0088, Loss2: 0.0087, Pure Ratio1: 10.2672, Pure Ratio2 10.2931 +Epoch [5/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0115, Loss2: 0.0109, Pure Ratio1: 10.2931, Pure Ratio2 10.3161 +Epoch [5/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0097, Loss2: 0.0090, Pure Ratio1: 10.1379, Pure Ratio2 10.1336 +Epoch [5/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0096, Loss2: 0.0095, Pure Ratio1: 10.1862, Pure Ratio2 10.1552 +Epoch [5/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0105, Loss2: 0.0103, Pure Ratio1: 10.0201, Pure Ratio2 9.9971 +Epoch [5/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0095, Loss2: 0.0100, Pure Ratio1: 10.0123, Pure Ratio2 9.9680 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 50.6410 % Model2 51.8630 %, Pure Ratio 1 9.9713 %, Pure Ratio 2 9.9469 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0110, Loss2: 0.0115, Pure Ratio1: 10.0708, Pure Ratio2 10.1947 +Epoch [6/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0113, Loss2: 0.0108, Pure Ratio1: 10.2832, Pure Ratio2 10.3274 +Epoch [6/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0092, Loss2: 0.0097, Pure Ratio1: 10.1711, Pure Ratio2 10.2183 +Epoch [6/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0089, Loss2: 0.0088, Pure Ratio1: 10.1062, Pure Ratio2 10.1416 +Epoch [6/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0093, Loss2: 0.0094, Pure Ratio1: 10.0035, Pure Ratio2 10.0177 +Epoch [6/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0098, Loss2: 0.0097, Pure Ratio1: 9.8496, Pure Ratio2 9.8968 +Epoch [6/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0092, Loss2: 0.0091, Pure Ratio1: 9.8559, Pure Ratio2 9.9115 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 53.1250 % Model2 51.6526 %, Pure Ratio 1 9.8911 %, Pure Ratio 2 9.9319 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0098, Loss2: 0.0093, Pure Ratio1: 9.8909, Pure Ratio2 10.0909 +Epoch [7/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0081, Loss2: 0.0083, Pure Ratio1: 9.9818, Pure Ratio2 10.0909 +Epoch [7/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0089, Loss2: 0.0084, Pure Ratio1: 10.0727, Pure Ratio2 10.1212 +Epoch [7/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0083, Loss2: 0.0087, Pure Ratio1: 9.9727, Pure Ratio2 10.0455 +Epoch [7/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0096, Loss2: 0.0097, Pure Ratio1: 10.0109, Pure Ratio2 10.0436 +Epoch [7/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0072, Loss2: 0.0075, Pure Ratio1: 10.0091, Pure Ratio2 10.0212 +Epoch [7/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0087, Loss2: 0.0095, Pure Ratio1: 9.9662, Pure Ratio2 9.9688 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 60.4367 % Model2 60.1663 %, Pure Ratio 1 9.9674 %, Pure Ratio 2 9.9627 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0075, Loss2: 0.0070, Pure Ratio1: 10.1852, Pure Ratio2 10.0370 +Epoch [8/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0073, Loss2: 0.0075, Pure Ratio1: 9.9630, Pure Ratio2 9.8611 +Epoch [8/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0078, Loss2: 0.0075, Pure Ratio1: 9.8951, Pure Ratio2 9.8395 +Epoch [8/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0076, Loss2: 0.0077, Pure Ratio1: 9.7685, Pure Ratio2 9.6574 +Epoch [8/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0075, Loss2: 0.0079, Pure Ratio1: 9.9370, Pure Ratio2 9.8519 +Epoch [8/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0080, Loss2: 0.0078, Pure Ratio1: 10.0463, Pure Ratio2 9.9753 +Epoch [8/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0088, Loss2: 0.0087, Pure Ratio1: 9.9339, Pure Ratio2 9.8519 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 53.3754 % Model2 51.6226 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 9.8837 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0074, Loss2: 0.0066, Pure Ratio1: 9.5048, Pure Ratio2 9.6381 +Epoch [9/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0072, Loss2: 0.0076, Pure Ratio1: 9.6381, Pure Ratio2 9.6095 +Epoch [9/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0059, Loss2: 0.0052, Pure Ratio1: 9.6889, Pure Ratio2 9.6127 +Epoch [9/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0088, Loss2: 0.0092, Pure Ratio1: 9.7476, Pure Ratio2 9.6905 +Epoch [9/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0073, Loss2: 0.0074, Pure Ratio1: 9.9543, Pure Ratio2 9.9124 +Epoch [9/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0077, Loss2: 0.0077, Pure Ratio1: 9.9333, Pure Ratio2 9.9429 +Epoch [9/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0080, Loss2: 0.0078, Pure Ratio1: 9.9238, Pure Ratio2 9.9483 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 63.8622 % Model2 63.6819 %, Pure Ratio 1 9.9316 %, Pure Ratio 2 9.9219 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0067, Loss2: 0.0074, Pure Ratio1: 10.2745, Pure Ratio2 10.1569 +Epoch [10/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0079, Loss2: 0.0076, Pure Ratio1: 10.0000, Pure Ratio2 9.9608 +Epoch [10/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0082, Loss2: 0.0075, Pure Ratio1: 9.7974, Pure Ratio2 9.6471 +Epoch [10/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0077, Loss2: 0.0075, Pure Ratio1: 9.8186, Pure Ratio2 9.7745 +Epoch [10/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0076, Loss2: 0.0074, Pure Ratio1: 9.8392, Pure Ratio2 9.8157 +Epoch [10/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0087, Loss2: 0.0089, Pure Ratio1: 9.8529, Pure Ratio2 9.8235 +Epoch [10/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0065, Loss2: 0.0070, Pure Ratio1: 9.8936, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 53.5457 % Model2 52.0232 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0061, Loss2: 0.0064, Pure Ratio1: 9.7255, Pure Ratio2 9.6078 +Epoch [11/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0073, Loss2: 0.0080, Pure Ratio1: 10.0784, Pure Ratio2 9.9608 +Epoch [11/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0068, Loss2: 0.0075, Pure Ratio1: 9.9804, Pure Ratio2 9.8627 +Epoch [11/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0079, Loss2: 0.0073, Pure Ratio1: 9.9902, Pure Ratio2 9.9020 +Epoch [11/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0066, Loss2: 0.0066, Pure Ratio1: 9.9608, Pure Ratio2 9.9059 +Epoch [11/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0078, Loss2: 0.0072, Pure Ratio1: 10.0098, Pure Ratio2 9.9477 +Epoch [11/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0077, Loss2: 0.0081, Pure Ratio1: 9.9720, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 59.0645 % Model2 58.5136 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0057, Loss2: 0.0060, Pure Ratio1: 9.4706, Pure Ratio2 9.6078 +Epoch [12/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0062, Loss2: 0.0053, Pure Ratio1: 9.9118, Pure Ratio2 10.1176 +Epoch [12/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0052, Loss2: 0.0054, Pure Ratio1: 10.0065, Pure Ratio2 10.1307 +Epoch [12/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0053, Loss2: 0.0055, Pure Ratio1: 9.9853, Pure Ratio2 10.0686 +Epoch [12/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0060, Loss2: 0.0057, Pure Ratio1: 9.9373, Pure Ratio2 9.9765 +Epoch [12/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0090, Loss2: 0.0091, Pure Ratio1: 9.8889, Pure Ratio2 9.9085 +Epoch [12/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0047, Loss2: 0.0051, Pure Ratio1: 9.9020, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 56.1498 % Model2 57.3718 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0069, Loss2: 0.0075, Pure Ratio1: 9.8039, Pure Ratio2 9.6275 +Epoch [13/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0060, Loss2: 0.0063, Pure Ratio1: 10.0000, Pure Ratio2 9.9902 +Epoch [13/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0058, Loss2: 0.0056, Pure Ratio1: 9.6863, Pure Ratio2 9.6797 +Epoch [13/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0068, Loss2: 0.0073, Pure Ratio1: 9.8284, Pure Ratio2 9.8627 +Epoch [13/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0072, Loss2: 0.0074, Pure Ratio1: 9.9373, Pure Ratio2 9.9490 +Epoch [13/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0060, Loss2: 0.0061, Pure Ratio1: 10.0882, Pure Ratio2 10.1176 +Epoch [13/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0079, Loss2: 0.0069, Pure Ratio1: 10.0700, Pure Ratio2 10.1120 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 58.1330 % Model2 59.0244 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0062, Loss2: 0.0058, Pure Ratio1: 9.8039, Pure Ratio2 9.8431 +Epoch [14/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0067, Loss2: 0.0071, Pure Ratio1: 9.6765, Pure Ratio2 9.7451 +Epoch [14/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0066, Loss2: 0.0067, Pure Ratio1: 9.8627, Pure Ratio2 9.8562 +Epoch [14/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0051, Loss2: 0.0054, Pure Ratio1: 9.8676, Pure Ratio2 9.8676 +Epoch [14/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0067, Loss2: 0.0075, Pure Ratio1: 9.8118, Pure Ratio2 9.8235 +Epoch [14/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0066, Loss2: 0.0067, Pure Ratio1: 9.7516, Pure Ratio2 9.7810 +Epoch [14/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0061, Loss2: 0.0057, Pure Ratio1: 9.8571, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 63.3013 % Model2 62.5200 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0064, Loss2: 0.0069, Pure Ratio1: 9.1176, Pure Ratio2 9.0980 +Epoch [15/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0045, Loss2: 0.0046, Pure Ratio1: 9.4804, Pure Ratio2 9.4902 +Epoch [15/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0066, Loss2: 0.0064, Pure Ratio1: 9.7386, Pure Ratio2 9.8039 +Epoch [15/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0072, Loss2: 0.0065, Pure Ratio1: 9.8284, Pure Ratio2 9.9363 +Epoch [15/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0066, Loss2: 0.0062, Pure Ratio1: 9.8667, Pure Ratio2 9.9373 +Epoch [15/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0069, Loss2: 0.0067, Pure Ratio1: 9.9085, Pure Ratio2 9.9869 +Epoch [15/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0061, Loss2: 0.0057, Pure Ratio1: 9.9636, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 60.8874 % Model2 60.6270 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0056, Loss2: 0.0052, Pure Ratio1: 9.3922, Pure Ratio2 9.2941 +Epoch [16/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0075, Loss2: 0.0068, Pure Ratio1: 9.3529, Pure Ratio2 9.2059 +Epoch [16/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0066, Loss2: 0.0056, Pure Ratio1: 9.8889, Pure Ratio2 9.7516 +Epoch [16/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 61.7188, Loss1: 0.0043, Loss2: 0.0054, Pure Ratio1: 9.9363, Pure Ratio2 9.8431 +Epoch [16/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0056, Loss2: 0.0062, Pure Ratio1: 9.9137, Pure Ratio2 9.8353 +Epoch [16/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0087, Loss2: 0.0083, Pure Ratio1: 9.8954, Pure Ratio2 9.8170 +Epoch [16/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0055, Loss2: 0.0053, Pure Ratio1: 9.9384, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 56.3101 % Model2 57.1915 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0053, Loss2: 0.0054, Pure Ratio1: 9.6078, Pure Ratio2 9.7647 +Epoch [17/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0045, Loss2: 0.0041, Pure Ratio1: 9.7353, Pure Ratio2 9.7941 +Epoch [17/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0045, Loss2: 0.0046, Pure Ratio1: 9.9869, Pure Ratio2 10.0980 +Epoch [17/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0045, Loss2: 0.0049, Pure Ratio1: 9.9314, Pure Ratio2 10.0392 +Epoch [17/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0049, Loss2: 0.0042, Pure Ratio1: 10.0000, Pure Ratio2 10.0667 +Epoch [17/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0067, Loss2: 0.0066, Pure Ratio1: 9.8791, Pure Ratio2 9.9771 +Epoch [17/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0052, Loss2: 0.0059, Pure Ratio1: 9.8852, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 59.9960 % Model2 59.1947 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0037, Loss2: 0.0036, Pure Ratio1: 9.8431, Pure Ratio2 9.6078 +Epoch [18/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0062, Loss2: 0.0065, Pure Ratio1: 9.9412, Pure Ratio2 9.8431 +Epoch [18/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0071, Loss2: 0.0056, Pure Ratio1: 10.1765, Pure Ratio2 10.1634 +Epoch [18/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0043, Loss2: 0.0036, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [18/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0069, Loss2: 0.0067, Pure Ratio1: 9.9922, Pure Ratio2 10.0431 +Epoch [18/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0045, Loss2: 0.0048, Pure Ratio1: 9.9412, Pure Ratio2 9.9510 +Epoch [18/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0061, Loss2: 0.0061, Pure Ratio1: 9.9776, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 58.0329 % Model2 59.5753 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0033, Loss2: 0.0029, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [19/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0043, Loss2: 0.0044, Pure Ratio1: 9.5490, Pure Ratio2 9.5294 +Epoch [19/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0043, Loss2: 0.0044, Pure Ratio1: 9.6863, Pure Ratio2 9.5948 +Epoch [19/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0061, Loss2: 0.0053, Pure Ratio1: 9.7598, Pure Ratio2 9.7059 +Epoch [19/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0049, Loss2: 0.0048, Pure Ratio1: 9.8118, Pure Ratio2 9.7765 +Epoch [19/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0045, Loss2: 0.0040, Pure Ratio1: 9.8268, Pure Ratio2 9.7647 +Epoch [19/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0048, Loss2: 0.0044, Pure Ratio1: 10.0028, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 63.3514 % Model2 61.1579 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0036, Loss2: 0.0031, Pure Ratio1: 10.3137, Pure Ratio2 10.4118 +Epoch [20/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [20/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0033, Loss2: 0.0042, Pure Ratio1: 10.0784, Pure Ratio2 10.1373 +Epoch [20/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0038, Loss2: 0.0036, Pure Ratio1: 9.9314, Pure Ratio2 9.9902 +Epoch [20/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0057, Loss2: 0.0049, Pure Ratio1: 9.9529, Pure Ratio2 9.9569 +Epoch [20/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0040, Loss2: 0.0043, Pure Ratio1: 9.9739, Pure Ratio2 9.9477 +Epoch [20/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.0035, Loss2: 0.0033, Pure Ratio1: 9.9020, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 60.1562 % Model2 61.6486 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 70.3125, Loss1: 0.0028, Loss2: 0.0039, Pure Ratio1: 10.5882, Pure Ratio2 10.5686 +Epoch [21/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.0026, Loss2: 0.0036, Pure Ratio1: 10.0490, Pure Ratio2 9.9804 +Epoch [21/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0040, Loss2: 0.0035, Pure Ratio1: 9.9346, Pure Ratio2 9.8562 +Epoch [21/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0042, Loss2: 0.0033, Pure Ratio1: 10.0735, Pure Ratio2 10.0343 +Epoch [21/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0037, Loss2: 0.0037, Pure Ratio1: 10.0745, Pure Ratio2 10.0275 +Epoch [21/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0039, Loss2: 0.0050, Pure Ratio1: 10.0000, Pure Ratio2 9.9935 +Epoch [21/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0041, Loss2: 0.0038, Pure Ratio1: 9.9944, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 58.9944 % Model2 57.6022 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0035, Loss2: 0.0044, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [22/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0033, Loss2: 0.0038, Pure Ratio1: 9.9804, Pure Ratio2 9.9510 +Epoch [22/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0024, Loss2: 0.0027, Pure Ratio1: 9.8235, Pure Ratio2 9.7908 +Epoch [22/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0055, Loss2: 0.0046, Pure Ratio1: 9.8824, Pure Ratio2 9.8039 +Epoch [22/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0040, Loss2: 0.0042, Pure Ratio1: 9.9765, Pure Ratio2 9.9137 +Epoch [22/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0038, Loss2: 0.0037, Pure Ratio1: 9.8693, Pure Ratio2 9.8366 +Epoch [22/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0042, Loss2: 0.0039, Pure Ratio1: 9.9664, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 59.2348 % Model2 58.0429 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0016, Loss2: 0.0017, Pure Ratio1: 10.4706, Pure Ratio2 10.3529 +Epoch [23/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.0022, Loss2: 0.0026, Pure Ratio1: 10.0882, Pure Ratio2 9.9608 +Epoch [23/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0031, Loss2: 0.0037, Pure Ratio1: 10.1765, Pure Ratio2 10.0131 +Epoch [23/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0048, Loss2: 0.0050, Pure Ratio1: 10.2402, Pure Ratio2 10.1029 +Epoch [23/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0033, Loss2: 0.0025, Pure Ratio1: 10.1804, Pure Ratio2 10.0392 +Epoch [23/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0034, Loss2: 0.0050, Pure Ratio1: 10.1699, Pure Ratio2 10.0523 +Epoch [23/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0040, Loss2: 0.0042, Pure Ratio1: 10.1148, Pure Ratio2 10.0140 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 59.5954 % Model2 63.5417 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0026, Loss2: 0.0027, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [24/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0033, Loss2: 0.0037, Pure Ratio1: 9.8431, Pure Ratio2 9.9020 +Epoch [24/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0020, Loss2: 0.0029, Pure Ratio1: 10.0654, Pure Ratio2 10.1569 +Epoch [24/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0041, Loss2: 0.0033, Pure Ratio1: 10.0196, Pure Ratio2 10.1275 +Epoch [24/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0034, Loss2: 0.0029, Pure Ratio1: 9.8980, Pure Ratio2 9.9725 +Epoch [24/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0020, Pure Ratio1: 9.8399, Pure Ratio2 9.9477 +Epoch [24/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0018, Loss2: 0.0018, Pure Ratio1: 9.8627, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 58.5236 % Model2 59.0545 %, Pure Ratio 1 9.8869 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0022, Pure Ratio1: 10.3137, Pure Ratio2 10.1961 +Epoch [25/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0032, Loss2: 0.0028, Pure Ratio1: 10.0784, Pure Ratio2 10.0000 +Epoch [25/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.0030, Loss2: 0.0021, Pure Ratio1: 10.1307, Pure Ratio2 10.0980 +Epoch [25/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 71.8750, Loss1: 0.0018, Loss2: 0.0027, Pure Ratio1: 9.8431, Pure Ratio2 9.8676 +Epoch [25/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0034, Loss2: 0.0034, Pure Ratio1: 9.8314, Pure Ratio2 9.8510 +Epoch [25/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0032, Loss2: 0.0025, Pure Ratio1: 9.7614, Pure Ratio2 9.7908 +Epoch [25/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0024, Loss2: 0.0023, Pure Ratio1: 9.8768, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 58.1931 % Model2 56.9712 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0019, Loss2: 0.0018, Pure Ratio1: 10.1765, Pure Ratio2 10.2745 +Epoch [26/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0017, Loss2: 0.0013, Pure Ratio1: 10.1176, Pure Ratio2 10.0882 +Epoch [26/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 76.5625, Loss1: 0.0010, Loss2: 0.0020, Pure Ratio1: 10.1242, Pure Ratio2 10.1503 +Epoch [26/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.0027, Loss2: 0.0027, Pure Ratio1: 10.2108, Pure Ratio2 10.2353 +Epoch [26/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0023, Loss2: 0.0025, Pure Ratio1: 10.0353, Pure Ratio2 10.0431 +Epoch [26/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0022, Loss2: 0.0022, Pure Ratio1: 10.0882, Pure Ratio2 10.1046 +Epoch [26/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0022, Loss2: 0.0022, Pure Ratio1: 10.0448, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 60.0260 % Model2 60.3666 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0016, Loss2: 0.0015, Pure Ratio1: 9.6275, Pure Ratio2 9.6078 +Epoch [27/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.7812, Loss1: 0.0015, Loss2: 0.0023, Pure Ratio1: 9.8333, Pure Ratio2 9.8333 +Epoch [27/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0021, Loss2: 0.0018, Pure Ratio1: 9.8627, Pure Ratio2 9.8170 +Epoch [27/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0020, Loss2: 0.0021, Pure Ratio1: 9.8431, Pure Ratio2 9.8284 +Epoch [27/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.8125, Loss1: 0.0022, Loss2: 0.0016, Pure Ratio1: 9.8510, Pure Ratio2 9.8588 +Epoch [27/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.0017, Loss2: 0.0023, Pure Ratio1: 9.9346, Pure Ratio2 9.9150 +Epoch [27/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0022, Pure Ratio1: 10.0448, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 58.8241 % Model2 58.9944 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 9.3922, Pure Ratio2 9.3529 +Epoch [28/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.9062, Loss1: 0.0026, Loss2: 0.0020, Pure Ratio1: 9.9412, Pure Ratio2 9.8431 +Epoch [28/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0012, Loss2: 0.0012, Pure Ratio1: 10.0850, Pure Ratio2 9.9673 +Epoch [28/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 10.0735, Pure Ratio2 9.9363 +Epoch [28/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0019, Loss2: 0.0017, Pure Ratio1: 9.9333, Pure Ratio2 9.8706 +Epoch [28/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0013, Loss2: 0.0017, Pure Ratio1: 9.9641, Pure Ratio2 9.9150 +Epoch [28/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 87.5000, Loss1: 0.0017, Loss2: 0.0012, Pure Ratio1: 10.0000, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 56.6907 % Model2 59.1446 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.0011, Loss2: 0.0018, Pure Ratio1: 9.4706, Pure Ratio2 9.4902 +Epoch [29/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0016, Loss2: 0.0014, Pure Ratio1: 9.9706, Pure Ratio2 9.9412 +Epoch [29/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0014, Loss2: 0.0023, Pure Ratio1: 9.9085, Pure Ratio2 9.9281 +Epoch [29/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.0019, Loss2: 0.0028, Pure Ratio1: 9.7941, Pure Ratio2 9.7843 +Epoch [29/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0017, Loss2: 0.0024, Pure Ratio1: 9.8157, Pure Ratio2 9.8314 +Epoch [29/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0016, Pure Ratio1: 9.8824, Pure Ratio2 9.8987 +Epoch [29/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 9.9636, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 56.2099 % Model2 56.0196 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9874 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0015, Loss2: 0.0012, Pure Ratio1: 9.7059, Pure Ratio2 9.6471 +Epoch [30/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 10.0294, Pure Ratio2 9.9412 +Epoch [30/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 80.4688, Loss1: 0.0011, Loss2: 0.0022, Pure Ratio1: 9.9477, Pure Ratio2 9.8627 +Epoch [30/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0015, Pure Ratio1: 9.8922, Pure Ratio2 9.8480 +Epoch [30/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.7765, Pure Ratio2 9.7137 +Epoch [30/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.8399, Pure Ratio2 9.8268 +Epoch [30/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0019, Loss2: 0.0012, Pure Ratio1: 9.9048, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 56.4103 % Model2 57.5421 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0015, Pure Ratio1: 9.8627, Pure Ratio2 9.6667 +Epoch [31/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.8529, Pure Ratio2 9.7647 +Epoch [31/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0017, Loss2: 0.0009, Pure Ratio1: 9.8693, Pure Ratio2 9.8039 +Epoch [31/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0022, Loss2: 0.0012, Pure Ratio1: 10.0833, Pure Ratio2 10.0637 +Epoch [31/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.0667, Pure Ratio2 10.0784 +Epoch [31/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0011, Pure Ratio1: 10.0163, Pure Ratio2 10.0654 +Epoch [31/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 10.0336, Pure Ratio2 10.0728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 56.7408 % Model2 57.9227 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0014, Loss2: 0.0009, Pure Ratio1: 10.5490, Pure Ratio2 10.3922 +Epoch [32/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.9804, Pure Ratio2 9.8235 +Epoch [32/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 10.0588, Pure Ratio2 9.9608 +Epoch [32/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0013, Pure Ratio1: 10.0686, Pure Ratio2 9.9755 +Epoch [32/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0014, Pure Ratio1: 10.0235, Pure Ratio2 9.9922 +Epoch [32/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0021, Loss2: 0.0013, Pure Ratio1: 10.0000, Pure Ratio2 10.0098 +Epoch [32/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0041, Loss2: 0.0044, Pure Ratio1: 10.0112, Pure Ratio2 10.0056 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 56.6006 % Model2 57.1314 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 8.8235, Pure Ratio2 8.8235 +Epoch [33/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.3529, Pure Ratio2 9.3431 +Epoch [33/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.9062, Loss1: 0.0023, Loss2: 0.0016, Pure Ratio1: 9.5033, Pure Ratio2 9.5359 +Epoch [33/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 9.7108, Pure Ratio2 9.7598 +Epoch [33/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.6941, Pure Ratio2 9.7529 +Epoch [33/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0010, Pure Ratio1: 9.7908, Pure Ratio2 9.8301 +Epoch [33/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.8627, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 58.1130 % Model2 57.7424 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0009, Loss2: 0.0012, Pure Ratio1: 10.1765, Pure Ratio2 10.3333 +Epoch [34/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0016, Loss2: 0.0008, Pure Ratio1: 10.0098, Pure Ratio2 10.0490 +Epoch [34/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0013, Pure Ratio1: 10.1699, Pure Ratio2 10.1765 +Epoch [34/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 84.3750, Loss1: 0.0012, Loss2: 0.0011, Pure Ratio1: 9.9559, Pure Ratio2 10.0000 +Epoch [34/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0019, Loss2: 0.0008, Pure Ratio1: 9.9647, Pure Ratio2 10.0039 +Epoch [34/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0023, Loss2: 0.0008, Pure Ratio1: 9.9477, Pure Ratio2 9.9739 +Epoch [34/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.9608, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 57.8626 % Model2 55.8494 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.1373, Pure Ratio2 10.0392 +Epoch [35/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.3431, Pure Ratio2 10.2353 +Epoch [35/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0025, Loss2: 0.0030, Pure Ratio1: 10.3791, Pure Ratio2 10.2157 +Epoch [35/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 10.1961, Pure Ratio2 9.9951 +Epoch [35/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0009, Pure Ratio1: 10.1490, Pure Ratio2 9.9451 +Epoch [35/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0017, Pure Ratio1: 10.0784, Pure Ratio2 9.9477 +Epoch [35/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 86.7188, Loss1: 0.0013, Loss2: 0.0007, Pure Ratio1: 10.0896, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 58.4335 % Model2 57.1414 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9070 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.6667, Pure Ratio2 9.4902 +Epoch [36/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0017, Pure Ratio1: 9.8137, Pure Ratio2 9.7451 +Epoch [36/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 10.1961, Pure Ratio2 10.1242 +Epoch [36/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.1029, Pure Ratio2 10.0098 +Epoch [36/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 10.1686, Pure Ratio2 10.0353 +Epoch [36/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 10.0359, Pure Ratio2 9.9379 +Epoch [36/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0011, Pure Ratio1: 9.9384, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 55.4087 % Model2 57.8125 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 10.3529, Pure Ratio2 10.0784 +Epoch [37/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 10.0588, Pure Ratio2 9.9804 +Epoch [37/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.0000, Pure Ratio2 9.8758 +Epoch [37/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 10.1029, Pure Ratio2 10.0147 +Epoch [37/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0012, Pure Ratio1: 9.9647, Pure Ratio2 9.8627 +Epoch [37/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 10.0458, Pure Ratio2 9.9608 +Epoch [37/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9748, Pure Ratio2 9.9048 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 57.6122 % Model2 57.2917 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [38/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 10.1471, Pure Ratio2 10.0686 +Epoch [38/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 82.8125, Loss1: 0.0014, Loss2: 0.0009, Pure Ratio1: 10.0327, Pure Ratio2 10.0261 +Epoch [38/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9951, Pure Ratio2 9.9461 +Epoch [38/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0022, Loss2: 0.0015, Pure Ratio1: 10.0863, Pure Ratio2 10.0275 +Epoch [38/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0020, Pure Ratio1: 10.0000, Pure Ratio2 9.9314 +Epoch [38/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9524, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 57.7224 % Model2 58.9543 %, Pure Ratio 1 10.0201 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0003, Pure Ratio1: 9.6078, Pure Ratio2 9.6471 +Epoch [39/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [39/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 90.6250, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 10.0654, Pure Ratio2 9.9935 +Epoch [39/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9902, Pure Ratio2 9.9118 +Epoch [39/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0013, Loss2: 0.0005, Pure Ratio1: 9.9451, Pure Ratio2 9.9137 +Epoch [39/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.8987 +Epoch [39/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.9468, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 57.1114 % Model2 58.2332 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 10.1373 +Epoch [40/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 9.9020 +Epoch [40/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.0392, Pure Ratio2 10.0915 +Epoch [40/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.9902, Pure Ratio2 9.9657 +Epoch [40/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0013, Loss2: 0.0012, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Epoch [40/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0009, Pure Ratio1: 9.8333, Pure Ratio2 9.8562 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.8739, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 57.4018 % Model2 57.4820 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 10.2353, Pure Ratio2 10.1765 +Epoch [41/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.1875, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.0000 +Epoch [41/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9869, Pure Ratio2 9.9542 +Epoch [41/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.9412, Pure Ratio2 9.8922 +Epoch [41/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9765, Pure Ratio2 9.9647 +Epoch [41/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.9248, Pure Ratio2 9.8627 +Epoch [41/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.9496, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 57.5421 % Model2 57.8425 %, Pure Ratio 1 9.9522 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.6078, Pure Ratio2 9.5686 +Epoch [42/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.0588, Pure Ratio2 9.9706 +Epoch [42/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 10.0065, Pure Ratio2 9.9020 +Epoch [42/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.9608, Pure Ratio2 9.8971 +Epoch [42/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0015, Loss2: 0.0009, Pure Ratio1: 10.0078, Pure Ratio2 9.9804 +Epoch [42/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 10.0033, Pure Ratio2 9.9804 +Epoch [42/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0011, Pure Ratio1: 9.9496, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 59.1747 % Model2 56.4603 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.4118, Pure Ratio2 10.2353 +Epoch [43/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 10.4804, Pure Ratio2 10.5686 +Epoch [43/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.2484, Pure Ratio2 10.3922 +Epoch [43/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 10.1078, Pure Ratio2 10.2108 +Epoch [43/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.2353, Pure Ratio2 10.3137 +Epoch [43/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 10.0261, Pure Ratio2 10.0654 +Epoch [43/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 10.0196, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 57.5020 % Model2 58.9143 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.4706, Pure Ratio2 9.2549 +Epoch [44/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [44/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.8693, Pure Ratio2 9.7712 +Epoch [44/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8676, Pure Ratio2 9.7892 +Epoch [44/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.9765, Pure Ratio2 9.9020 +Epoch [44/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.0719, Pure Ratio2 9.9542 +Epoch [44/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.9972, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 57.4920 % Model2 56.9211 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9020 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0013, Pure Ratio1: 9.7255, Pure Ratio2 10.0000 +Epoch [45/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 85.1562, Loss1: 0.0015, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.7549 +Epoch [45/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.8954 +Epoch [45/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.7794, Pure Ratio2 9.8235 +Epoch [45/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 9.9059, Pure Ratio2 9.9608 +Epoch [45/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8464, Pure Ratio2 9.9346 +Epoch [45/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0011, Loss2: 0.0009, Pure Ratio1: 9.8179, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 56.9311 % Model2 57.9127 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 10.0050 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0015, Loss2: 0.0006, Pure Ratio1: 10.1176, Pure Ratio2 10.0980 +Epoch [46/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.8824, Pure Ratio2 9.8333 +Epoch [46/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 9.9935 +Epoch [46/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8333, Pure Ratio2 9.8775 +Epoch [46/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0471, Pure Ratio2 10.0902 +Epoch [46/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.9379, Pure Ratio2 9.9967 +Epoch [46/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0039, Loss2: 0.0041, Pure Ratio1: 9.9468, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 57.3317 % Model2 58.0329 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0427 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.0196, Pure Ratio2 9.9216 +Epoch [47/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9510, Pure Ratio2 9.8137 +Epoch [47/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.9542, Pure Ratio2 9.8366 +Epoch [47/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.8137, Pure Ratio2 9.7647 +Epoch [47/200], Iter [250/390] Training Accuracy1: 94.5312, Training Accuracy2: 93.7500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.8588 +Epoch [47/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.9510, Pure Ratio2 9.8954 +Epoch [47/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.9328, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 58.7640 % Model2 58.4535 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.4706, Pure Ratio2 9.3137 +Epoch [48/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.5392 +Epoch [48/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.6275, Pure Ratio2 9.5294 +Epoch [48/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0007, Pure Ratio1: 9.7206, Pure Ratio2 9.6961 +Epoch [48/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.7569, Pure Ratio2 9.7922 +Epoch [48/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8203, Pure Ratio2 9.8824 +Epoch [48/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8768, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 58.6138 % Model2 58.1831 %, Pure Ratio 1 9.9095 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.7647, Pure Ratio2 9.8431 +Epoch [49/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.0294 +Epoch [49/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9477 +Epoch [49/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0735 +Epoch [49/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 10.1255, Pure Ratio2 10.1216 +Epoch [49/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0719, Pure Ratio2 10.0621 +Epoch [49/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.0616, Pure Ratio2 10.0700 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 59.0044 % Model2 56.6907 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1569, Pure Ratio2 9.9020 +Epoch [50/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 10.1961, Pure Ratio2 10.0392 +Epoch [50/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 10.2484, Pure Ratio2 10.1176 +Epoch [50/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 10.2304, Pure Ratio2 10.1127 +Epoch [50/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 10.0902, Pure Ratio2 10.0314 +Epoch [50/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 82.0312, Loss1: 0.0021, Loss2: 0.0007, Pure Ratio1: 9.9314, Pure Ratio2 9.9085 +Epoch [50/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9664, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 57.1615 % Model2 58.6038 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.3529, Pure Ratio2 10.1569 +Epoch [51/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.1471, Pure Ratio2 10.0588 +Epoch [51/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9542, Pure Ratio2 9.9608 +Epoch [51/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.9951 +Epoch [51/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9451, Pure Ratio2 9.9529 +Epoch [51/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9935, Pure Ratio2 9.9967 +Epoch [51/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9804, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 58.1530 % Model2 57.3718 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.9020, Pure Ratio2 10.2353 +Epoch [52/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0011, Pure Ratio1: 9.9118, Pure Ratio2 9.9216 +Epoch [52/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9477 +Epoch [52/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.9510 +Epoch [52/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Epoch [52/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.9575, Pure Ratio2 9.9608 +Epoch [52/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.9328, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 57.1214 % Model2 57.9627 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9296 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [53/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0196 +Epoch [53/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0011, Pure Ratio1: 9.6667, Pure Ratio2 9.7843 +Epoch [53/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6275, Pure Ratio2 9.7500 +Epoch [53/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.5922, Pure Ratio2 9.7294 +Epoch [53/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7778, Pure Ratio2 9.9281 +Epoch [53/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 9.8095, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 58.4635 % Model2 54.3069 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.7647, Pure Ratio2 10.6667 +Epoch [54/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.3824, Pure Ratio2 10.2647 +Epoch [54/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0018, Loss2: 0.0008, Pure Ratio1: 10.3333, Pure Ratio2 10.2418 +Epoch [54/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.1520, Pure Ratio2 10.0588 +Epoch [54/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 9.9373 +Epoch [54/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.0392, Pure Ratio2 9.9118 +Epoch [54/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 86.7188, Loss1: 0.0010, Loss2: 0.0004, Pure Ratio1: 10.0084, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 57.3017 % Model2 57.0813 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0588, Pure Ratio2 10.0980 +Epoch [55/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.1961, Pure Ratio2 10.2157 +Epoch [55/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 10.0980, Pure Ratio2 10.0261 +Epoch [55/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.1127 +Epoch [55/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 10.0353, Pure Ratio2 10.0118 +Epoch [55/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8791, Pure Ratio2 9.9248 +Epoch [55/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9860, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 56.6306 % Model2 56.6907 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0007, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [56/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8431, Pure Ratio2 9.8627 +Epoch [56/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8105, Pure Ratio2 9.7712 +Epoch [56/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.7745, Pure Ratio2 9.7549 +Epoch [56/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8627, Pure Ratio2 9.8275 +Epoch [56/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.9739, Pure Ratio2 9.9608 +Epoch [56/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0010, Pure Ratio1: 10.0224, Pure Ratio2 9.9944 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 58.5837 % Model2 57.6322 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.5686 +Epoch [57/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9706, Pure Ratio2 10.0098 +Epoch [57/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 88.2812, Loss1: 0.0018, Loss2: 0.0003, Pure Ratio1: 9.8693, Pure Ratio2 9.9412 +Epoch [57/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0007, Pure Ratio1: 9.7304, Pure Ratio2 9.8627 +Epoch [57/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8980, Pure Ratio2 9.9961 +Epoch [57/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8497, Pure Ratio2 9.9641 +Epoch [57/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8487, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 59.1546 % Model2 54.9079 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 10.0654 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 10.0392, Pure Ratio2 10.0392 +Epoch [58/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9608, Pure Ratio2 10.0392 +Epoch [58/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.9085, Pure Ratio2 9.9281 +Epoch [58/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 10.0245 +Epoch [58/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9255, Pure Ratio2 9.9647 +Epoch [58/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.8758, Pure Ratio2 9.8562 +Epoch [58/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9104, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 58.0128 % Model2 56.6406 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8824 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.1961, Pure Ratio2 10.4020 +Epoch [59/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 10.2549 +Epoch [59/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.9559, Pure Ratio2 10.0833 +Epoch [59/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0004, Pure Ratio1: 9.9255, Pure Ratio2 10.0039 +Epoch [59/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9248, Pure Ratio2 9.9935 +Epoch [59/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 58.3433 % Model2 56.9311 %, Pure Ratio 1 9.9246 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.6275 +Epoch [60/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.8529 +Epoch [60/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.1046, Pure Ratio2 10.1307 +Epoch [60/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0392, Pure Ratio2 10.0490 +Epoch [60/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0510, Pure Ratio2 9.9804 +Epoch [60/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0621, Pure Ratio2 10.0131 +Epoch [60/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0476, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 57.1514 % Model2 55.9095 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0196, Pure Ratio2 9.7059 +Epoch [61/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0012, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.0098 +Epoch [61/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 10.2353, Pure Ratio2 10.1046 +Epoch [61/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 10.2402, Pure Ratio2 10.1373 +Epoch [61/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 9.9059 +Epoch [61/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8856, Pure Ratio2 9.8399 +Epoch [61/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9580, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 59.5152 % Model2 58.1631 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.8627 +Epoch [62/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.5980, Pure Ratio2 9.7843 +Epoch [62/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6993, Pure Ratio2 9.8301 +Epoch [62/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.7941, Pure Ratio2 9.8676 +Epoch [62/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8784, Pure Ratio2 9.9020 +Epoch [62/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.9444, Pure Ratio2 9.9837 +Epoch [62/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8992, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 57.2015 % Model2 57.8526 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 10.9020, Pure Ratio2 11.0196 +Epoch [63/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.4804 +Epoch [63/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0980 +Epoch [63/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0049, Pure Ratio2 10.0980 +Epoch [63/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0863, Pure Ratio2 10.1451 +Epoch [63/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0490, Pure Ratio2 10.0850 +Epoch [63/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0700, Pure Ratio2 10.0896 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 59.0946 % Model2 56.3702 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8039, Pure Ratio2 9.8824 +Epoch [64/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.7941, Pure Ratio2 9.9608 +Epoch [64/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.7320, Pure Ratio2 9.9281 +Epoch [64/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9069, Pure Ratio2 9.9902 +Epoch [64/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 10.0431, Pure Ratio2 10.0824 +Epoch [64/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0556 +Epoch [64/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9860, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 56.8510 % Model2 54.2368 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 10.0779 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.8039, Pure Ratio2 9.8627 +Epoch [65/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.8725 +Epoch [65/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8889, Pure Ratio2 9.9477 +Epoch [65/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8873, Pure Ratio2 9.9167 +Epoch [65/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.9725, Pure Ratio2 9.9608 +Epoch [65/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9444 +Epoch [65/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.9888, Pure Ratio2 9.9580 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 56.4103 % Model2 55.8193 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 10.6863, Pure Ratio2 10.4902 +Epoch [66/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 10.3431, Pure Ratio2 10.1961 +Epoch [66/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.3333, Pure Ratio2 10.3072 +Epoch [66/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9657, Pure Ratio2 9.9755 +Epoch [66/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.1020, Pure Ratio2 10.1529 +Epoch [66/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0294, Pure Ratio2 10.1013 +Epoch [66/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0756, Pure Ratio2 10.1148 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 58.5036 % Model2 58.2833 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [67/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.9688, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.9510 +Epoch [67/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9869, Pure Ratio2 9.9542 +Epoch [67/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.1127, Pure Ratio2 10.1373 +Epoch [67/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 10.0039, Pure Ratio2 10.0118 +Epoch [67/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0294, Pure Ratio2 10.0294 +Epoch [67/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 58.5136 % Model2 56.6106 %, Pure Ratio 1 10.0151 %, Pure Ratio 2 9.9975 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.6667 +Epoch [68/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0686, Pure Ratio2 9.9608 +Epoch [68/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9542 +Epoch [68/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7745 +Epoch [68/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9804, Pure Ratio2 10.0275 +Epoch [68/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 10.0621, Pure Ratio2 10.0784 +Epoch [68/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0056, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 56.1599 % Model2 56.4002 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0352 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.3725, Pure Ratio2 10.1765 +Epoch [69/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.3431, Pure Ratio2 10.1569 +Epoch [69/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 10.1569, Pure Ratio2 9.9804 +Epoch [69/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.9559 +Epoch [69/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8745, Pure Ratio2 9.8980 +Epoch [69/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8007, Pure Ratio2 9.8922 +Epoch [69/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8964, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 59.4852 % Model2 57.6723 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.5294, Pure Ratio2 9.9020 +Epoch [70/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6176, Pure Ratio2 9.8922 +Epoch [70/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.8824 +Epoch [70/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6961, Pure Ratio2 9.8088 +Epoch [70/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.7529, Pure Ratio2 9.8706 +Epoch [70/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.8987 +Epoch [70/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.9104, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 57.7023 % Model2 57.3417 %, Pure Ratio 1 9.8944 %, Pure Ratio 2 9.9950 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.3529, Pure Ratio2 10.5098 +Epoch [71/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.5000, Pure Ratio2 10.6373 +Epoch [71/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 10.2026, Pure Ratio2 10.3399 +Epoch [71/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.2696 +Epoch [71/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 10.0235 +Epoch [71/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8333, Pure Ratio2 9.9739 +Epoch [71/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 9.8739, Pure Ratio2 10.0336 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 56.7408 % Model2 57.6222 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.2157 +Epoch [72/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.4216, Pure Ratio2 10.3431 +Epoch [72/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.3595, Pure Ratio2 10.3203 +Epoch [72/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0441, Pure Ratio2 10.0392 +Epoch [72/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0510 +Epoch [72/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 10.0033 +Epoch [72/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8571, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 58.3834 % Model2 57.8325 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0392 +Epoch [73/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.7549 +Epoch [73/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9869, Pure Ratio2 10.0065 +Epoch [73/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.0049, Pure Ratio2 9.9853 +Epoch [73/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 10.1059, Pure Ratio2 10.0314 +Epoch [73/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0882, Pure Ratio2 10.0065 +Epoch [73/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.0924, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 57.5321 % Model2 58.4736 %, Pure Ratio 1 10.1031 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 10.4706 +Epoch [74/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0686 +Epoch [74/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 10.0131, Pure Ratio2 9.9935 +Epoch [74/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0637, Pure Ratio2 10.0833 +Epoch [74/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.0118, Pure Ratio2 9.9843 +Epoch [74/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 10.0229, Pure Ratio2 9.9641 +Epoch [74/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9860, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 57.6823 % Model2 57.8626 %, Pure Ratio 1 10.0000 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8824, Pure Ratio2 9.5098 +Epoch [75/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 92.9688, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.3235, Pure Ratio2 9.9608 +Epoch [75/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2353, Pure Ratio2 10.0654 +Epoch [75/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.1961, Pure Ratio2 10.1373 +Epoch [75/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.1804, Pure Ratio2 10.1373 +Epoch [75/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.1667, Pure Ratio2 10.0686 +Epoch [75/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0504, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 56.6907 % Model2 56.9010 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0196, Pure Ratio2 9.7843 +Epoch [76/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9510, Pure Ratio2 9.8235 +Epoch [76/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9150, Pure Ratio2 9.8235 +Epoch [76/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9755, Pure Ratio2 9.8873 +Epoch [76/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9059, Pure Ratio2 9.8118 +Epoch [76/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8333, Pure Ratio2 9.7059 +Epoch [76/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8683, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 58.1731 % Model2 58.1430 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 9.8824 +Epoch [77/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0196, Pure Ratio2 10.0196 +Epoch [77/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8497, Pure Ratio2 9.8693 +Epoch [77/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8922, Pure Ratio2 9.8725 +Epoch [77/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0039, Pure Ratio2 9.9961 +Epoch [77/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0327, Pure Ratio2 10.0425 +Epoch [77/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0924, Pure Ratio2 10.1008 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 58.9243 % Model2 57.7324 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9849 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.1569, Pure Ratio2 9.3333 +Epoch [78/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0033, Loss2: 0.0029, Pure Ratio1: 9.4706, Pure Ratio2 9.4902 +Epoch [78/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7516 +Epoch [78/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.0014, Loss2: 0.0001, Pure Ratio1: 9.7304, Pure Ratio2 9.6863 +Epoch [78/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.7216 +Epoch [78/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7908 +Epoch [78/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8880, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 59.8057 % Model2 56.4002 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.2745, Pure Ratio2 10.0588 +Epoch [79/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.8235 +Epoch [79/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 10.0131, Pure Ratio2 9.8824 +Epoch [79/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7304 +Epoch [79/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7569 +Epoch [79/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8725, Pure Ratio2 9.8268 +Epoch [79/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9272, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 59.8257 % Model2 58.3734 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.2157, Pure Ratio2 10.2157 +Epoch [80/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.6765 +Epoch [80/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.7582, Pure Ratio2 9.6993 +Epoch [80/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8235, Pure Ratio2 9.6814 +Epoch [80/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.9686, Pure Ratio2 9.8784 +Epoch [80/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 9.9477 +Epoch [80/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0392, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 58.1130 % Model2 55.1983 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [81/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.0294 +Epoch [81/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.2549 +Epoch [81/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0000, Pure Ratio1: 9.9559, Pure Ratio2 9.9412 +Epoch [81/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1020, Pure Ratio2 10.1020 +Epoch [81/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0490, Pure Ratio2 10.0000 +Epoch [81/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.9524, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 59.0244 % Model2 58.3734 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9824 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [82/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6569, Pure Ratio2 9.8333 +Epoch [82/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.7451, Pure Ratio2 9.8889 +Epoch [82/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.8676 +Epoch [82/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9098, Pure Ratio2 10.0314 +Epoch [82/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 10.0588 +Epoch [82/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9832, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 57.0713 % Model2 56.7308 %, Pure Ratio 1 9.9296 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.4510, Pure Ratio2 10.3529 +Epoch [83/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1078, Pure Ratio2 10.0882 +Epoch [83/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9542, Pure Ratio2 10.0523 +Epoch [83/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8137, Pure Ratio2 9.8039 +Epoch [83/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8314, Pure Ratio2 9.8353 +Epoch [83/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.8791, Pure Ratio2 9.8987 +Epoch [83/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.9776, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 58.4535 % Model2 56.5204 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.1765 +Epoch [84/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0011, Pure Ratio1: 9.8137, Pure Ratio2 9.6176 +Epoch [84/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6536, Pure Ratio2 9.4771 +Epoch [84/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7696, Pure Ratio2 9.6618 +Epoch [84/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7529, Pure Ratio2 9.6275 +Epoch [84/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8791, Pure Ratio2 9.7810 +Epoch [84/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9748, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 58.4135 % Model2 56.8610 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9271 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.1373, Pure Ratio2 10.2353 +Epoch [85/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8725, Pure Ratio2 10.0686 +Epoch [85/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.9150 +Epoch [85/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0010, Pure Ratio1: 9.8922, Pure Ratio2 9.8284 +Epoch [85/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9569, Pure Ratio2 9.8588 +Epoch [85/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.9673, Pure Ratio2 9.8856 +Epoch [85/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9524, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 58.2332 % Model2 56.9812 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.0784 +Epoch [86/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.9216 +Epoch [86/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8105, Pure Ratio2 9.7778 +Epoch [86/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9951, Pure Ratio2 9.9314 +Epoch [86/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9569 +Epoch [86/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.8595, Pure Ratio2 9.8170 +Epoch [86/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9048, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 58.0429 % Model2 59.4251 %, Pure Ratio 1 10.0126 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.4510, Pure Ratio2 9.5294 +Epoch [87/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 9.7353 +Epoch [87/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9150, Pure Ratio2 9.8693 +Epoch [87/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8725, Pure Ratio2 9.9412 +Epoch [87/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.0078 +Epoch [87/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.9641, Pure Ratio2 10.0163 +Epoch [87/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 91.4062, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.9664, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 57.9227 % Model2 57.2716 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.4510 +Epoch [88/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.3137 +Epoch [88/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.2157, Pure Ratio2 10.3529 +Epoch [88/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.3235 +Epoch [88/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0627, Pure Ratio2 10.0824 +Epoch [88/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9183, Pure Ratio2 9.9346 +Epoch [88/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9496, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 58.0028 % Model2 58.5337 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0000 +Epoch [89/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8529, Pure Ratio2 9.8725 +Epoch [89/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2614, Pure Ratio2 10.2614 +Epoch [89/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0033, Loss2: 0.0008, Pure Ratio1: 10.1275, Pure Ratio2 10.1029 +Epoch [89/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 10.0235, Pure Ratio2 9.9765 +Epoch [89/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0294, Pure Ratio2 10.0229 +Epoch [89/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0756, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 56.5104 % Model2 56.4904 %, Pure Ratio 1 10.0980 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.5490, Pure Ratio2 10.6863 +Epoch [90/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.2255, Pure Ratio2 10.4216 +Epoch [90/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.3725 +Epoch [90/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 10.1961 +Epoch [90/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1294, Pure Ratio2 10.2196 +Epoch [90/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1111, Pure Ratio2 10.1699 +Epoch [90/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9804, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 58.3534 % Model2 56.7608 %, Pure Ratio 1 9.9573 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.4118, Pure Ratio2 10.3137 +Epoch [91/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.4020, Pure Ratio2 10.3137 +Epoch [91/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1830, Pure Ratio2 10.0980 +Epoch [91/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.2304, Pure Ratio2 10.1520 +Epoch [91/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9686, Pure Ratio2 9.9176 +Epoch [91/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.8529 +Epoch [91/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0224, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 57.3518 % Model2 57.5020 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 10.0196 +Epoch [92/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.2745 +Epoch [92/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1176, Pure Ratio2 10.2157 +Epoch [92/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.1814 +Epoch [92/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0020, Loss2: 0.0013, Pure Ratio1: 9.8902, Pure Ratio2 9.9529 +Epoch [92/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9510, Pure Ratio2 9.9739 +Epoch [92/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9496, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 57.1014 % Model2 58.0729 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.3137, Pure Ratio2 10.6275 +Epoch [93/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8922, Pure Ratio2 10.1961 +Epoch [93/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9346, Pure Ratio2 10.0654 +Epoch [93/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9853, Pure Ratio2 10.1029 +Epoch [93/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 10.0118 +Epoch [93/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9248 +Epoch [93/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8964, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 58.4535 % Model2 58.0128 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.0588, Pure Ratio2 9.2157 +Epoch [94/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4314, Pure Ratio2 9.5588 +Epoch [94/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.4706, Pure Ratio2 9.5229 +Epoch [94/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.7304 +Epoch [94/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6549, Pure Ratio2 9.7529 +Epoch [94/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.6928, Pure Ratio2 9.7549 +Epoch [94/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 9.8487, Pure Ratio2 9.9244 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 57.9627 % Model2 57.0312 %, Pure Ratio 1 9.9070 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.2157 +Epoch [95/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.3824, Pure Ratio2 10.3627 +Epoch [95/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.1307, Pure Ratio2 10.0850 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.2990, Pure Ratio2 10.2647 +Epoch [95/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [95/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1503, Pure Ratio2 10.1601 +Epoch [95/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0924, Pure Ratio2 10.1232 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 55.9896 % Model2 57.3818 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 10.0302 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2549, Pure Ratio2 9.5294 +Epoch [96/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.8627 +Epoch [96/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.7908 +Epoch [96/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 9.6324, Pure Ratio2 9.7598 +Epoch [96/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8078, Pure Ratio2 9.8706 +Epoch [96/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8856, Pure Ratio2 9.8954 +Epoch [96/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9300, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 59.6354 % Model2 57.8325 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.5490 +Epoch [97/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8039, Pure Ratio2 9.6765 +Epoch [97/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.9804 +Epoch [97/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9020 +Epoch [97/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8549, Pure Ratio2 9.8549 +Epoch [97/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8856, Pure Ratio2 9.9379 +Epoch [97/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9188, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 57.6222 % Model2 57.6823 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7843, Pure Ratio2 9.7647 +Epoch [98/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 9.9706 +Epoch [98/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8758, Pure Ratio2 9.7778 +Epoch [98/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.0343, Pure Ratio2 9.9069 +Epoch [98/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 9.9294, Pure Ratio2 9.8627 +Epoch [98/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 9.9248 +Epoch [98/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0308, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 56.2099 % Model2 57.9327 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1569, Pure Ratio2 10.0784 +Epoch [99/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0490, Pure Ratio2 10.1275 +Epoch [99/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.1830 +Epoch [99/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0784 +Epoch [99/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.1451 +Epoch [99/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.9281, Pure Ratio2 10.0490 +Epoch [99/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8123, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 55.3385 % Model2 57.3017 %, Pure Ratio 1 9.8140 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 10.0000 +Epoch [100/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.1275 +Epoch [100/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.8235, Pure Ratio2 9.9412 +Epoch [100/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7206, Pure Ratio2 9.8039 +Epoch [100/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8667, Pure Ratio2 9.9020 +Epoch [100/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8791, Pure Ratio2 9.9510 +Epoch [100/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8936, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 58.2833 % Model2 56.8610 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.1765, Pure Ratio2 9.8824 +Epoch [101/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.8137 +Epoch [101/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.0523 +Epoch [101/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [101/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9176, Pure Ratio2 9.9882 +Epoch [101/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9967, Pure Ratio2 9.9804 +Epoch [101/200], Iter [350/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 57.6422 % Model2 56.6206 %, Pure Ratio 1 9.9723 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2549, Pure Ratio2 10.0196 +Epoch [102/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.7157 +Epoch [102/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.9085 +Epoch [102/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0490, Pure Ratio2 9.9706 +Epoch [102/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0078, Pure Ratio2 10.0118 +Epoch [102/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9673, Pure Ratio2 9.9771 +Epoch [102/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0336, Pure Ratio2 10.0756 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 58.1631 % Model2 57.9627 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9412 +Epoch [103/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.9510 +Epoch [103/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9085, Pure Ratio2 10.0196 +Epoch [103/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0882, Pure Ratio2 10.1127 +Epoch [103/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9059, Pure Ratio2 9.9412 +Epoch [103/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9477 +Epoch [103/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8768, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 56.4804 % Model2 58.5637 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9120 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 10.0000 +Epoch [104/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.7843 +Epoch [104/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8170, Pure Ratio2 10.0458 +Epoch [104/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.1863 +Epoch [104/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0627, Pure Ratio2 10.1529 +Epoch [104/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9935, Pure Ratio2 10.0784 +Epoch [104/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 10.0560 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 55.9295 % Model2 56.2400 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 10.0553 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.4314 +Epoch [105/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.1176 +Epoch [105/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0523, Pure Ratio2 10.0850 +Epoch [105/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8480, Pure Ratio2 9.9314 +Epoch [105/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9961, Pure Ratio2 10.0784 +Epoch [105/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9967 +Epoch [105/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9468, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 57.9227 % Model2 58.3333 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.6078 +Epoch [106/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.5686 +Epoch [106/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8301, Pure Ratio2 9.7516 +Epoch [106/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.7941 +Epoch [106/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8118 +Epoch [106/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 10.0098, Pure Ratio2 9.9412 +Epoch [106/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0308, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 57.3217 % Model2 57.3518 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.5686 +Epoch [107/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.5490, Pure Ratio2 10.5980 +Epoch [107/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1634 +Epoch [107/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.2990, Pure Ratio2 10.3382 +Epoch [107/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2000, Pure Ratio2 10.1843 +Epoch [107/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.1275 +Epoch [107/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 9.9776 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 57.6122 % Model2 57.2917 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.8235 +Epoch [108/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [108/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7516, Pure Ratio2 9.7451 +Epoch [108/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7206, Pure Ratio2 9.7892 +Epoch [108/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.8157 +Epoch [108/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8987, Pure Ratio2 9.9510 +Epoch [108/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9076, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 58.6639 % Model2 56.6807 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.5098, Pure Ratio2 10.3333 +Epoch [109/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2451 +Epoch [109/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 10.2614, Pure Ratio2 10.2288 +Epoch [109/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2304, Pure Ratio2 10.1667 +Epoch [109/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 10.0275 +Epoch [109/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0850, Pure Ratio2 10.0294 +Epoch [109/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0448, Pure Ratio2 10.0252 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 58.0629 % Model2 58.6238 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.5490, Pure Ratio2 10.3922 +Epoch [110/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.3824, Pure Ratio2 10.3333 +Epoch [110/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1046, Pure Ratio2 9.9869 +Epoch [110/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.0784 +Epoch [110/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1882, Pure Ratio2 10.1020 +Epoch [110/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0752, Pure Ratio2 9.9967 +Epoch [110/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0028, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 57.7224 % Model2 56.3802 %, Pure Ratio 1 10.0075 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Epoch [111/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1176 +Epoch [111/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9085, Pure Ratio2 9.9085 +Epoch [111/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.9510, Pure Ratio2 9.9804 +Epoch [111/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9373, Pure Ratio2 9.9255 +Epoch [111/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9314, Pure Ratio2 9.9281 +Epoch [111/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9468, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 58.8442 % Model2 58.7039 %, Pure Ratio 1 9.9372 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 9.9216 +Epoch [112/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.9118 +Epoch [112/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.2941 +Epoch [112/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.2598, Pure Ratio2 10.3627 +Epoch [112/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0824, Pure Ratio2 10.1765 +Epoch [112/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 10.0719 +Epoch [112/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9720, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 57.2416 % Model2 57.9627 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.5686, Pure Ratio2 10.4706 +Epoch [113/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.0882 +Epoch [113/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0719, Pure Ratio2 9.9216 +Epoch [113/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.1422, Pure Ratio2 10.0000 +Epoch [113/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 9.9333 +Epoch [113/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.9118 +Epoch [113/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9944, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 57.6222 % Model2 56.9411 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.0980, Pure Ratio2 9.8627 +Epoch [114/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 92.1875, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9412 +Epoch [114/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.1634, Pure Ratio2 10.0196 +Epoch [114/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1716, Pure Ratio2 10.0490 +Epoch [114/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2392, Pure Ratio2 10.0902 +Epoch [114/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0010, Pure Ratio1: 10.1242, Pure Ratio2 10.0000 +Epoch [114/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0644, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 57.4219 % Model2 58.3133 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.8039 +Epoch [115/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0098, Pure Ratio2 9.8235 +Epoch [115/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2614, Pure Ratio2 10.1699 +Epoch [115/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0147 +Epoch [115/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.9686, Pure Ratio2 9.9804 +Epoch [115/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8987 +Epoch [115/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 57.3317 % Model2 58.6438 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.6275, Pure Ratio2 9.4902 +Epoch [116/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0294, Pure Ratio2 9.9902 +Epoch [116/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6797, Pure Ratio2 9.6340 +Epoch [116/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.7598 +Epoch [116/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9373, Pure Ratio2 9.8745 +Epoch [116/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.9510 +Epoch [116/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9384, Pure Ratio2 9.9748 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 57.5020 % Model2 58.2732 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7843 +Epoch [117/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8922 +Epoch [117/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 9.9804 +Epoch [117/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0539, Pure Ratio2 9.9608 +Epoch [117/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.8627 +Epoch [117/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 9.8922 +Epoch [117/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0672, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 58.4736 % Model2 57.9026 %, Pure Ratio 1 10.0654 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8039 +Epoch [118/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 10.0686 +Epoch [118/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.0588 +Epoch [118/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.0735 +Epoch [118/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9569, Pure Ratio2 10.1020 +Epoch [118/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0327, Pure Ratio2 10.1569 +Epoch [118/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 10.1289 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 58.5236 % Model2 57.4219 %, Pure Ratio 1 9.9598 %, Pure Ratio 2 10.0754 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3333 +Epoch [119/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.5196, Pure Ratio2 10.5588 +Epoch [119/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2745, Pure Ratio2 10.2941 +Epoch [119/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2206 +Epoch [119/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1804, Pure Ratio2 10.1647 +Epoch [119/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1307, Pure Ratio2 10.1111 +Epoch [119/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0476, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 58.4235 % Model2 57.1114 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.5098 +Epoch [120/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 10.0392 +Epoch [120/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 10.0980 +Epoch [120/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9167, Pure Ratio2 10.0343 +Epoch [120/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8471, Pure Ratio2 10.0000 +Epoch [120/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.0915 +Epoch [120/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 59.4251 % Model2 58.6138 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 10.0679 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [121/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8235 +Epoch [121/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.6928 +Epoch [121/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.8873 +Epoch [121/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.7922 +Epoch [121/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9608 +Epoch [121/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9888, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 56.3401 % Model2 57.8025 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 9.8431 +Epoch [122/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0294, Pure Ratio2 9.9412 +Epoch [122/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0327 +Epoch [122/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7892, Pure Ratio2 9.7892 +Epoch [122/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.8078, Pure Ratio2 9.8275 +Epoch [122/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8954 +Epoch [122/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 58.8141 % Model2 58.0329 %, Pure Ratio 1 9.9899 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5882, Pure Ratio2 9.8039 +Epoch [123/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 10.3235 +Epoch [123/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 10.2549 +Epoch [123/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9902, Pure Ratio2 10.2255 +Epoch [123/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8941, Pure Ratio2 10.0392 +Epoch [123/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.9641 +Epoch [123/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8992, Pure Ratio2 10.0672 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 58.6538 % Model2 58.2933 %, Pure Ratio 1 9.8894 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 10.2157 +Epoch [124/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 10.0294 +Epoch [124/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1634 +Epoch [124/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2059, Pure Ratio2 10.2108 +Epoch [124/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1294, Pure Ratio2 10.1608 +Epoch [124/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0425, Pure Ratio2 10.0458 +Epoch [124/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9636, Pure Ratio2 9.9468 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 57.4820 % Model2 57.7524 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0000, Pure Ratio2 10.1176 +Epoch [125/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.1373 +Epoch [125/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2092, Pure Ratio2 10.1046 +Epoch [125/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0686, Pure Ratio2 9.9657 +Epoch [125/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9922 +Epoch [125/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 10.1667, Pure Ratio2 10.0882 +Epoch [125/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0140, Pure Ratio2 9.9188 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 58.9643 % Model2 57.8425 %, Pure Ratio 1 10.0478 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4118, Pure Ratio2 9.4118 +Epoch [126/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5000, Pure Ratio2 9.3922 +Epoch [126/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.4052, Pure Ratio2 9.3856 +Epoch [126/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6127, Pure Ratio2 9.6176 +Epoch [126/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.6902 +Epoch [126/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8072 +Epoch [126/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 58.3634 % Model2 56.5605 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.4902, Pure Ratio2 10.2157 +Epoch [127/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7843 +Epoch [127/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 9.9608 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9020 +Epoch [127/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8392, Pure Ratio2 9.8353 +Epoch [127/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8889, Pure Ratio2 9.9150 +Epoch [127/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9160, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 56.4103 % Model2 57.5521 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.8627 +Epoch [128/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9314, Pure Ratio2 9.9314 +Epoch [128/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9869 +Epoch [128/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1324, Pure Ratio2 10.1471 +Epoch [128/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0745, Pure Ratio2 10.0824 +Epoch [128/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.1242 +Epoch [128/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 59.1046 % Model2 56.9611 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9723 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0007, Pure Ratio1: 9.8627, Pure Ratio2 9.5882 +Epoch [129/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.6373 +Epoch [129/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8301 +Epoch [129/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8775 +Epoch [129/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8588 +Epoch [129/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8954, Pure Ratio2 9.8954 +Epoch [129/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9776, Pure Ratio2 9.9608 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 58.1931 % Model2 58.6639 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9170 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.3725 +Epoch [130/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.8431 +Epoch [130/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.8105 +Epoch [130/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.6324 +Epoch [130/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.1490, Pure Ratio2 9.8941 +Epoch [130/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.8137 +Epoch [130/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0012, Pure Ratio1: 10.1092, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 57.7524 % Model2 57.2817 %, Pure Ratio 1 10.0905 %, Pure Ratio 2 9.8743 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.5686 +Epoch [131/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.7941 +Epoch [131/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0784, Pure Ratio2 9.9804 +Epoch [131/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.1127, Pure Ratio2 10.0245 +Epoch [131/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0627, Pure Ratio2 9.9843 +Epoch [131/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9935 +Epoch [131/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 57.4820 % Model2 58.5938 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5294 +Epoch [132/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.6275 +Epoch [132/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.6275 +Epoch [132/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9559, Pure Ratio2 9.8431 +Epoch [132/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8078 +Epoch [132/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8203 +Epoch [132/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9440, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 57.6322 % Model2 56.6506 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.1373 +Epoch [133/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.2255 +Epoch [133/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0196 +Epoch [133/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.8971 +Epoch [133/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0353, Pure Ratio2 10.0353 +Epoch [133/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9183 +Epoch [133/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9944, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 57.5321 % Model2 57.4720 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.6863 +Epoch [134/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0980 +Epoch [134/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.8562 +Epoch [134/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.8480 +Epoch [134/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 10.0314 +Epoch [134/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.9837 +Epoch [134/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 57.7424 % Model2 58.1230 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.9774 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.3922 +Epoch [135/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2843, Pure Ratio2 10.4118 +Epoch [135/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2026, Pure Ratio2 10.2353 +Epoch [135/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 10.0637 +Epoch [135/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9451 +Epoch [135/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0163, Pure Ratio2 10.0327 +Epoch [135/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 57.8125 % Model2 57.2316 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.4118, Pure Ratio2 10.2745 +Epoch [136/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 9.9412 +Epoch [136/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7974 +Epoch [136/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.8382 +Epoch [136/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7647 +Epoch [136/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8758 +Epoch [136/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9328, Pure Ratio2 9.9440 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 57.3818 % Model2 58.3534 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0980 +Epoch [137/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [137/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.0719 +Epoch [137/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0000 +Epoch [137/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0017, Pure Ratio1: 9.9804, Pure Ratio2 10.0353 +Epoch [137/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0294, Pure Ratio2 10.0850 +Epoch [137/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9972, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 56.1999 % Model2 56.6807 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 10.0126 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 10.2941, Pure Ratio2 10.5098 +Epoch [138/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0015, Pure Ratio1: 9.7843, Pure Ratio2 10.0588 +Epoch [138/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 10.0131 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9706, Pure Ratio2 10.0882 +Epoch [138/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 10.0902 +Epoch [138/200], Iter [300/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 10.1275 +Epoch [138/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8683, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 57.3618 % Model2 56.9611 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 10.0327 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 10.2157 +Epoch [139/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 9.6078, Pure Ratio2 9.9118 +Epoch [139/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.8954 +Epoch [139/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7892, Pure Ratio2 9.9314 +Epoch [139/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 10.0784 +Epoch [139/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8464, Pure Ratio2 9.9967 +Epoch [139/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9832, Pure Ratio2 10.1513 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 57.9026 % Model2 57.5020 %, Pure Ratio 1 9.8919 %, Pure Ratio 2 10.0528 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.6667 +Epoch [140/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9510 +Epoch [140/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 10.0523 +Epoch [140/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8725 +Epoch [140/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9098, Pure Ratio2 9.9647 +Epoch [140/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.9608 +Epoch [140/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 58.0128 % Model2 57.5020 %, Pure Ratio 1 9.9221 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5882, Pure Ratio2 10.3725 +Epoch [141/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2059, Pure Ratio2 10.0490 +Epoch [141/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.2680, Pure Ratio2 10.1569 +Epoch [141/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 9.8971 +Epoch [141/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 9.8235 +Epoch [141/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9673, Pure Ratio2 9.8529 +Epoch [141/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 57.3417 % Model2 59.4752 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1176 +Epoch [142/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 10.0196 +Epoch [142/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 10.0392 +Epoch [142/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9755, Pure Ratio2 9.9804 +Epoch [142/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9647, Pure Ratio2 10.0000 +Epoch [142/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 10.0098 +Epoch [142/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.9524 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 57.4820 % Model2 57.9026 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 10.0392 +Epoch [143/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.9118 +Epoch [143/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9281 +Epoch [143/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0294 +Epoch [143/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9569, Pure Ratio2 10.0667 +Epoch [143/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9150, Pure Ratio2 9.9902 +Epoch [143/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 59.8758 % Model2 59.0244 %, Pure Ratio 1 9.9346 %, Pure Ratio 2 10.0402 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0784 +Epoch [144/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5392, Pure Ratio2 10.4216 +Epoch [144/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2680, Pure Ratio2 10.2484 +Epoch [144/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0833, Pure Ratio2 10.1176 +Epoch [144/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 10.0941, Pure Ratio2 10.1098 +Epoch [144/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.1078 +Epoch [144/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9720, Pure Ratio2 10.0476 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 56.1198 % Model2 57.3518 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 10.0201 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.0784 +Epoch [145/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.5294 +Epoch [145/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.8431 +Epoch [145/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8873, Pure Ratio2 9.9363 +Epoch [145/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 10.0392 +Epoch [145/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9444, Pure Ratio2 9.9967 +Epoch [145/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9244, Pure Ratio2 9.9832 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 59.3450 % Model2 59.5453 %, Pure Ratio 1 9.9497 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0980, Pure Ratio2 9.7451 +Epoch [146/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 10.0686, Pure Ratio2 9.8922 +Epoch [146/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.8562 +Epoch [146/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9559 +Epoch [146/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9882, Pure Ratio2 9.9529 +Epoch [146/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9346 +Epoch [146/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 57.8526 % Model2 57.9527 %, Pure Ratio 1 9.9170 %, Pure Ratio 2 9.9573 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.3529 +Epoch [147/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6765, Pure Ratio2 9.6275 +Epoch [147/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.8627 +Epoch [147/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9020 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 9.8980 +Epoch [147/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0163, Pure Ratio2 9.9608 +Epoch [147/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.9720 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 58.1731 % Model2 57.6322 %, Pure Ratio 1 9.9774 %, Pure Ratio 2 9.9623 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0000 +Epoch [148/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8725 +Epoch [148/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7778 +Epoch [148/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.8676 +Epoch [148/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0353 +Epoch [148/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 10.0327 +Epoch [148/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8992, Pure Ratio2 9.9860 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 57.2817 % Model2 58.4535 %, Pure Ratio 1 9.9749 %, Pure Ratio 2 10.0578 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2353, Pure Ratio2 9.2745 +Epoch [149/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5784, Pure Ratio2 9.5882 +Epoch [149/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.7059 +Epoch [149/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7696, Pure Ratio2 9.8480 +Epoch [149/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9451, Pure Ratio2 9.9529 +Epoch [149/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8497 +Epoch [149/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0022, Loss2: 0.0001, Pure Ratio1: 9.8459, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 57.6122 % Model2 58.5136 %, Pure Ratio 1 9.9196 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.3529 +Epoch [150/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.2059, Pure Ratio2 10.1667 +Epoch [150/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.8758 +Epoch [150/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 9.9902 +Epoch [150/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0275, Pure Ratio2 9.9490 +Epoch [150/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9739 +Epoch [150/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 56.4904 % Model2 58.3834 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.0784 +Epoch [151/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0196 +Epoch [151/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.1830 +Epoch [151/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0147 +Epoch [151/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0392, Pure Ratio2 10.0627 +Epoch [151/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9641, Pure Ratio2 9.9869 +Epoch [151/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 58.2031 % Model2 57.6623 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8431 +Epoch [152/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7353, Pure Ratio2 9.8137 +Epoch [152/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9477, Pure Ratio2 9.9477 +Epoch [152/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8431 +Epoch [152/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7333 +Epoch [152/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7647 +Epoch [152/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8964, Pure Ratio2 9.9020 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 57.3117 % Model2 57.4219 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.5294 +Epoch [153/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3922, Pure Ratio2 10.4412 +Epoch [153/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3333, Pure Ratio2 10.3137 +Epoch [153/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3284, Pure Ratio2 10.2598 +Epoch [153/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1608, Pure Ratio2 10.0784 +Epoch [153/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0948, Pure Ratio2 10.0131 +Epoch [153/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0616, Pure Ratio2 9.9692 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 56.3502 % Model2 56.6006 %, Pure Ratio 1 10.0804 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 10.0000 +Epoch [154/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.1765 +Epoch [154/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.1373 +Epoch [154/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0441 +Epoch [154/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9882 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9575 +Epoch [154/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9496, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 57.8425 % Model2 58.5036 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 10.0628 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.2157 +Epoch [155/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 10.0294 +Epoch [155/200], Iter [150/390] Training Accuracy1: 96.8750, Training Accuracy2: 96.8750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.8170 +Epoch [155/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8578 +Epoch [155/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9765, Pure Ratio2 10.0275 +Epoch [155/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0556 +Epoch [155/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 57.3618 % Model2 58.3834 %, Pure Ratio 1 9.9397 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8431 +Epoch [156/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.0196 +Epoch [156/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1503, Pure Ratio2 10.2026 +Epoch [156/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3088, Pure Ratio2 10.2843 +Epoch [156/200], Iter [250/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2039, Pure Ratio2 10.2745 +Epoch [156/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.2157 +Epoch [156/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0140, Pure Ratio2 10.0868 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 58.4836 % Model2 58.3534 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9548 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6078 +Epoch [157/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.6765 +Epoch [157/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7843 +Epoch [157/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.7990 +Epoch [157/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.8431 +Epoch [157/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8399 +Epoch [157/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9748, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 57.3718 % Model2 57.4119 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9045 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.1176 +Epoch [158/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2647, Pure Ratio2 10.0392 +Epoch [158/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.9020 +Epoch [158/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8676 +Epoch [158/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.8745 +Epoch [158/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8431 +Epoch [158/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8599, Pure Ratio2 9.8431 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 56.5705 % Model2 58.5337 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9447 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2157 +Epoch [159/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 9.9608 +Epoch [159/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.8431 +Epoch [159/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.9657 +Epoch [159/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8863, Pure Ratio2 9.9647 +Epoch [159/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9444, Pure Ratio2 10.0098 +Epoch [159/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8880, Pure Ratio2 9.9328 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 58.3433 % Model2 57.9728 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.9397 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.4706 +Epoch [160/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.3824 +Epoch [160/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1895, Pure Ratio2 10.2288 +Epoch [160/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2451, Pure Ratio2 10.1569 +Epoch [160/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1647, Pure Ratio2 10.1098 +Epoch [160/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 9.9869 +Epoch [160/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0056, Pure Ratio2 9.9636 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 58.3934 % Model2 57.9928 %, Pure Ratio 1 10.0402 %, Pure Ratio 2 10.0251 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0784 +Epoch [161/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0294 +Epoch [161/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7778 +Epoch [161/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8578 +Epoch [161/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 10.0392 +Epoch [161/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9837, Pure Ratio2 9.9902 +Epoch [161/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 58.4936 % Model2 57.5921 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 9.9522 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8039 +Epoch [162/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.6961 +Epoch [162/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 9.8562 +Epoch [162/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 9.9657 +Epoch [162/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1333, Pure Ratio2 10.1451 +Epoch [162/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0556, Pure Ratio2 10.0654 +Epoch [162/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0644, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 58.1931 % Model2 57.6522 %, Pure Ratio 1 10.0327 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8627 +Epoch [163/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7549 +Epoch [163/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9150, Pure Ratio2 9.9216 +Epoch [163/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.9265 +Epoch [163/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8667, Pure Ratio2 9.8157 +Epoch [163/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9412 +Epoch [163/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9888, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 58.5737 % Model2 57.8526 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7843, Pure Ratio2 9.7647 +Epoch [164/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8922 +Epoch [164/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8366 +Epoch [164/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7598, Pure Ratio2 9.7549 +Epoch [164/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9333 +Epoch [164/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8693, Pure Ratio2 9.8693 +Epoch [164/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 57.9227 % Model2 58.3033 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 10.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 9.9216 +Epoch [165/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.8235 +Epoch [165/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0010, Pure Ratio1: 9.8170, Pure Ratio2 9.7320 +Epoch [165/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8039 +Epoch [165/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8941 +Epoch [165/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9150 +Epoch [165/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9188, Pure Ratio2 9.9356 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 57.5721 % Model2 58.6639 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6667 +Epoch [166/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 10.0686 +Epoch [166/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.8693 +Epoch [166/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0392 +Epoch [166/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 10.0196 +Epoch [166/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.9902 +Epoch [166/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 10.0280 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 57.9527 % Model2 58.1731 %, Pure Ratio 1 9.9271 %, Pure Ratio 2 10.0151 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.7059 +Epoch [167/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0196, Pure Ratio2 9.7843 +Epoch [167/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8758 +Epoch [167/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.6618 +Epoch [167/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.8745 +Epoch [167/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.9150 +Epoch [167/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0084, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 58.1130 % Model2 56.6506 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9321 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6275, Pure Ratio2 10.5882 +Epoch [168/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.2059 +Epoch [168/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1830, Pure Ratio2 10.1438 +Epoch [168/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1716, Pure Ratio2 10.2059 +Epoch [168/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2392, Pure Ratio2 10.2314 +Epoch [168/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2778, Pure Ratio2 10.2386 +Epoch [168/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1821, Pure Ratio2 10.1457 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 58.7640 % Model2 59.2949 %, Pure Ratio 1 10.0176 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1569 +Epoch [169/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0490, Pure Ratio2 10.1078 +Epoch [169/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0458, Pure Ratio2 10.0784 +Epoch [169/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 9.9902 +Epoch [169/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.8980 +Epoch [169/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9314 +Epoch [169/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9132, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 57.8025 % Model2 58.1330 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.9749 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.2745, Pure Ratio2 8.9216 +Epoch [170/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.6765 +Epoch [170/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8301 +Epoch [170/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8775 +Epoch [170/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9098 +Epoch [170/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9379, Pure Ratio2 9.9542 +Epoch [170/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0084 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 57.7624 % Model2 58.6639 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 10.0226 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.5882 +Epoch [171/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.6569 +Epoch [171/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7516, Pure Ratio2 9.7974 +Epoch [171/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8775 +Epoch [171/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9176, Pure Ratio2 9.9804 +Epoch [171/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9248, Pure Ratio2 9.9216 +Epoch [171/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9664 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 58.4635 % Model2 58.0429 %, Pure Ratio 1 9.9145 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.5882, Pure Ratio2 10.4510 +Epoch [172/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4020, Pure Ratio2 10.3922 +Epoch [172/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.2157 +Epoch [172/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0686, Pure Ratio2 10.0245 +Epoch [172/200], Iter [250/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0745 +Epoch [172/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9967, Pure Ratio2 9.9869 +Epoch [172/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9356, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 59.2648 % Model2 58.1731 %, Pure Ratio 1 9.9045 %, Pure Ratio 2 9.8793 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7059 +Epoch [173/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.4608 +Epoch [173/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.8497 +Epoch [173/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9216 +Epoch [173/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8353, Pure Ratio2 9.7412 +Epoch [173/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.7810 +Epoch [173/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8179, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 58.0329 % Model2 58.7139 %, Pure Ratio 1 9.9321 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.3725 +Epoch [174/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.7549 +Epoch [174/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.8693 +Epoch [174/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9853 +Epoch [174/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0549, Pure Ratio2 10.0157 +Epoch [174/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.8922 +Epoch [174/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9048, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 57.8125 % Model2 58.0128 %, Pure Ratio 1 9.9925 %, Pure Ratio 2 9.9095 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.8627 +Epoch [175/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8137 +Epoch [175/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.7974 +Epoch [175/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.9167 +Epoch [175/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.9490 +Epoch [175/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0490 +Epoch [175/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.0112 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 58.0329 % Model2 58.0929 %, Pure Ratio 1 10.0101 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.0196 +Epoch [176/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.0882 +Epoch [176/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1699, Pure Ratio2 10.0327 +Epoch [176/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1618, Pure Ratio2 10.0490 +Epoch [176/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0275 +Epoch [176/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8595 +Epoch [176/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9692, Pure Ratio2 9.9412 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 58.4335 % Model2 58.8141 %, Pure Ratio 1 9.9824 %, Pure Ratio 2 9.9346 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.1176 +Epoch [177/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.7059 +Epoch [177/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6209 +Epoch [177/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9118, Pure Ratio2 9.7843 +Epoch [177/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7098 +Epoch [177/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8431 +Epoch [177/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8263, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 57.7724 % Model2 58.1530 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3137, Pure Ratio2 10.2745 +Epoch [178/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2255, Pure Ratio2 10.1863 +Epoch [178/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 10.0523 +Epoch [178/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9951, Pure Ratio2 10.0980 +Epoch [178/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 10.0157 +Epoch [178/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.9150 +Epoch [178/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9076, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 58.3233 % Model2 58.5437 %, Pure Ratio 1 9.9623 %, Pure Ratio 2 9.9598 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.8235 +Epoch [179/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5392, Pure Ratio2 9.6667 +Epoch [179/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6928, Pure Ratio2 9.7255 +Epoch [179/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7843 +Epoch [179/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0353, Pure Ratio2 10.0745 +Epoch [179/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1863 +Epoch [179/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 10.0588 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 59.4251 % Model2 58.6338 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 9.8627 +Epoch [180/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1078 +Epoch [180/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9673, Pure Ratio2 9.9869 +Epoch [180/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.9020 +Epoch [180/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8902, Pure Ratio2 9.8745 +Epoch [180/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8627 +Epoch [180/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9104, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 58.0629 % Model2 58.7240 %, Pure Ratio 1 9.9648 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0784 +Epoch [181/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1765 +Epoch [181/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1438, Pure Ratio2 10.1569 +Epoch [181/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9265, Pure Ratio2 9.9363 +Epoch [181/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0471, Pure Ratio2 10.0314 +Epoch [181/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9641, Pure Ratio2 10.0065 +Epoch [181/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9776, Pure Ratio2 9.9804 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 58.3534 % Model2 58.4235 %, Pure Ratio 1 9.9698 %, Pure Ratio 2 9.9698 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6275, Pure Ratio2 10.5882 +Epoch [182/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2745, Pure Ratio2 10.1765 +Epoch [182/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2288, Pure Ratio2 10.1830 +Epoch [182/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0931, Pure Ratio2 10.0931 +Epoch [182/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9922, Pure Ratio2 9.9961 +Epoch [182/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8758, Pure Ratio2 9.8693 +Epoch [182/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9580, Pure Ratio2 9.9552 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 59.3650 % Model2 58.0529 %, Pure Ratio 1 9.9849 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5686 +Epoch [183/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.5490 +Epoch [183/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.7386 +Epoch [183/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0735, Pure Ratio2 10.0049 +Epoch [183/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0000 +Epoch [183/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0392 +Epoch [183/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0168, Pure Ratio2 9.9972 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 57.7123 % Model2 58.1731 %, Pure Ratio 1 9.9673 %, Pure Ratio 2 9.9372 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.1373 +Epoch [184/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.9216 +Epoch [184/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [184/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7353 +Epoch [184/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.8588 +Epoch [184/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 10.0327 +Epoch [184/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9748, Pure Ratio2 9.9916 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 58.5337 % Model2 57.9728 %, Pure Ratio 1 9.9422 %, Pure Ratio 2 9.9925 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.5686 +Epoch [185/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6863 +Epoch [185/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 10.0196 +Epoch [185/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9755, Pure Ratio2 10.0637 +Epoch [185/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8745, Pure Ratio2 9.9608 +Epoch [185/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9641 +Epoch [185/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 10.0420 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 58.5938 % Model2 58.0929 %, Pure Ratio 1 9.9950 %, Pure Ratio 2 10.0704 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.7647 +Epoch [186/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3725, Pure Ratio2 9.4510 +Epoch [186/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.6993 +Epoch [186/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.7990 +Epoch [186/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0863, Pure Ratio2 9.9569 +Epoch [186/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 9.9575 +Epoch [186/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1317, Pure Ratio2 10.0196 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 58.6639 % Model2 57.7724 %, Pure Ratio 1 10.0427 %, Pure Ratio 2 9.9497 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.6078, Pure Ratio2 10.8431 +Epoch [187/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4804, Pure Ratio2 10.4608 +Epoch [187/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0392 +Epoch [187/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1422 +Epoch [187/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0863 +Epoch [187/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0752, Pure Ratio2 9.9542 +Epoch [187/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0028, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 57.6522 % Model2 57.6122 %, Pure Ratio 1 9.9874 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.2353 +Epoch [188/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.0784 +Epoch [188/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9281 +Epoch [188/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8480, Pure Ratio2 9.8235 +Epoch [188/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9098 +Epoch [188/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 9.9575 +Epoch [188/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9944, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 57.9026 % Model2 57.9026 %, Pure Ratio 1 9.9447 %, Pure Ratio 2 9.9422 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.2549 +Epoch [189/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.0980 +Epoch [189/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0065, Pure Ratio2 9.9739 +Epoch [189/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9657, Pure Ratio2 9.9216 +Epoch [189/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9137, Pure Ratio2 9.9059 +Epoch [189/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.9085 +Epoch [189/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9888 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 58.2131 % Model2 57.5621 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9899 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.5098 +Epoch [190/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3333, Pure Ratio2 9.5294 +Epoch [190/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.9804 +Epoch [190/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9902 +Epoch [190/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.8588 +Epoch [190/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.9608 +Epoch [190/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 58.3534 % Model2 58.2833 %, Pure Ratio 1 9.9975 %, Pure Ratio 2 10.0075 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.1176 +Epoch [191/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.7647 +Epoch [191/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.9412 +Epoch [191/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.9951 +Epoch [191/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9294 +Epoch [191/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9739, Pure Ratio2 9.9281 +Epoch [191/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9832, Pure Ratio2 9.9272 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 58.5737 % Model2 58.3133 %, Pure Ratio 1 9.9799 %, Pure Ratio 2 9.9221 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5686, Pure Ratio2 10.4510 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9902, Pure Ratio2 10.0588 +Epoch [192/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8301, Pure Ratio2 9.9216 +Epoch [192/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8627 +Epoch [192/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7725, Pure Ratio2 9.8196 +Epoch [192/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8072, Pure Ratio2 9.9183 +Epoch [192/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 10.0532 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 58.9443 % Model2 59.2548 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 10.0101 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1961 +Epoch [193/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0588 +Epoch [193/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0196 +Epoch [193/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8186 +Epoch [193/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8863, Pure Ratio2 9.8784 +Epoch [193/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9575, Pure Ratio2 9.9085 +Epoch [193/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 58.4235 % Model2 59.0445 %, Pure Ratio 1 10.0050 %, Pure Ratio 2 9.9648 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.5882 +Epoch [194/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.0882 +Epoch [194/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0654, Pure Ratio2 10.0392 +Epoch [194/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0196, Pure Ratio2 9.9755 +Epoch [194/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0235, Pure Ratio2 10.0314 +Epoch [194/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0817, Pure Ratio2 10.1144 +Epoch [194/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 58.0829 % Model2 58.3734 %, Pure Ratio 1 9.9548 %, Pure Ratio 2 9.9673 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4314, Pure Ratio2 9.1569 +Epoch [195/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.2451 +Epoch [195/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.6144 +Epoch [195/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.7892 +Epoch [195/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7098, Pure Ratio2 9.6588 +Epoch [195/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7059 +Epoch [195/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9384, Pure Ratio2 9.9104 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 58.3233 % Model2 58.5537 %, Pure Ratio 1 10.0452 %, Pure Ratio 2 9.9799 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5490, Pure Ratio2 10.0980 +Epoch [196/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.6078, Pure Ratio2 10.2941 +Epoch [196/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3203, Pure Ratio2 10.0523 +Epoch [196/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 9.8725 +Epoch [196/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1451, Pure Ratio2 9.9804 +Epoch [196/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 9.9935 +Epoch [196/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1457, Pure Ratio2 10.0028 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 58.1831 % Model2 58.4535 %, Pure Ratio 1 10.0226 %, Pure Ratio 2 9.9246 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0000 +Epoch [197/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1863, Pure Ratio2 10.1863 +Epoch [197/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0588 +Epoch [197/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 10.0098 +Epoch [197/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0431, Pure Ratio2 10.0431 +Epoch [197/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0131, Pure Ratio2 9.9804 +Epoch [197/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9524, Pure Ratio2 9.9384 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 58.0829 % Model2 58.3433 %, Pure Ratio 1 9.8969 %, Pure Ratio 2 9.8969 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2157, Pure Ratio2 10.1373 +Epoch [198/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9020 +Epoch [198/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.8693 +Epoch [198/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 9.8186 +Epoch [198/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.7608 +Epoch [198/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8170, Pure Ratio2 9.8072 +Epoch [198/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8796, Pure Ratio2 9.8824 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 58.9744 % Model2 58.5938 %, Pure Ratio 1 9.9472 %, Pure Ratio 2 9.9472 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7059, Pure Ratio2 10.5882 +Epoch [199/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1471 +Epoch [199/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0784, Pure Ratio2 10.0065 +Epoch [199/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9706 +Epoch [199/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0118, Pure Ratio2 9.9961 +Epoch [199/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 9.9706 +Epoch [199/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0532, Pure Ratio2 10.0000 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 58.4635 % Model2 57.7724 %, Pure Ratio 1 10.0277 %, Pure Ratio 2 9.9145 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.7059, Pure Ratio2 10.6471 +Epoch [200/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5294, Pure Ratio2 10.5980 +Epoch [200/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.5098, Pure Ratio2 10.4118 +Epoch [200/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3627, Pure Ratio2 10.2941 +Epoch [200/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1490, Pure Ratio2 10.1333 +Epoch [200/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.1176 +Epoch [200/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0504, Pure Ratio2 10.0812 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 58.1430 % Model2 57.6923 %, Pure Ratio 1 10.0025 %, Pure Ratio 2 9.9899 % diff --git a/other_methods/coteaching/coteaching_results/out_6_6.log b/other_methods/coteaching/coteaching_results/out_6_6.log new file mode 100644 index 0000000..0e1f6e9 --- /dev/null +++ b/other_methods/coteaching/coteaching_results/out_6_6.log @@ -0,0 +1,2041 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test images: Model1 10.0060 % Model2 9.9860 % Pure Ratio1 0.0000 % Pure Ratio2 0.0000 % +Training cifar10_coteaching_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0153, Loss2: 0.0154, Pure Ratio1: 9.3120, Pure Ratio2 9.3760 +Epoch [2/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 27.3438, Loss1: 0.0159, Loss2: 0.0157, Pure Ratio1: 9.4400, Pure Ratio2 9.4640 +Epoch [2/200], Iter [150/390] Training Accuracy1: 22.6562, Training Accuracy2: 27.3438, Loss1: 0.0161, Loss2: 0.0162, Pure Ratio1: 9.4453, Pure Ratio2 9.4720 +Epoch [2/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0152, Loss2: 0.0151, Pure Ratio1: 9.6400, Pure Ratio2 9.6520 +Epoch [2/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0141, Loss2: 0.0144, Pure Ratio1: 9.6192, Pure Ratio2 9.6320 +Epoch [2/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0134, Loss2: 0.0139, Pure Ratio1: 9.6907, Pure Ratio2 9.6933 +Epoch [2/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0144, Loss2: 0.0141, Pure Ratio1: 9.7440, Pure Ratio2 9.7463 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test images: Model1 22.6462 % Model2 24.0585 %, Pure Ratio 1 9.7538 %, Pure Ratio 2 9.7621 % +Training cifar10_coteaching_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0128, Loss2: 0.0130, Pure Ratio1: 9.4098, Pure Ratio2 9.4590 +Epoch [3/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0144, Loss2: 0.0143, Pure Ratio1: 9.1557, Pure Ratio2 9.2213 +Epoch [3/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0144, Loss2: 0.0147, Pure Ratio1: 9.5902, Pure Ratio2 9.6503 +Epoch [3/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0134, Loss2: 0.0132, Pure Ratio1: 9.6393, Pure Ratio2 9.7049 +Epoch [3/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 32.8125, Loss1: 0.0143, Loss2: 0.0147, Pure Ratio1: 9.6131, Pure Ratio2 9.6885 +Epoch [3/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0135, Loss2: 0.0136, Pure Ratio1: 9.7322, Pure Ratio2 9.8060 +Epoch [3/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0137, Loss2: 0.0136, Pure Ratio1: 9.7119, Pure Ratio2 9.7728 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test images: Model1 21.2340 % Model2 19.2408 %, Pure Ratio 1 9.7205 %, Pure Ratio 2 9.7709 % +Training cifar10_coteaching_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0135, Loss2: 0.0135, Pure Ratio1: 8.8403, Pure Ratio2 8.8571 +Epoch [4/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0118, Loss2: 0.0121, Pure Ratio1: 9.3950, Pure Ratio2 9.4202 +Epoch [4/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0130, Pure Ratio1: 9.4454, Pure Ratio2 9.4398 +Epoch [4/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0144, Loss2: 0.0142, Pure Ratio1: 9.3992, Pure Ratio2 9.4202 +Epoch [4/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0124, Loss2: 0.0133, Pure Ratio1: 9.4723, Pure Ratio2 9.4824 +Epoch [4/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0133, Loss2: 0.0135, Pure Ratio1: 9.5322, Pure Ratio2 9.5294 +Epoch [4/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0123, Loss2: 0.0124, Pure Ratio1: 9.6639, Pure Ratio2 9.6447 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test images: Model1 20.1823 % Model2 18.8301 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7242 % +Training cifar10_coteaching_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0118, Loss2: 0.0118, Pure Ratio1: 9.7069, Pure Ratio2 9.7241 +Epoch [5/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0129, Loss2: 0.0133, Pure Ratio1: 10.0948, Pure Ratio2 10.0948 +Epoch [5/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0111, Loss2: 0.0113, Pure Ratio1: 10.1034, Pure Ratio2 10.1034 +Epoch [5/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0114, Loss2: 0.0122, Pure Ratio1: 10.0302, Pure Ratio2 10.0000 +Epoch [5/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0129, Pure Ratio1: 10.0310, Pure Ratio2 10.0345 +Epoch [5/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0112, Loss2: 0.0116, Pure Ratio1: 9.8678, Pure Ratio2 9.8621 +Epoch [5/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0143, Loss2: 0.0144, Pure Ratio1: 9.8177, Pure Ratio2 9.7956 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test images: Model1 24.1587 % Model2 22.2256 %, Pure Ratio 1 9.8099 %, Pure Ratio 2 9.7834 % +Training cifar10_coteaching_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0117, Loss2: 0.0119, Pure Ratio1: 9.7522, Pure Ratio2 9.8230 +Epoch [6/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0127, Loss2: 0.0132, Pure Ratio1: 10.3009, Pure Ratio2 10.2832 +Epoch [6/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 45.3125, Loss1: 0.0101, Loss2: 0.0107, Pure Ratio1: 9.9705, Pure Ratio2 10.0177 +Epoch [6/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0104, Loss2: 0.0109, Pure Ratio1: 9.8451, Pure Ratio2 9.8805 +Epoch [6/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0101, Loss2: 0.0101, Pure Ratio1: 9.7345, Pure Ratio2 9.7770 +Epoch [6/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0118, Loss2: 0.0119, Pure Ratio1: 9.8230, Pure Ratio2 9.8348 +Epoch [6/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0103, Loss2: 0.0106, Pure Ratio1: 9.7219, Pure Ratio2 9.7295 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test images: Model1 21.6747 % Model2 22.4359 %, Pure Ratio 1 9.6937 %, Pure Ratio 2 9.7141 % +Training cifar10_coteaching_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0107, Loss2: 0.0111, Pure Ratio1: 10.0000, Pure Ratio2 10.1273 +Epoch [7/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 31.2500, Loss1: 0.0119, Loss2: 0.0120, Pure Ratio1: 9.9182, Pure Ratio2 9.8273 +Epoch [7/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0116, Loss2: 0.0117, Pure Ratio1: 9.9758, Pure Ratio2 9.8727 +Epoch [7/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0109, Loss2: 0.0113, Pure Ratio1: 9.6500, Pure Ratio2 9.6318 +Epoch [7/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0112, Loss2: 0.0104, Pure Ratio1: 9.6509, Pure Ratio2 9.6655 +Epoch [7/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0094, Loss2: 0.0095, Pure Ratio1: 9.7212, Pure Ratio2 9.7242 +Epoch [7/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0115, Loss2: 0.0117, Pure Ratio1: 9.7351, Pure Ratio2 9.7351 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test images: Model1 24.1386 % Model2 21.9852 %, Pure Ratio 1 9.7949 %, Pure Ratio 2 9.8065 % +Training cifar10_coteaching_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0097, Loss2: 0.0094, Pure Ratio1: 9.8704, Pure Ratio2 9.9074 +Epoch [8/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0098, Loss2: 0.0098, Pure Ratio1: 9.3426, Pure Ratio2 9.4074 +Epoch [8/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0092, Loss2: 0.0098, Pure Ratio1: 9.2037, Pure Ratio2 9.2531 +Epoch [8/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0094, Loss2: 0.0091, Pure Ratio1: 9.1991, Pure Ratio2 9.1898 +Epoch [8/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0110, Loss2: 0.0101, Pure Ratio1: 9.3963, Pure Ratio2 9.3963 +Epoch [8/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0108, Loss2: 0.0111, Pure Ratio1: 9.5710, Pure Ratio2 9.5772 +Epoch [8/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0100, Loss2: 0.0101, Pure Ratio1: 9.6693, Pure Ratio2 9.6825 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test images: Model1 24.9099 % Model2 25.2103 %, Pure Ratio 1 9.7246 %, Pure Ratio 2 9.7270 % +Training cifar10_coteaching_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0101, Loss2: 0.0092, Pure Ratio1: 9.2762, Pure Ratio2 9.3143 +Epoch [9/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0096, Loss2: 0.0094, Pure Ratio1: 9.3048, Pure Ratio2 9.2667 +Epoch [9/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0093, Loss2: 0.0095, Pure Ratio1: 9.5873, Pure Ratio2 9.5937 +Epoch [9/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0117, Loss2: 0.0116, Pure Ratio1: 9.6238, Pure Ratio2 9.6190 +Epoch [9/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0100, Loss2: 0.0098, Pure Ratio1: 9.7676, Pure Ratio2 9.7448 +Epoch [9/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0091, Loss2: 0.0087, Pure Ratio1: 9.6317, Pure Ratio2 9.6254 +Epoch [9/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0101, Loss2: 0.0099, Pure Ratio1: 9.7061, Pure Ratio2 9.7088 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test images: Model1 22.4659 % Model2 22.7564 %, Pure Ratio 1 9.7729 %, Pure Ratio 2 9.7631 % +Training cifar10_coteaching_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0089, Loss2: 0.0091, Pure Ratio1: 10.1176, Pure Ratio2 10.1176 +Epoch [10/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0095, Loss2: 0.0097, Pure Ratio1: 9.5196, Pure Ratio2 9.5392 +Epoch [10/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.8438, Loss1: 0.0098, Loss2: 0.0104, Pure Ratio1: 9.4641, Pure Ratio2 9.4641 +Epoch [10/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0105, Loss2: 0.0103, Pure Ratio1: 9.5392, Pure Ratio2 9.5735 +Epoch [10/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0085, Loss2: 0.0083, Pure Ratio1: 9.6314, Pure Ratio2 9.6510 +Epoch [10/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0092, Loss2: 0.0089, Pure Ratio1: 9.7941, Pure Ratio2 9.8203 +Epoch [10/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0095, Loss2: 0.0102, Pure Ratio1: 9.7311, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test images: Model1 26.0216 % Model2 26.2019 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0072, Loss2: 0.0070, Pure Ratio1: 10.0784, Pure Ratio2 9.9804 +Epoch [11/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0089, Loss2: 0.0091, Pure Ratio1: 10.0784, Pure Ratio2 10.0490 +Epoch [11/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0083, Loss2: 0.0082, Pure Ratio1: 10.0784, Pure Ratio2 10.0392 +Epoch [11/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0080, Loss2: 0.0082, Pure Ratio1: 10.0098, Pure Ratio2 10.0049 +Epoch [11/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0084, Loss2: 0.0080, Pure Ratio1: 9.8196, Pure Ratio2 9.8549 +Epoch [11/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0095, Loss2: 0.0098, Pure Ratio1: 9.7353, Pure Ratio2 9.7712 +Epoch [11/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0083, Loss2: 0.0086, Pure Ratio1: 9.7031, Pure Ratio2 9.7003 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test images: Model1 23.6178 % Model2 25.9716 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0074, Loss2: 0.0078, Pure Ratio1: 9.2745, Pure Ratio2 9.5098 +Epoch [12/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0081, Loss2: 0.0078, Pure Ratio1: 9.5196, Pure Ratio2 9.6471 +Epoch [12/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0070, Loss2: 0.0067, Pure Ratio1: 9.4314, Pure Ratio2 9.5294 +Epoch [12/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0084, Loss2: 0.0079, Pure Ratio1: 9.5931, Pure Ratio2 9.6667 +Epoch [12/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0091, Loss2: 0.0086, Pure Ratio1: 9.5686, Pure Ratio2 9.6588 +Epoch [12/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0094, Loss2: 0.0093, Pure Ratio1: 9.6209, Pure Ratio2 9.6961 +Epoch [12/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0084, Loss2: 0.0079, Pure Ratio1: 9.6919, Pure Ratio2 9.7367 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test images: Model1 24.7296 % Model2 24.1987 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0077, Loss2: 0.0091, Pure Ratio1: 9.9020, Pure Ratio2 10.0392 +Epoch [13/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0088, Loss2: 0.0086, Pure Ratio1: 9.8824, Pure Ratio2 9.8922 +Epoch [13/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0086, Loss2: 0.0079, Pure Ratio1: 9.7451, Pure Ratio2 9.7516 +Epoch [13/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0084, Loss2: 0.0087, Pure Ratio1: 9.9216, Pure Ratio2 9.9118 +Epoch [13/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0077, Loss2: 0.0078, Pure Ratio1: 9.7843, Pure Ratio2 9.7843 +Epoch [13/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0119, Loss2: 0.0115, Pure Ratio1: 9.7680, Pure Ratio2 9.8235 +Epoch [13/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0076, Loss2: 0.0077, Pure Ratio1: 9.7871, Pure Ratio2 9.8599 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test images: Model1 23.8482 % Model2 23.6278 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0074, Loss2: 0.0080, Pure Ratio1: 8.9804, Pure Ratio2 8.9020 +Epoch [14/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0059, Loss2: 0.0062, Pure Ratio1: 9.3137, Pure Ratio2 9.3235 +Epoch [14/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0115, Loss2: 0.0112, Pure Ratio1: 9.4314, Pure Ratio2 9.4641 +Epoch [14/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0080, Loss2: 0.0078, Pure Ratio1: 9.7304, Pure Ratio2 9.7598 +Epoch [14/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0090, Loss2: 0.0086, Pure Ratio1: 9.6824, Pure Ratio2 9.6784 +Epoch [14/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0076, Loss2: 0.0078, Pure Ratio1: 9.6340, Pure Ratio2 9.6209 +Epoch [14/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0087, Loss2: 0.0084, Pure Ratio1: 9.6807, Pure Ratio2 9.6751 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test images: Model1 23.9984 % Model2 24.3790 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0065, Loss2: 0.0067, Pure Ratio1: 9.3137, Pure Ratio2 9.1569 +Epoch [15/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0067, Loss2: 0.0060, Pure Ratio1: 9.7353, Pure Ratio2 9.6176 +Epoch [15/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0065, Loss2: 0.0070, Pure Ratio1: 9.7778, Pure Ratio2 9.6928 +Epoch [15/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0093, Loss2: 0.0101, Pure Ratio1: 9.9755, Pure Ratio2 9.9069 +Epoch [15/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0085, Loss2: 0.0091, Pure Ratio1: 10.0000, Pure Ratio2 9.9333 +Epoch [15/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0073, Loss2: 0.0077, Pure Ratio1: 9.8660, Pure Ratio2 9.8203 +Epoch [15/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0099, Loss2: 0.0097, Pure Ratio1: 9.7731, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test images: Model1 29.2568 % Model2 27.3337 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0086, Loss2: 0.0079, Pure Ratio1: 9.6667, Pure Ratio2 9.6275 +Epoch [16/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0067, Loss2: 0.0069, Pure Ratio1: 9.6275, Pure Ratio2 9.6275 +Epoch [16/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0083, Loss2: 0.0079, Pure Ratio1: 9.6667, Pure Ratio2 9.6601 +Epoch [16/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0070, Loss2: 0.0065, Pure Ratio1: 9.7647, Pure Ratio2 9.7157 +Epoch [16/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0075, Loss2: 0.0078, Pure Ratio1: 9.6706, Pure Ratio2 9.6431 +Epoch [16/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0076, Loss2: 0.0082, Pure Ratio1: 9.8105, Pure Ratio2 9.7680 +Epoch [16/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0071, Loss2: 0.0076, Pure Ratio1: 9.7899, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test images: Model1 24.0986 % Model2 23.5477 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.7964 % +Training cifar10_coteaching_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0056, Loss2: 0.0055, Pure Ratio1: 9.8627, Pure Ratio2 9.9020 +Epoch [17/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0067, Loss2: 0.0072, Pure Ratio1: 9.7451, Pure Ratio2 9.7255 +Epoch [17/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0068, Loss2: 0.0069, Pure Ratio1: 10.0261, Pure Ratio2 10.0327 +Epoch [17/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0064, Loss2: 0.0071, Pure Ratio1: 9.8137, Pure Ratio2 9.8039 +Epoch [17/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0060, Loss2: 0.0061, Pure Ratio1: 9.8588, Pure Ratio2 9.8588 +Epoch [17/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0070, Loss2: 0.0069, Pure Ratio1: 9.7418, Pure Ratio2 9.7451 +Epoch [17/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0068, Loss2: 0.0076, Pure Ratio1: 9.7703, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test images: Model1 24.7997 % Model2 26.8830 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0065, Loss2: 0.0063, Pure Ratio1: 9.2745, Pure Ratio2 9.1176 +Epoch [18/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0061, Loss2: 0.0066, Pure Ratio1: 9.5098, Pure Ratio2 9.5196 +Epoch [18/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0073, Loss2: 0.0080, Pure Ratio1: 9.6405, Pure Ratio2 9.6340 +Epoch [18/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0082, Loss2: 0.0079, Pure Ratio1: 9.5098, Pure Ratio2 9.5147 +Epoch [18/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0057, Loss2: 0.0069, Pure Ratio1: 9.5216, Pure Ratio2 9.5529 +Epoch [18/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0065, Loss2: 0.0068, Pure Ratio1: 9.6699, Pure Ratio2 9.6699 +Epoch [18/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0087, Loss2: 0.0083, Pure Ratio1: 9.7927, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test images: Model1 24.5994 % Model2 23.3674 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0053, Loss2: 0.0057, Pure Ratio1: 9.0980, Pure Ratio2 9.0196 +Epoch [19/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0062, Loss2: 0.0058, Pure Ratio1: 9.2843, Pure Ratio2 9.2451 +Epoch [19/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0045, Loss2: 0.0047, Pure Ratio1: 9.5621, Pure Ratio2 9.5163 +Epoch [19/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0057, Loss2: 0.0068, Pure Ratio1: 9.6078, Pure Ratio2 9.5980 +Epoch [19/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0051, Loss2: 0.0052, Pure Ratio1: 9.7647, Pure Ratio2 9.7804 +Epoch [19/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0057, Loss2: 0.0050, Pure Ratio1: 9.7582, Pure Ratio2 9.8039 +Epoch [19/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0046, Loss2: 0.0048, Pure Ratio1: 9.7843, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test images: Model1 25.1302 % Model2 24.5292 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0045, Loss2: 0.0043, Pure Ratio1: 9.8039, Pure Ratio2 9.5098 +Epoch [20/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0051, Loss2: 0.0053, Pure Ratio1: 9.7255, Pure Ratio2 9.6471 +Epoch [20/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0055, Loss2: 0.0046, Pure Ratio1: 9.8301, Pure Ratio2 9.7712 +Epoch [20/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0054, Loss2: 0.0052, Pure Ratio1: 9.9216, Pure Ratio2 9.9020 +Epoch [20/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0048, Loss2: 0.0046, Pure Ratio1: 9.7804, Pure Ratio2 9.7608 +Epoch [20/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0063, Loss2: 0.0053, Pure Ratio1: 9.7941, Pure Ratio2 9.7843 +Epoch [20/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 54.6875, Loss1: 0.0063, Loss2: 0.0067, Pure Ratio1: 9.8403, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test images: Model1 26.5325 % Model2 25.8814 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0050, Loss2: 0.0050, Pure Ratio1: 9.3137, Pure Ratio2 9.2549 +Epoch [21/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0043, Loss2: 0.0049, Pure Ratio1: 9.5392, Pure Ratio2 9.5784 +Epoch [21/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0047, Loss2: 0.0042, Pure Ratio1: 9.8889, Pure Ratio2 9.8627 +Epoch [21/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0063, Loss2: 0.0056, Pure Ratio1: 9.9951, Pure Ratio2 9.9461 +Epoch [21/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0044, Loss2: 0.0042, Pure Ratio1: 10.0353, Pure Ratio2 10.0039 +Epoch [21/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0048, Loss2: 0.0048, Pure Ratio1: 9.8987, Pure Ratio2 9.8660 +Epoch [21/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0053, Loss2: 0.0059, Pure Ratio1: 9.9384, Pure Ratio2 9.9160 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test images: Model1 25.1002 % Model2 25.0601 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 60.1562, Loss1: 0.0039, Loss2: 0.0050, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [22/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0039, Loss2: 0.0039, Pure Ratio1: 9.7843, Pure Ratio2 9.7941 +Epoch [22/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0043, Loss2: 0.0046, Pure Ratio1: 9.7451, Pure Ratio2 9.7712 +Epoch [22/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0043, Loss2: 0.0048, Pure Ratio1: 9.6127, Pure Ratio2 9.6275 +Epoch [22/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 9.5647, Pure Ratio2 9.5608 +Epoch [22/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0053, Loss2: 0.0061, Pure Ratio1: 9.6176, Pure Ratio2 9.5915 +Epoch [22/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0037, Loss2: 0.0034, Pure Ratio1: 9.7983, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test images: Model1 26.3121 % Model2 26.6026 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0047, Loss2: 0.0052, Pure Ratio1: 9.5294, Pure Ratio2 9.6667 +Epoch [23/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0055, Loss2: 0.0045, Pure Ratio1: 9.3137, Pure Ratio2 9.4608 +Epoch [23/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0054, Loss2: 0.0042, Pure Ratio1: 9.5882, Pure Ratio2 9.6078 +Epoch [23/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0036, Loss2: 0.0037, Pure Ratio1: 9.5980, Pure Ratio2 9.5735 +Epoch [23/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 9.5608, Pure Ratio2 9.5412 +Epoch [23/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 66.4062, Loss1: 0.0030, Loss2: 0.0042, Pure Ratio1: 9.6961, Pure Ratio2 9.6765 +Epoch [23/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0035, Loss2: 0.0036, Pure Ratio1: 9.8403, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test images: Model1 26.1518 % Model2 26.6627 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 66.4062, Loss1: 0.0028, Loss2: 0.0039, Pure Ratio1: 9.6471, Pure Ratio2 9.6667 +Epoch [24/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0041, Loss2: 0.0048, Pure Ratio1: 9.5000, Pure Ratio2 9.4216 +Epoch [24/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 69.5312, Loss1: 0.0026, Loss2: 0.0039, Pure Ratio1: 9.7778, Pure Ratio2 9.7778 +Epoch [24/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0031, Loss2: 0.0028, Pure Ratio1: 9.6814, Pure Ratio2 9.7255 +Epoch [24/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0026, Loss2: 0.0027, Pure Ratio1: 9.7922, Pure Ratio2 9.8039 +Epoch [24/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0035, Loss2: 0.0044, Pure Ratio1: 9.7157, Pure Ratio2 9.7549 +Epoch [24/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 67.1875, Loss1: 0.0034, Loss2: 0.0039, Pure Ratio1: 9.6975, Pure Ratio2 9.7199 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test images: Model1 25.2103 % Model2 24.1787 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.0024, Loss2: 0.0035, Pure Ratio1: 9.4118, Pure Ratio2 9.5098 +Epoch [25/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0026, Loss2: 0.0027, Pure Ratio1: 9.6863, Pure Ratio2 9.7549 +Epoch [25/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0050, Loss2: 0.0044, Pure Ratio1: 9.7451, Pure Ratio2 9.8039 +Epoch [25/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.0029, Loss2: 0.0027, Pure Ratio1: 9.7892, Pure Ratio2 9.8235 +Epoch [25/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0036, Loss2: 0.0039, Pure Ratio1: 9.7059, Pure Ratio2 9.7529 +Epoch [25/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0047, Loss2: 0.0036, Pure Ratio1: 9.8007, Pure Ratio2 9.8268 +Epoch [25/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0034, Loss2: 0.0029, Pure Ratio1: 9.8095, Pure Ratio2 9.8487 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test images: Model1 28.3454 % Model2 27.5341 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0016, Loss2: 0.0021, Pure Ratio1: 9.7843, Pure Ratio2 9.7059 +Epoch [26/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0034, Loss2: 0.0036, Pure Ratio1: 9.7941, Pure Ratio2 9.6765 +Epoch [26/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0022, Loss2: 0.0019, Pure Ratio1: 9.6993, Pure Ratio2 9.6340 +Epoch [26/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.0000, Loss1: 0.0028, Loss2: 0.0038, Pure Ratio1: 9.9657, Pure Ratio2 9.8873 +Epoch [26/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0025, Loss2: 0.0032, Pure Ratio1: 9.8980, Pure Ratio2 9.8039 +Epoch [26/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0031, Loss2: 0.0023, Pure Ratio1: 9.8627, Pure Ratio2 9.8137 +Epoch [26/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0033, Loss2: 0.0036, Pure Ratio1: 9.7647, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test images: Model1 26.3421 % Model2 26.4323 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.7562 % +Training cifar10_coteaching_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0021, Loss2: 0.0027, Pure Ratio1: 9.4706, Pure Ratio2 9.4314 +Epoch [27/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.0000, Loss1: 0.0022, Loss2: 0.0024, Pure Ratio1: 9.8627, Pure Ratio2 9.7255 +Epoch [27/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0036, Loss2: 0.0032, Pure Ratio1: 10.0065, Pure Ratio2 9.9673 +Epoch [27/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.0019, Loss2: 0.0022, Pure Ratio1: 9.8676, Pure Ratio2 9.8775 +Epoch [27/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0039, Loss2: 0.0023, Pure Ratio1: 9.8745, Pure Ratio2 9.8588 +Epoch [27/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0025, Loss2: 0.0034, Pure Ratio1: 9.8889, Pure Ratio2 9.8791 +Epoch [27/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0022, Pure Ratio1: 9.8487, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test images: Model1 25.6210 % Model2 27.3738 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0018, Loss2: 0.0019, Pure Ratio1: 9.3333, Pure Ratio2 9.4314 +Epoch [28/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 82.8125, Loss1: 0.0026, Loss2: 0.0012, Pure Ratio1: 9.7647, Pure Ratio2 9.8137 +Epoch [28/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0018, Loss2: 0.0022, Pure Ratio1: 10.1373, Pure Ratio2 10.0850 +Epoch [28/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0018, Loss2: 0.0020, Pure Ratio1: 10.0441, Pure Ratio2 10.0196 +Epoch [28/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0022, Loss2: 0.0020, Pure Ratio1: 9.7569, Pure Ratio2 9.6745 +Epoch [28/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0026, Loss2: 0.0024, Pure Ratio1: 9.7810, Pure Ratio2 9.7059 +Epoch [28/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0031, Loss2: 0.0023, Pure Ratio1: 9.8711, Pure Ratio2 9.8151 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test images: Model1 26.2821 % Model2 25.9415 %, Pure Ratio 1 9.8467 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0024, Pure Ratio1: 9.3529, Pure Ratio2 9.3137 +Epoch [29/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0021, Loss2: 0.0032, Pure Ratio1: 9.8824, Pure Ratio2 9.9216 +Epoch [29/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0012, Loss2: 0.0012, Pure Ratio1: 9.6863, Pure Ratio2 9.6536 +Epoch [29/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.6716, Pure Ratio2 9.6618 +Epoch [29/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0023, Pure Ratio1: 9.6902, Pure Ratio2 9.6431 +Epoch [29/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0022, Loss2: 0.0017, Pure Ratio1: 9.7549, Pure Ratio2 9.7092 +Epoch [29/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0034, Loss2: 0.0020, Pure Ratio1: 9.7843, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test images: Model1 26.4423 % Model2 25.1803 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0015, Loss2: 0.0022, Pure Ratio1: 9.8039, Pure Ratio2 9.6667 +Epoch [30/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 10.1765, Pure Ratio2 10.1078 +Epoch [30/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0012, Pure Ratio1: 9.9020, Pure Ratio2 9.9085 +Epoch [30/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.0016, Loss2: 0.0028, Pure Ratio1: 10.0294, Pure Ratio2 10.0490 +Epoch [30/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0019, Loss2: 0.0023, Pure Ratio1: 9.9725, Pure Ratio2 9.9804 +Epoch [30/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0017, Loss2: 0.0015, Pure Ratio1: 9.9085, Pure Ratio2 9.9346 +Epoch [30/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0035, Loss2: 0.0037, Pure Ratio1: 9.7871, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test images: Model1 25.9816 % Model2 27.1735 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0017, Loss2: 0.0013, Pure Ratio1: 9.6863, Pure Ratio2 9.7843 +Epoch [31/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0016, Loss2: 0.0021, Pure Ratio1: 9.3627, Pure Ratio2 9.3627 +Epoch [31/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 9.7059, Pure Ratio2 9.7451 +Epoch [31/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.9020, Pure Ratio2 9.9412 +Epoch [31/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.8588, Pure Ratio2 9.9451 +Epoch [31/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.0033, Loss2: 0.0025, Pure Ratio1: 9.7941, Pure Ratio2 9.8562 +Epoch [31/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0035, Loss2: 0.0021, Pure Ratio1: 9.8515, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test images: Model1 27.0533 % Model2 27.5441 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.8542 % +Training cifar10_coteaching_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.4902, Pure Ratio2 9.5294 +Epoch [32/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.6275, Pure Ratio2 9.6078 +Epoch [32/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0013, Loss2: 0.0013, Pure Ratio1: 9.8693, Pure Ratio2 9.8431 +Epoch [32/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0011, Pure Ratio1: 9.7598, Pure Ratio2 9.7647 +Epoch [32/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0008, Pure Ratio1: 9.7569, Pure Ratio2 9.7608 +Epoch [32/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0030, Loss2: 0.0021, Pure Ratio1: 9.7092, Pure Ratio2 9.7190 +Epoch [32/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0016, Loss2: 0.0017, Pure Ratio1: 9.8207, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test images: Model1 25.6911 % Model2 24.7897 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.8165 % +Training cifar10_coteaching_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.0011, Loss2: 0.0011, Pure Ratio1: 9.3725, Pure Ratio2 9.2941 +Epoch [33/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 73.4375, Loss1: 0.0012, Loss2: 0.0016, Pure Ratio1: 9.8137, Pure Ratio2 9.7843 +Epoch [33/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.0018, Loss2: 0.0022, Pure Ratio1: 9.7647, Pure Ratio2 9.8105 +Epoch [33/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.0008, Loss2: 0.0013, Pure Ratio1: 9.8431, Pure Ratio2 9.9167 +Epoch [33/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.8157, Pure Ratio2 9.8549 +Epoch [33/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0021, Loss2: 0.0013, Pure Ratio1: 9.7778, Pure Ratio2 9.7941 +Epoch [33/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.0010, Loss2: 0.0017, Pure Ratio1: 9.8095, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test images: Model1 26.7628 % Model2 24.2588 %, Pure Ratio 1 9.8768 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0008, Pure Ratio1: 9.9804, Pure Ratio2 9.9412 +Epoch [34/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.0018, Loss2: 0.0009, Pure Ratio1: 9.7647, Pure Ratio2 9.7255 +Epoch [34/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0016, Pure Ratio1: 9.9608, Pure Ratio2 9.9477 +Epoch [34/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0012, Loss2: 0.0018, Pure Ratio1: 9.7892, Pure Ratio2 9.8186 +Epoch [34/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.7333, Pure Ratio2 9.7255 +Epoch [34/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 82.0312, Loss1: 0.0016, Loss2: 0.0011, Pure Ratio1: 9.8039, Pure Ratio2 9.8268 +Epoch [34/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.0012, Loss2: 0.0006, Pure Ratio1: 9.9020, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test images: Model1 25.4006 % Model2 26.8930 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.0027, Loss2: 0.0023, Pure Ratio1: 9.0980, Pure Ratio2 9.1961 +Epoch [35/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7549, Pure Ratio2 9.8627 +Epoch [35/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8562, Pure Ratio2 10.0196 +Epoch [35/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0010, Pure Ratio1: 9.8824, Pure Ratio2 9.9853 +Epoch [35/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0026, Loss2: 0.0029, Pure Ratio1: 9.8118, Pure Ratio2 9.8863 +Epoch [35/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0019, Pure Ratio1: 9.8824, Pure Ratio2 9.9183 +Epoch [35/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0011, Loss2: 0.0006, Pure Ratio1: 9.8067, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test images: Model1 26.0717 % Model2 26.0116 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8643 % +Training cifar10_coteaching_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [36/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0008, Loss2: 0.0012, Pure Ratio1: 9.8725, Pure Ratio2 9.8431 +Epoch [36/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0018, Pure Ratio1: 9.8562, Pure Ratio2 9.8431 +Epoch [36/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.6912, Pure Ratio2 9.7696 +Epoch [36/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0016, Loss2: 0.0012, Pure Ratio1: 9.8000, Pure Ratio2 9.8745 +Epoch [36/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.7876, Pure Ratio2 9.8366 +Epoch [36/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0008, Pure Ratio1: 9.8655, Pure Ratio2 9.9132 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test images: Model1 26.3822 % Model2 26.5224 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 9.7059, Pure Ratio2 9.8235 +Epoch [37/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.6863, Pure Ratio2 9.7353 +Epoch [37/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0012, Pure Ratio1: 9.8693, Pure Ratio2 9.8627 +Epoch [37/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8088, Pure Ratio2 9.7892 +Epoch [37/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0016, Loss2: 0.0007, Pure Ratio1: 9.7216, Pure Ratio2 9.7137 +Epoch [37/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0011, Pure Ratio1: 9.7516, Pure Ratio2 9.7680 +Epoch [37/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.7899, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test images: Model1 25.4207 % Model2 27.5140 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0008, Pure Ratio1: 9.8627, Pure Ratio2 9.6667 +Epoch [38/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 10.0784, Pure Ratio2 9.9804 +Epoch [38/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9542, Pure Ratio2 9.8235 +Epoch [38/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0010, Pure Ratio1: 9.8088, Pure Ratio2 9.6520 +Epoch [38/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0024, Loss2: 0.0012, Pure Ratio1: 9.8392, Pure Ratio2 9.6863 +Epoch [38/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.7680, Pure Ratio2 9.6863 +Epoch [38/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 78.9062, Loss1: 0.0008, Loss2: 0.0017, Pure Ratio1: 9.7815, Pure Ratio2 9.7227 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test images: Model1 28.8261 % Model2 26.1518 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0019, Pure Ratio1: 9.8824, Pure Ratio2 9.9412 +Epoch [39/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 10.0490, Pure Ratio2 10.0784 +Epoch [39/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 10.0980, Pure Ratio2 10.0915 +Epoch [39/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0018, Pure Ratio1: 10.0049, Pure Ratio2 9.9461 +Epoch [39/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9373, Pure Ratio2 9.9294 +Epoch [39/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.8268, Pure Ratio2 9.8235 +Epoch [39/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0011, Pure Ratio1: 9.8263, Pure Ratio2 9.8403 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test images: Model1 26.3522 % Model2 26.9732 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0011, Pure Ratio1: 9.6471, Pure Ratio2 9.7843 +Epoch [40/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0012, Pure Ratio1: 9.7059, Pure Ratio2 9.6765 +Epoch [40/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0010, Loss2: 0.0014, Pure Ratio1: 10.0131, Pure Ratio2 10.0065 +Epoch [40/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0015, Loss2: 0.0011, Pure Ratio1: 9.7990, Pure Ratio2 9.7549 +Epoch [40/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0012, Loss2: 0.0019, Pure Ratio1: 9.7569, Pure Ratio2 9.6510 +Epoch [40/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0016, Loss2: 0.0005, Pure Ratio1: 9.8268, Pure Ratio2 9.7190 +Epoch [40/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0012, Pure Ratio1: 9.8487, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test images: Model1 24.6895 % Model2 25.7712 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0009, Pure Ratio1: 9.5490, Pure Ratio2 9.5490 +Epoch [41/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0010, Pure Ratio1: 9.4314, Pure Ratio2 9.4314 +Epoch [41/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.6601, Pure Ratio2 9.5359 +Epoch [41/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.5686, Pure Ratio2 9.4559 +Epoch [41/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0010, Pure Ratio1: 9.7059, Pure Ratio2 9.6353 +Epoch [41/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.0008, Loss2: 0.0016, Pure Ratio1: 9.7745, Pure Ratio2 9.6732 +Epoch [41/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.0013, Loss2: 0.0015, Pure Ratio1: 9.7927, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test images: Model1 25.8113 % Model2 25.0901 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 10.3922, Pure Ratio2 10.1569 +Epoch [42/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0008, Loss2: 0.0009, Pure Ratio1: 9.8039, Pure Ratio2 9.6078 +Epoch [42/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.8366, Pure Ratio2 9.6797 +Epoch [42/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9657, Pure Ratio2 9.8578 +Epoch [42/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.9490, Pure Ratio2 9.8667 +Epoch [42/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9118, Pure Ratio2 9.8268 +Epoch [42/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0014, Pure Ratio1: 9.8964, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test images: Model1 25.2103 % Model2 25.4607 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0012, Loss2: 0.0008, Pure Ratio1: 9.2353, Pure Ratio2 9.2157 +Epoch [43/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0009, Loss2: 0.0009, Pure Ratio1: 9.5882, Pure Ratio2 9.6373 +Epoch [43/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7255, Pure Ratio2 9.7582 +Epoch [43/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.7108, Pure Ratio2 9.8088 +Epoch [43/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 80.4688, Loss1: 0.0007, Loss2: 0.0014, Pure Ratio1: 9.7451, Pure Ratio2 9.7882 +Epoch [43/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.7255, Pure Ratio2 9.7647 +Epoch [43/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test images: Model1 25.6510 % Model2 26.7027 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.8240 % +Training cifar10_coteaching_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.8039, Pure Ratio2 9.7255 +Epoch [44/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0014, Loss2: 0.0011, Pure Ratio1: 9.9118, Pure Ratio2 9.8431 +Epoch [44/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0013, Loss2: 0.0015, Pure Ratio1: 9.8824, Pure Ratio2 9.8497 +Epoch [44/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0010, Pure Ratio1: 9.7598, Pure Ratio2 9.7843 +Epoch [44/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0013, Loss2: 0.0010, Pure Ratio1: 9.8000, Pure Ratio2 9.8235 +Epoch [44/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8431 +Epoch [44/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.7927, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test images: Model1 27.2636 % Model2 25.8914 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0007, Pure Ratio1: 9.7059, Pure Ratio2 9.7647 +Epoch [45/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.0014, Loss2: 0.0007, Pure Ratio1: 10.0784, Pure Ratio2 10.1667 +Epoch [45/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7386, Pure Ratio2 9.7647 +Epoch [45/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [45/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0016, Pure Ratio1: 9.6784, Pure Ratio2 9.6549 +Epoch [45/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0009, Pure Ratio1: 9.7549, Pure Ratio2 9.7320 +Epoch [45/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0005, Pure Ratio1: 9.7759, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test images: Model1 26.0817 % Model2 26.0617 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 10.2353, Pure Ratio2 10.4314 +Epoch [46/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.7255, Pure Ratio2 9.9412 +Epoch [46/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7255, Pure Ratio2 9.9085 +Epoch [46/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 10.0539 +Epoch [46/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.0024, Loss2: 0.0019, Pure Ratio1: 9.8863, Pure Ratio2 10.0157 +Epoch [46/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 9.7712, Pure Ratio2 9.8791 +Epoch [46/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.7563, Pure Ratio2 9.8880 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test images: Model1 25.8113 % Model2 26.5224 %, Pure Ratio 1 9.7310 %, Pure Ratio 2 9.8693 % +Training cifar10_coteaching_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6078, Pure Ratio2 9.6078 +Epoch [47/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.5000, Pure Ratio2 9.4608 +Epoch [47/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.8824, Pure Ratio2 9.8824 +Epoch [47/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.8382 +Epoch [47/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.7882 +Epoch [47/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7582, Pure Ratio2 9.7647 +Epoch [47/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8347, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test images: Model1 26.7228 % Model2 26.8429 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.8438, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.2353, Pure Ratio2 9.4314 +Epoch [48/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.4118, Pure Ratio2 9.5196 +Epoch [48/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0005, Pure Ratio1: 9.6797, Pure Ratio2 9.8039 +Epoch [48/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0009, Pure Ratio1: 9.6863, Pure Ratio2 9.7108 +Epoch [48/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.7804, Pure Ratio2 9.7961 +Epoch [48/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.8301, Pure Ratio2 9.8039 +Epoch [48/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.8235, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test images: Model1 26.8630 % Model2 26.7328 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1569, Pure Ratio2 9.9216 +Epoch [49/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 10.3431, Pure Ratio2 10.0980 +Epoch [49/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 10.0261, Pure Ratio2 9.8235 +Epoch [49/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 10.0539, Pure Ratio2 9.8775 +Epoch [49/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0010, Loss2: 0.0005, Pure Ratio1: 9.9373, Pure Ratio2 9.8118 +Epoch [49/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0010, Pure Ratio1: 9.8072, Pure Ratio2 9.7516 +Epoch [49/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0016, Loss2: 0.0007, Pure Ratio1: 9.8431, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test images: Model1 25.9215 % Model2 25.7412 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.1765, Pure Ratio2 10.0588 +Epoch [50/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0002, Loss2: 0.0021, Pure Ratio1: 10.1569, Pure Ratio2 10.0000 +Epoch [50/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 83.5938, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 9.9935, Pure Ratio2 9.8824 +Epoch [50/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8137 +Epoch [50/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.7804, Pure Ratio2 9.6902 +Epoch [50/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0003, Pure Ratio1: 9.8170, Pure Ratio2 9.7484 +Epoch [50/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0010, Pure Ratio1: 9.9328, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test images: Model1 26.2220 % Model2 26.1118 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.6078, Pure Ratio2 10.3922 +Epoch [51/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0011, Loss2: 0.0002, Pure Ratio1: 10.1863, Pure Ratio2 10.0882 +Epoch [51/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 10.0327, Pure Ratio2 9.9412 +Epoch [51/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 9.9951, Pure Ratio2 9.8873 +Epoch [51/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.9686, Pure Ratio2 9.8196 +Epoch [51/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [51/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0021, Loss2: 0.0007, Pure Ratio1: 9.8936, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test images: Model1 24.6595 % Model2 25.7412 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0008, Pure Ratio1: 9.4118, Pure Ratio2 9.6667 +Epoch [52/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8039, Pure Ratio2 9.9510 +Epoch [52/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.0015, Loss2: 0.0008, Pure Ratio1: 9.7908, Pure Ratio2 9.8824 +Epoch [52/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6324, Pure Ratio2 9.6765 +Epoch [52/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6863, Pure Ratio2 9.7176 +Epoch [52/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7124, Pure Ratio2 9.7712 +Epoch [52/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7731, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test images: Model1 28.0849 % Model2 26.4423 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8467 % +Training cifar10_coteaching_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.4510, Pure Ratio2 9.7059 +Epoch [53/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0014, Pure Ratio1: 9.5392, Pure Ratio2 9.7745 +Epoch [53/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.5752, Pure Ratio2 9.7843 +Epoch [53/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0014, Loss2: 0.0006, Pure Ratio1: 9.5441, Pure Ratio2 9.7206 +Epoch [53/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.5373, Pure Ratio2 9.6824 +Epoch [53/200], Iter [300/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.5621, Pure Ratio2 9.6895 +Epoch [53/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.6919, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test images: Model1 26.1418 % Model2 25.3305 %, Pure Ratio 1 9.7461 %, Pure Ratio 2 9.8617 % +Training cifar10_coteaching_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.1373, Pure Ratio2 10.3725 +Epoch [54/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8627, Pure Ratio2 9.8725 +Epoch [54/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8301, Pure Ratio2 9.8366 +Epoch [54/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.7990, Pure Ratio2 9.7304 +Epoch [54/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.0010, Loss2: 0.0012, Pure Ratio1: 9.6667, Pure Ratio2 9.6039 +Epoch [54/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0018, Loss2: 0.0004, Pure Ratio1: 9.7549, Pure Ratio2 9.6601 +Epoch [54/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7759, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test images: Model1 26.4123 % Model2 25.1302 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.8431, Pure Ratio2 9.9216 +Epoch [55/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7745, Pure Ratio2 9.7941 +Epoch [55/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.8824, Pure Ratio2 9.9020 +Epoch [55/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9265, Pure Ratio2 9.9706 +Epoch [55/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0004, Pure Ratio1: 9.8314, Pure Ratio2 9.9098 +Epoch [55/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8660, Pure Ratio2 9.9346 +Epoch [55/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8095, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test images: Model1 26.5625 % Model2 26.6727 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0006, Pure Ratio1: 10.4118, Pure Ratio2 10.3529 +Epoch [56/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0006, Pure Ratio1: 10.4314, Pure Ratio2 10.2843 +Epoch [56/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.3072, Pure Ratio2 10.1961 +Epoch [56/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.0882, Pure Ratio2 10.0686 +Epoch [56/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 10.0588, Pure Ratio2 10.0196 +Epoch [56/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.9967, Pure Ratio2 9.9902 +Epoch [56/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.9300, Pure Ratio2 9.9300 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test images: Model1 27.4940 % Model2 25.2704 %, Pure Ratio 1 9.8592 %, Pure Ratio 2 9.8768 % +Training cifar10_coteaching_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.1569, Pure Ratio2 9.1373 +Epoch [57/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.7647, Pure Ratio2 9.7549 +Epoch [57/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8366 +Epoch [57/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.9314, Pure Ratio2 9.9020 +Epoch [57/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.9098 +Epoch [57/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0017, Loss2: 0.0016, Pure Ratio1: 9.8725, Pure Ratio2 9.8464 +Epoch [57/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.8515, Pure Ratio2 9.8011 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test images: Model1 26.5525 % Model2 26.7127 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.9020, Pure Ratio2 9.9216 +Epoch [58/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.2941, Pure Ratio2 10.3039 +Epoch [58/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.2614, Pure Ratio2 10.3137 +Epoch [58/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0049, Pure Ratio2 10.0294 +Epoch [58/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.8471, Pure Ratio2 9.7922 +Epoch [58/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.7320, Pure Ratio2 9.6667 +Epoch [58/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.7759, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test images: Model1 25.3606 % Model2 26.8730 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0011, Loss2: 0.0005, Pure Ratio1: 9.7647, Pure Ratio2 10.1176 +Epoch [59/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9902, Pure Ratio2 10.2059 +Epoch [59/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0008, Loss2: 0.0004, Pure Ratio1: 9.9412, Pure Ratio2 10.0654 +Epoch [59/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0001, Pure Ratio1: 9.7157, Pure Ratio2 9.7892 +Epoch [59/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7294, Pure Ratio2 9.8118 +Epoch [59/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.0016, Loss2: 0.0002, Pure Ratio1: 9.7124, Pure Ratio2 9.7582 +Epoch [59/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0008, Loss2: 0.0005, Pure Ratio1: 9.7619, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test images: Model1 26.4423 % Model2 27.3938 %, Pure Ratio 1 9.8014 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7059, Pure Ratio2 10.0000 +Epoch [60/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.9412 +Epoch [60/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.7320, Pure Ratio2 9.8562 +Epoch [60/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.8137, Pure Ratio2 9.9314 +Epoch [60/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8941, Pure Ratio2 9.9804 +Epoch [60/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0012, Loss2: 0.0005, Pure Ratio1: 9.8497, Pure Ratio2 9.9346 +Epoch [60/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0006, Pure Ratio1: 9.7619, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test images: Model1 26.2921 % Model2 26.6126 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8994 % +Training cifar10_coteaching_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 10.2941, Pure Ratio2 10.0588 +Epoch [61/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 10.2549, Pure Ratio2 10.1961 +Epoch [61/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0016, Pure Ratio1: 10.1438, Pure Ratio2 10.1046 +Epoch [61/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9853, Pure Ratio2 9.9608 +Epoch [61/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 82.8125, Loss1: 0.0013, Loss2: 0.0001, Pure Ratio1: 9.8706, Pure Ratio2 9.8431 +Epoch [61/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7778, Pure Ratio2 9.7843 +Epoch [61/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.8375, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test images: Model1 27.0733 % Model2 25.8413 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.8416 % +Training cifar10_coteaching_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.0009, Loss2: 0.0002, Pure Ratio1: 9.3725, Pure Ratio2 9.2157 +Epoch [62/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7059, Pure Ratio2 9.6176 +Epoch [62/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0004, Pure Ratio1: 9.7647, Pure Ratio2 9.6601 +Epoch [62/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7696, Pure Ratio2 9.6422 +Epoch [62/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.8980, Pure Ratio2 9.7725 +Epoch [62/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.0014, Loss2: 0.0014, Pure Ratio1: 9.9412, Pure Ratio2 9.8399 +Epoch [62/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0004, Pure Ratio1: 9.9216, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test images: Model1 26.0016 % Model2 27.4139 %, Pure Ratio 1 9.8542 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.7059, Pure Ratio2 9.5294 +Epoch [63/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8529 +Epoch [63/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.7190, Pure Ratio2 9.6928 +Epoch [63/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8137, Pure Ratio2 9.8333 +Epoch [63/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9804, Pure Ratio2 9.9490 +Epoch [63/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.8725, Pure Ratio2 9.8758 +Epoch [63/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8067, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test images: Model1 27.5341 % Model2 27.2236 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 8.9020, Pure Ratio2 9.1176 +Epoch [64/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.0686, Pure Ratio2 9.2157 +Epoch [64/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6993, Pure Ratio2 9.7712 +Epoch [64/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.8039, Pure Ratio2 9.8725 +Epoch [64/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0003, Loss2: 0.0005, Pure Ratio1: 9.9255, Pure Ratio2 9.9961 +Epoch [64/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.8366, Pure Ratio2 9.8889 +Epoch [64/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0004, Pure Ratio1: 9.7787, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test images: Model1 26.8730 % Model2 28.0749 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.4118, Pure Ratio2 9.4510 +Epoch [65/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.6569 +Epoch [65/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.6209, Pure Ratio2 9.6078 +Epoch [65/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.5735, Pure Ratio2 9.6029 +Epoch [65/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.7216, Pure Ratio2 9.7451 +Epoch [65/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7516, Pure Ratio2 9.7876 +Epoch [65/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8011, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test images: Model1 26.0016 % Model2 25.6210 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.4706, Pure Ratio2 10.4314 +Epoch [66/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8235, Pure Ratio2 9.8235 +Epoch [66/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.9542, Pure Ratio2 9.8889 +Epoch [66/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8873, Pure Ratio2 9.8775 +Epoch [66/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8863, Pure Ratio2 9.8824 +Epoch [66/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8464, Pure Ratio2 9.8431 +Epoch [66/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8067, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test images: Model1 26.3421 % Model2 25.9816 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0013, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [67/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 9.7941, Pure Ratio2 9.6961 +Epoch [67/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.8105 +Epoch [67/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.0539, Pure Ratio2 9.9608 +Epoch [67/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.9882, Pure Ratio2 9.8941 +Epoch [67/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0327, Pure Ratio2 9.9379 +Epoch [67/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.9692, Pure Ratio2 9.8908 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test images: Model1 25.8814 % Model2 26.7228 %, Pure Ratio 1 9.8668 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.0588, Pure Ratio2 9.0392 +Epoch [68/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.1961, Pure Ratio2 9.2745 +Epoch [68/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 92.1875, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 9.3399, Pure Ratio2 9.4967 +Epoch [68/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.4216, Pure Ratio2 9.5147 +Epoch [68/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.4039, Pure Ratio2 9.4706 +Epoch [68/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.6111, Pure Ratio2 9.6503 +Epoch [68/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7227, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test images: Model1 25.0901 % Model2 26.8429 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9020, Pure Ratio2 9.8235 +Epoch [69/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.1373, Pure Ratio2 10.0294 +Epoch [69/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0004, Loss2: 0.0009, Pure Ratio1: 9.8301, Pure Ratio2 9.8235 +Epoch [69/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8676, Pure Ratio2 9.8480 +Epoch [69/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.7882, Pure Ratio2 9.8039 +Epoch [69/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.0003, Loss2: 0.0012, Pure Ratio1: 9.8170, Pure Ratio2 9.8235 +Epoch [69/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0014, Pure Ratio1: 9.7955, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test images: Model1 26.4323 % Model2 26.3321 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0006, Pure Ratio1: 9.5882, Pure Ratio2 9.8824 +Epoch [70/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9804 +Epoch [70/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.0009, Loss2: 0.0016, Pure Ratio1: 9.7451, Pure Ratio2 9.9020 +Epoch [70/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0006, Loss2: 0.0008, Pure Ratio1: 9.8578, Pure Ratio2 9.9069 +Epoch [70/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9137, Pure Ratio2 9.8824 +Epoch [70/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9444, Pure Ratio2 9.8889 +Epoch [70/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.9104, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test images: Model1 27.2436 % Model2 27.6643 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.0010, Loss2: 0.0006, Pure Ratio1: 10.2157, Pure Ratio2 10.0784 +Epoch [71/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 10.1275, Pure Ratio2 10.1275 +Epoch [71/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 10.1503, Pure Ratio2 10.1438 +Epoch [71/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 10.0539, Pure Ratio2 10.0931 +Epoch [71/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 10.0510, Pure Ratio2 10.0627 +Epoch [71/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0003, Pure Ratio1: 9.8758, Pure Ratio2 9.8627 +Epoch [71/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0005, Loss2: 0.0009, Pure Ratio1: 9.7927, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test images: Model1 26.0617 % Model2 25.6410 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 10.1765, Pure Ratio2 10.2941 +Epoch [72/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 10.0294 +Epoch [72/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9346, Pure Ratio2 10.0850 +Epoch [72/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 9.9118 +Epoch [72/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0005, Pure Ratio1: 9.7922, Pure Ratio2 9.9059 +Epoch [72/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.6895, Pure Ratio2 9.8105 +Epoch [72/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6331, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test images: Model1 26.1018 % Model2 26.6326 %, Pure Ratio 1 9.7034 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0784, Pure Ratio2 9.9412 +Epoch [73/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 10.0000, Pure Ratio2 10.0686 +Epoch [73/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0008, Pure Ratio1: 9.9150, Pure Ratio2 9.9477 +Epoch [73/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7892, Pure Ratio2 9.8529 +Epoch [73/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8471, Pure Ratio2 9.9569 +Epoch [73/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.0007, Loss2: 0.0024, Pure Ratio1: 9.7941, Pure Ratio2 9.9444 +Epoch [73/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.8852 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test images: Model1 26.5725 % Model2 27.2536 %, Pure Ratio 1 9.7461 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.8824 +Epoch [74/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.7647, Pure Ratio2 9.6569 +Epoch [74/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.7647, Pure Ratio2 9.6601 +Epoch [74/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0013, Loss2: 0.0004, Pure Ratio1: 9.8235, Pure Ratio2 9.7990 +Epoch [74/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0008, Loss2: 0.0006, Pure Ratio1: 9.8667, Pure Ratio2 9.8510 +Epoch [74/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0014, Loss2: 0.0002, Pure Ratio1: 9.6601, Pure Ratio2 9.7026 +Epoch [74/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7003, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test images: Model1 26.0717 % Model2 25.7612 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6275, Pure Ratio2 9.7647 +Epoch [75/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0013, Pure Ratio1: 9.5686, Pure Ratio2 9.6667 +Epoch [75/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.0005, Loss2: 0.0005, Pure Ratio1: 9.6536, Pure Ratio2 9.7255 +Epoch [75/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8971 +Epoch [75/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8863, Pure Ratio2 9.9333 +Epoch [75/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9412, Pure Ratio2 9.9739 +Epoch [75/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8992, Pure Ratio2 9.8964 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test images: Model1 25.6911 % Model2 26.4824 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.8668 % +Training cifar10_coteaching_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0016, Loss2: 0.0003, Pure Ratio1: 10.6667, Pure Ratio2 10.5882 +Epoch [76/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 10.3725, Pure Ratio2 10.2549 +Epoch [76/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.0458, Pure Ratio2 9.9608 +Epoch [76/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0007, Loss2: 0.0002, Pure Ratio1: 10.0049, Pure Ratio2 9.9461 +Epoch [76/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.8431, Pure Ratio2 9.7922 +Epoch [76/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.8170, Pure Ratio2 9.7843 +Epoch [76/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7535, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test images: Model1 26.5024 % Model2 26.3622 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9216, Pure Ratio2 9.8431 +Epoch [77/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9510, Pure Ratio2 9.9118 +Epoch [77/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0010, Loss2: 0.0003, Pure Ratio1: 10.0719, Pure Ratio2 9.9608 +Epoch [77/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 9.9412 +Epoch [77/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 89.0625, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.0157, Pure Ratio2 9.9373 +Epoch [77/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.8758 +Epoch [77/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9244, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test images: Model1 25.8013 % Model2 25.7312 %, Pure Ratio 1 9.8366 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0012, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.4706 +Epoch [78/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0008, Pure Ratio1: 9.8137, Pure Ratio2 9.6275 +Epoch [78/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.9739, Pure Ratio2 9.7974 +Epoch [78/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0003, Loss2: 0.0006, Pure Ratio1: 9.8333, Pure Ratio2 9.6324 +Epoch [78/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.8039, Pure Ratio2 9.6863 +Epoch [78/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7418, Pure Ratio2 9.6601 +Epoch [78/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0013, Loss2: 0.0004, Pure Ratio1: 9.7339, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test images: Model1 25.7612 % Model2 26.3021 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7647, Pure Ratio2 9.7255 +Epoch [79/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7451 +Epoch [79/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7059, Pure Ratio2 9.7974 +Epoch [79/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6373, Pure Ratio2 9.7059 +Epoch [79/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7216, Pure Ratio2 9.8000 +Epoch [79/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0009, Pure Ratio1: 9.6732, Pure Ratio2 9.7288 +Epoch [79/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7675, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test images: Model1 27.5841 % Model2 27.5841 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.0784, Pure Ratio2 9.2157 +Epoch [80/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.5588, Pure Ratio2 9.6078 +Epoch [80/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.4248, Pure Ratio2 9.5621 +Epoch [80/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0010, Loss2: 0.0002, Pure Ratio1: 9.5441, Pure Ratio2 9.6814 +Epoch [80/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6353, Pure Ratio2 9.7765 +Epoch [80/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0005, Loss2: 0.0013, Pure Ratio1: 9.7092, Pure Ratio2 9.8464 +Epoch [80/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6947, Pure Ratio2 9.8039 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test images: Model1 26.6126 % Model2 25.9916 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.4902, Pure Ratio2 9.3922 +Epoch [81/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.5686, Pure Ratio2 9.5000 +Epoch [81/200], Iter [150/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6013, Pure Ratio2 9.5882 +Epoch [81/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0011, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.5637 +Epoch [81/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.7176, Pure Ratio2 9.6549 +Epoch [81/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 88.2812, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.8072, Pure Ratio2 9.6830 +Epoch [81/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 9.7927, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test images: Model1 26.8630 % Model2 25.5108 %, Pure Ratio 1 9.8819 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.8431 +Epoch [82/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.6373, Pure Ratio2 9.7157 +Epoch [82/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.8039 +Epoch [82/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8333, Pure Ratio2 9.9314 +Epoch [82/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8118, Pure Ratio2 9.8863 +Epoch [82/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.6830, Pure Ratio2 9.7876 +Epoch [82/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0011, Loss2: 0.0001, Pure Ratio1: 9.8207, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test images: Model1 25.8113 % Model2 25.7512 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.8039, Pure Ratio2 9.6863 +Epoch [83/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7353, Pure Ratio2 9.7843 +Epoch [83/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7451, Pure Ratio2 9.7516 +Epoch [83/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6618, Pure Ratio2 9.6716 +Epoch [83/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7176, Pure Ratio2 9.7216 +Epoch [83/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0006, Loss2: 0.0007, Pure Ratio1: 9.7516, Pure Ratio2 9.7582 +Epoch [83/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7339, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test images: Model1 26.3822 % Model2 25.7412 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7989 % +Training cifar10_coteaching_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0008, Pure Ratio1: 9.9804, Pure Ratio2 9.5882 +Epoch [84/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7843, Pure Ratio2 9.4314 +Epoch [84/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.7974, Pure Ratio2 9.4837 +Epoch [84/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 9.7941, Pure Ratio2 9.5245 +Epoch [84/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8549, Pure Ratio2 9.5882 +Epoch [84/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8137, Pure Ratio2 9.5686 +Epoch [84/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0008, Pure Ratio1: 9.8347, Pure Ratio2 9.6162 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test images: Model1 25.7512 % Model2 27.6242 %, Pure Ratio 1 9.9020 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7255, Pure Ratio2 10.0000 +Epoch [85/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7549, Pure Ratio2 10.0000 +Epoch [85/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.8497, Pure Ratio2 10.0327 +Epoch [85/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6814, Pure Ratio2 9.8578 +Epoch [85/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.8353 +Epoch [85/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 85.1562, Loss1: 0.0016, Loss2: 0.0002, Pure Ratio1: 9.5817, Pure Ratio2 9.7680 +Epoch [85/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5518, Pure Ratio2 9.6863 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test images: Model1 26.5124 % Model2 26.5425 %, Pure Ratio 1 9.6405 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 10.0588, Pure Ratio2 10.0392 +Epoch [86/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.7451 +Epoch [86/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7124, Pure Ratio2 9.6209 +Epoch [86/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 87.5000, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.6029, Pure Ratio2 9.5539 +Epoch [86/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8157, Pure Ratio2 9.7608 +Epoch [86/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.7190, Pure Ratio2 9.6797 +Epoch [86/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7395, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test images: Model1 26.6226 % Model2 26.0917 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.4706, Pure Ratio2 9.3529 +Epoch [87/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.5686 +Epoch [87/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 90.6250, Loss1: 0.0011, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 9.9935 +Epoch [87/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9167, Pure Ratio2 9.9314 +Epoch [87/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8824 +Epoch [87/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.0010, Loss2: 0.0001, Pure Ratio1: 9.8954, Pure Ratio2 9.8497 +Epoch [87/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8207 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test images: Model1 26.0016 % Model2 24.9700 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.5490 +Epoch [88/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0009, Loss2: 0.0007, Pure Ratio1: 9.5980, Pure Ratio2 9.6667 +Epoch [88/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6993, Pure Ratio2 9.7320 +Epoch [88/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.6716, Pure Ratio2 9.7108 +Epoch [88/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.7137, Pure Ratio2 9.7765 +Epoch [88/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.6993, Pure Ratio2 9.7418 +Epoch [88/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0009, Loss2: 0.0001, Pure Ratio1: 9.7423, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test images: Model1 26.1118 % Model2 25.5008 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.6275 +Epoch [89/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0074, Loss2: 0.0063, Pure Ratio1: 9.7843, Pure Ratio2 9.8333 +Epoch [89/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.9542, Pure Ratio2 9.9739 +Epoch [89/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9461, Pure Ratio2 9.8824 +Epoch [89/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9373, Pure Ratio2 9.8510 +Epoch [89/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7941, Pure Ratio2 9.7255 +Epoch [89/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0006, Loss2: 0.0004, Pure Ratio1: 9.7507, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test images: Model1 27.4740 % Model2 24.3790 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7084 % +Training cifar10_coteaching_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.3333, Pure Ratio2 10.4706 +Epoch [90/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.5588, Pure Ratio2 10.5098 +Epoch [90/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 10.3987, Pure Ratio2 10.3595 +Epoch [90/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 10.0000, Pure Ratio2 9.9706 +Epoch [90/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.9843, Pure Ratio2 9.9725 +Epoch [90/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.9902, Pure Ratio2 9.9771 +Epoch [90/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.9328, Pure Ratio2 9.9496 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test images: Model1 26.1518 % Model2 24.4792 %, Pure Ratio 1 9.8316 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.7059 +Epoch [91/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.0005, Loss2: 0.0007, Pure Ratio1: 9.7451, Pure Ratio2 9.6275 +Epoch [91/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0006, Pure Ratio1: 9.8039, Pure Ratio2 9.7778 +Epoch [91/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.8235, Pure Ratio2 9.8578 +Epoch [91/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9333, Pure Ratio2 9.9333 +Epoch [91/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8660 +Epoch [91/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.0008, Loss2: 0.0002, Pure Ratio1: 9.8487, Pure Ratio2 9.8796 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test images: Model1 25.8213 % Model2 26.0817 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.4118, Pure Ratio2 9.3725 +Epoch [92/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6569, Pure Ratio2 9.8039 +Epoch [92/200], Iter [150/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0007, Loss2: 0.0001, Pure Ratio1: 9.9739, Pure Ratio2 9.9346 +Epoch [92/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.8775 +Epoch [92/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8549, Pure Ratio2 9.7804 +Epoch [92/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8758, Pure Ratio2 9.8235 +Epoch [92/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.7871, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test images: Model1 26.1719 % Model2 26.2420 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9216, Pure Ratio2 9.7255 +Epoch [93/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0294, Pure Ratio2 9.9216 +Epoch [93/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8039, Pure Ratio2 9.7974 +Epoch [93/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7059, Pure Ratio2 9.7108 +Epoch [93/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7373, Pure Ratio2 9.6941 +Epoch [93/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8072 +Epoch [93/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7507, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test images: Model1 27.5942 % Model2 27.0533 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.9216 +Epoch [94/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.0048, Loss2: 0.0059, Pure Ratio1: 9.7255, Pure Ratio2 9.7059 +Epoch [94/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6405, Pure Ratio2 9.6340 +Epoch [94/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6275, Pure Ratio2 9.6176 +Epoch [94/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.5569, Pure Ratio2 9.5098 +Epoch [94/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6601, Pure Ratio2 9.6013 +Epoch [94/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7283, Pure Ratio2 9.6779 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test images: Model1 25.7011 % Model2 26.7228 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.4902, Pure Ratio2 9.7451 +Epoch [95/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8333, Pure Ratio2 9.8137 +Epoch [95/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0004, Pure Ratio1: 9.9935, Pure Ratio2 10.0000 +Epoch [95/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 9.9706 +Epoch [95/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8353, Pure Ratio2 9.8275 +Epoch [95/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0003, Loss2: 0.0002, Pure Ratio1: 9.9183, Pure Ratio2 9.9052 +Epoch [95/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.8515, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test images: Model1 25.6310 % Model2 24.9399 %, Pure Ratio 1 9.8341 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 93.7500, Training Accuracy2: 92.1875, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.6078 +Epoch [96/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4020, Pure Ratio2 9.5000 +Epoch [96/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.3856, Pure Ratio2 9.4248 +Epoch [96/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4265, Pure Ratio2 9.4657 +Epoch [96/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5412, Pure Ratio2 9.5451 +Epoch [96/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6503, Pure Ratio2 9.6601 +Epoch [96/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7535 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test images: Model1 26.4022 % Model2 27.4239 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.3922, Pure Ratio2 9.2549 +Epoch [97/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 93.7500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4902, Pure Ratio2 9.3529 +Epoch [97/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4771, Pure Ratio2 9.4052 +Epoch [97/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.6373, Pure Ratio2 9.5882 +Epoch [97/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.7725, Pure Ratio2 9.7059 +Epoch [97/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7745, Pure Ratio2 9.7026 +Epoch [97/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8179, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test images: Model1 26.3221 % Model2 24.5593 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.4118, Pure Ratio2 9.2941 +Epoch [98/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8922, Pure Ratio2 9.8627 +Epoch [98/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7843 +Epoch [98/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7990 +Epoch [98/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8196, Pure Ratio2 9.7412 +Epoch [98/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8366 +Epoch [98/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8796, Pure Ratio2 9.8263 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test images: Model1 26.7228 % Model2 26.3321 %, Pure Ratio 1 9.8617 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 10.3922, Pure Ratio2 10.3922 +Epoch [99/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.2353, Pure Ratio2 10.1471 +Epoch [99/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1961, Pure Ratio2 10.0719 +Epoch [99/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0686, Pure Ratio2 9.9755 +Epoch [99/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 10.0235, Pure Ratio2 9.9647 +Epoch [99/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.7810 +Epoch [99/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8627, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test images: Model1 26.2420 % Model2 25.2704 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7914 % +Training cifar10_coteaching_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.8824, Pure Ratio2 9.9412 +Epoch [100/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7549 +Epoch [100/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 93.7500, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6209, Pure Ratio2 9.5033 +Epoch [100/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0006, Pure Ratio1: 9.6422, Pure Ratio2 9.5098 +Epoch [100/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7529, Pure Ratio2 9.6824 +Epoch [100/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7222, Pure Ratio2 9.6667 +Epoch [100/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test images: Model1 27.8946 % Model2 25.0200 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.0392, Pure Ratio2 10.3137 +Epoch [101/200], Iter [100/390] Training Accuracy1: 94.5312, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.5588 +Epoch [101/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 10.2810, Pure Ratio2 10.4706 +Epoch [101/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 10.0294, Pure Ratio2 10.1176 +Epoch [101/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0005, Pure Ratio1: 9.9216, Pure Ratio2 10.0471 +Epoch [101/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0007, Pure Ratio1: 9.8235, Pure Ratio2 9.9118 +Epoch [101/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7507, Pure Ratio2 9.8515 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test images: Model1 26.6126 % Model2 25.1002 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.9020 +Epoch [102/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7353 +Epoch [102/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.8824 +Epoch [102/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8676, Pure Ratio2 9.8431 +Epoch [102/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0006, Loss2: 0.0005, Pure Ratio1: 9.9020, Pure Ratio2 9.9020 +Epoch [102/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.8268, Pure Ratio2 9.8170 +Epoch [102/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8571, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test images: Model1 26.6226 % Model2 26.1118 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6471 +Epoch [103/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0003, Pure Ratio1: 9.6765, Pure Ratio2 9.6275 +Epoch [103/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.6667 +Epoch [103/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7549, Pure Ratio2 9.7745 +Epoch [103/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6588, Pure Ratio2 9.7490 +Epoch [103/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0006, Pure Ratio1: 9.6405, Pure Ratio2 9.7418 +Epoch [103/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6583, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test images: Model1 27.0633 % Model2 26.6526 %, Pure Ratio 1 9.7059 %, Pure Ratio 2 9.7034 % +Training cifar10_coteaching_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.4118 +Epoch [104/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.1863, Pure Ratio2 9.1863 +Epoch [104/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.4248, Pure Ratio2 9.3856 +Epoch [104/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6127, Pure Ratio2 9.6275 +Epoch [104/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0009, Pure Ratio1: 9.5569, Pure Ratio2 9.5765 +Epoch [104/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0012, Pure Ratio1: 9.6405, Pure Ratio2 9.6438 +Epoch [104/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test images: Model1 25.8313 % Model2 26.7829 %, Pure Ratio 1 9.7511 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.3137 +Epoch [105/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.8333 +Epoch [105/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8562, Pure Ratio2 9.9281 +Epoch [105/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7206, Pure Ratio2 9.8137 +Epoch [105/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.7686, Pure Ratio2 9.8784 +Epoch [105/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.7876, Pure Ratio2 9.8758 +Epoch [105/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0004, Loss2: 0.0007, Pure Ratio1: 9.7759, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test images: Model1 27.0032 % Model2 26.1318 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.8391 % +Training cifar10_coteaching_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.4314, Pure Ratio2 9.0980 +Epoch [106/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.6176, Pure Ratio2 9.6275 +Epoch [106/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5229, Pure Ratio2 9.5229 +Epoch [106/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.5833, Pure Ratio2 9.5735 +Epoch [106/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7098, Pure Ratio2 9.6863 +Epoch [106/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.8039, Pure Ratio2 9.7418 +Epoch [106/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8291, Pure Ratio2 9.7759 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test images: Model1 28.0950 % Model2 25.6110 %, Pure Ratio 1 9.8743 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.6078, Pure Ratio2 9.5490 +Epoch [107/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7451 +Epoch [107/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6732, Pure Ratio2 9.6797 +Epoch [107/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9167, Pure Ratio2 9.9265 +Epoch [107/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8314, Pure Ratio2 9.8706 +Epoch [107/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.8693, Pure Ratio2 9.9575 +Epoch [107/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0004, Pure Ratio1: 9.8067, Pure Ratio2 9.9076 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test images: Model1 28.2252 % Model2 26.3121 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.2745, Pure Ratio2 9.4510 +Epoch [108/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0002, Loss2: 0.0003, Pure Ratio1: 9.6078, Pure Ratio2 9.7451 +Epoch [108/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.6471, Pure Ratio2 9.7386 +Epoch [108/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.7059, Pure Ratio2 9.7843 +Epoch [108/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7176, Pure Ratio2 9.7961 +Epoch [108/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0007, Pure Ratio1: 9.8268, Pure Ratio2 9.8889 +Epoch [108/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7675, Pure Ratio2 9.8179 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test images: Model1 26.9631 % Model2 26.5124 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.8190 % +Training cifar10_coteaching_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 10.5294, Pure Ratio2 10.5686 +Epoch [109/200], Iter [100/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3235, Pure Ratio2 10.4412 +Epoch [109/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0065, Pure Ratio2 10.1961 +Epoch [109/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 10.0294 +Epoch [109/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9843, Pure Ratio2 9.9922 +Epoch [109/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.9477, Pure Ratio2 9.9739 +Epoch [109/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8880, Pure Ratio2 9.8992 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test images: Model1 26.1819 % Model2 26.2720 %, Pure Ratio 1 9.7838 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [110/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.1765 +Epoch [110/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.9869, Pure Ratio2 10.1176 +Epoch [110/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9608, Pure Ratio2 10.0490 +Epoch [110/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0003, Loss2: 0.0003, Pure Ratio1: 9.9882, Pure Ratio2 10.0471 +Epoch [110/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8203, Pure Ratio2 9.8954 +Epoch [110/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.0006, Loss2: 0.0001, Pure Ratio1: 9.7871, Pure Ratio2 9.8347 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test images: Model1 25.8313 % Model2 26.4623 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.8064 % +Training cifar10_coteaching_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.5098 +Epoch [111/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.9020 +Epoch [111/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9216, Pure Ratio2 9.8693 +Epoch [111/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7500 +Epoch [111/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 88.2812, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7725, Pure Ratio2 9.7961 +Epoch [111/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8072, Pure Ratio2 9.8529 +Epoch [111/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.7115, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test images: Model1 26.6226 % Model2 26.1819 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1765, Pure Ratio2 10.4118 +Epoch [112/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1275, Pure Ratio2 10.1765 +Epoch [112/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0654 +Epoch [112/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8873 +Epoch [112/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8941, Pure Ratio2 9.9373 +Epoch [112/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8627, Pure Ratio2 9.9412 +Epoch [112/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7759, Pure Ratio2 9.8291 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test images: Model1 25.9415 % Model2 26.8029 %, Pure Ratio 1 9.7159 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2941, Pure Ratio2 10.2353 +Epoch [113/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.9314 +Epoch [113/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.7386, Pure Ratio2 9.8039 +Epoch [113/200], Iter [200/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.7157, Pure Ratio2 9.8676 +Epoch [113/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.8157 +Epoch [113/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7908 +Epoch [113/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test images: Model1 24.8998 % Model2 25.2504 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7763 % +Training cifar10_coteaching_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0004, Pure Ratio1: 9.9608, Pure Ratio2 9.6275 +Epoch [114/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.3922, Pure Ratio2 9.2745 +Epoch [114/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5425, Pure Ratio2 9.4314 +Epoch [114/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9167, Pure Ratio2 9.7549 +Epoch [114/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7922, Pure Ratio2 9.6980 +Epoch [114/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0006, Loss2: 0.0002, Pure Ratio1: 9.7745, Pure Ratio2 9.7255 +Epoch [114/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 86.7188, Loss1: 0.0005, Loss2: 0.0003, Pure Ratio1: 9.7955, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test images: Model1 26.5625 % Model2 27.2536 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 84.3750, Loss1: 0.0008, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.3333 +Epoch [115/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.9216, Pure Ratio2 10.0000 +Epoch [115/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 10.1242, Pure Ratio2 10.1830 +Epoch [115/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 10.0539, Pure Ratio2 10.0980 +Epoch [115/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1176 +Epoch [115/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0098, Pure Ratio2 10.0392 +Epoch [115/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0006, Loss2: 0.0003, Pure Ratio1: 10.0112, Pure Ratio2 10.0448 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test images: Model1 27.1534 % Model2 25.3506 %, Pure Ratio 1 9.8291 %, Pure Ratio 2 9.8567 % +Training cifar10_coteaching_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.1569, Pure Ratio2 9.1569 +Epoch [116/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4804, Pure Ratio2 9.4216 +Epoch [116/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.5882 +Epoch [116/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7402 +Epoch [116/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7529, Pure Ratio2 9.6275 +Epoch [116/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7484, Pure Ratio2 9.6176 +Epoch [116/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8095, Pure Ratio2 9.6583 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test images: Model1 26.5525 % Model2 25.7612 %, Pure Ratio 1 9.8643 %, Pure Ratio 2 9.7411 % +Training cifar10_coteaching_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2157, Pure Ratio2 9.1765 +Epoch [117/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.3137, Pure Ratio2 9.2157 +Epoch [117/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.6078, Pure Ratio2 9.5686 +Epoch [117/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.7059, Pure Ratio2 9.6422 +Epoch [117/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7059, Pure Ratio2 9.7255 +Epoch [117/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0005, Loss2: 0.0002, Pure Ratio1: 9.7386, Pure Ratio2 9.7484 +Epoch [117/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7507, Pure Ratio2 9.7619 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test images: Model1 26.5224 % Model2 26.7829 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.7662 % +Training cifar10_coteaching_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.2745, Pure Ratio2 10.1176 +Epoch [118/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 10.0490 +Epoch [118/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8562, Pure Ratio2 9.9085 +Epoch [118/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8088, Pure Ratio2 9.8676 +Epoch [118/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8706, Pure Ratio2 9.8510 +Epoch [118/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7680, Pure Ratio2 9.7582 +Epoch [118/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0001, Pure Ratio1: 9.8599, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test images: Model1 26.0317 % Model2 25.8113 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1569, Pure Ratio2 10.1176 +Epoch [119/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.3725, Pure Ratio2 10.2059 +Epoch [119/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0915, Pure Ratio2 10.0131 +Epoch [119/200], Iter [200/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 10.1078, Pure Ratio2 10.0000 +Epoch [119/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 10.0235, Pure Ratio2 9.9725 +Epoch [119/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9052, Pure Ratio2 9.8366 +Epoch [119/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8263, Pure Ratio2 9.7283 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test images: Model1 25.5809 % Model2 26.3321 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7486 % +Training cifar10_coteaching_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6863, Pure Ratio2 9.6078 +Epoch [120/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.6961 +Epoch [120/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8889, Pure Ratio2 9.7516 +Epoch [120/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7647, Pure Ratio2 9.6814 +Epoch [120/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.7059 +Epoch [120/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8072, Pure Ratio2 9.7320 +Epoch [120/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8207, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test images: Model1 26.2320 % Model2 26.6827 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.7662 % +Training cifar10_coteaching_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 8.8235, Pure Ratio2 9.1765 +Epoch [121/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4412, Pure Ratio2 9.6961 +Epoch [121/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4379, Pure Ratio2 9.7124 +Epoch [121/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.3873, Pure Ratio2 9.5833 +Epoch [121/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.4941, Pure Ratio2 9.6706 +Epoch [121/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6634, Pure Ratio2 9.8105 +Epoch [121/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7423, Pure Ratio2 9.8936 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test images: Model1 25.5909 % Model2 26.7728 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 9.8627 +Epoch [122/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.6373 +Epoch [122/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0523, Pure Ratio2 9.8824 +Epoch [122/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.6765 +Epoch [122/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8000, Pure Ratio2 9.7059 +Epoch [122/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6797 +Epoch [122/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8655, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test images: Model1 26.0417 % Model2 26.6126 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.7562 % +Training cifar10_coteaching_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8824, Pure Ratio2 9.9608 +Epoch [123/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9412, Pure Ratio2 9.9314 +Epoch [123/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8105 +Epoch [123/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7794, Pure Ratio2 9.7157 +Epoch [123/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.8039 +Epoch [123/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.6928 +Epoch [123/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test images: Model1 27.2736 % Model2 25.7011 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9020, Pure Ratio2 10.1961 +Epoch [124/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Epoch [124/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8366, Pure Ratio2 9.7974 +Epoch [124/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8775 +Epoch [124/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8314, Pure Ratio2 9.8353 +Epoch [124/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8856, Pure Ratio2 9.8922 +Epoch [124/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.7227, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test images: Model1 25.5509 % Model2 25.8113 %, Pure Ratio 1 9.8064 %, Pure Ratio 2 9.8215 % +Training cifar10_coteaching_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 8.7059, Pure Ratio2 8.6863 +Epoch [125/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.0980, Pure Ratio2 9.1765 +Epoch [125/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4248, Pure Ratio2 9.4902 +Epoch [125/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7304, Pure Ratio2 9.7402 +Epoch [125/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0005, Pure Ratio1: 9.7843, Pure Ratio2 9.7333 +Epoch [125/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0018, Loss2: 0.0006, Pure Ratio1: 9.8399, Pure Ratio2 9.7810 +Epoch [125/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7591, Pure Ratio2 9.7255 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test images: Model1 25.3506 % Model2 26.0116 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 8.8235, Pure Ratio2 8.6667 +Epoch [126/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 8.9902, Pure Ratio2 8.8431 +Epoch [126/200], Iter [150/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2353, Pure Ratio2 9.1438 +Epoch [126/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3824, Pure Ratio2 9.3431 +Epoch [126/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4980, Pure Ratio2 9.4196 +Epoch [126/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5065, Pure Ratio2 9.4869 +Epoch [126/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5994, Pure Ratio2 9.5602 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test images: Model1 26.7428 % Model2 26.3421 %, Pure Ratio 1 9.7587 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1569, Pure Ratio2 9.8431 +Epoch [127/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.3039 +Epoch [127/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6405, Pure Ratio2 9.4706 +Epoch [127/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7696, Pure Ratio2 9.6029 +Epoch [127/200], Iter [250/390] Training Accuracy1: 95.3125, Training Accuracy2: 95.3125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.6157 +Epoch [127/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7222, Pure Ratio2 9.6046 +Epoch [127/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0004, Pure Ratio1: 9.7451, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test images: Model1 26.3021 % Model2 27.0433 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.1961, Pure Ratio2 9.3333 +Epoch [128/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5784, Pure Ratio2 9.5980 +Epoch [128/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6144, Pure Ratio2 9.6209 +Epoch [128/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.7843 +Epoch [128/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.8314 +Epoch [128/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8366, Pure Ratio2 9.8725 +Epoch [128/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test images: Model1 26.2921 % Model2 26.7328 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0784 +Epoch [129/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7549 +Epoch [129/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8301, Pure Ratio2 9.8627 +Epoch [129/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.6961 +Epoch [129/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7333, Pure Ratio2 9.6863 +Epoch [129/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7124, Pure Ratio2 9.6242 +Epoch [129/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test images: Model1 26.7127 % Model2 26.6727 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8824, Pure Ratio2 10.0196 +Epoch [130/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.7647 +Epoch [130/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0002, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.9739 +Epoch [130/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6520, Pure Ratio2 9.6716 +Epoch [130/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8431, Pure Ratio2 9.8510 +Epoch [130/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.8105 +Epoch [130/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8487, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test images: Model1 26.7728 % Model2 25.9615 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.4902 +Epoch [131/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5686 +Epoch [131/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8693 +Epoch [131/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9608, Pure Ratio2 9.9069 +Epoch [131/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7686, Pure Ratio2 9.7333 +Epoch [131/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.8562, Pure Ratio2 9.8137 +Epoch [131/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8319, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test images: Model1 26.4523 % Model2 26.7328 %, Pure Ratio 1 9.8115 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7255 +Epoch [132/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8529, Pure Ratio2 9.6765 +Epoch [132/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.6797 +Epoch [132/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.7451, Pure Ratio2 9.7353 +Epoch [132/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6980, Pure Ratio2 9.6941 +Epoch [132/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.7876, Pure Ratio2 9.7974 +Epoch [132/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test images: Model1 26.7528 % Model2 26.2720 %, Pure Ratio 1 9.8240 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4118, Pure Ratio2 10.5490 +Epoch [133/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 10.0588, Pure Ratio2 10.1765 +Epoch [133/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.9281, Pure Ratio2 9.9477 +Epoch [133/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6814, Pure Ratio2 9.7696 +Epoch [133/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0001, Pure Ratio1: 9.7020, Pure Ratio2 9.7804 +Epoch [133/200], Iter [300/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6765, Pure Ratio2 9.7418 +Epoch [133/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6919, Pure Ratio2 9.7451 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test images: Model1 26.8129 % Model2 26.7127 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.8316 % +Training cifar10_coteaching_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.3137, Pure Ratio2 9.4118 +Epoch [134/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8333 +Epoch [134/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7059 +Epoch [134/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9167, Pure Ratio2 9.8873 +Epoch [134/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8275, Pure Ratio2 9.8000 +Epoch [134/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6732, Pure Ratio2 9.6667 +Epoch [134/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test images: Model1 26.4824 % Model2 26.8229 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.8431, Pure Ratio2 8.9020 +Epoch [135/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.4510, Pure Ratio2 9.6176 +Epoch [135/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.0131, Pure Ratio2 10.0392 +Epoch [135/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8971, Pure Ratio2 9.9804 +Epoch [135/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.8118 +Epoch [135/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7680, Pure Ratio2 9.8170 +Epoch [135/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7731, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test images: Model1 26.4022 % Model2 26.3221 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7255, Pure Ratio2 9.6863 +Epoch [136/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.5392 +Epoch [136/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7386, Pure Ratio2 9.5817 +Epoch [136/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.7353 +Epoch [136/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7961, Pure Ratio2 9.6824 +Epoch [136/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.7157 +Epoch [136/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0003, Pure Ratio1: 9.8095, Pure Ratio2 9.6975 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test images: Model1 26.4523 % Model2 26.4022 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0004, Pure Ratio1: 9.5294, Pure Ratio2 9.5882 +Epoch [137/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.6275 +Epoch [137/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.6601 +Epoch [137/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6520, Pure Ratio2 9.5686 +Epoch [137/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0002, Loss2: 0.0001, Pure Ratio1: 9.8392, Pure Ratio2 9.7647 +Epoch [137/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.7516 +Epoch [137/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.8067, Pure Ratio2 9.7591 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test images: Model1 26.9732 % Model2 26.2720 %, Pure Ratio 1 9.8014 %, Pure Ratio 2 9.7511 % +Training cifar10_coteaching_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3725, Pure Ratio2 10.3922 +Epoch [138/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7647 +Epoch [138/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7320, Pure Ratio2 9.7647 +Epoch [138/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6814, Pure Ratio2 9.7157 +Epoch [138/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 87.5000, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.7176 +Epoch [138/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6797, Pure Ratio2 9.7614 +Epoch [138/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0006, Pure Ratio1: 9.7227, Pure Ratio2 9.8123 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test images: Model1 25.5308 % Model2 26.2520 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.8492 % +Training cifar10_coteaching_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.3922 +Epoch [139/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.2157 +Epoch [139/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8235 +Epoch [139/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6618, Pure Ratio2 9.6863 +Epoch [139/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8275 +Epoch [139/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7614, Pure Ratio2 9.8137 +Epoch [139/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7367, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test images: Model1 25.7512 % Model2 25.7512 %, Pure Ratio 1 9.6833 %, Pure Ratio 2 9.7587 % +Training cifar10_coteaching_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.6667 +Epoch [140/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8137 +Epoch [140/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9346, Pure Ratio2 9.9412 +Epoch [140/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7941, Pure Ratio2 9.7990 +Epoch [140/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6745, Pure Ratio2 9.7020 +Epoch [140/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6013, Pure Ratio2 9.5752 +Epoch [140/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6555, Pure Ratio2 9.5854 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test images: Model1 26.2019 % Model2 26.3021 %, Pure Ratio 1 9.7562 %, Pure Ratio 2 9.6757 % +Training cifar10_coteaching_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0980, Pure Ratio2 9.2157 +Epoch [141/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2843, Pure Ratio2 9.2451 +Epoch [141/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5621, Pure Ratio2 9.6209 +Epoch [141/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5686, Pure Ratio2 9.5735 +Epoch [141/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6745, Pure Ratio2 9.7333 +Epoch [141/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.6503 +Epoch [141/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test images: Model1 26.8129 % Model2 26.9131 %, Pure Ratio 1 9.6707 %, Pure Ratio 2 9.6757 % +Training cifar10_coteaching_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.8824 +Epoch [142/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8333 +Epoch [142/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.8693, Pure Ratio2 9.8562 +Epoch [142/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.9510, Pure Ratio2 9.9559 +Epoch [142/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8941, Pure Ratio2 9.9059 +Epoch [142/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8464, Pure Ratio2 9.8595 +Epoch [142/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.8459, Pure Ratio2 9.8655 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test images: Model1 26.7228 % Model2 26.9832 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6471, Pure Ratio2 9.6667 +Epoch [143/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.5196 +Epoch [143/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.5490 +Epoch [143/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7696, Pure Ratio2 9.6176 +Epoch [143/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8196, Pure Ratio2 9.7412 +Epoch [143/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7092 +Epoch [143/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0005, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test images: Model1 24.6194 % Model2 26.1518 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [144/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1275, Pure Ratio2 10.1471 +Epoch [144/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8954, Pure Ratio2 9.9150 +Epoch [144/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8186 +Epoch [144/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7294, Pure Ratio2 9.7647 +Epoch [144/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.7222, Pure Ratio2 9.7320 +Epoch [144/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.9375, Loss1: 0.0004, Loss2: 0.0000, Pure Ratio1: 9.7563, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test images: Model1 25.7913 % Model2 26.5925 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.8140 % +Training cifar10_coteaching_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1961, Pure Ratio2 10.2353 +Epoch [145/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.0007, Loss2: 0.0003, Pure Ratio1: 9.8137, Pure Ratio2 9.9118 +Epoch [145/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.8627 +Epoch [145/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8039, Pure Ratio2 9.7843 +Epoch [145/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7490 +Epoch [145/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7059 +Epoch [145/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7339, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test images: Model1 25.9615 % Model2 27.7845 %, Pure Ratio 1 9.7989 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.9216 +Epoch [146/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9020 +Epoch [146/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.8170 +Epoch [146/200], Iter [200/390] Training Accuracy1: 94.5312, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.6814, Pure Ratio2 9.8382 +Epoch [146/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6706, Pure Ratio2 9.8275 +Epoch [146/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.7418 +Epoch [146/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.7927 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test images: Model1 26.4223 % Model2 25.2404 %, Pure Ratio 1 9.7084 %, Pure Ratio 2 9.7888 % +Training cifar10_coteaching_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.6275 +Epoch [147/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.9118 +Epoch [147/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7386, Pure Ratio2 9.8105 +Epoch [147/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.7304 +Epoch [147/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7569 +Epoch [147/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7745 +Epoch [147/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test images: Model1 27.2336 % Model2 25.5208 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [148/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.7451 +Epoch [148/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7974, Pure Ratio2 9.7582 +Epoch [148/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.8725 +Epoch [148/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8000, Pure Ratio2 9.7608 +Epoch [148/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7876, Pure Ratio2 9.7843 +Epoch [148/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test images: Model1 26.8730 % Model2 26.8129 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.9020, Pure Ratio2 9.9804 +Epoch [149/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.7941 +Epoch [149/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.1307, Pure Ratio2 10.0784 +Epoch [149/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0147 +Epoch [149/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9255, Pure Ratio2 9.8824 +Epoch [149/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.9688, Loss1: 0.0006, Loss2: 0.0000, Pure Ratio1: 9.9150, Pure Ratio2 9.8791 +Epoch [149/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.8852, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test images: Model1 26.5625 % Model2 26.4724 %, Pure Ratio 1 9.8441 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0980 +Epoch [150/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0000 +Epoch [150/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8170, Pure Ratio2 9.8431 +Epoch [150/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8284, Pure Ratio2 9.8627 +Epoch [150/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6667, Pure Ratio2 9.6902 +Epoch [150/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7157, Pure Ratio2 9.7320 +Epoch [150/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6919, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test images: Model1 26.5425 % Model2 26.5925 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8235 +Epoch [151/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 82.8125, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.3529, Pure Ratio2 9.5294 +Epoch [151/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6797 +Epoch [151/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.6471, Pure Ratio2 9.6716 +Epoch [151/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5843, Pure Ratio2 9.6196 +Epoch [151/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6307, Pure Ratio2 9.6699 +Epoch [151/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test images: Model1 26.2320 % Model2 26.3421 %, Pure Ratio 1 9.7360 %, Pure Ratio 2 9.7612 % +Training cifar10_coteaching_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.3725 +Epoch [152/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.4314 +Epoch [152/200], Iter [150/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6797 +Epoch [152/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.5441 +Epoch [152/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5059, Pure Ratio2 9.5098 +Epoch [152/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6438, Pure Ratio2 9.6471 +Epoch [152/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7199, Pure Ratio2 9.7171 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test images: Model1 26.5625 % Model2 25.8814 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8824 +Epoch [153/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6961, Pure Ratio2 9.8627 +Epoch [153/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7908 +Epoch [153/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.9412 +Epoch [153/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.8235, Pure Ratio2 9.9569 +Epoch [153/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8072, Pure Ratio2 9.9314 +Epoch [153/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.9216 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test images: Model1 26.5525 % Model2 27.6042 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.8894 % +Training cifar10_coteaching_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 10.0000, Pure Ratio2 9.7843 +Epoch [154/200], Iter [100/390] Training Accuracy1: 95.3125, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 10.1078, Pure Ratio2 10.0980 +Epoch [154/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8431 +Epoch [154/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8775, Pure Ratio2 9.7843 +Epoch [154/200], Iter [250/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9294, Pure Ratio2 9.8471 +Epoch [154/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8366, Pure Ratio2 9.7745 +Epoch [154/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.8459 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test images: Model1 26.0917 % Model2 26.8229 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.5294, Pure Ratio2 9.3333 +Epoch [155/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.5588, Pure Ratio2 9.5196 +Epoch [155/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.5425, Pure Ratio2 9.4575 +Epoch [155/200], Iter [200/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.4657 +Epoch [155/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7725, Pure Ratio2 9.7098 +Epoch [155/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7876, Pure Ratio2 9.7582 +Epoch [155/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0003, Loss2: 0.0000, Pure Ratio1: 9.7871, Pure Ratio2 9.7423 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test images: Model1 26.2921 % Model2 26.3722 %, Pure Ratio 1 9.8165 %, Pure Ratio 2 9.7637 % +Training cifar10_coteaching_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.4902 +Epoch [156/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9706, Pure Ratio2 9.8431 +Epoch [156/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.8301 +Epoch [156/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9902, Pure Ratio2 9.9755 +Epoch [156/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.9490 +Epoch [156/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8725 +Epoch [156/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7283, Pure Ratio2 9.7731 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test images: Model1 26.1018 % Model2 26.8930 %, Pure Ratio 1 9.7863 %, Pure Ratio 2 9.8441 % +Training cifar10_coteaching_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2549, Pure Ratio2 9.3529 +Epoch [157/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6667 +Epoch [157/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.6405 +Epoch [157/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0001, Loss2: 0.0010, Pure Ratio1: 9.7892, Pure Ratio2 9.7549 +Epoch [157/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8078, Pure Ratio2 9.7529 +Epoch [157/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8203, Pure Ratio2 9.7484 +Epoch [157/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8151, Pure Ratio2 9.7899 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test images: Model1 26.7127 % Model2 25.7612 %, Pure Ratio 1 9.8014 %, Pure Ratio 2 9.7863 % +Training cifar10_coteaching_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.2745, Pure Ratio2 8.9216 +Epoch [158/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.3529 +Epoch [158/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.4510 +Epoch [158/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.4755, Pure Ratio2 9.4265 +Epoch [158/200], Iter [250/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.6275 +Epoch [158/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6307, Pure Ratio2 9.6209 +Epoch [158/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0001, Pure Ratio1: 9.6443, Pure Ratio2 9.6331 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test images: Model1 26.9631 % Model2 26.4824 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.5882 +Epoch [159/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.4314 +Epoch [159/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.4967 +Epoch [159/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.4461 +Epoch [159/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5412, Pure Ratio2 9.5216 +Epoch [159/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.7059 +Epoch [159/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7479, Pure Ratio2 9.7563 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test images: Model1 25.3606 % Model2 27.0032 %, Pure Ratio 1 9.7712 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.4314, Pure Ratio2 10.1373 +Epoch [160/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0004, Loss2: 0.0002, Pure Ratio1: 9.9314, Pure Ratio2 9.8137 +Epoch [160/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8235 +Epoch [160/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.8578 +Epoch [160/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8157, Pure Ratio2 9.8588 +Epoch [160/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7516 +Epoch [160/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7647 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test images: Model1 27.3538 % Model2 27.1735 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.2549, Pure Ratio2 10.6863 +Epoch [161/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9314, Pure Ratio2 10.1961 +Epoch [161/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.6340 +Epoch [161/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5588, Pure Ratio2 9.7794 +Epoch [161/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6745, Pure Ratio2 9.8000 +Epoch [161/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6307, Pure Ratio2 9.7712 +Epoch [161/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6751, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test images: Model1 26.5325 % Model2 25.4908 %, Pure Ratio 1 9.7260 %, Pure Ratio 2 9.8341 % +Training cifar10_coteaching_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.8235 +Epoch [162/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0008, Pure Ratio1: 9.6471, Pure Ratio2 9.7647 +Epoch [162/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.6601 +Epoch [162/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0003, Pure Ratio1: 9.4951, Pure Ratio2 9.5343 +Epoch [162/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7490, Pure Ratio2 9.7882 +Epoch [162/200], Iter [300/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7876, Pure Ratio2 9.8007 +Epoch [162/200], Iter [350/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8207, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test images: Model1 26.9331 % Model2 26.4623 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.8014 % +Training cifar10_coteaching_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.2941 +Epoch [163/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1275, Pure Ratio2 9.1765 +Epoch [163/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2288, Pure Ratio2 9.2941 +Epoch [163/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.3284, Pure Ratio2 9.4657 +Epoch [163/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.4353, Pure Ratio2 9.5333 +Epoch [163/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5850, Pure Ratio2 9.6438 +Epoch [163/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6667, Pure Ratio2 9.7087 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test images: Model1 26.0817 % Model2 26.3221 %, Pure Ratio 1 9.7411 %, Pure Ratio 2 9.7738 % +Training cifar10_coteaching_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5490, Pure Ratio2 9.5098 +Epoch [164/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6275 +Epoch [164/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4510, Pure Ratio2 9.3987 +Epoch [164/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5049, Pure Ratio2 9.5245 +Epoch [164/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5843, Pure Ratio2 9.5294 +Epoch [164/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6373, Pure Ratio2 9.6275 +Epoch [164/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6947, Pure Ratio2 9.6779 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test images: Model1 26.8530 % Model2 27.8245 %, Pure Ratio 1 9.7461 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.3137 +Epoch [165/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8137 +Epoch [165/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7843 +Epoch [165/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6863 +Epoch [165/200], Iter [250/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5647, Pure Ratio2 9.5608 +Epoch [165/200], Iter [300/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7778 +Epoch [165/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7675 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test images: Model1 26.4123 % Model2 26.7328 %, Pure Ratio 1 9.7964 %, Pure Ratio 2 9.8115 % +Training cifar10_coteaching_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.2353 +Epoch [166/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7549, Pure Ratio2 9.6863 +Epoch [166/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.6209 +Epoch [166/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0010, Pure Ratio1: 9.9020, Pure Ratio2 9.9559 +Epoch [166/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7922, Pure Ratio2 9.8235 +Epoch [166/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8562 +Epoch [166/200], Iter [350/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8319, Pure Ratio2 9.8683 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test images: Model1 26.7929 % Model2 26.6226 %, Pure Ratio 1 9.7360 %, Pure Ratio 2 9.7813 % +Training cifar10_coteaching_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.4510, Pure Ratio2 9.3529 +Epoch [167/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6373 +Epoch [167/200], Iter [150/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6209, Pure Ratio2 9.5556 +Epoch [167/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5980, Pure Ratio2 9.4951 +Epoch [167/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6941, Pure Ratio2 9.5843 +Epoch [167/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7059 +Epoch [167/200], Iter [350/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7311, Pure Ratio2 9.6639 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test images: Model1 26.9030 % Model2 26.7127 %, Pure Ratio 1 9.7788 %, Pure Ratio 2 9.7260 % +Training cifar10_coteaching_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0588, Pure Ratio2 9.7059 +Epoch [168/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.4118 +Epoch [168/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6797, Pure Ratio2 9.5294 +Epoch [168/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9118, Pure Ratio2 9.7892 +Epoch [168/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.7098 +Epoch [168/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [168/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8908, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test images: Model1 25.7312 % Model2 26.2420 %, Pure Ratio 1 9.8391 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.5294 +Epoch [169/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.7745 +Epoch [169/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7778, Pure Ratio2 9.7516 +Epoch [169/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8137, Pure Ratio2 9.8186 +Epoch [169/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6824, Pure Ratio2 9.6863 +Epoch [169/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.7712 +Epoch [169/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test images: Model1 25.3305 % Model2 26.6526 %, Pure Ratio 1 9.7813 %, Pure Ratio 2 9.7712 % +Training cifar10_coteaching_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.6275 +Epoch [170/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5000, Pure Ratio2 9.4902 +Epoch [170/200], Iter [150/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3595, Pure Ratio2 9.3922 +Epoch [170/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5147, Pure Ratio2 9.5686 +Epoch [170/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6314, Pure Ratio2 9.6745 +Epoch [170/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.6307 +Epoch [170/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7535, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test images: Model1 26.9331 % Model2 27.4139 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.3333, Pure Ratio2 10.2549 +Epoch [171/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0098 +Epoch [171/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8497, Pure Ratio2 9.9216 +Epoch [171/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.9461 +Epoch [171/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8549, Pure Ratio2 9.8824 +Epoch [171/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8791, Pure Ratio2 9.9052 +Epoch [171/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7199, Pure Ratio2 9.7703 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test images: Model1 26.0617 % Model2 26.7428 %, Pure Ratio 1 9.7285 %, Pure Ratio 2 9.7511 % +Training cifar10_coteaching_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0001, Loss2: 0.0002, Pure Ratio1: 9.9412, Pure Ratio2 9.8824 +Epoch [172/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5196, Pure Ratio2 9.4510 +Epoch [172/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.5621 +Epoch [172/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.6029 +Epoch [172/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6314, Pure Ratio2 9.6235 +Epoch [172/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.6797 +Epoch [172/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7199, Pure Ratio2 9.6891 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test images: Model1 26.6727 % Model2 26.5525 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.7310 % +Training cifar10_coteaching_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.5098 +Epoch [173/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2941, Pure Ratio2 9.3627 +Epoch [173/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4444, Pure Ratio2 9.3987 +Epoch [173/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5245, Pure Ratio2 9.4657 +Epoch [173/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5686, Pure Ratio2 9.5647 +Epoch [173/200], Iter [300/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.6373 +Epoch [173/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7227, Pure Ratio2 9.6947 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test images: Model1 26.7127 % Model2 26.7929 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 10.0980 +Epoch [174/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.8922 +Epoch [174/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5817, Pure Ratio2 9.8301 +Epoch [174/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.8137 +Epoch [174/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.8314 +Epoch [174/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8203 +Epoch [174/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8011, Pure Ratio2 9.8067 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test images: Model1 26.3922 % Model2 27.1334 %, Pure Ratio 1 9.7536 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.9804 +Epoch [175/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4902, Pure Ratio2 9.5588 +Epoch [175/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6732, Pure Ratio2 9.6536 +Epoch [175/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5245, Pure Ratio2 9.5539 +Epoch [175/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.6588 +Epoch [175/200], Iter [300/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6993, Pure Ratio2 9.7255 +Epoch [175/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7171, Pure Ratio2 9.7339 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test images: Model1 25.8914 % Model2 26.3722 %, Pure Ratio 1 9.7612 %, Pure Ratio 2 9.7838 % +Training cifar10_coteaching_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5294, Pure Ratio2 9.3529 +Epoch [176/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6176, Pure Ratio2 9.3039 +Epoch [176/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6667, Pure Ratio2 9.4314 +Epoch [176/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.5980 +Epoch [176/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7804, Pure Ratio2 9.6039 +Epoch [176/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7190, Pure Ratio2 9.5425 +Epoch [176/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.6331 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test images: Model1 26.3121 % Model2 25.7913 %, Pure Ratio 1 9.8416 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.4706 +Epoch [177/200], Iter [100/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.0098 +Epoch [177/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.6340 +Epoch [177/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9510, Pure Ratio2 9.8578 +Epoch [177/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8588, Pure Ratio2 9.8118 +Epoch [177/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.7484 +Epoch [177/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7787, Pure Ratio2 9.7395 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test images: Model1 26.2620 % Model2 26.7328 %, Pure Ratio 1 9.8215 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0392, Pure Ratio2 10.0588 +Epoch [178/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.1078 +Epoch [178/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0980, Pure Ratio2 10.0523 +Epoch [178/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0049, Pure Ratio2 9.9804 +Epoch [178/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8784, Pure Ratio2 9.8784 +Epoch [178/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8562, Pure Ratio2 9.8333 +Epoch [178/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8067, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test images: Model1 26.7728 % Model2 26.3121 %, Pure Ratio 1 9.8793 %, Pure Ratio 2 9.8265 % +Training cifar10_coteaching_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1373, Pure Ratio2 9.0392 +Epoch [179/200], Iter [100/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2647, Pure Ratio2 9.2157 +Epoch [179/200], Iter [150/390] Training Accuracy1: 89.0625, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5752, Pure Ratio2 9.4771 +Epoch [179/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6814, Pure Ratio2 9.6716 +Epoch [179/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7490, Pure Ratio2 9.7961 +Epoch [179/200], Iter [300/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7778 +Epoch [179/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8599, Pure Ratio2 9.8375 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test images: Model1 26.6727 % Model2 26.4924 %, Pure Ratio 1 9.7914 %, Pure Ratio 2 9.7788 % +Training cifar10_coteaching_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6863 +Epoch [180/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 9.8725 +Epoch [180/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7516 +Epoch [180/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6912, Pure Ratio2 9.6275 +Epoch [180/200], Iter [250/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6078, Pure Ratio2 9.5412 +Epoch [180/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6275 +Epoch [180/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6359, Pure Ratio2 9.6106 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test images: Model1 26.6026 % Model2 25.9115 %, Pure Ratio 1 9.7461 %, Pure Ratio 2 9.7360 % +Training cifar10_coteaching_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 10.1569 +Epoch [181/200], Iter [100/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.2059 +Epoch [181/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6536, Pure Ratio2 9.7712 +Epoch [181/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.5784 +Epoch [181/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.7490 +Epoch [181/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6503, Pure Ratio2 9.6569 +Epoch [181/200], Iter [350/390] Training Accuracy1: 94.5312, Training Accuracy2: 94.5312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.8095 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test images: Model1 26.6026 % Model2 26.7929 %, Pure Ratio 1 9.7436 %, Pure Ratio 2 9.7939 % +Training cifar10_coteaching_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.2745, Pure Ratio2 9.1961 +Epoch [182/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.4804 +Epoch [182/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6536 +Epoch [182/200], Iter [200/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6667 +Epoch [182/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.7020 +Epoch [182/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8105, Pure Ratio2 9.7516 +Epoch [182/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7899, Pure Ratio2 9.7507 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test images: Model1 26.3121 % Model2 25.9615 %, Pure Ratio 1 9.7637 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5098, Pure Ratio2 9.7059 +Epoch [183/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.8431, Pure Ratio2 9.8627 +Epoch [183/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0327, Pure Ratio2 10.0915 +Epoch [183/200], Iter [200/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.9020, Pure Ratio2 9.9363 +Epoch [183/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7843 +Epoch [183/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.8203 +Epoch [183/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7759, Pure Ratio2 9.7787 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test images: Model1 26.5825 % Model2 26.2520 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7687 % +Training cifar10_coteaching_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 8.9804, Pure Ratio2 8.7647 +Epoch [184/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.1961, Pure Ratio2 9.0196 +Epoch [184/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5556, Pure Ratio2 9.4183 +Epoch [184/200], Iter [200/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.4804 +Epoch [184/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.7294, Pure Ratio2 9.6510 +Epoch [184/200], Iter [300/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.7941 +Epoch [184/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7871, Pure Ratio2 9.7311 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test images: Model1 26.5224 % Model2 26.3622 %, Pure Ratio 1 9.7888 %, Pure Ratio 2 9.7134 % +Training cifar10_coteaching_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4118, Pure Ratio2 9.4118 +Epoch [185/200], Iter [100/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0002, Pure Ratio1: 9.5686, Pure Ratio2 9.5196 +Epoch [185/200], Iter [150/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 9.7843 +Epoch [185/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8873, Pure Ratio2 9.7402 +Epoch [185/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7569, Pure Ratio2 9.6275 +Epoch [185/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8007, Pure Ratio2 9.6863 +Epoch [185/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7423, Pure Ratio2 9.6611 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test images: Model1 26.7528 % Model2 26.7328 %, Pure Ratio 1 9.8039 %, Pure Ratio 2 9.7059 % +Training cifar10_coteaching_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.0980, Pure Ratio2 9.2745 +Epoch [186/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6471, Pure Ratio2 9.7059 +Epoch [186/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8235 +Epoch [186/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6912 +Epoch [186/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8118, Pure Ratio2 9.8078 +Epoch [186/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8333, Pure Ratio2 9.8529 +Epoch [186/200], Iter [350/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8291, Pure Ratio2 9.8571 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test images: Model1 26.7728 % Model2 26.5525 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8089 % +Training cifar10_coteaching_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7255 +Epoch [187/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8529, Pure Ratio2 9.9804 +Epoch [187/200], Iter [150/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.9739 +Epoch [187/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9363, Pure Ratio2 10.0392 +Epoch [187/200], Iter [250/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9647, Pure Ratio2 10.0314 +Epoch [187/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 9.9052 +Epoch [187/200], Iter [350/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8711 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test images: Model1 26.5925 % Model2 26.3822 %, Pure Ratio 1 9.7763 %, Pure Ratio 2 9.8366 % +Training cifar10_coteaching_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.6078 +Epoch [188/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6863 +Epoch [188/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.7516, Pure Ratio2 9.6732 +Epoch [188/200], Iter [200/390] Training Accuracy1: 83.5938, Training Accuracy2: 84.3750, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7010 +Epoch [188/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7922, Pure Ratio2 9.7333 +Epoch [188/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8595, Pure Ratio2 9.8333 +Epoch [188/200], Iter [350/390] Training Accuracy1: 89.8438, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8095, Pure Ratio2 9.7843 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test images: Model1 26.3421 % Model2 26.3221 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9804, Pure Ratio2 10.0588 +Epoch [189/200], Iter [100/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8627, Pure Ratio2 9.8824 +Epoch [189/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.7778 +Epoch [189/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7451, Pure Ratio2 9.6765 +Epoch [189/200], Iter [250/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.7137 +Epoch [189/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8399, Pure Ratio2 9.7059 +Epoch [189/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8347, Pure Ratio2 9.7059 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test images: Model1 26.6727 % Model2 26.4423 %, Pure Ratio 1 9.8190 %, Pure Ratio 2 9.7009 % +Training cifar10_coteaching_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3529, Pure Ratio2 9.4118 +Epoch [190/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.8627 +Epoch [190/200], Iter [150/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8039, Pure Ratio2 9.8235 +Epoch [190/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8725, Pure Ratio2 9.8824 +Epoch [190/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7333, Pure Ratio2 9.7451 +Epoch [190/200], Iter [300/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7418, Pure Ratio2 9.7451 +Epoch [190/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7255, Pure Ratio2 9.7479 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test images: Model1 26.4724 % Model2 26.3221 %, Pure Ratio 1 9.7210 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0000 +Epoch [191/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.7745 +Epoch [191/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8889, Pure Ratio2 9.9020 +Epoch [191/200], Iter [200/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7990, Pure Ratio2 9.8627 +Epoch [191/200], Iter [250/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7216, Pure Ratio2 9.8039 +Epoch [191/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.0312, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8235, Pure Ratio2 9.8758 +Epoch [191/200], Iter [350/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7815 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test images: Model1 26.2320 % Model2 26.4223 %, Pure Ratio 1 9.7939 %, Pure Ratio 2 9.8039 % +Training cifar10_coteaching_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.3922, Pure Ratio2 9.3922 +Epoch [192/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4804, Pure Ratio2 9.5196 +Epoch [192/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6797 +Epoch [192/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5882, Pure Ratio2 9.6667 +Epoch [192/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6039, Pure Ratio2 9.7137 +Epoch [192/200], Iter [300/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.8889 +Epoch [192/200], Iter [350/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7703, Pure Ratio2 9.8543 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test images: Model1 26.3822 % Model2 26.6126 %, Pure Ratio 1 9.7662 %, Pure Ratio 2 9.8592 % +Training cifar10_coteaching_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0002, Loss2: 0.0000, Pure Ratio1: 9.9608, Pure Ratio2 9.8039 +Epoch [193/200], Iter [100/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7451 +Epoch [193/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.6013 +Epoch [193/200], Iter [200/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.1562, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6471 +Epoch [193/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7608, Pure Ratio2 9.7137 +Epoch [193/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6340, Pure Ratio2 9.6046 +Epoch [193/200], Iter [350/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7059, Pure Ratio2 9.6863 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test images: Model1 25.8814 % Model2 26.2620 %, Pure Ratio 1 9.7335 %, Pure Ratio 2 9.7461 % +Training cifar10_coteaching_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.8235 +Epoch [194/200], Iter [100/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 10.0098, Pure Ratio2 9.9608 +Epoch [194/200], Iter [150/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.0458 +Epoch [194/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 9.8676 +Epoch [194/200], Iter [250/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9333, Pure Ratio2 9.8784 +Epoch [194/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9183, Pure Ratio2 9.8954 +Epoch [194/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8768, Pure Ratio2 9.8768 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test images: Model1 26.1619 % Model2 26.0417 %, Pure Ratio 1 9.8089 %, Pure Ratio 2 9.8291 % +Training cifar10_coteaching_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1765, Pure Ratio2 10.3137 +Epoch [195/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0000, Pure Ratio2 10.0784 +Epoch [195/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9935, Pure Ratio2 10.0196 +Epoch [195/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8922, Pure Ratio2 9.8873 +Epoch [195/200], Iter [250/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7882, Pure Ratio2 9.7725 +Epoch [195/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7810, Pure Ratio2 9.7549 +Epoch [195/200], Iter [350/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7367, Pure Ratio2 9.7031 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test images: Model1 25.7612 % Model2 25.9515 %, Pure Ratio 1 9.7687 %, Pure Ratio 2 9.7436 % +Training cifar10_coteaching_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6863, Pure Ratio2 9.6471 +Epoch [196/200], Iter [100/390] Training Accuracy1: 85.1562, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5000, Pure Ratio2 9.6471 +Epoch [196/200], Iter [150/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6601, Pure Ratio2 9.7320 +Epoch [196/200], Iter [200/390] Training Accuracy1: 87.5000, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6716, Pure Ratio2 9.7500 +Epoch [196/200], Iter [250/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6588, Pure Ratio2 9.7373 +Epoch [196/200], Iter [300/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6438, Pure Ratio2 9.7157 +Epoch [196/200], Iter [350/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6639, Pure Ratio2 9.7115 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test images: Model1 26.1118 % Model2 25.6611 %, Pure Ratio 1 9.6782 %, Pure Ratio 2 9.7386 % +Training cifar10_coteaching_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1176, Pure Ratio2 10.1569 +Epoch [197/200], Iter [100/390] Training Accuracy1: 87.5000, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1471, Pure Ratio2 10.1961 +Epoch [197/200], Iter [150/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0261, Pure Ratio2 10.0131 +Epoch [197/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 84.3750, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.0343, Pure Ratio2 9.9657 +Epoch [197/200], Iter [250/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0001, Loss2: 0.0000, Pure Ratio1: 9.9490, Pure Ratio2 9.8549 +Epoch [197/200], Iter [300/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0005, Pure Ratio1: 10.0065, Pure Ratio2 9.9020 +Epoch [197/200], Iter [350/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9216, Pure Ratio2 9.8235 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test images: Model1 25.7011 % Model2 26.3622 %, Pure Ratio 1 9.8492 %, Pure Ratio 2 9.7562 % +Training cifar10_coteaching_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 89.8438, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8824, Pure Ratio2 9.7647 +Epoch [198/200], Iter [100/390] Training Accuracy1: 93.7500, Training Accuracy2: 93.7500, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.5392, Pure Ratio2 9.4804 +Epoch [198/200], Iter [150/390] Training Accuracy1: 90.6250, Training Accuracy2: 90.6250, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6275, Pure Ratio2 9.6013 +Epoch [198/200], Iter [200/390] Training Accuracy1: 90.6250, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.6618, Pure Ratio2 9.6520 +Epoch [198/200], Iter [250/390] Training Accuracy1: 85.1562, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0001, Pure Ratio1: 9.6000, Pure Ratio2 9.5137 +Epoch [198/200], Iter [300/390] Training Accuracy1: 88.2812, Training Accuracy2: 88.2812, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7353, Pure Ratio2 9.6471 +Epoch [198/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 87.5000, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7843, Pure Ratio2 9.7143 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test images: Model1 25.9215 % Model2 26.0717 %, Pure Ratio 1 9.8265 %, Pure Ratio 2 9.7536 % +Training cifar10_coteaching_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.8438, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8431, Pure Ratio2 10.0000 +Epoch [199/200], Iter [100/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9412, Pure Ratio2 10.0686 +Epoch [199/200], Iter [150/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7582, Pure Ratio2 9.7908 +Epoch [199/200], Iter [200/390] Training Accuracy1: 86.7188, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.4706, Pure Ratio2 9.5294 +Epoch [199/200], Iter [250/390] Training Accuracy1: 83.5938, Training Accuracy2: 83.5938, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7490, Pure Ratio2 9.7569 +Epoch [199/200], Iter [300/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8595, Pure Ratio2 9.8660 +Epoch [199/200], Iter [350/390] Training Accuracy1: 87.5000, Training Accuracy2: 86.7188, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8067, Pure Ratio2 9.8319 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test images: Model1 26.5124 % Model2 25.8514 %, Pure Ratio 1 9.7738 %, Pure Ratio 2 9.7562 % +Training cifar10_coteaching_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 91.4062, Training Accuracy2: 91.4062, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 10.1373, Pure Ratio2 10.5098 +Epoch [200/200], Iter [100/390] Training Accuracy1: 88.2812, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7941, Pure Ratio2 10.0294 +Epoch [200/200], Iter [150/390] Training Accuracy1: 92.1875, Training Accuracy2: 92.1875, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.9346, Pure Ratio2 10.0850 +Epoch [200/200], Iter [200/390] Training Accuracy1: 92.9688, Training Accuracy2: 92.9688, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7157, Pure Ratio2 9.8480 +Epoch [200/200], Iter [250/390] Training Accuracy1: 85.9375, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.8000, Pure Ratio2 9.8549 +Epoch [200/200], Iter [300/390] Training Accuracy1: 86.7188, Training Accuracy2: 85.9375, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7745, Pure Ratio2 9.8268 +Epoch [200/200], Iter [350/390] Training Accuracy1: 89.0625, Training Accuracy2: 89.0625, Loss1: 0.0000, Loss2: 0.0000, Pure Ratio1: 9.7647, Pure Ratio2 9.7955 +Evaluating cifar10_coteaching_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test images: Model1 25.9916 % Model2 25.8614 %, Pure Ratio 1 9.7134 %, Pure Ratio 2 9.7235 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_0_2.log b/other_methods/coteaching_plus/coteaching_plus_results/out_0_2.log new file mode 100644 index 0000000..1abbb9b --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_0_2.log @@ -0,0 +1,2015 @@ +Files already downloaded and verified +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 18.7500, Training Accuracy2: 21.0938, Loss1: 0.0177, Loss2: 0.0175 +Epoch [2/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 24.2188, Loss1: 0.0161, Loss2: 0.0159 +Epoch [2/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0158, Loss2: 0.0159 +Epoch [2/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 36.7188, Loss1: 0.0162, Loss2: 0.0164 +Epoch [2/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0143, Loss2: 0.0149 +Epoch [2/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0138, Loss2: 0.0138 +Epoch [2/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 32.7724 % Model2 35.2163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0150 +Epoch [3/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0153 +Epoch [3/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0161, Loss2: 0.0156 +Epoch [3/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0143, Loss2: 0.0145 +Epoch [3/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0135, Loss2: 0.0133 +Epoch [3/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 39.8438, Loss1: 0.0142, Loss2: 0.0139 +Epoch [3/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0130, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 40.6150 % Model2 42.0172 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0136, Loss2: 0.0134 +Epoch [4/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0136, Loss2: 0.0134 +Epoch [4/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0135, Loss2: 0.0133 +Epoch [4/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0125, Loss2: 0.0123 +Epoch [4/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0130, Loss2: 0.0124 +Epoch [4/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0128, Loss2: 0.0129 +Epoch [4/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 34.3750, Loss1: 0.0136, Loss2: 0.0132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 44.3209 % Model2 45.8634 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0129, Loss2: 0.0127 +Epoch [5/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0132, Loss2: 0.0123 +Epoch [5/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0122, Loss2: 0.0117 +Epoch [5/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0141, Loss2: 0.0139 +Epoch [5/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.8438, Loss1: 0.0120, Loss2: 0.0127 +Epoch [5/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0139 +Epoch [5/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0132, Loss2: 0.0133 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 45.0321 % Model2 47.3658 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0113, Loss2: 0.0112 +Epoch [6/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0117, Loss2: 0.0119 +Epoch [6/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0124, Loss2: 0.0118 +Epoch [6/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0126, Loss2: 0.0127 +Epoch [6/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0132, Loss2: 0.0126 +Epoch [6/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0114, Loss2: 0.0111 +Epoch [6/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0121, Loss2: 0.0124 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 48.5176 % Model2 48.7079 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0110, Loss2: 0.0106 +Epoch [7/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0106, Loss2: 0.0109 +Epoch [7/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0122, Loss2: 0.0128 +Epoch [7/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0123, Loss2: 0.0123 +Epoch [7/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0125, Loss2: 0.0120 +Epoch [7/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0115, Loss2: 0.0117 +Epoch [7/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0123, Loss2: 0.0115 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 50.2804 % Model2 51.5725 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0112, Loss2: 0.0107 +Epoch [8/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0122, Loss2: 0.0116 +Epoch [8/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0103, Loss2: 0.0096 +Epoch [8/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0104, Loss2: 0.0106 +Epoch [8/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0096, Loss2: 0.0092 +Epoch [8/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0126, Loss2: 0.0130 +Epoch [8/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0109, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 50.9215 % Model2 53.5557 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0105, Loss2: 0.0103 +Epoch [9/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0110, Loss2: 0.0117 +Epoch [9/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0115, Loss2: 0.0106 +Epoch [9/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0105, Loss2: 0.0100 +Epoch [9/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0119, Loss2: 0.0116 +Epoch [9/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0114, Loss2: 0.0110 +Epoch [9/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0116, Loss2: 0.0114 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 52.2135 % Model2 53.3954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0113, Loss2: 0.0112 +Epoch [10/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0094, Loss2: 0.0091 +Epoch [10/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0103, Loss2: 0.0102 +Epoch [10/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0098, Loss2: 0.0108 +Epoch [10/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 42.9688, Loss1: 0.0092, Loss2: 0.0101 +Epoch [10/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0119 +Epoch [10/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0096, Loss2: 0.0097 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 53.5857 % Model2 54.2067 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0106, Loss2: 0.0102 +Epoch [11/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0084, Loss2: 0.0092 +Epoch [11/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0137, Loss2: 0.0126 +Epoch [11/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0083, Loss2: 0.0089 +Epoch [11/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0091, Loss2: 0.0092 +Epoch [11/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.0312, Loss1: 0.0096, Loss2: 0.0087 +Epoch [11/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0100, Loss2: 0.0089 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 54.1066 % Model2 54.9579 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0088, Loss2: 0.0080 +Epoch [12/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0114, Loss2: 0.0107 +Epoch [12/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0093, Loss2: 0.0093 +Epoch [12/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0110, Loss2: 0.0105 +Epoch [12/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0116, Loss2: 0.0105 +Epoch [12/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0087, Loss2: 0.0074 +Epoch [12/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0097, Loss2: 0.0088 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 54.4872 % Model2 56.4403 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0101, Loss2: 0.0097 +Epoch [13/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0103, Loss2: 0.0103 +Epoch [13/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0102, Loss2: 0.0097 +Epoch [13/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0103, Loss2: 0.0105 +Epoch [13/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0074, Loss2: 0.0070 +Epoch [13/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0070, Loss2: 0.0070 +Epoch [13/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0087, Loss2: 0.0086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 55.3886 % Model2 57.2716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0088, Loss2: 0.0082 +Epoch [14/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0117, Loss2: 0.0104 +Epoch [14/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0088, Loss2: 0.0081 +Epoch [14/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0106, Loss2: 0.0097 +Epoch [14/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0094, Loss2: 0.0084 +Epoch [14/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0104, Loss2: 0.0092 +Epoch [14/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0116, Loss2: 0.0116 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 54.7576 % Model2 56.6206 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0087, Loss2: 0.0081 +Epoch [15/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0096, Loss2: 0.0094 +Epoch [15/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0098, Loss2: 0.0101 +Epoch [15/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0095, Loss2: 0.0095 +Epoch [15/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0098, Loss2: 0.0088 +Epoch [15/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0105, Loss2: 0.0093 +Epoch [15/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0106, Loss2: 0.0106 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 55.5288 % Model2 56.7308 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0115, Loss2: 0.0109 +Epoch [16/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0093, Loss2: 0.0083 +Epoch [16/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0089, Loss2: 0.0079 +Epoch [16/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0091, Loss2: 0.0085 +Epoch [16/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0091, Loss2: 0.0093 +Epoch [16/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0103, Loss2: 0.0086 +Epoch [16/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0090, Loss2: 0.0082 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 56.0797 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0085, Loss2: 0.0082 +Epoch [17/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0095, Loss2: 0.0078 +Epoch [17/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 44.5312, Loss1: 0.0105, Loss2: 0.0108 +Epoch [17/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0111, Loss2: 0.0090 +Epoch [17/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0097, Loss2: 0.0088 +Epoch [17/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0090, Loss2: 0.0085 +Epoch [17/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0077, Loss2: 0.0078 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 55.9896 % Model2 57.6923 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0094, Loss2: 0.0082 +Epoch [18/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0092, Loss2: 0.0091 +Epoch [18/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0094, Loss2: 0.0087 +Epoch [18/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0083 +Epoch [18/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0103, Loss2: 0.0095 +Epoch [18/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0092, Loss2: 0.0088 +Epoch [18/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0077, Loss2: 0.0072 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 54.8377 % Model2 56.8309 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0101, Loss2: 0.0097 +Epoch [19/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0089, Loss2: 0.0080 +Epoch [19/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0074, Loss2: 0.0084 +Epoch [19/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0081, Loss2: 0.0082 +Epoch [19/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0075, Loss2: 0.0069 +Epoch [19/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0098, Loss2: 0.0085 +Epoch [19/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0079, Loss2: 0.0086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 56.0296 % Model2 57.4920 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0075, Loss2: 0.0067 +Epoch [20/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0096, Loss2: 0.0097 +Epoch [20/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0062, Loss2: 0.0057 +Epoch [20/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0095, Loss2: 0.0087 +Epoch [20/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0085, Loss2: 0.0081 +Epoch [20/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0093, Loss2: 0.0073 +Epoch [20/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0076, Loss2: 0.0080 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 56.3101 % Model2 57.5220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 54.6875, Loss1: 0.0585, Loss2: 0.0552 +Epoch [21/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0627, Loss2: 0.0618 +Epoch [21/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0564, Loss2: 0.0558 +Epoch [21/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0645, Loss2: 0.0641 +Epoch [21/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0540, Loss2: 0.0533 +Epoch [21/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0764, Loss2: 0.0777 +Epoch [21/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0546, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 53.2853 % Model2 54.8878 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 61.7188, Loss1: 0.0824, Loss2: 0.0770 +Epoch [22/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0554, Loss2: 0.0545 +Epoch [22/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.9062, Loss1: 0.0713, Loss2: 0.0663 +Epoch [22/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0582, Loss2: 0.0572 +Epoch [22/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0632, Loss2: 0.0636 +Epoch [22/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0494, Loss2: 0.0501 +Epoch [22/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0508, Loss2: 0.0529 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 53.9563 % Model2 53.7260 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0559, Loss2: 0.0558 +Epoch [23/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0561, Loss2: 0.0558 +Epoch [23/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0590, Loss2: 0.0596 +Epoch [23/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0861, Loss2: 0.0891 +Epoch [23/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0762, Loss2: 0.0772 +Epoch [23/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0638, Loss2: 0.0585 +Epoch [23/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0602, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 53.1951 % Model2 55.9495 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0769, Loss2: 0.0768 +Epoch [24/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 51.5625, Loss1: 0.0591, Loss2: 0.0603 +Epoch [24/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0597 +Epoch [24/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0712, Loss2: 0.0669 +Epoch [24/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0723, Loss2: 0.0687 +Epoch [24/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 46.8750, Loss1: 0.0679, Loss2: 0.0621 +Epoch [24/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0666, Loss2: 0.0618 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 55.2784 % Model2 57.3518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0700, Loss2: 0.0654 +Epoch [25/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0739, Loss2: 0.0707 +Epoch [25/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0743, Loss2: 0.0730 +Epoch [25/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0670, Loss2: 0.0638 +Epoch [25/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0636, Loss2: 0.0603 +Epoch [25/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0591, Loss2: 0.0565 +Epoch [25/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0700, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 55.6290 % Model2 56.7308 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0516, Loss2: 0.0489 +Epoch [26/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0585, Loss2: 0.0571 +Epoch [26/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0657, Loss2: 0.0635 +Epoch [26/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0653, Loss2: 0.0644 +Epoch [26/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0573, Loss2: 0.0553 +Epoch [26/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0614, Loss2: 0.0631 +Epoch [26/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0648, Loss2: 0.0641 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 55.5889 % Model2 57.8425 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0743, Loss2: 0.0734 +Epoch [27/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0546, Loss2: 0.0545 +Epoch [27/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0536, Loss2: 0.0521 +Epoch [27/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0689, Loss2: 0.0679 +Epoch [27/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0706, Loss2: 0.0681 +Epoch [27/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0672, Loss2: 0.0619 +Epoch [27/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0577, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 55.5288 % Model2 56.5805 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0709, Loss2: 0.0689 +Epoch [28/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0754, Loss2: 0.0749 +Epoch [28/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0583, Loss2: 0.0568 +Epoch [28/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 60.9375, Loss1: 0.0555, Loss2: 0.0512 +Epoch [28/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0911, Loss2: 0.0919 +Epoch [28/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0596, Loss2: 0.0584 +Epoch [28/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0552, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 54.5172 % Model2 55.9996 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0668, Loss2: 0.0654 +Epoch [29/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0768, Loss2: 0.0716 +Epoch [29/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0581, Loss2: 0.0570 +Epoch [29/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0702, Loss2: 0.0657 +Epoch [29/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0737, Loss2: 0.0714 +Epoch [29/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0611, Loss2: 0.0601 +Epoch [29/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0565, Loss2: 0.0561 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 55.8994 % Model2 57.5020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0881, Loss2: 0.0844 +Epoch [30/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0662, Loss2: 0.0682 +Epoch [30/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.0312, Loss1: 0.0651, Loss2: 0.0608 +Epoch [30/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0646, Loss2: 0.0640 +Epoch [30/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0619, Loss2: 0.0645 +Epoch [30/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 60.9375, Loss1: 0.0575, Loss2: 0.0534 +Epoch [30/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0765, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 55.8894 % Model2 57.6422 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0725, Loss2: 0.0742 +Epoch [31/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0673, Loss2: 0.0608 +Epoch [31/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0788, Loss2: 0.0793 +Epoch [31/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0580, Loss2: 0.0560 +Epoch [31/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0596, Loss2: 0.0565 +Epoch [31/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0860, Loss2: 0.0850 +Epoch [31/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0642, Loss2: 0.0627 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 55.3486 % Model2 54.1867 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0701, Loss2: 0.0744 +Epoch [32/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0590, Loss2: 0.0552 +Epoch [32/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0616, Loss2: 0.0597 +Epoch [32/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0718, Loss2: 0.0743 +Epoch [32/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0713, Loss2: 0.0687 +Epoch [32/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0668, Loss2: 0.0672 +Epoch [32/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0587, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 56.8409 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0508, Loss2: 0.0516 +Epoch [33/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0697, Loss2: 0.0710 +Epoch [33/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0500, Loss2: 0.0518 +Epoch [33/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0643, Loss2: 0.0679 +Epoch [33/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0556, Loss2: 0.0535 +Epoch [33/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0720, Loss2: 0.0736 +Epoch [33/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0675, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 55.2985 % Model2 56.3401 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0657, Loss2: 0.0612 +Epoch [34/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0715, Loss2: 0.0723 +Epoch [34/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0633, Loss2: 0.0628 +Epoch [34/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.1094, Loss2: 0.1007 +Epoch [34/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0711, Loss2: 0.0760 +Epoch [34/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0699, Loss2: 0.0692 +Epoch [34/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0707, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 55.5789 % Model2 59.0345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0878, Loss2: 0.0845 +Epoch [35/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0737, Loss2: 0.0708 +Epoch [35/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0539, Loss2: 0.0514 +Epoch [35/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0737, Loss2: 0.0725 +Epoch [35/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.9375, Loss1: 0.0679, Loss2: 0.0624 +Epoch [35/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0710, Loss2: 0.0706 +Epoch [35/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0735, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 55.7893 % Model2 56.9812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0980, Loss2: 0.1023 +Epoch [36/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0655, Loss2: 0.0638 +Epoch [36/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0659, Loss2: 0.0693 +Epoch [36/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0622, Loss2: 0.0625 +Epoch [36/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0526, Loss2: 0.0522 +Epoch [36/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 69.5312, Loss1: 0.0640, Loss2: 0.0556 +Epoch [36/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0631, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 56.0797 % Model2 58.2232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0863, Loss2: 0.0766 +Epoch [37/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0953, Loss2: 0.0934 +Epoch [37/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0619, Loss2: 0.0610 +Epoch [37/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0658, Loss2: 0.0648 +Epoch [37/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0595, Loss2: 0.0587 +Epoch [37/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0677, Loss2: 0.0624 +Epoch [37/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0645, Loss2: 0.0620 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 55.7692 % Model2 57.8626 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0713, Loss2: 0.0740 +Epoch [38/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0645, Loss2: 0.0660 +Epoch [38/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0633, Loss2: 0.0629 +Epoch [38/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0743, Loss2: 0.0755 +Epoch [38/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0659, Loss2: 0.0646 +Epoch [38/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0690, Loss2: 0.0724 +Epoch [38/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0739, Loss2: 0.0736 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 56.9611 % Model2 55.6090 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0665, Loss2: 0.0654 +Epoch [39/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0898, Loss2: 0.0916 +Epoch [39/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0568, Loss2: 0.0581 +Epoch [39/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0583, Loss2: 0.0617 +Epoch [39/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0571, Loss2: 0.0575 +Epoch [39/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0550, Loss2: 0.0601 +Epoch [39/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0838, Loss2: 0.0731 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 56.8910 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0657, Loss2: 0.0651 +Epoch [40/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0726, Loss2: 0.0706 +Epoch [40/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0670, Loss2: 0.0706 +Epoch [40/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 60.1562, Loss1: 0.0618, Loss2: 0.0554 +Epoch [40/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0573, Loss2: 0.0568 +Epoch [40/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0732, Loss2: 0.0732 +Epoch [40/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0825, Loss2: 0.0737 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 56.1899 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 64.8438, Loss1: 0.0620, Loss2: 0.0574 +Epoch [41/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0867, Loss2: 0.0887 +Epoch [41/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0749, Loss2: 0.0714 +Epoch [41/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0518, Loss2: 0.0493 +Epoch [41/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0566, Loss2: 0.0572 +Epoch [41/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0800, Loss2: 0.0720 +Epoch [41/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0651, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 57.4319 % Model2 57.6222 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0747, Loss2: 0.0718 +Epoch [42/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0643, Loss2: 0.0639 +Epoch [42/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0521, Loss2: 0.0506 +Epoch [42/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0698 +Epoch [42/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0615, Loss2: 0.0598 +Epoch [42/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0620, Loss2: 0.0582 +Epoch [42/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0681, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 57.8626 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0644, Loss2: 0.0642 +Epoch [43/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0832, Loss2: 0.0838 +Epoch [43/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0718, Loss2: 0.0726 +Epoch [43/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0741, Loss2: 0.0699 +Epoch [43/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0583, Loss2: 0.0608 +Epoch [43/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0621, Loss2: 0.0640 +Epoch [43/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0608, Loss2: 0.0626 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 56.6807 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0587, Loss2: 0.0546 +Epoch [44/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0901, Loss2: 0.0810 +Epoch [44/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0771, Loss2: 0.0777 +Epoch [44/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0697, Loss2: 0.0684 +Epoch [44/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0792, Loss2: 0.0714 +Epoch [44/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0678, Loss2: 0.0621 +Epoch [44/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0563, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 56.1198 % Model2 57.5521 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0844, Loss2: 0.0785 +Epoch [45/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0644, Loss2: 0.0602 +Epoch [45/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0717, Loss2: 0.0734 +Epoch [45/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0674, Loss2: 0.0628 +Epoch [45/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0584, Loss2: 0.0553 +Epoch [45/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0882, Loss2: 0.0845 +Epoch [45/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0710, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 55.6190 % Model2 57.6723 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0627, Loss2: 0.0593 +Epoch [46/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0625, Loss2: 0.0642 +Epoch [46/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0887, Loss2: 0.0852 +Epoch [46/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0939, Loss2: 0.0907 +Epoch [46/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0820, Loss2: 0.0795 +Epoch [46/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0784, Loss2: 0.0813 +Epoch [46/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0801, Loss2: 0.0762 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 56.2700 % Model2 57.1815 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0737, Loss2: 0.0743 +Epoch [47/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0768, Loss2: 0.0763 +Epoch [47/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0820, Loss2: 0.0724 +Epoch [47/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0592, Loss2: 0.0583 +Epoch [47/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0559, Loss2: 0.0569 +Epoch [47/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0780, Loss2: 0.0808 +Epoch [47/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0657, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 56.9812 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0800, Loss2: 0.0805 +Epoch [48/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.0827, Loss2: 0.0750 +Epoch [48/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0808, Loss2: 0.0821 +Epoch [48/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0635, Loss2: 0.0618 +Epoch [48/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0690, Loss2: 0.0684 +Epoch [48/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0605, Loss2: 0.0602 +Epoch [48/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0556, Loss2: 0.0567 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 56.8810 % Model2 57.2416 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0719, Loss2: 0.0662 +Epoch [49/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0713, Loss2: 0.0726 +Epoch [49/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0701, Loss2: 0.0697 +Epoch [49/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0731, Loss2: 0.0700 +Epoch [49/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0608, Loss2: 0.0641 +Epoch [49/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0780, Loss2: 0.0711 +Epoch [49/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0767, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 55.8193 % Model2 57.1214 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0575, Loss2: 0.0577 +Epoch [50/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0614, Loss2: 0.0581 +Epoch [50/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 50.7812, Loss1: 0.0586, Loss2: 0.0658 +Epoch [50/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0743, Loss2: 0.0779 +Epoch [50/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0798, Loss2: 0.0734 +Epoch [50/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0792, Loss2: 0.0811 +Epoch [50/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.0816, Loss2: 0.0891 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 56.9010 % Model2 56.7808 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0565, Loss2: 0.0560 +Epoch [51/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0745, Loss2: 0.0750 +Epoch [51/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 59.3750, Loss1: 0.0742, Loss2: 0.0799 +Epoch [51/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0673, Loss2: 0.0684 +Epoch [51/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0862, Loss2: 0.0794 +Epoch [51/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0745, Loss2: 0.0711 +Epoch [51/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0672, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 55.3886 % Model2 57.3618 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0603, Loss2: 0.0629 +Epoch [52/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0695, Loss2: 0.0716 +Epoch [52/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0676, Loss2: 0.0615 +Epoch [52/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0848, Loss2: 0.0876 +Epoch [52/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0775, Loss2: 0.0729 +Epoch [52/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0763, Loss2: 0.0726 +Epoch [52/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0720, Loss2: 0.0766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 56.4403 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.0583, Loss2: 0.0546 +Epoch [53/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0740, Loss2: 0.0685 +Epoch [53/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0675, Loss2: 0.0651 +Epoch [53/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0777, Loss2: 0.0724 +Epoch [53/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0847, Loss2: 0.0839 +Epoch [53/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0625, Loss2: 0.0609 +Epoch [53/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0628, Loss2: 0.0579 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 55.8193 % Model2 56.0797 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0708, Loss2: 0.0676 +Epoch [54/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0622, Loss2: 0.0628 +Epoch [54/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0638, Loss2: 0.0609 +Epoch [54/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0630, Loss2: 0.0555 +Epoch [54/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0833, Loss2: 0.0766 +Epoch [54/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0727, Loss2: 0.0740 +Epoch [54/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0596, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 57.5621 % Model2 57.5020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0523, Loss2: 0.0550 +Epoch [55/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0688, Loss2: 0.0753 +Epoch [55/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0711, Loss2: 0.0658 +Epoch [55/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.9375, Loss1: 0.0697, Loss2: 0.0787 +Epoch [55/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.1077, Loss2: 0.1035 +Epoch [55/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0741, Loss2: 0.0685 +Epoch [55/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0804, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 56.7909 % Model2 57.8626 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0618, Loss2: 0.0594 +Epoch [56/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0644, Loss2: 0.0637 +Epoch [56/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0774, Loss2: 0.0741 +Epoch [56/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0748, Loss2: 0.0784 +Epoch [56/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0700, Loss2: 0.0659 +Epoch [56/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0743, Loss2: 0.0667 +Epoch [56/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0688, Loss2: 0.0663 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 56.7308 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0619, Loss2: 0.0572 +Epoch [57/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0762, Loss2: 0.0760 +Epoch [57/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0535, Loss2: 0.0553 +Epoch [57/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0633, Loss2: 0.0668 +Epoch [57/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0588, Loss2: 0.0571 +Epoch [57/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0732, Loss2: 0.0751 +Epoch [57/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0715, Loss2: 0.0703 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 56.7208 % Model2 57.8025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0633, Loss2: 0.0608 +Epoch [58/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0681, Loss2: 0.0689 +Epoch [58/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0709, Loss2: 0.0656 +Epoch [58/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0542, Loss2: 0.0539 +Epoch [58/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0739, Loss2: 0.0712 +Epoch [58/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.0312, Loss1: 0.0686, Loss2: 0.0802 +Epoch [58/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0650, Loss2: 0.0640 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 56.9010 % Model2 57.6923 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0895, Loss2: 0.0869 +Epoch [59/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0872, Loss2: 0.0887 +Epoch [59/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0526, Loss2: 0.0501 +Epoch [59/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.8125, Loss1: 0.0605, Loss2: 0.0643 +Epoch [59/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0674, Loss2: 0.0665 +Epoch [59/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0624, Loss2: 0.0602 +Epoch [59/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0590, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 55.5489 % Model2 55.8193 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0931, Loss2: 0.0923 +Epoch [60/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0709, Loss2: 0.0663 +Epoch [60/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0757, Loss2: 0.0743 +Epoch [60/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0619, Loss2: 0.0561 +Epoch [60/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0781, Loss2: 0.0730 +Epoch [60/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0763, Loss2: 0.0796 +Epoch [60/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 72.6562, Loss1: 0.1168, Loss2: 0.1028 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 56.7608 % Model2 57.5020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0595, Loss2: 0.0545 +Epoch [61/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0759, Loss2: 0.0768 +Epoch [61/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0630, Loss2: 0.0626 +Epoch [61/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0561, Loss2: 0.0587 +Epoch [61/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0763, Loss2: 0.0692 +Epoch [61/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0627, Loss2: 0.0620 +Epoch [61/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0673, Loss2: 0.0644 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 55.9896 % Model2 57.2516 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0703 +Epoch [62/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 69.5312, Loss1: 0.0700, Loss2: 0.0624 +Epoch [62/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0768, Loss2: 0.0756 +Epoch [62/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 61.7188, Loss1: 0.0585, Loss2: 0.0638 +Epoch [62/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0750, Loss2: 0.0809 +Epoch [62/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0760, Loss2: 0.0691 +Epoch [62/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0765, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 56.2700 % Model2 57.9828 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0794, Loss2: 0.0795 +Epoch [63/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0743, Loss2: 0.0711 +Epoch [63/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0503, Loss2: 0.0487 +Epoch [63/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0770, Loss2: 0.0776 +Epoch [63/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 76.5625, Loss1: 0.0744, Loss2: 0.0634 +Epoch [63/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0827, Loss2: 0.0804 +Epoch [63/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0812, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 55.8694 % Model2 55.7592 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0695, Loss2: 0.0676 +Epoch [64/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0728, Loss2: 0.0717 +Epoch [64/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0756, Loss2: 0.0735 +Epoch [64/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0690, Loss2: 0.0718 +Epoch [64/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0556, Loss2: 0.0528 +Epoch [64/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0759, Loss2: 0.0717 +Epoch [64/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0769, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 56.5204 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0984, Loss2: 0.0941 +Epoch [65/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0635, Loss2: 0.0610 +Epoch [65/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0686, Loss2: 0.0610 +Epoch [65/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0709, Loss2: 0.0724 +Epoch [65/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0915, Loss2: 0.0826 +Epoch [65/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.8125, Loss1: 0.0556, Loss2: 0.0581 +Epoch [65/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0946, Loss2: 0.1047 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 56.5705 % Model2 57.9928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0766, Loss2: 0.0781 +Epoch [66/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0649, Loss2: 0.0651 +Epoch [66/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0774, Loss2: 0.0769 +Epoch [66/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0662, Loss2: 0.0649 +Epoch [66/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0776, Loss2: 0.0699 +Epoch [66/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0570, Loss2: 0.0582 +Epoch [66/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 63.2812, Loss1: 0.0788, Loss2: 0.0834 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 56.3001 % Model2 56.9812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0650, Loss2: 0.0641 +Epoch [67/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0765, Loss2: 0.0782 +Epoch [67/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0740, Loss2: 0.0685 +Epoch [67/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0936, Loss2: 0.0913 +Epoch [67/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0550, Loss2: 0.0565 +Epoch [67/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0571, Loss2: 0.0574 +Epoch [67/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0751, Loss2: 0.0705 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 56.8710 % Model2 57.7424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0694, Loss2: 0.0752 +Epoch [68/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0590, Loss2: 0.0545 +Epoch [68/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0676, Loss2: 0.0679 +Epoch [68/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.1038, Loss2: 0.1044 +Epoch [68/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0839, Loss2: 0.0848 +Epoch [68/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0636, Loss2: 0.0647 +Epoch [68/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0709, Loss2: 0.0714 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 55.0080 % Model2 56.7909 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0641, Loss2: 0.0584 +Epoch [69/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 75.7812, Loss1: 0.0725, Loss2: 0.0604 +Epoch [69/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0746, Loss2: 0.0706 +Epoch [69/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0672, Loss2: 0.0620 +Epoch [69/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0607, Loss2: 0.0594 +Epoch [69/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0867, Loss2: 0.0773 +Epoch [69/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0612, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 56.2500 % Model2 57.2115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0596, Loss2: 0.0615 +Epoch [70/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0748, Loss2: 0.0745 +Epoch [70/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0789, Loss2: 0.0765 +Epoch [70/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 53.9062, Loss1: 0.0602, Loss2: 0.0689 +Epoch [70/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0640, Loss2: 0.0575 +Epoch [70/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0965, Loss2: 0.0997 +Epoch [70/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0766, Loss2: 0.0792 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 56.6607 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0644, Loss2: 0.0648 +Epoch [71/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0988, Loss2: 0.0905 +Epoch [71/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0580, Loss2: 0.0578 +Epoch [71/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0769, Loss2: 0.0807 +Epoch [71/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0811, Loss2: 0.0788 +Epoch [71/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0816, Loss2: 0.0776 +Epoch [71/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0762, Loss2: 0.0720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 56.5405 % Model2 57.1915 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0776, Loss2: 0.0700 +Epoch [72/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 70.3125, Loss1: 0.0916, Loss2: 0.0845 +Epoch [72/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 55.4688, Loss1: 0.0881, Loss2: 0.0941 +Epoch [72/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0781, Loss2: 0.0771 +Epoch [72/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0610, Loss2: 0.0585 +Epoch [72/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0738, Loss2: 0.0765 +Epoch [72/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0802, Loss2: 0.0821 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 54.6675 % Model2 56.4403 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0926, Loss2: 0.0824 +Epoch [73/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0663, Loss2: 0.0669 +Epoch [73/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0815, Loss2: 0.0795 +Epoch [73/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0714, Loss2: 0.0726 +Epoch [73/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0793, Loss2: 0.0731 +Epoch [73/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0777, Loss2: 0.0718 +Epoch [73/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0682, Loss2: 0.0708 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 55.8894 % Model2 57.5921 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0756, Loss2: 0.0728 +Epoch [74/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0846, Loss2: 0.0829 +Epoch [74/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0718, Loss2: 0.0661 +Epoch [74/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0836, Loss2: 0.0774 +Epoch [74/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0690, Loss2: 0.0702 +Epoch [74/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0570, Loss2: 0.0569 +Epoch [74/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0705, Loss2: 0.0624 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 54.8177 % Model2 57.1815 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0704, Loss2: 0.0759 +Epoch [75/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0686, Loss2: 0.0680 +Epoch [75/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 63.2812, Loss1: 0.0711, Loss2: 0.0617 +Epoch [75/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0783, Loss2: 0.0771 +Epoch [75/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1124, Loss2: 0.1110 +Epoch [75/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0808, Loss2: 0.0754 +Epoch [75/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0793, Loss2: 0.0738 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 55.3185 % Model2 56.9511 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0548, Loss2: 0.0530 +Epoch [76/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0675, Loss2: 0.0660 +Epoch [76/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0926, Loss2: 0.0957 +Epoch [76/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0727, Loss2: 0.0695 +Epoch [76/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0643, Loss2: 0.0625 +Epoch [76/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0664, Loss2: 0.0670 +Epoch [76/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0795, Loss2: 0.0794 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 56.8409 % Model2 56.6306 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0740, Loss2: 0.0766 +Epoch [77/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0922 +Epoch [77/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0962, Loss2: 0.0987 +Epoch [77/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0739, Loss2: 0.0691 +Epoch [77/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0785, Loss2: 0.0823 +Epoch [77/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1097, Loss2: 0.1007 +Epoch [77/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0746, Loss2: 0.0778 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 56.0196 % Model2 56.3902 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0970, Loss2: 0.0995 +Epoch [78/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.0312, Loss1: 0.0704, Loss2: 0.0783 +Epoch [78/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0662, Loss2: 0.0658 +Epoch [78/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 68.7500, Loss1: 0.0751, Loss2: 0.0831 +Epoch [78/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0693, Loss2: 0.0719 +Epoch [78/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0756, Loss2: 0.0775 +Epoch [78/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0858, Loss2: 0.0775 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 55.8894 % Model2 56.9111 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0681, Loss2: 0.0648 +Epoch [79/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0792, Loss2: 0.0794 +Epoch [79/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0564, Loss2: 0.0584 +Epoch [79/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0881, Loss2: 0.0870 +Epoch [79/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0648, Loss2: 0.0584 +Epoch [79/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 74.2188, Loss1: 0.0939, Loss2: 0.0779 +Epoch [79/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0687, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 55.2284 % Model2 56.9712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0809, Loss2: 0.0856 +Epoch [80/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0974, Loss2: 0.0988 +Epoch [80/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0856, Loss2: 0.0792 +Epoch [80/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0897, Loss2: 0.0856 +Epoch [80/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 64.0625, Loss1: 0.0664, Loss2: 0.0586 +Epoch [80/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0821, Loss2: 0.0835 +Epoch [80/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0788, Loss2: 0.0698 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 55.3986 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 73.4375, Loss1: 0.0924, Loss2: 0.0786 +Epoch [81/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0641, Loss2: 0.0602 +Epoch [81/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0586, Loss2: 0.0612 +Epoch [81/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0664, Loss2: 0.0678 +Epoch [81/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0774, Loss2: 0.0787 +Epoch [81/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0887, Loss2: 0.0781 +Epoch [81/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0540, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 55.3986 % Model2 56.4804 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0735, Loss2: 0.0701 +Epoch [82/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1001, Loss2: 0.1066 +Epoch [82/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0798, Loss2: 0.0739 +Epoch [82/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0890, Loss2: 0.0819 +Epoch [82/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0616, Loss2: 0.0681 +Epoch [82/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0642, Loss2: 0.0611 +Epoch [82/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0829, Loss2: 0.0828 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 57.2216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0581, Loss2: 0.0606 +Epoch [83/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0864, Loss2: 0.0893 +Epoch [83/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0670, Loss2: 0.0666 +Epoch [83/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0643, Loss2: 0.0654 +Epoch [83/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0737, Loss2: 0.0774 +Epoch [83/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0894, Loss2: 0.0867 +Epoch [83/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0682, Loss2: 0.0647 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 55.9495 % Model2 57.0312 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 73.4375, Loss1: 0.0924, Loss2: 0.0769 +Epoch [84/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0776, Loss2: 0.0695 +Epoch [84/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0855, Loss2: 0.0883 +Epoch [84/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0779, Loss2: 0.0714 +Epoch [84/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0614, Loss2: 0.0636 +Epoch [84/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0669, Loss2: 0.0610 +Epoch [84/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0522, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 56.5304 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0830, Loss2: 0.0859 +Epoch [85/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 69.5312, Loss1: 0.0854, Loss2: 0.0734 +Epoch [85/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0805, Loss2: 0.0880 +Epoch [85/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0879, Loss2: 0.0953 +Epoch [85/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0785, Loss2: 0.0751 +Epoch [85/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0840, Loss2: 0.0805 +Epoch [85/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0805, Loss2: 0.0805 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 57.4820 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0906, Loss2: 0.0838 +Epoch [86/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0767, Loss2: 0.0773 +Epoch [86/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0716, Loss2: 0.0717 +Epoch [86/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.0701, Loss2: 0.0750 +Epoch [86/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0733, Loss2: 0.0767 +Epoch [86/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0973, Loss2: 0.0926 +Epoch [86/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0710, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 57.5721 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0685, Loss2: 0.0669 +Epoch [87/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0778, Loss2: 0.0782 +Epoch [87/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0707, Loss2: 0.0742 +Epoch [87/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1118, Loss2: 0.1074 +Epoch [87/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0857, Loss2: 0.0855 +Epoch [87/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0679, Loss2: 0.0651 +Epoch [87/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0799, Loss2: 0.0817 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 53.7961 % Model2 55.9295 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0685, Loss2: 0.0638 +Epoch [88/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 75.0000, Loss1: 0.1075, Loss2: 0.0909 +Epoch [88/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0851, Loss2: 0.0844 +Epoch [88/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0694, Loss2: 0.0697 +Epoch [88/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0793, Loss2: 0.0774 +Epoch [88/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0752, Loss2: 0.0746 +Epoch [88/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0631, Loss2: 0.0658 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 55.0581 % Model2 56.8710 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0960, Loss2: 0.0922 +Epoch [89/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0777, Loss2: 0.0745 +Epoch [89/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0784, Loss2: 0.0812 +Epoch [89/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0587, Loss2: 0.0606 +Epoch [89/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0787, Loss2: 0.0740 +Epoch [89/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0745, Loss2: 0.0730 +Epoch [89/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0672, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 56.0096 % Model2 55.3285 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 62.5000, Loss1: 0.0573, Loss2: 0.0626 +Epoch [90/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0749, Loss2: 0.0707 +Epoch [90/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0892, Loss2: 0.0948 +Epoch [90/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0697, Loss2: 0.0679 +Epoch [90/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0960, Loss2: 0.0961 +Epoch [90/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 66.4062, Loss1: 0.0699, Loss2: 0.0624 +Epoch [90/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0676, Loss2: 0.0627 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 54.9279 % Model2 55.8794 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0617, Loss2: 0.0644 +Epoch [91/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.0927, Loss2: 0.0968 +Epoch [91/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0988, Loss2: 0.0920 +Epoch [91/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0622, Loss2: 0.0627 +Epoch [91/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0919, Loss2: 0.0910 +Epoch [91/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0712, Loss2: 0.0677 +Epoch [91/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0856, Loss2: 0.0896 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 54.8478 % Model2 56.2901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0675, Loss2: 0.0709 +Epoch [92/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0730, Loss2: 0.0674 +Epoch [92/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0778, Loss2: 0.0728 +Epoch [92/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.9375, Loss1: 0.0667, Loss2: 0.0724 +Epoch [92/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.0938, Loss1: 0.0847, Loss2: 0.0766 +Epoch [92/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0665, Loss2: 0.0664 +Epoch [92/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0678, Loss2: 0.0660 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 55.4087 % Model2 56.5505 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0733, Loss2: 0.0671 +Epoch [93/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0684, Loss2: 0.0712 +Epoch [93/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0769, Loss2: 0.0727 +Epoch [93/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0838, Loss2: 0.0805 +Epoch [93/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1038, Loss2: 0.1059 +Epoch [93/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0894, Loss2: 0.0828 +Epoch [93/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0788, Loss2: 0.0784 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 55.2985 % Model2 55.5689 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0594, Loss2: 0.0546 +Epoch [94/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0872, Loss2: 0.0784 +Epoch [94/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0865, Loss2: 0.0877 +Epoch [94/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0797, Loss2: 0.0700 +Epoch [94/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0906, Loss2: 0.0943 +Epoch [94/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0863, Loss2: 0.0910 +Epoch [94/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0572, Loss2: 0.0552 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 56.1599 % Model2 56.3301 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0834, Loss2: 0.0882 +Epoch [95/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0863, Loss2: 0.0863 +Epoch [95/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1256, Loss2: 0.1150 +Epoch [95/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0771, Loss2: 0.0766 +Epoch [95/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0771, Loss2: 0.0765 +Epoch [95/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0923, Loss2: 0.1015 +Epoch [95/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0870, Loss2: 0.0950 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 55.2584 % Model2 57.1314 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0844, Loss2: 0.0806 +Epoch [96/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.0938, Loss1: 0.0746, Loss2: 0.0663 +Epoch [96/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0785, Loss2: 0.0731 +Epoch [96/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0708, Loss2: 0.0697 +Epoch [96/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 68.7500, Loss1: 0.0735, Loss2: 0.0637 +Epoch [96/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.1077, Loss2: 0.0901 +Epoch [96/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0920, Loss2: 0.0822 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 55.3786 % Model2 56.9912 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0792, Loss2: 0.0695 +Epoch [97/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0746, Loss2: 0.0741 +Epoch [97/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0910, Loss2: 0.0887 +Epoch [97/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0789, Loss2: 0.0744 +Epoch [97/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0772, Loss2: 0.0728 +Epoch [97/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0837, Loss2: 0.0883 +Epoch [97/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0750, Loss2: 0.0730 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 55.7792 % Model2 55.9796 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0949, Loss2: 0.0926 +Epoch [98/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0884, Loss2: 0.0890 +Epoch [98/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0850, Loss2: 0.0815 +Epoch [98/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0995, Loss2: 0.1046 +Epoch [98/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0754, Loss2: 0.0733 +Epoch [98/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0859, Loss2: 0.0870 +Epoch [98/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0685, Loss2: 0.0641 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 55.0180 % Model2 56.3702 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0676, Loss2: 0.0725 +Epoch [99/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0741, Loss2: 0.0707 +Epoch [99/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1068, Loss2: 0.1032 +Epoch [99/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0885, Loss2: 0.0861 +Epoch [99/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0773, Loss2: 0.0820 +Epoch [99/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 75.7812, Loss1: 0.0704, Loss2: 0.0597 +Epoch [99/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1074, Loss2: 0.0959 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 56.0697 % Model2 56.0397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1049, Loss2: 0.1083 +Epoch [100/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.1076, Loss2: 0.1003 +Epoch [100/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0939, Loss2: 0.0881 +Epoch [100/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1010, Loss2: 0.1020 +Epoch [100/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0767, Loss2: 0.0761 +Epoch [100/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0645, Loss2: 0.0605 +Epoch [100/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 75.0000, Loss1: 0.0931, Loss2: 0.0839 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 56.1699 % Model2 56.3602 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0920, Loss2: 0.0861 +Epoch [101/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0739, Loss2: 0.0744 +Epoch [101/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0678, Loss2: 0.0688 +Epoch [101/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0828, Loss2: 0.0914 +Epoch [101/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.1562, Loss1: 0.0667, Loss2: 0.0763 +Epoch [101/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0915, Loss2: 0.0957 +Epoch [101/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0754, Loss2: 0.0847 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 55.8894 % Model2 55.6691 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.1010, Loss2: 0.1105 +Epoch [102/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1020, Loss2: 0.0971 +Epoch [102/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0726, Loss2: 0.0751 +Epoch [102/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0743, Loss2: 0.0755 +Epoch [102/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.0891, Loss2: 0.0959 +Epoch [102/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0765, Loss2: 0.0755 +Epoch [102/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 55.7292 % Model2 56.5505 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0819, Loss2: 0.0821 +Epoch [103/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 60.9375, Loss1: 0.0597, Loss2: 0.0661 +Epoch [103/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0770, Loss2: 0.0796 +Epoch [103/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1069, Loss2: 0.1032 +Epoch [103/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0976, Loss2: 0.0904 +Epoch [103/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0789, Loss2: 0.0755 +Epoch [103/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 61.7188, Loss1: 0.0744, Loss2: 0.0817 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 55.0581 % Model2 56.2800 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.0810, Loss2: 0.0833 +Epoch [104/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.0881, Loss2: 0.0783 +Epoch [104/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0884, Loss2: 0.0939 +Epoch [104/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1103, Loss2: 0.1038 +Epoch [104/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0889, Loss2: 0.0882 +Epoch [104/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0724, Loss2: 0.0722 +Epoch [104/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0712, Loss2: 0.0725 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 55.9996 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1129, Loss2: 0.1080 +Epoch [105/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0886, Loss2: 0.0873 +Epoch [105/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0750, Loss2: 0.0726 +Epoch [105/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0745, Loss2: 0.0740 +Epoch [105/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0755, Loss2: 0.0696 +Epoch [105/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0678, Loss2: 0.0681 +Epoch [105/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0937, Loss2: 0.0979 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 55.2284 % Model2 57.0413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0714, Loss2: 0.0691 +Epoch [106/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1181, Loss2: 0.1156 +Epoch [106/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.1078, Loss2: 0.1151 +Epoch [106/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0833, Loss2: 0.0891 +Epoch [106/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1054, Loss2: 0.1021 +Epoch [106/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1002, Loss2: 0.0970 +Epoch [106/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0600, Loss2: 0.0596 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 56.5905 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1266, Loss2: 0.1248 +Epoch [107/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0802, Loss2: 0.0788 +Epoch [107/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0775, Loss2: 0.0806 +Epoch [107/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.1013, Loss2: 0.0910 +Epoch [107/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0938, Loss2: 0.0974 +Epoch [107/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 68.7500, Loss1: 0.0930, Loss2: 0.1047 +Epoch [107/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1056, Loss2: 0.1118 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 55.4287 % Model2 56.2901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1086, Loss2: 0.1029 +Epoch [108/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0725, Loss2: 0.0682 +Epoch [108/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0658, Loss2: 0.0618 +Epoch [108/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0617, Loss2: 0.0649 +Epoch [108/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1283, Loss2: 0.1209 +Epoch [108/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0849, Loss2: 0.0892 +Epoch [108/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0900, Loss2: 0.0842 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 55.2684 % Model2 56.5905 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0762, Loss2: 0.0822 +Epoch [109/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0880, Loss2: 0.0862 +Epoch [109/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0821, Loss2: 0.0760 +Epoch [109/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0817, Loss2: 0.0717 +Epoch [109/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0925, Loss2: 0.0891 +Epoch [109/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0757, Loss2: 0.0799 +Epoch [109/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1038, Loss2: 0.1070 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 55.8093 % Model2 56.5505 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0784, Loss2: 0.0808 +Epoch [110/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0993, Loss2: 0.1069 +Epoch [110/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0768, Loss2: 0.0801 +Epoch [110/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0755, Loss2: 0.0701 +Epoch [110/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0944, Loss2: 0.0857 +Epoch [110/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0803, Loss2: 0.0796 +Epoch [110/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1036, Loss2: 0.1062 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 54.5473 % Model2 56.6607 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0811, Loss2: 0.0770 +Epoch [111/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1032, Loss2: 0.1005 +Epoch [111/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0968, Loss2: 0.0918 +Epoch [111/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0685, Loss2: 0.0654 +Epoch [111/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0842, Loss2: 0.0835 +Epoch [111/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0938, Loss2: 0.0928 +Epoch [111/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0786, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 56.6206 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0894, Loss2: 0.0947 +Epoch [112/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0736, Loss2: 0.0788 +Epoch [112/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 60.1562, Loss1: 0.0693, Loss2: 0.0835 +Epoch [112/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0819, Loss2: 0.0839 +Epoch [112/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0965, Loss2: 0.0917 +Epoch [112/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0902, Loss2: 0.0778 +Epoch [112/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1603, Loss2: 0.1296 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 54.7175 % Model2 55.9896 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1189, Loss2: 0.1094 +Epoch [113/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.0936, Loss2: 0.0888 +Epoch [113/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1295, Loss2: 0.1298 +Epoch [113/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.1027, Loss2: 0.1053 +Epoch [113/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1338, Loss2: 0.1335 +Epoch [113/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1053, Loss2: 0.0997 +Epoch [113/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0656, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 55.2083 % Model2 56.0397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 64.8438, Loss1: 0.0823, Loss2: 0.0915 +Epoch [114/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 71.0938, Loss1: 0.0686, Loss2: 0.0656 +Epoch [114/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0851 +Epoch [114/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0937, Loss2: 0.0917 +Epoch [114/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0938, Loss2: 0.0924 +Epoch [114/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0789, Loss2: 0.0811 +Epoch [114/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0856, Loss2: 0.0883 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 55.3786 % Model2 56.4804 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0818, Loss2: 0.0825 +Epoch [115/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.1147, Loss2: 0.1225 +Epoch [115/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 68.7500, Loss1: 0.0755, Loss2: 0.0844 +Epoch [115/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0980, Loss2: 0.1019 +Epoch [115/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0804, Loss2: 0.0832 +Epoch [115/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0664, Loss2: 0.0725 +Epoch [115/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0867, Loss2: 0.0834 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 55.6891 % Model2 56.2800 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1110, Loss2: 0.1177 +Epoch [116/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0825, Loss2: 0.0795 +Epoch [116/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0754, Loss2: 0.0797 +Epoch [116/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0981, Loss2: 0.0977 +Epoch [116/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0901, Loss2: 0.0904 +Epoch [116/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0801, Loss2: 0.0898 +Epoch [116/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0764, Loss2: 0.0790 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 55.5689 % Model2 56.1498 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0731, Loss2: 0.0764 +Epoch [117/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0893, Loss2: 0.0918 +Epoch [117/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0821, Loss2: 0.0784 +Epoch [117/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1059, Loss2: 0.1043 +Epoch [117/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.1237, Loss2: 0.1116 +Epoch [117/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0854, Loss2: 0.0854 +Epoch [117/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0905, Loss2: 0.0925 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 55.0481 % Model2 56.5505 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0780, Loss2: 0.0741 +Epoch [118/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1103, Loss2: 0.1013 +Epoch [118/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.0918, Loss2: 0.0886 +Epoch [118/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1150, Loss2: 0.1282 +Epoch [118/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0783, Loss2: 0.0794 +Epoch [118/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0779, Loss2: 0.0792 +Epoch [118/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0713, Loss2: 0.0735 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 55.3786 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.1034, Loss2: 0.1070 +Epoch [119/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0946, Loss2: 0.0876 +Epoch [119/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0968, Loss2: 0.1007 +Epoch [119/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1093, Loss2: 0.1119 +Epoch [119/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0754, Loss2: 0.0750 +Epoch [119/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0663 +Epoch [119/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0776, Loss2: 0.0712 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 55.1482 % Model2 57.2817 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0891, Loss2: 0.0813 +Epoch [120/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1019, Loss2: 0.0996 +Epoch [120/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.0813, Loss2: 0.0740 +Epoch [120/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0914, Loss2: 0.0893 +Epoch [120/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0751, Loss2: 0.0752 +Epoch [120/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1093, Loss2: 0.1033 +Epoch [120/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0703, Loss2: 0.0657 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 54.4071 % Model2 55.9295 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0840, Loss2: 0.0812 +Epoch [121/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1148, Loss2: 0.1165 +Epoch [121/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0649, Loss2: 0.0636 +Epoch [121/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1239, Loss2: 0.1058 +Epoch [121/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1243, Loss2: 0.1134 +Epoch [121/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0808, Loss2: 0.0793 +Epoch [121/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1182, Loss2: 0.1171 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 55.2885 % Model2 56.6106 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0922, Loss2: 0.0841 +Epoch [122/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0896, Loss2: 0.0910 +Epoch [122/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.1044, Loss2: 0.0875 +Epoch [122/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1005, Loss2: 0.0936 +Epoch [122/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1106, Loss2: 0.1044 +Epoch [122/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1291, Loss2: 0.1285 +Epoch [122/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0744, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 55.1282 % Model2 56.4002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0948, Loss2: 0.0903 +Epoch [123/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1053, Loss2: 0.1042 +Epoch [123/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0823, Loss2: 0.0841 +Epoch [123/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 65.6250, Loss1: 0.0880, Loss2: 0.0963 +Epoch [123/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0864, Loss2: 0.0905 +Epoch [123/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.0762, Loss2: 0.0709 +Epoch [123/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0946, Loss2: 0.0851 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 55.1282 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0758, Loss2: 0.0745 +Epoch [124/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0972, Loss2: 0.0897 +Epoch [124/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1032, Loss2: 0.1058 +Epoch [124/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 73.4375, Loss1: 0.1093, Loss2: 0.1292 +Epoch [124/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0703, Loss2: 0.0702 +Epoch [124/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0805, Loss2: 0.0911 +Epoch [124/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0921, Loss2: 0.0904 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 55.1082 % Model2 56.2901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0611, Loss2: 0.0638 +Epoch [125/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1240, Loss2: 0.1295 +Epoch [125/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0908, Loss2: 0.0892 +Epoch [125/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0829, Loss2: 0.0873 +Epoch [125/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0909, Loss2: 0.0901 +Epoch [125/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 59.3750, Loss1: 0.0744, Loss2: 0.0843 +Epoch [125/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1078, Loss2: 0.1091 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 54.7276 % Model2 55.6290 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0777, Loss2: 0.0721 +Epoch [126/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 75.0000, Loss1: 0.1139, Loss2: 0.0961 +Epoch [126/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0902, Loss2: 0.0884 +Epoch [126/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 72.6562, Loss1: 0.0954, Loss2: 0.0846 +Epoch [126/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0765, Loss2: 0.0756 +Epoch [126/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0793, Loss2: 0.0765 +Epoch [126/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 72.6562, Loss1: 0.0788, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 56.9111 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1049, Loss2: 0.1158 +Epoch [127/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1033, Loss2: 0.1069 +Epoch [127/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0889, Loss2: 0.0877 +Epoch [127/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1253, Loss2: 0.1207 +Epoch [127/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1149, Loss2: 0.1181 +Epoch [127/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1705, Loss2: 0.1616 +Epoch [127/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0817, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 55.5088 % Model2 56.4403 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0920, Loss2: 0.0928 +Epoch [128/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0888, Loss2: 0.0933 +Epoch [128/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0707, Loss2: 0.0714 +Epoch [128/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0837, Loss2: 0.0792 +Epoch [128/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.1096, Loss2: 0.0981 +Epoch [128/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1169, Loss2: 0.1236 +Epoch [128/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.1404, Loss2: 0.1328 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 55.6490 % Model2 56.1699 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0740, Loss2: 0.0790 +Epoch [129/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1171, Loss2: 0.1143 +Epoch [129/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0899, Loss2: 0.0849 +Epoch [129/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0808, Loss2: 0.0796 +Epoch [129/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1159, Loss2: 0.1194 +Epoch [129/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.1173, Loss2: 0.1090 +Epoch [129/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0791, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 55.2384 % Model2 56.2500 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1649, Loss2: 0.1548 +Epoch [130/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0923, Loss2: 0.0888 +Epoch [130/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1029, Loss2: 0.1038 +Epoch [130/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0883, Loss2: 0.0844 +Epoch [130/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0817, Loss2: 0.0791 +Epoch [130/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0997, Loss2: 0.0986 +Epoch [130/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1311, Loss2: 0.1235 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 55.3686 % Model2 56.2600 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0973, Loss2: 0.1012 +Epoch [131/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0911, Loss2: 0.0881 +Epoch [131/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0733, Loss2: 0.0754 +Epoch [131/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0843, Loss2: 0.0856 +Epoch [131/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1662, Loss2: 0.1542 +Epoch [131/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1243, Loss2: 0.1196 +Epoch [131/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1022, Loss2: 0.1008 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 54.4471 % Model2 56.4904 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0978, Loss2: 0.0952 +Epoch [132/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1021, Loss2: 0.0929 +Epoch [132/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0883, Loss2: 0.0865 +Epoch [132/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0916, Loss2: 0.0897 +Epoch [132/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1424, Loss2: 0.1396 +Epoch [132/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0795, Loss2: 0.0753 +Epoch [132/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.1051, Loss2: 0.1111 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 55.1583 % Model2 56.3401 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0773, Loss2: 0.0743 +Epoch [133/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.1419, Loss2: 0.1265 +Epoch [133/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1400, Loss2: 0.1456 +Epoch [133/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1020, Loss2: 0.1105 +Epoch [133/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0992, Loss2: 0.1040 +Epoch [133/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1183, Loss2: 0.1185 +Epoch [133/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0991, Loss2: 0.0994 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 55.1282 % Model2 56.2700 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0834, Loss2: 0.0872 +Epoch [134/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.0968, Loss2: 0.0971 +Epoch [134/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 63.2812, Loss1: 0.0950, Loss2: 0.1109 +Epoch [134/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0940, Loss2: 0.1006 +Epoch [134/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0774, Loss2: 0.0722 +Epoch [134/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1065, Loss2: 0.1142 +Epoch [134/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0976, Loss2: 0.0964 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 54.9780 % Model2 56.4002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0796, Loss2: 0.0818 +Epoch [135/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0986, Loss2: 0.0976 +Epoch [135/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0910, Loss2: 0.0963 +Epoch [135/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1077, Loss2: 0.1085 +Epoch [135/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1203, Loss2: 0.1140 +Epoch [135/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1049, Loss2: 0.1049 +Epoch [135/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0872, Loss2: 0.0893 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 54.6575 % Model2 56.2600 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1667, Loss2: 0.1614 +Epoch [136/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 80.4688, Loss1: 0.1505, Loss2: 0.1383 +Epoch [136/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1073, Loss2: 0.1102 +Epoch [136/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0954, Loss2: 0.0874 +Epoch [136/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1024, Loss2: 0.0946 +Epoch [136/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0875, Loss2: 0.0880 +Epoch [136/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1135, Loss2: 0.1161 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 55.3886 % Model2 56.3802 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1125, Loss2: 0.1033 +Epoch [137/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1158, Loss2: 0.1207 +Epoch [137/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1134, Loss2: 0.1123 +Epoch [137/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0887, Loss2: 0.0871 +Epoch [137/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1496, Loss2: 0.1489 +Epoch [137/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1108, Loss2: 0.1086 +Epoch [137/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1222, Loss2: 0.1198 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 55.0982 % Model2 56.1999 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0962, Loss2: 0.1024 +Epoch [138/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.1029, Loss2: 0.1053 +Epoch [138/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1462, Loss2: 0.1475 +Epoch [138/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0957, Loss2: 0.0907 +Epoch [138/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 68.7500, Loss1: 0.1012, Loss2: 0.1164 +Epoch [138/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1018, Loss2: 0.1097 +Epoch [138/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 78.1250, Loss1: 0.1335, Loss2: 0.1109 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 54.9579 % Model2 56.0897 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0893, Loss2: 0.0890 +Epoch [139/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0923, Loss2: 0.0920 +Epoch [139/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1140, Loss2: 0.1125 +Epoch [139/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1374, Loss2: 0.1306 +Epoch [139/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0832, Loss2: 0.0883 +Epoch [139/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0730, Loss2: 0.0661 +Epoch [139/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1032, Loss2: 0.1005 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 55.0381 % Model2 55.9395 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.0867, Loss2: 0.0840 +Epoch [140/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0964, Loss2: 0.0970 +Epoch [140/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1024, Loss2: 0.1054 +Epoch [140/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0987, Loss2: 0.1024 +Epoch [140/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1001, Loss2: 0.0940 +Epoch [140/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1084, Loss2: 0.1156 +Epoch [140/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1188, Loss2: 0.1243 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 55.2183 % Model2 56.1799 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0744, Loss2: 0.0729 +Epoch [141/200], Iter [100/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.1753, Loss2: 0.1696 +Epoch [141/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.1183, Loss2: 0.1289 +Epoch [141/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1126, Loss2: 0.1110 +Epoch [141/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0996, Loss2: 0.1008 +Epoch [141/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0934, Loss2: 0.0925 +Epoch [141/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.1104, Loss2: 0.0949 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 54.5573 % Model2 56.6607 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1336, Loss2: 0.1307 +Epoch [142/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.0000, Loss1: 0.0945, Loss2: 0.0823 +Epoch [142/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0817, Loss2: 0.0783 +Epoch [142/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1201, Loss2: 0.1177 +Epoch [142/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0956, Loss2: 0.0931 +Epoch [142/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0956, Loss2: 0.0964 +Epoch [142/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0950, Loss2: 0.0988 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 54.8478 % Model2 56.1498 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1319, Loss2: 0.1378 +Epoch [143/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.7812, Loss1: 0.1228, Loss2: 0.1019 +Epoch [143/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.1002, Loss2: 0.1050 +Epoch [143/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1390, Loss2: 0.1320 +Epoch [143/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1312, Loss2: 0.1270 +Epoch [143/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.1071, Loss2: 0.0987 +Epoch [143/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0936, Loss2: 0.0958 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 55.3486 % Model2 55.8494 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1081, Loss2: 0.1084 +Epoch [144/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1034, Loss2: 0.1043 +Epoch [144/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0778, Loss2: 0.0798 +Epoch [144/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.1051, Loss2: 0.1157 +Epoch [144/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1096, Loss2: 0.1065 +Epoch [144/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1192, Loss2: 0.1159 +Epoch [144/200], Iter [350/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.1549, Loss2: 0.1676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 54.8678 % Model2 56.4403 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1618, Loss2: 0.1780 +Epoch [145/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1207, Loss2: 0.1180 +Epoch [145/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1025, Loss2: 0.1042 +Epoch [145/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0959, Loss2: 0.0878 +Epoch [145/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1201, Loss2: 0.1274 +Epoch [145/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1122, Loss2: 0.1063 +Epoch [145/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.1290, Loss2: 0.1331 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 54.7075 % Model2 55.7192 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.1081, Loss2: 0.1124 +Epoch [146/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1204, Loss2: 0.1106 +Epoch [146/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.1843, Loss2: 0.1729 +Epoch [146/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0996, Loss2: 0.0957 +Epoch [146/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1222, Loss2: 0.1243 +Epoch [146/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0882, Loss2: 0.0911 +Epoch [146/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0888, Loss2: 0.0882 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 55.3486 % Model2 56.0697 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1330, Loss2: 0.1408 +Epoch [147/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.1037, Loss2: 0.0908 +Epoch [147/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0767, Loss2: 0.0786 +Epoch [147/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.1252, Loss2: 0.1252 +Epoch [147/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0974, Loss2: 0.0996 +Epoch [147/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.1092, Loss2: 0.1219 +Epoch [147/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1116, Loss2: 0.1025 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 54.7676 % Model2 56.7208 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1236, Loss2: 0.1144 +Epoch [148/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1086, Loss2: 0.1164 +Epoch [148/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0979, Loss2: 0.1023 +Epoch [148/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1048, Loss2: 0.1093 +Epoch [148/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0994, Loss2: 0.1001 +Epoch [148/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1332, Loss2: 0.1325 +Epoch [148/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1196, Loss2: 0.1130 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 54.8878 % Model2 56.1098 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1060, Loss2: 0.1103 +Epoch [149/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0898, Loss2: 0.0955 +Epoch [149/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1317, Loss2: 0.1332 +Epoch [149/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1002, Loss2: 0.0944 +Epoch [149/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0966, Loss2: 0.0874 +Epoch [149/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0970, Loss2: 0.0908 +Epoch [149/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0964, Loss2: 0.1043 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 55.3686 % Model2 56.1599 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0917, Loss2: 0.0903 +Epoch [150/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0742, Loss2: 0.0736 +Epoch [150/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.0946, Loss2: 0.0915 +Epoch [150/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 75.7812, Loss1: 0.1038, Loss2: 0.0891 +Epoch [150/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0965, Loss2: 0.0958 +Epoch [150/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0914, Loss2: 0.0927 +Epoch [150/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1119, Loss2: 0.1060 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 55.2584 % Model2 56.6006 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0909, Loss2: 0.0991 +Epoch [151/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0846, Loss2: 0.0832 +Epoch [151/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0900, Loss2: 0.0952 +Epoch [151/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1342, Loss2: 0.1360 +Epoch [151/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1577, Loss2: 0.1649 +Epoch [151/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1058, Loss2: 0.1080 +Epoch [151/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1324, Loss2: 0.1373 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 55.0781 % Model2 56.4303 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1071, Loss2: 0.1041 +Epoch [152/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0847, Loss2: 0.0795 +Epoch [152/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1432, Loss2: 0.1318 +Epoch [152/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.1062, Loss2: 0.0996 +Epoch [152/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1121, Loss2: 0.1105 +Epoch [152/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.0880, Loss2: 0.0981 +Epoch [152/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1293, Loss2: 0.1402 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 54.6875 % Model2 55.9696 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0907, Loss2: 0.1016 +Epoch [153/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1463, Loss2: 0.1549 +Epoch [153/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1165, Loss2: 0.1224 +Epoch [153/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1042, Loss2: 0.1033 +Epoch [153/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.1228, Loss2: 0.1035 +Epoch [153/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1469, Loss2: 0.1421 +Epoch [153/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1101, Loss2: 0.1040 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 54.3870 % Model2 56.1298 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1389, Loss2: 0.1268 +Epoch [154/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1382, Loss2: 0.1546 +Epoch [154/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 78.1250, Loss1: 0.0918, Loss2: 0.0864 +Epoch [154/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1231, Loss2: 0.1089 +Epoch [154/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1318, Loss2: 0.1232 +Epoch [154/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0970, Loss2: 0.0960 +Epoch [154/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1272, Loss2: 0.1230 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 54.7476 % Model2 56.3902 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1180, Loss2: 0.1126 +Epoch [155/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1019, Loss2: 0.0945 +Epoch [155/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 75.0000, Loss1: 0.1170, Loss2: 0.1012 +Epoch [155/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.1478, Loss2: 0.1445 +Epoch [155/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1279, Loss2: 0.1262 +Epoch [155/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 80.4688, Loss1: 0.1735, Loss2: 0.1919 +Epoch [155/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1121, Loss2: 0.1174 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 54.7977 % Model2 56.2500 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.1004, Loss2: 0.1088 +Epoch [156/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1043, Loss2: 0.0973 +Epoch [156/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.1325, Loss2: 0.1395 +Epoch [156/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.1156, Loss2: 0.1266 +Epoch [156/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1607, Loss2: 0.1647 +Epoch [156/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1308, Loss2: 0.1155 +Epoch [156/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1285, Loss2: 0.1236 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 54.8377 % Model2 56.3702 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1200, Loss2: 0.1249 +Epoch [157/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0959, Loss2: 0.0957 +Epoch [157/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.0962, Loss2: 0.1003 +Epoch [157/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1219, Loss2: 0.1216 +Epoch [157/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0877, Loss2: 0.0926 +Epoch [157/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.1268, Loss2: 0.1428 +Epoch [157/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1023, Loss2: 0.0980 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 54.8077 % Model2 56.3201 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1020, Loss2: 0.1060 +Epoch [158/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.1485, Loss2: 0.1261 +Epoch [158/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.1090, Loss2: 0.0976 +Epoch [158/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1022, Loss2: 0.0973 +Epoch [158/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1423, Loss2: 0.1244 +Epoch [158/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 81.2500, Loss1: 0.1283, Loss2: 0.1120 +Epoch [158/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0867, Loss2: 0.0871 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 54.5072 % Model2 55.5288 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1541, Loss2: 0.1378 +Epoch [159/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1166, Loss2: 0.1271 +Epoch [159/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1082, Loss2: 0.1041 +Epoch [159/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0838, Loss2: 0.0885 +Epoch [159/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.1357, Loss2: 0.1506 +Epoch [159/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1126, Loss2: 0.1137 +Epoch [159/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1914, Loss2: 0.1910 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 54.4872 % Model2 55.5689 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.0992, Loss2: 0.1017 +Epoch [160/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1164, Loss2: 0.1150 +Epoch [160/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1292, Loss2: 0.1164 +Epoch [160/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0928, Loss2: 0.0876 +Epoch [160/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.1532, Loss2: 0.1766 +Epoch [160/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1220, Loss2: 0.1246 +Epoch [160/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1556, Loss2: 0.1461 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 54.6675 % Model2 55.9896 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1825, Loss2: 0.1851 +Epoch [161/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0943, Loss2: 0.0889 +Epoch [161/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1287, Loss2: 0.1228 +Epoch [161/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0976, Loss2: 0.0971 +Epoch [161/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.1596, Loss2: 0.1735 +Epoch [161/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1636, Loss2: 0.1516 +Epoch [161/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1325, Loss2: 0.1237 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 54.7877 % Model2 55.7292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1142, Loss2: 0.1043 +Epoch [162/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.1562, Loss2: 0.1328 +Epoch [162/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1153, Loss2: 0.1058 +Epoch [162/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1309, Loss2: 0.1214 +Epoch [162/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1440, Loss2: 0.1411 +Epoch [162/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 69.5312, Loss1: 0.0942, Loss2: 0.1109 +Epoch [162/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.1184, Loss2: 0.1260 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 54.7476 % Model2 55.9896 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1101, Loss2: 0.1085 +Epoch [163/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1371, Loss2: 0.1519 +Epoch [163/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1076, Loss2: 0.1079 +Epoch [163/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1080, Loss2: 0.1067 +Epoch [163/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 80.4688, Loss1: 0.1185, Loss2: 0.1050 +Epoch [163/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0967, Loss2: 0.0991 +Epoch [163/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1140, Loss2: 0.1143 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 54.8978 % Model2 55.2985 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1172, Loss2: 0.1098 +Epoch [164/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1222, Loss2: 0.1248 +Epoch [164/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.1243, Loss2: 0.1290 +Epoch [164/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1240, Loss2: 0.1158 +Epoch [164/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0953, Loss2: 0.0956 +Epoch [164/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1318, Loss2: 0.1237 +Epoch [164/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1155, Loss2: 0.1123 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 54.7776 % Model2 55.8994 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1085, Loss2: 0.1100 +Epoch [165/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.1603, Loss2: 0.1504 +Epoch [165/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1036, Loss2: 0.1079 +Epoch [165/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1330, Loss2: 0.1250 +Epoch [165/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1041, Loss2: 0.1039 +Epoch [165/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0887, Loss2: 0.0882 +Epoch [165/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1459, Loss2: 0.1464 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 54.7276 % Model2 56.2300 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1264, Loss2: 0.1393 +Epoch [166/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1111, Loss2: 0.1055 +Epoch [166/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1716, Loss2: 0.1878 +Epoch [166/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1833, Loss2: 0.1746 +Epoch [166/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 79.6875, Loss1: 0.1327, Loss2: 0.1189 +Epoch [166/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1384, Loss2: 0.1298 +Epoch [166/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1308, Loss2: 0.1258 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 54.9579 % Model2 55.8694 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1076, Loss2: 0.1063 +Epoch [167/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0922, Loss2: 0.0881 +Epoch [167/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1901, Loss2: 0.1717 +Epoch [167/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1597, Loss2: 0.1460 +Epoch [167/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0960, Loss2: 0.0993 +Epoch [167/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1449, Loss2: 0.1336 +Epoch [167/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0912, Loss2: 0.0897 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 54.4571 % Model2 56.0196 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 77.3438, Loss1: 0.1320, Loss2: 0.1458 +Epoch [168/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.2025, Loss2: 0.1636 +Epoch [168/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1384, Loss2: 0.1268 +Epoch [168/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1883, Loss2: 0.1843 +Epoch [168/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1296, Loss2: 0.1433 +Epoch [168/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 69.5312, Loss1: 0.1159, Loss2: 0.1316 +Epoch [168/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1554, Loss2: 0.1500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 54.5773 % Model2 55.7592 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1107, Loss2: 0.1139 +Epoch [169/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1442, Loss2: 0.1246 +Epoch [169/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1346, Loss2: 0.1322 +Epoch [169/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1208, Loss2: 0.1052 +Epoch [169/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1902, Loss2: 0.1922 +Epoch [169/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.1121, Loss2: 0.0998 +Epoch [169/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1433, Loss2: 0.1451 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 54.5172 % Model2 55.4187 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1424, Loss2: 0.1567 +Epoch [170/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.1597, Loss2: 0.1652 +Epoch [170/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.1230, Loss2: 0.1107 +Epoch [170/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1039, Loss2: 0.1065 +Epoch [170/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1307, Loss2: 0.1402 +Epoch [170/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1249, Loss2: 0.1248 +Epoch [170/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1272, Loss2: 0.1283 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 54.9379 % Model2 55.6591 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0986, Loss2: 0.1022 +Epoch [171/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1236, Loss2: 0.1208 +Epoch [171/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1179, Loss2: 0.1101 +Epoch [171/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.1473, Loss2: 0.1553 +Epoch [171/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1270, Loss2: 0.1234 +Epoch [171/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1379, Loss2: 0.1412 +Epoch [171/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1241, Loss2: 0.1099 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 54.4371 % Model2 55.5589 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.1304, Loss2: 0.1317 +Epoch [172/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1507, Loss2: 0.1593 +Epoch [172/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1281, Loss2: 0.1264 +Epoch [172/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1816, Loss2: 0.1764 +Epoch [172/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.1322, Loss2: 0.1458 +Epoch [172/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1339, Loss2: 0.1374 +Epoch [172/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.1224, Loss2: 0.1436 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 54.7676 % Model2 55.8293 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1581, Loss2: 0.1598 +Epoch [173/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1343, Loss2: 0.1454 +Epoch [173/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.1378, Loss2: 0.1501 +Epoch [173/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1503, Loss2: 0.1458 +Epoch [173/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 80.4688, Loss1: 0.1136, Loss2: 0.0957 +Epoch [173/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.1559, Loss2: 0.1899 +Epoch [173/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0914, Loss2: 0.0931 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 54.6775 % Model2 55.4187 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.1250, Loss1: 0.2017, Loss2: 0.1700 +Epoch [174/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1622, Loss2: 0.1659 +Epoch [174/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1349, Loss2: 0.1341 +Epoch [174/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.1075, Loss2: 0.1175 +Epoch [174/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1454, Loss2: 0.1455 +Epoch [174/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1293, Loss2: 0.1367 +Epoch [174/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1133, Loss2: 0.1200 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 54.3670 % Model2 55.9796 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1310, Loss2: 0.1306 +Epoch [175/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1367, Loss2: 0.1202 +Epoch [175/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1528, Loss2: 0.1504 +Epoch [175/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1158, Loss2: 0.1105 +Epoch [175/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1581, Loss2: 0.1497 +Epoch [175/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1143, Loss2: 0.1259 +Epoch [175/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1483, Loss2: 0.1591 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 54.6875 % Model2 55.4387 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1030, Loss2: 0.1104 +Epoch [176/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.0898, Loss2: 0.1025 +Epoch [176/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1115, Loss2: 0.1031 +Epoch [176/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1270, Loss2: 0.1395 +Epoch [176/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.2804, Loss2: 0.2443 +Epoch [176/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1630, Loss2: 0.1593 +Epoch [176/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1423, Loss2: 0.1452 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 54.3470 % Model2 55.7292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1531, Loss2: 0.1408 +Epoch [177/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.2113, Loss2: 0.2112 +Epoch [177/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1381, Loss2: 0.1398 +Epoch [177/200], Iter [200/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.2039, Loss2: 0.2253 +Epoch [177/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1634, Loss2: 0.1803 +Epoch [177/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.1525, Loss2: 0.1373 +Epoch [177/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.2328, Loss2: 0.2104 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 54.2668 % Model2 55.3085 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1128, Loss2: 0.1122 +Epoch [178/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 84.3750, Loss1: 0.1420, Loss2: 0.1225 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.1422, Loss2: 0.1417 +Epoch [178/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.2909, Loss2: 0.2748 +Epoch [178/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1352, Loss2: 0.1203 +Epoch [178/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1344, Loss2: 0.1263 +Epoch [178/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1967, Loss2: 0.1902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 54.2268 % Model2 55.9395 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1362, Loss2: 0.1546 +Epoch [179/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.1875, Loss1: 0.1464, Loss2: 0.1675 +Epoch [179/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1067, Loss2: 0.1115 +Epoch [179/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.1746, Loss2: 0.1713 +Epoch [179/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1501, Loss2: 0.1397 +Epoch [179/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1098, Loss2: 0.1222 +Epoch [179/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1064, Loss2: 0.1006 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 54.3470 % Model2 55.4688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1307, Loss2: 0.1159 +Epoch [180/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1487, Loss2: 0.1440 +Epoch [180/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1766, Loss2: 0.1683 +Epoch [180/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1127, Loss2: 0.1151 +Epoch [180/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1396, Loss2: 0.1467 +Epoch [180/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1588, Loss2: 0.1579 +Epoch [180/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1151, Loss2: 0.1183 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 53.8762 % Model2 55.7692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1146, Loss2: 0.1198 +Epoch [181/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 75.7812, Loss1: 0.1697, Loss2: 0.1939 +Epoch [181/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1249, Loss2: 0.1206 +Epoch [181/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.2118, Loss2: 0.1985 +Epoch [181/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.1982, Loss2: 0.1986 +Epoch [181/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1496, Loss2: 0.1459 +Epoch [181/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 76.5625, Loss1: 0.1241, Loss2: 0.1083 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 54.1466 % Model2 55.7392 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 81.2500, Loss1: 0.2709, Loss2: 0.2361 +Epoch [182/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.1303, Loss2: 0.1487 +Epoch [182/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1349, Loss2: 0.1395 +Epoch [182/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1623, Loss2: 0.1761 +Epoch [182/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1020, Loss2: 0.0965 +Epoch [182/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1365, Loss2: 0.1368 +Epoch [182/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0912, Loss2: 0.0951 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 54.3470 % Model2 55.7091 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1982, Loss2: 0.2137 +Epoch [183/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1374, Loss2: 0.1492 +Epoch [183/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.1129, Loss2: 0.1176 +Epoch [183/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.1339, Loss2: 0.1262 +Epoch [183/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1340, Loss2: 0.1370 +Epoch [183/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1744, Loss2: 0.1697 +Epoch [183/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.1286, Loss2: 0.1101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 54.2568 % Model2 55.7692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1906, Loss2: 0.1904 +Epoch [184/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1126, Loss2: 0.1097 +Epoch [184/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.1437, Loss2: 0.1283 +Epoch [184/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1479, Loss2: 0.1553 +Epoch [184/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1359, Loss2: 0.1302 +Epoch [184/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 70.3125, Loss1: 0.1088, Loss2: 0.1298 +Epoch [184/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1757, Loss2: 0.1762 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 54.0966 % Model2 55.5689 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1260, Loss2: 0.1276 +Epoch [185/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1406, Loss2: 0.1410 +Epoch [185/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1438, Loss2: 0.1306 +Epoch [185/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1675, Loss2: 0.1769 +Epoch [185/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.2011, Loss2: 0.2153 +Epoch [185/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 85.1562, Loss1: 0.2301, Loss2: 0.1951 +Epoch [185/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1144, Loss2: 0.1081 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 54.2568 % Model2 55.7192 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 83.5938, Training Accuracy2: 81.2500, Loss1: 0.2611, Loss2: 0.2840 +Epoch [186/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1362, Loss2: 0.1375 +Epoch [186/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.1202, Loss2: 0.1357 +Epoch [186/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.1117, Loss2: 0.1143 +Epoch [186/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1585, Loss2: 0.1503 +Epoch [186/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1877, Loss2: 0.1655 +Epoch [186/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1407, Loss2: 0.1315 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 54.5473 % Model2 55.5589 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.1771, Loss2: 0.1704 +Epoch [187/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1444, Loss2: 0.1457 +Epoch [187/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1978, Loss2: 0.2229 +Epoch [187/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.2142, Loss2: 0.1900 +Epoch [187/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1900, Loss2: 0.1697 +Epoch [187/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1178, Loss2: 0.1338 +Epoch [187/200], Iter [350/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.1495, Loss2: 0.1547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 54.2668 % Model2 55.2584 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.1205, Loss2: 0.1284 +Epoch [188/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 73.4375, Loss1: 0.1190, Loss2: 0.1335 +Epoch [188/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.1034, Loss2: 0.1149 +Epoch [188/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1118, Loss2: 0.1076 +Epoch [188/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1357, Loss2: 0.1344 +Epoch [188/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1646, Loss2: 0.1819 +Epoch [188/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1078, Loss2: 0.1174 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 54.2768 % Model2 55.6591 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1503, Loss2: 0.1598 +Epoch [189/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1363, Loss2: 0.1276 +Epoch [189/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1125, Loss2: 0.1016 +Epoch [189/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1496, Loss2: 0.1589 +Epoch [189/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1406, Loss2: 0.1376 +Epoch [189/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1287, Loss2: 0.1187 +Epoch [189/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1338, Loss2: 0.1375 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 54.2268 % Model2 55.9195 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.2032, Loss2: 0.1900 +Epoch [190/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1257, Loss2: 0.1256 +Epoch [190/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1506, Loss2: 0.1501 +Epoch [190/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.2141, Loss2: 0.2062 +Epoch [190/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1901, Loss2: 0.1697 +Epoch [190/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1169, Loss2: 0.1092 +Epoch [190/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 83.5938, Loss1: 0.1419, Loss2: 0.1336 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 54.0164 % Model2 55.4988 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.1337, Loss2: 0.1354 +Epoch [191/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.1771, Loss2: 0.1483 +Epoch [191/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.1409, Loss2: 0.1362 +Epoch [191/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1500, Loss2: 0.1597 +Epoch [191/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1393, Loss2: 0.1299 +Epoch [191/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1187, Loss2: 0.1335 +Epoch [191/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1360, Loss2: 0.1307 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 53.9663 % Model2 55.6390 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 81.2500, Loss1: 0.2486, Loss2: 0.2355 +Epoch [192/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1234, Loss2: 0.1212 +Epoch [192/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1548, Loss2: 0.1291 +Epoch [192/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1410, Loss2: 0.1458 +Epoch [192/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.2265, Loss2: 0.2651 +Epoch [192/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1080, Loss2: 0.1042 +Epoch [192/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.1092, Loss2: 0.1200 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 53.9864 % Model2 55.3185 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1417, Loss2: 0.1451 +Epoch [193/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0746, Loss2: 0.0784 +Epoch [193/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1960, Loss2: 0.1766 +Epoch [193/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.1939, Loss2: 0.1973 +Epoch [193/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.3049, Loss2: 0.2764 +Epoch [193/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1756, Loss2: 0.1651 +Epoch [193/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.1071, Loss2: 0.0987 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 54.1466 % Model2 55.4688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 72.6562, Loss1: 0.1082, Loss2: 0.1190 +Epoch [194/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.1366, Loss2: 0.1413 +Epoch [194/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1263, Loss2: 0.1235 +Epoch [194/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1625, Loss2: 0.1549 +Epoch [194/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.1186, Loss2: 0.1208 +Epoch [194/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1645, Loss2: 0.1645 +Epoch [194/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.1701, Loss2: 0.1592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 54.0365 % Model2 55.4688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1705, Loss2: 0.1646 +Epoch [195/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.1250, Loss1: 0.1581, Loss2: 0.1419 +Epoch [195/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.1250, Loss1: 0.1897, Loss2: 0.2052 +Epoch [195/200], Iter [200/390] Training Accuracy1: 84.3750, Training Accuracy2: 81.2500, Loss1: 0.1596, Loss2: 0.1709 +Epoch [195/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1335, Loss2: 0.1280 +Epoch [195/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1439, Loss2: 0.1496 +Epoch [195/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1242, Loss2: 0.1355 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 53.7360 % Model2 55.3886 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.1180, Loss2: 0.1295 +Epoch [196/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1541, Loss2: 0.1498 +Epoch [196/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1092, Loss2: 0.1035 +Epoch [196/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1051, Loss2: 0.1119 +Epoch [196/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0936, Loss2: 0.0886 +Epoch [196/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1608, Loss2: 0.1445 +Epoch [196/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1900, Loss2: 0.1850 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 53.8261 % Model2 55.4087 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1271, Loss2: 0.1337 +Epoch [197/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1507, Loss2: 0.1725 +Epoch [197/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1408, Loss2: 0.1389 +Epoch [197/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1302, Loss2: 0.1270 +Epoch [197/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1603, Loss2: 0.1542 +Epoch [197/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1411, Loss2: 0.1543 +Epoch [197/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.1148, Loss2: 0.1169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 53.8261 % Model2 55.4688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1361, Loss2: 0.1268 +Epoch [198/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1443, Loss2: 0.1451 +Epoch [198/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.1235, Loss2: 0.1350 +Epoch [198/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1915, Loss2: 0.1903 +Epoch [198/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.1685, Loss2: 0.1642 +Epoch [198/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.1993, Loss2: 0.2137 +Epoch [198/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1495, Loss2: 0.1675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 53.8161 % Model2 55.3886 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 67.9688, Loss1: 0.0909, Loss2: 0.1089 +Epoch [199/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1891, Loss2: 0.1770 +Epoch [199/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1044, Loss2: 0.0971 +Epoch [199/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.0000, Loss1: 0.1227, Loss2: 0.1040 +Epoch [199/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.2158, Loss2: 0.2151 +Epoch [199/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1182, Loss2: 0.1207 +Epoch [199/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 82.0312, Loss1: 0.1644, Loss2: 0.1493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 53.7460 % Model2 55.2985 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.1591, Loss2: 0.1645 +Epoch [200/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1694, Loss2: 0.1499 +Epoch [200/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1834, Loss2: 0.1835 +Epoch [200/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1052, Loss2: 0.1029 +Epoch [200/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1301, Loss2: 0.1269 +Epoch [200/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.9062, Loss1: 0.1298, Loss2: 0.1448 +Epoch [200/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.1764, Loss2: 0.1826 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 53.7961 % Model2 55.3085 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_0_4.log b/other_methods/coteaching_plus/coteaching_plus_results/out_0_4.log new file mode 100644 index 0000000..a20ee4f --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_0_4.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 13.2812, Training Accuracy2: 13.2812, Loss1: 0.0180, Loss2: 0.0180 +Epoch [2/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 17.9688, Loss1: 0.0172, Loss2: 0.0173 +Epoch [2/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 21.0938, Loss1: 0.0160, Loss2: 0.0163 +Epoch [2/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 31.2500, Loss1: 0.0167, Loss2: 0.0166 +Epoch [2/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 23.4375, Loss1: 0.0158, Loss2: 0.0157 +Epoch [2/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0155, Loss2: 0.0156 +Epoch [2/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 22.6562, Loss1: 0.0168, Loss2: 0.0172 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 33.2332 % Model2 33.4836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0155, Loss2: 0.0158 +Epoch [3/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.7812, Loss1: 0.0160, Loss2: 0.0160 +Epoch [3/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 21.8750, Loss1: 0.0157, Loss2: 0.0158 +Epoch [3/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0157, Loss2: 0.0158 +Epoch [3/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0155 +Epoch [3/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0162, Loss2: 0.0161 +Epoch [3/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0152 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 41.2861 % Model2 39.4832 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0153, Loss2: 0.0154 +Epoch [4/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0142, Loss2: 0.0144 +Epoch [4/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 24.2188, Loss1: 0.0156, Loss2: 0.0163 +Epoch [4/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.0312, Loss1: 0.0155, Loss2: 0.0157 +Epoch [4/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0155, Loss2: 0.0153 +Epoch [4/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0148, Loss2: 0.0149 +Epoch [4/200], Iter [350/390] Training Accuracy1: 20.3125, Training Accuracy2: 25.7812, Loss1: 0.0168, Loss2: 0.0162 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 40.2845 % Model2 38.1510 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 32.0312, Loss1: 0.0154, Loss2: 0.0154 +Epoch [5/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0156, Loss2: 0.0153 +Epoch [5/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0141, Loss2: 0.0145 +Epoch [5/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 26.5625, Loss1: 0.0171, Loss2: 0.0169 +Epoch [5/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0152, Loss2: 0.0151 +Epoch [5/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 29.6875, Loss1: 0.0152, Loss2: 0.0155 +Epoch [5/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0149, Loss2: 0.0154 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 42.5080 % Model2 42.2676 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0142, Loss2: 0.0136 +Epoch [6/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0133, Loss2: 0.0132 +Epoch [6/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0142, Loss2: 0.0139 +Epoch [6/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0161, Loss2: 0.0162 +Epoch [6/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0144, Loss2: 0.0145 +Epoch [6/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0149, Loss2: 0.0149 +Epoch [6/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 36.7188, Loss1: 0.0159, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 44.5312 % Model2 47.4659 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0129, Loss2: 0.0132 +Epoch [7/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0149, Loss2: 0.0142 +Epoch [7/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0149, Loss2: 0.0153 +Epoch [7/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0141, Loss2: 0.0135 +Epoch [7/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0153, Loss2: 0.0151 +Epoch [7/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0155, Loss2: 0.0155 +Epoch [7/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0148, Loss2: 0.0153 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 47.5361 % Model2 49.3990 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0139, Loss2: 0.0137 +Epoch [8/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0151, Loss2: 0.0151 +Epoch [8/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0134, Loss2: 0.0125 +Epoch [8/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0126, Loss2: 0.0124 +Epoch [8/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0132, Loss2: 0.0135 +Epoch [8/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0154 +Epoch [8/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 32.0312, Loss1: 0.0155, Loss2: 0.0161 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 48.8682 % Model2 50.6611 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0156, Loss2: 0.0145 +Epoch [9/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 33.5938, Loss1: 0.0165, Loss2: 0.0157 +Epoch [9/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0140, Loss2: 0.0132 +Epoch [9/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0137, Loss2: 0.0137 +Epoch [9/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0137, Loss2: 0.0133 +Epoch [9/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0129, Loss2: 0.0128 +Epoch [9/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0134, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 50.1002 % Model2 50.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0131, Loss2: 0.0136 +Epoch [10/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0136, Loss2: 0.0134 +Epoch [10/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0151, Loss2: 0.0150 +Epoch [10/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.1562, Loss1: 0.0138, Loss2: 0.0139 +Epoch [10/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0138, Loss2: 0.0143 +Epoch [10/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0137, Loss2: 0.0127 +Epoch [10/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0145, Loss2: 0.0139 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 48.4976 % Model2 49.9399 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0124, Loss2: 0.0115 +Epoch [11/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0128, Loss2: 0.0131 +Epoch [11/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0160, Loss2: 0.0153 +Epoch [11/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0127, Loss2: 0.0125 +Epoch [11/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0143, Loss2: 0.0149 +Epoch [11/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0156, Loss2: 0.0157 +Epoch [11/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0142, Loss2: 0.0130 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 48.7280 % Model2 50.6310 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0129, Loss2: 0.0131 +Epoch [12/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0152, Loss2: 0.0157 +Epoch [12/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0130, Loss2: 0.0126 +Epoch [12/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0138, Loss2: 0.0140 +Epoch [12/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0134, Loss2: 0.0133 +Epoch [12/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0109, Loss2: 0.0105 +Epoch [12/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0128, Loss2: 0.0121 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 50.6010 % Model2 52.5040 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0117, Loss2: 0.0113 +Epoch [13/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0137, Loss2: 0.0133 +Epoch [13/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0139, Loss2: 0.0133 +Epoch [13/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0134, Loss2: 0.0141 +Epoch [13/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0130, Loss2: 0.0124 +Epoch [13/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0123, Loss2: 0.0121 +Epoch [13/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0140, Loss2: 0.0138 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 51.0717 % Model2 52.2736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0124, Loss2: 0.0124 +Epoch [14/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.9375, Loss1: 0.0143, Loss2: 0.0139 +Epoch [14/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0138, Loss2: 0.0136 +Epoch [14/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0136, Loss2: 0.0142 +Epoch [14/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0137, Loss2: 0.0139 +Epoch [14/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0145, Loss2: 0.0153 +Epoch [14/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0133, Loss2: 0.0142 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 47.8365 % Model2 50.4107 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0145, Loss2: 0.0140 +Epoch [15/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0121, Loss2: 0.0125 +Epoch [15/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 33.5938, Loss1: 0.0135, Loss2: 0.0130 +Epoch [15/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0113, Loss2: 0.0121 +Epoch [15/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0145, Loss2: 0.0133 +Epoch [15/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0138, Loss2: 0.0129 +Epoch [15/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0127, Loss2: 0.0128 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 49.9700 % Model2 52.8345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0136, Loss2: 0.0135 +Epoch [16/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0121, Loss2: 0.0128 +Epoch [16/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0121, Loss2: 0.0116 +Epoch [16/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0149, Loss2: 0.0143 +Epoch [16/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 47.6562, Loss1: 0.0118, Loss2: 0.0107 +Epoch [16/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0129, Loss2: 0.0134 +Epoch [16/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0117, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 53.5256 % Model2 53.7861 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0119, Loss2: 0.0112 +Epoch [17/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0140, Loss2: 0.0130 +Epoch [17/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0137, Loss2: 0.0134 +Epoch [17/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0127, Loss2: 0.0126 +Epoch [17/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0119, Loss2: 0.0115 +Epoch [17/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0130, Loss2: 0.0122 +Epoch [17/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0130, Loss2: 0.0123 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 52.4539 % Model2 53.2452 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0113, Loss2: 0.0108 +Epoch [18/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0155, Loss2: 0.0156 +Epoch [18/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0124, Loss2: 0.0118 +Epoch [18/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0132, Loss2: 0.0123 +Epoch [18/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0114, Loss2: 0.0120 +Epoch [18/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0102, Loss2: 0.0109 +Epoch [18/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0117, Loss2: 0.0121 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 51.3021 % Model2 50.5108 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0128 +Epoch [19/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0123, Loss2: 0.0124 +Epoch [19/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0091, Loss2: 0.0092 +Epoch [19/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0118, Loss2: 0.0123 +Epoch [19/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0125 +Epoch [19/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0115, Loss2: 0.0117 +Epoch [19/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 34.3750, Loss1: 0.0133, Loss2: 0.0135 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 51.1218 % Model2 51.7829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0114, Loss2: 0.0117 +Epoch [20/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0144, Loss2: 0.0143 +Epoch [20/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0116, Loss2: 0.0113 +Epoch [20/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0128, Loss2: 0.0126 +Epoch [20/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0115, Loss2: 0.0115 +Epoch [20/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0107, Loss2: 0.0110 +Epoch [20/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0120, Loss2: 0.0114 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 51.9331 % Model2 51.5825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0683, Loss2: 0.0678 +Epoch [21/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0581, Loss2: 0.0571 +Epoch [21/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 41.4062, Loss1: 0.0716, Loss2: 0.0664 +Epoch [21/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.8438, Loss1: 0.0542, Loss2: 0.0569 +Epoch [21/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0635, Loss2: 0.0611 +Epoch [21/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0452, Loss2: 0.0451 +Epoch [21/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0602, Loss2: 0.0617 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 52.4439 % Model2 52.7344 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0492, Loss2: 0.0478 +Epoch [22/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0631, Loss2: 0.0622 +Epoch [22/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0511, Loss2: 0.0503 +Epoch [22/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0506, Loss2: 0.0532 +Epoch [22/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0649, Loss2: 0.0650 +Epoch [22/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0488, Loss2: 0.0477 +Epoch [22/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0639, Loss2: 0.0626 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 47.2857 % Model2 52.5341 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0569, Loss2: 0.0562 +Epoch [23/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0539, Loss2: 0.0542 +Epoch [23/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0553, Loss2: 0.0563 +Epoch [23/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0615, Loss2: 0.0609 +Epoch [23/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0657, Loss2: 0.0654 +Epoch [23/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0578, Loss2: 0.0576 +Epoch [23/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0696, Loss2: 0.0671 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 51.9030 % Model2 51.4022 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0827, Loss2: 0.0858 +Epoch [24/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0547, Loss2: 0.0563 +Epoch [24/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0612, Loss2: 0.0615 +Epoch [24/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0551, Loss2: 0.0552 +Epoch [24/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0729, Loss2: 0.0717 +Epoch [24/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0543, Loss2: 0.0524 +Epoch [24/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0557, Loss2: 0.0551 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 51.3722 % Model2 52.9647 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 32.8125, Loss1: 0.0596, Loss2: 0.0613 +Epoch [25/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0513, Loss2: 0.0512 +Epoch [25/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0808, Loss2: 0.0812 +Epoch [25/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0602, Loss2: 0.0590 +Epoch [25/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0514, Loss2: 0.0498 +Epoch [25/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0581, Loss2: 0.0563 +Epoch [25/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0634, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 51.6827 % Model2 54.5373 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0541, Loss2: 0.0542 +Epoch [26/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0700, Loss2: 0.0666 +Epoch [26/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0557, Loss2: 0.0543 +Epoch [26/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0589, Loss2: 0.0593 +Epoch [26/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0758, Loss2: 0.0777 +Epoch [26/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0667, Loss2: 0.0663 +Epoch [26/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0543, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 52.0533 % Model2 52.0333 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0481, Loss2: 0.0463 +Epoch [27/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0523, Loss2: 0.0499 +Epoch [27/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0639, Loss2: 0.0654 +Epoch [27/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0526, Loss2: 0.0532 +Epoch [27/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0492, Loss2: 0.0481 +Epoch [27/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0618, Loss2: 0.0596 +Epoch [27/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0521, Loss2: 0.0495 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 52.6242 % Model2 54.1767 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.9375, Loss1: 0.0548, Loss2: 0.0550 +Epoch [28/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0607, Loss2: 0.0608 +Epoch [28/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0575, Loss2: 0.0560 +Epoch [28/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0527, Loss2: 0.0522 +Epoch [28/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0752, Loss2: 0.0764 +Epoch [28/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0601, Loss2: 0.0620 +Epoch [28/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0694, Loss2: 0.0723 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 52.9848 % Model2 53.4956 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0563, Loss2: 0.0536 +Epoch [29/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0542, Loss2: 0.0526 +Epoch [29/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 45.3125, Loss1: 0.0682, Loss2: 0.0620 +Epoch [29/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0612, Loss2: 0.0610 +Epoch [29/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0665, Loss2: 0.0642 +Epoch [29/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0656, Loss2: 0.0651 +Epoch [29/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0564, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 52.1534 % Model2 53.8361 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0639, Loss2: 0.0631 +Epoch [30/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0509, Loss2: 0.0515 +Epoch [30/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0550, Loss2: 0.0563 +Epoch [30/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0568, Loss2: 0.0567 +Epoch [30/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0607, Loss2: 0.0625 +Epoch [30/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0772, Loss2: 0.0729 +Epoch [30/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0571, Loss2: 0.0585 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 53.9363 % Model2 52.4940 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0603, Loss2: 0.0603 +Epoch [31/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0544, Loss2: 0.0525 +Epoch [31/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0712, Loss2: 0.0692 +Epoch [31/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0574, Loss2: 0.0567 +Epoch [31/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0520, Loss2: 0.0501 +Epoch [31/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0644, Loss2: 0.0628 +Epoch [31/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0518, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 49.4491 % Model2 53.6258 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0529, Loss2: 0.0522 +Epoch [32/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0650, Loss2: 0.0632 +Epoch [32/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0613, Loss2: 0.0607 +Epoch [32/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0551, Loss2: 0.0532 +Epoch [32/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0616, Loss2: 0.0623 +Epoch [32/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0590, Loss2: 0.0571 +Epoch [32/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0552, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 52.5240 % Model2 52.2536 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0553, Loss2: 0.0537 +Epoch [33/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0745, Loss2: 0.0688 +Epoch [33/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0533, Loss2: 0.0507 +Epoch [33/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0693, Loss2: 0.0690 +Epoch [33/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0612, Loss2: 0.0578 +Epoch [33/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0607, Loss2: 0.0614 +Epoch [33/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0577, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 52.5541 % Model2 55.2284 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0504, Loss2: 0.0478 +Epoch [34/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0498, Loss2: 0.0485 +Epoch [34/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0698, Loss2: 0.0667 +Epoch [34/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0624, Loss2: 0.0639 +Epoch [34/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 44.5312, Loss1: 0.0467, Loss2: 0.0491 +Epoch [34/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 54.6875, Loss1: 0.0557, Loss2: 0.0519 +Epoch [34/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0518, Loss2: 0.0516 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 52.2837 % Model2 52.7043 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0726, Loss2: 0.0693 +Epoch [35/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0518, Loss2: 0.0532 +Epoch [35/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0651 +Epoch [35/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0518, Loss2: 0.0524 +Epoch [35/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0510, Loss2: 0.0492 +Epoch [35/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0501, Loss2: 0.0509 +Epoch [35/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0683, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 52.7043 % Model2 54.1466 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0535, Loss2: 0.0528 +Epoch [36/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0615, Loss2: 0.0578 +Epoch [36/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0685, Loss2: 0.0678 +Epoch [36/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0771, Loss2: 0.0733 +Epoch [36/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0569, Loss2: 0.0557 +Epoch [36/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0618, Loss2: 0.0601 +Epoch [36/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0540, Loss2: 0.0547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 53.5357 % Model2 53.1150 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.0625, Loss1: 0.0591, Loss2: 0.0619 +Epoch [37/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0638, Loss2: 0.0628 +Epoch [37/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0620, Loss2: 0.0637 +Epoch [37/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0690, Loss2: 0.0672 +Epoch [37/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0530, Loss2: 0.0508 +Epoch [37/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0527, Loss2: 0.0525 +Epoch [37/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 49.2188, Loss1: 0.0479, Loss2: 0.0451 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 51.2520 % Model2 54.2869 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0586, Loss2: 0.0588 +Epoch [38/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0631, Loss2: 0.0597 +Epoch [38/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0497, Loss2: 0.0480 +Epoch [38/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0591, Loss2: 0.0606 +Epoch [38/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0777, Loss2: 0.0756 +Epoch [38/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.0000, Loss1: 0.0507, Loss2: 0.0549 +Epoch [38/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 51.5625, Loss1: 0.0602, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 51.3622 % Model2 53.0248 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0540, Loss2: 0.0546 +Epoch [39/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0544, Loss2: 0.0533 +Epoch [39/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0597, Loss2: 0.0613 +Epoch [39/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0562, Loss2: 0.0583 +Epoch [39/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0593, Loss2: 0.0592 +Epoch [39/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0579, Loss2: 0.0561 +Epoch [39/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0700, Loss2: 0.0720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 52.8245 % Model2 52.4940 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0452, Loss2: 0.0456 +Epoch [40/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0604, Loss2: 0.0592 +Epoch [40/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0846, Loss2: 0.0811 +Epoch [40/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.9688, Loss1: 0.0524, Loss2: 0.0546 +Epoch [40/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0582, Loss2: 0.0562 +Epoch [40/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0659, Loss2: 0.0662 +Epoch [40/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.8750, Loss1: 0.0473, Loss2: 0.0479 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 51.0917 % Model2 51.6226 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0583, Loss2: 0.0587 +Epoch [41/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0536, Loss2: 0.0538 +Epoch [41/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0617, Loss2: 0.0610 +Epoch [41/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0488, Loss2: 0.0474 +Epoch [41/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0513, Loss2: 0.0506 +Epoch [41/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0612, Loss2: 0.0582 +Epoch [41/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0646, Loss2: 0.0620 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 51.1619 % Model2 51.4623 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0557, Loss2: 0.0566 +Epoch [42/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0669, Loss2: 0.0679 +Epoch [42/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0574, Loss2: 0.0556 +Epoch [42/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0705, Loss2: 0.0696 +Epoch [42/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0541, Loss2: 0.0522 +Epoch [42/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0524, Loss2: 0.0521 +Epoch [42/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0698, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 50.8013 % Model2 52.4639 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0502, Loss2: 0.0509 +Epoch [43/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0542, Loss2: 0.0560 +Epoch [43/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0602, Loss2: 0.0606 +Epoch [43/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0744, Loss2: 0.0719 +Epoch [43/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0634, Loss2: 0.0629 +Epoch [43/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0758, Loss2: 0.0757 +Epoch [43/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0494, Loss2: 0.0479 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 48.9984 % Model2 52.8345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0645, Loss2: 0.0646 +Epoch [44/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0496, Loss2: 0.0498 +Epoch [44/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0571, Loss2: 0.0573 +Epoch [44/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0771, Loss2: 0.0731 +Epoch [44/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0758, Loss2: 0.0762 +Epoch [44/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0605, Loss2: 0.0605 +Epoch [44/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0621, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 52.7143 % Model2 52.7544 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0565, Loss2: 0.0568 +Epoch [45/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0512, Loss2: 0.0533 +Epoch [45/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0468, Loss2: 0.0467 +Epoch [45/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0627, Loss2: 0.0639 +Epoch [45/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0455, Loss2: 0.0460 +Epoch [45/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0664, Loss2: 0.0666 +Epoch [45/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0657, Loss2: 0.0635 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 50.8614 % Model2 52.2736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0598, Loss2: 0.0591 +Epoch [46/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0576, Loss2: 0.0584 +Epoch [46/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0750, Loss2: 0.0725 +Epoch [46/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0721, Loss2: 0.0741 +Epoch [46/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0636, Loss2: 0.0617 +Epoch [46/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0601, Loss2: 0.0590 +Epoch [46/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0551, Loss2: 0.0506 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 50.1102 % Model2 51.6026 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0486, Loss2: 0.0496 +Epoch [47/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0536, Loss2: 0.0534 +Epoch [47/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0586, Loss2: 0.0582 +Epoch [47/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0473, Loss2: 0.0470 +Epoch [47/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0577, Loss2: 0.0556 +Epoch [47/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0618, Loss2: 0.0636 +Epoch [47/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0646, Loss2: 0.0641 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 50.4507 % Model2 52.2436 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0597, Loss2: 0.0567 +Epoch [48/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0545, Loss2: 0.0524 +Epoch [48/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0619, Loss2: 0.0614 +Epoch [48/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0611, Loss2: 0.0594 +Epoch [48/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0596, Loss2: 0.0585 +Epoch [48/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0536, Loss2: 0.0519 +Epoch [48/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0700, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 49.7196 % Model2 52.1735 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0674, Loss2: 0.0663 +Epoch [49/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0516, Loss2: 0.0507 +Epoch [49/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0814, Loss2: 0.0756 +Epoch [49/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0552, Loss2: 0.0572 +Epoch [49/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0540, Loss2: 0.0511 +Epoch [49/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0529, Loss2: 0.0522 +Epoch [49/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0522, Loss2: 0.0540 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 51.4223 % Model2 52.1735 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0524, Loss2: 0.0527 +Epoch [50/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0527, Loss2: 0.0511 +Epoch [50/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0675, Loss2: 0.0673 +Epoch [50/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0699, Loss2: 0.0683 +Epoch [50/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0552, Loss2: 0.0519 +Epoch [50/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0523, Loss2: 0.0548 +Epoch [50/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0593, Loss2: 0.0580 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 49.7997 % Model2 51.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0592, Loss2: 0.0598 +Epoch [51/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0591, Loss2: 0.0592 +Epoch [51/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0631, Loss2: 0.0600 +Epoch [51/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0531, Loss2: 0.0509 +Epoch [51/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0632, Loss2: 0.0626 +Epoch [51/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 40.6250, Loss1: 0.0463, Loss2: 0.0488 +Epoch [51/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0801, Loss2: 0.0729 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 50.4207 % Model2 53.1250 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0527, Loss2: 0.0531 +Epoch [52/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0753, Loss2: 0.0734 +Epoch [52/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0584, Loss2: 0.0614 +Epoch [52/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0552, Loss2: 0.0583 +Epoch [52/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0605, Loss2: 0.0619 +Epoch [52/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 48.4375, Loss1: 0.0681, Loss2: 0.0762 +Epoch [52/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0760, Loss2: 0.0792 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 50.8113 % Model2 51.3121 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0637, Loss2: 0.0607 +Epoch [53/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0635, Loss2: 0.0612 +Epoch [53/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0550, Loss2: 0.0531 +Epoch [53/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0557, Loss2: 0.0571 +Epoch [53/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0521, Loss2: 0.0501 +Epoch [53/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0556, Loss2: 0.0536 +Epoch [53/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0531, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 50.7612 % Model2 52.7043 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0593, Loss2: 0.0594 +Epoch [54/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0536, Loss2: 0.0509 +Epoch [54/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0643, Loss2: 0.0644 +Epoch [54/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0691, Loss2: 0.0696 +Epoch [54/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0601, Loss2: 0.0603 +Epoch [54/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0557, Loss2: 0.0535 +Epoch [54/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0731, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 50.0401 % Model2 51.9331 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 44.5312, Loss1: 0.0504, Loss2: 0.0590 +Epoch [55/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0532, Loss2: 0.0530 +Epoch [55/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0623, Loss2: 0.0608 +Epoch [55/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0643, Loss2: 0.0622 +Epoch [55/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0564, Loss2: 0.0574 +Epoch [55/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0591, Loss2: 0.0590 +Epoch [55/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0468, Loss2: 0.0472 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 50.1002 % Model2 51.6326 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0622, Loss2: 0.0635 +Epoch [56/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0548, Loss2: 0.0541 +Epoch [56/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0683, Loss2: 0.0619 +Epoch [56/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 52.3438, Loss1: 0.0570, Loss2: 0.0664 +Epoch [56/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0590, Loss2: 0.0636 +Epoch [56/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0714, Loss2: 0.0724 +Epoch [56/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0608, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 50.6811 % Model2 52.5942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0495, Loss2: 0.0489 +Epoch [57/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 67.9688, Loss1: 0.0766, Loss2: 0.0692 +Epoch [57/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0486, Loss2: 0.0475 +Epoch [57/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0726, Loss2: 0.0692 +Epoch [57/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0596, Loss2: 0.0605 +Epoch [57/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0537, Loss2: 0.0543 +Epoch [57/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0536, Loss2: 0.0536 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 51.4824 % Model2 51.7728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0585, Loss2: 0.0574 +Epoch [58/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0650, Loss2: 0.0584 +Epoch [58/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0534, Loss2: 0.0507 +Epoch [58/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0623, Loss2: 0.0605 +Epoch [58/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0575, Loss2: 0.0584 +Epoch [58/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0499, Loss2: 0.0480 +Epoch [58/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0584, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 51.4423 % Model2 51.5124 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0525, Loss2: 0.0522 +Epoch [59/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0595, Loss2: 0.0638 +Epoch [59/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0631, Loss2: 0.0582 +Epoch [59/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0550, Loss2: 0.0534 +Epoch [59/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0616, Loss2: 0.0604 +Epoch [59/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 44.5312, Loss1: 0.0551, Loss2: 0.0602 +Epoch [59/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0530, Loss2: 0.0531 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 50.6611 % Model2 51.9131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0654, Loss2: 0.0633 +Epoch [60/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0622, Loss2: 0.0613 +Epoch [60/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0607, Loss2: 0.0602 +Epoch [60/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0567, Loss2: 0.0588 +Epoch [60/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0610, Loss2: 0.0644 +Epoch [60/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0490, Loss2: 0.0494 +Epoch [60/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0603, Loss2: 0.0604 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 49.9399 % Model2 51.6226 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0701, Loss2: 0.0715 +Epoch [61/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0530, Loss2: 0.0544 +Epoch [61/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0506, Loss2: 0.0491 +Epoch [61/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 65.6250, Loss1: 0.0657, Loss2: 0.0559 +Epoch [61/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0504, Loss2: 0.0513 +Epoch [61/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0738, Loss2: 0.0728 +Epoch [61/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0555, Loss2: 0.0535 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 49.3790 % Model2 49.9299 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0622, Loss2: 0.0643 +Epoch [62/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0685, Loss2: 0.0696 +Epoch [62/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0573, Loss2: 0.0616 +Epoch [62/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0583, Loss2: 0.0578 +Epoch [62/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0586, Loss2: 0.0569 +Epoch [62/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0544, Loss2: 0.0522 +Epoch [62/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0503, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 49.9399 % Model2 50.8113 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 56.2500, Loss1: 0.0563, Loss2: 0.0539 +Epoch [63/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 59.3750, Loss1: 0.0631, Loss2: 0.0702 +Epoch [63/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0544, Loss2: 0.0521 +Epoch [63/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0718, Loss2: 0.0683 +Epoch [63/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0541, Loss2: 0.0517 +Epoch [63/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0675, Loss2: 0.0649 +Epoch [63/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0556, Loss2: 0.0579 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 48.9683 % Model2 51.2320 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0596, Loss2: 0.0595 +Epoch [64/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0556, Loss2: 0.0569 +Epoch [64/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0829, Loss2: 0.0860 +Epoch [64/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0514, Loss2: 0.0536 +Epoch [64/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 59.3750, Loss1: 0.0582, Loss2: 0.0518 +Epoch [64/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0470, Loss2: 0.0445 +Epoch [64/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0609, Loss2: 0.0593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 50.0801 % Model2 51.9030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 57.8125, Loss1: 0.0539, Loss2: 0.0498 +Epoch [65/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0602, Loss2: 0.0594 +Epoch [65/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0685, Loss2: 0.0715 +Epoch [65/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0588, Loss2: 0.0585 +Epoch [65/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0620, Loss2: 0.0633 +Epoch [65/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0559, Loss2: 0.0561 +Epoch [65/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0603, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 50.1703 % Model2 51.5725 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0532, Loss2: 0.0530 +Epoch [66/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0640, Loss2: 0.0589 +Epoch [66/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0552, Loss2: 0.0544 +Epoch [66/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0511, Loss2: 0.0469 +Epoch [66/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0518, Loss2: 0.0529 +Epoch [66/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0642, Loss2: 0.0631 +Epoch [66/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0595, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 49.5793 % Model2 50.4307 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0474, Loss2: 0.0477 +Epoch [67/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0665, Loss2: 0.0667 +Epoch [67/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0564, Loss2: 0.0533 +Epoch [67/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0785, Loss2: 0.0757 +Epoch [67/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0558, Loss2: 0.0555 +Epoch [67/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0497, Loss2: 0.0509 +Epoch [67/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0542, Loss2: 0.0556 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 50.2304 % Model2 50.7913 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0699, Loss2: 0.0713 +Epoch [68/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0515, Loss2: 0.0486 +Epoch [68/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0496, Loss2: 0.0542 +Epoch [68/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0582, Loss2: 0.0535 +Epoch [68/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0656, Loss2: 0.0615 +Epoch [68/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0696, Loss2: 0.0644 +Epoch [68/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0843, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 49.7596 % Model2 50.5409 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0591, Loss2: 0.0597 +Epoch [69/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0544, Loss2: 0.0569 +Epoch [69/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0650, Loss2: 0.0651 +Epoch [69/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0551, Loss2: 0.0553 +Epoch [69/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 55.4688, Loss1: 0.0657, Loss2: 0.0577 +Epoch [69/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0654, Loss2: 0.0654 +Epoch [69/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0704, Loss2: 0.0627 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 49.8097 % Model2 51.4223 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0699, Loss2: 0.0695 +Epoch [70/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0585, Loss2: 0.0578 +Epoch [70/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0643, Loss2: 0.0597 +Epoch [70/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.9375, Loss1: 0.0703, Loss2: 0.0644 +Epoch [70/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 58.5938, Loss1: 0.0519, Loss2: 0.0476 +Epoch [70/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0854, Loss2: 0.0780 +Epoch [70/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0594, Loss2: 0.0574 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 50.3105 % Model2 50.4607 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0716, Loss2: 0.0729 +Epoch [71/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0718, Loss2: 0.0761 +Epoch [71/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0647, Loss2: 0.0623 +Epoch [71/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0563, Loss2: 0.0536 +Epoch [71/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 44.5312, Loss1: 0.0521, Loss2: 0.0567 +Epoch [71/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0551, Loss2: 0.0561 +Epoch [71/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0552, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 49.9399 % Model2 52.0833 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0587, Loss2: 0.0599 +Epoch [72/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0635, Loss2: 0.0623 +Epoch [72/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0548, Loss2: 0.0534 +Epoch [72/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0731, Loss2: 0.0737 +Epoch [72/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0482, Loss2: 0.0497 +Epoch [72/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0644, Loss2: 0.0653 +Epoch [72/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0631, Loss2: 0.0612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 49.9099 % Model2 50.6711 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0610, Loss2: 0.0601 +Epoch [73/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0600, Loss2: 0.0585 +Epoch [73/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0590, Loss2: 0.0571 +Epoch [73/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0601, Loss2: 0.0606 +Epoch [73/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0593, Loss2: 0.0610 +Epoch [73/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0547, Loss2: 0.0550 +Epoch [73/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0601, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 49.0385 % Model2 50.6611 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0839, Loss2: 0.0841 +Epoch [74/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0770, Loss2: 0.0745 +Epoch [74/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0665, Loss2: 0.0715 +Epoch [74/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0730, Loss2: 0.0715 +Epoch [74/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0540 +Epoch [74/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0569, Loss2: 0.0557 +Epoch [74/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0610, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 49.5793 % Model2 51.7428 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0473, Loss2: 0.0455 +Epoch [75/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0759, Loss2: 0.0808 +Epoch [75/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0712, Loss2: 0.0754 +Epoch [75/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0677, Loss2: 0.0694 +Epoch [75/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0686, Loss2: 0.0676 +Epoch [75/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0623, Loss2: 0.0564 +Epoch [75/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0641, Loss2: 0.0617 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 49.1186 % Model2 51.4423 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0568, Loss2: 0.0611 +Epoch [76/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0546, Loss2: 0.0541 +Epoch [76/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0514, Loss2: 0.0506 +Epoch [76/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0533, Loss2: 0.0543 +Epoch [76/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0831, Loss2: 0.0769 +Epoch [76/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0600, Loss2: 0.0603 +Epoch [76/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0569, Loss2: 0.0548 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 49.3289 % Model2 49.4792 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0576, Loss2: 0.0593 +Epoch [77/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0734, Loss2: 0.0699 +Epoch [77/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0635, Loss2: 0.0576 +Epoch [77/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0496, Loss2: 0.0488 +Epoch [77/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0631, Loss2: 0.0579 +Epoch [77/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0624, Loss2: 0.0610 +Epoch [77/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0659, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 48.8682 % Model2 50.2704 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0565, Loss2: 0.0575 +Epoch [78/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0512, Loss2: 0.0518 +Epoch [78/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0599, Loss2: 0.0622 +Epoch [78/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0637, Loss2: 0.0654 +Epoch [78/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0525, Loss2: 0.0568 +Epoch [78/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0603, Loss2: 0.0616 +Epoch [78/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0636, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 49.4892 % Model2 50.4407 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0613, Loss2: 0.0614 +Epoch [79/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0594, Loss2: 0.0584 +Epoch [79/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0676, Loss2: 0.0688 +Epoch [79/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0569, Loss2: 0.0570 +Epoch [79/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0642, Loss2: 0.0665 +Epoch [79/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0630, Loss2: 0.0627 +Epoch [79/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0555, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 49.3089 % Model2 50.7913 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0520, Loss2: 0.0490 +Epoch [80/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0683, Loss2: 0.0651 +Epoch [80/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0632, Loss2: 0.0596 +Epoch [80/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0565, Loss2: 0.0560 +Epoch [80/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0688, Loss2: 0.0692 +Epoch [80/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0717, Loss2: 0.0689 +Epoch [80/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0625, Loss2: 0.0567 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 48.3574 % Model2 50.8413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0578, Loss2: 0.0555 +Epoch [81/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0571, Loss2: 0.0602 +Epoch [81/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0555, Loss2: 0.0536 +Epoch [81/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0675, Loss2: 0.0678 +Epoch [81/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0684, Loss2: 0.0686 +Epoch [81/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0664, Loss2: 0.0683 +Epoch [81/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0580, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 50.0100 % Model2 51.1218 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0603, Loss2: 0.0609 +Epoch [82/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0783, Loss2: 0.0749 +Epoch [82/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0653, Loss2: 0.0610 +Epoch [82/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0560, Loss2: 0.0543 +Epoch [82/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0583, Loss2: 0.0539 +Epoch [82/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0582, Loss2: 0.0629 +Epoch [82/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0543, Loss2: 0.0521 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 49.3289 % Model2 50.6711 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0538, Loss2: 0.0526 +Epoch [83/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0631, Loss2: 0.0622 +Epoch [83/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0674, Loss2: 0.0663 +Epoch [83/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0606, Loss2: 0.0661 +Epoch [83/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0679, Loss2: 0.0655 +Epoch [83/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0588, Loss2: 0.0590 +Epoch [83/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0614, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 49.0385 % Model2 50.1903 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0673, Loss2: 0.0673 +Epoch [84/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0680, Loss2: 0.0689 +Epoch [84/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0555, Loss2: 0.0594 +Epoch [84/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0552, Loss2: 0.0531 +Epoch [84/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 48.4375, Loss1: 0.0500, Loss2: 0.0557 +Epoch [84/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0626, Loss2: 0.0573 +Epoch [84/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0496, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 49.2889 % Model2 50.3305 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0628, Loss2: 0.0631 +Epoch [85/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0601, Loss2: 0.0597 +Epoch [85/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0738, Loss2: 0.0692 +Epoch [85/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0570, Loss2: 0.0593 +Epoch [85/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0535, Loss2: 0.0586 +Epoch [85/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0524, Loss2: 0.0488 +Epoch [85/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0606, Loss2: 0.0590 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 48.9683 % Model2 48.6879 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0605, Loss2: 0.0588 +Epoch [86/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0639, Loss2: 0.0593 +Epoch [86/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0521 +Epoch [86/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0483, Loss2: 0.0462 +Epoch [86/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0597, Loss2: 0.0603 +Epoch [86/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0609, Loss2: 0.0648 +Epoch [86/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0555, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 49.4491 % Model2 51.1018 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0646, Loss2: 0.0643 +Epoch [87/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0550, Loss2: 0.0575 +Epoch [87/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0692, Loss2: 0.0691 +Epoch [87/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0697, Loss2: 0.0733 +Epoch [87/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0669, Loss2: 0.0658 +Epoch [87/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0729, Loss2: 0.0725 +Epoch [87/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0574, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 48.2772 % Model2 50.4207 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0636, Loss2: 0.0669 +Epoch [88/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0605, Loss2: 0.0607 +Epoch [88/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0511, Loss2: 0.0505 +Epoch [88/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0600, Loss2: 0.0549 +Epoch [88/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0499, Loss2: 0.0463 +Epoch [88/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0671, Loss2: 0.0663 +Epoch [88/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0697, Loss2: 0.0716 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 49.8498 % Model2 50.8313 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0729, Loss2: 0.0754 +Epoch [89/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0791, Loss2: 0.0770 +Epoch [89/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0509, Loss2: 0.0503 +Epoch [89/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0543, Loss2: 0.0518 +Epoch [89/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0628, Loss2: 0.0608 +Epoch [89/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0924, Loss2: 0.0886 +Epoch [89/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0604, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 48.4976 % Model2 49.4992 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0590, Loss2: 0.0588 +Epoch [90/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0600, Loss2: 0.0590 +Epoch [90/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0503, Loss2: 0.0498 +Epoch [90/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0608, Loss2: 0.0614 +Epoch [90/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0696, Loss2: 0.0662 +Epoch [90/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0608, Loss2: 0.0605 +Epoch [90/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0597, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 49.2288 % Model2 50.2404 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0567, Loss2: 0.0633 +Epoch [91/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0564, Loss2: 0.0557 +Epoch [91/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0699, Loss2: 0.0685 +Epoch [91/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0578, Loss2: 0.0557 +Epoch [91/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0838, Loss2: 0.0830 +Epoch [91/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0567, Loss2: 0.0548 +Epoch [91/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0582, Loss2: 0.0545 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 49.0585 % Model2 50.2003 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0531, Loss2: 0.0551 +Epoch [92/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0614, Loss2: 0.0621 +Epoch [92/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0644, Loss2: 0.0651 +Epoch [92/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0633, Loss2: 0.0615 +Epoch [92/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0587 +Epoch [92/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 48.4375, Loss1: 0.0451, Loss2: 0.0524 +Epoch [92/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0579, Loss2: 0.0570 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 49.0284 % Model2 50.4307 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0556, Loss2: 0.0556 +Epoch [93/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0705, Loss2: 0.0675 +Epoch [93/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0577, Loss2: 0.0603 +Epoch [93/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0659, Loss2: 0.0649 +Epoch [93/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0682, Loss2: 0.0697 +Epoch [93/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0667, Loss2: 0.0666 +Epoch [93/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0516, Loss2: 0.0527 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 48.8782 % Model2 51.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0642, Loss2: 0.0645 +Epoch [94/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0828, Loss2: 0.0784 +Epoch [94/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0656, Loss2: 0.0681 +Epoch [94/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0618, Loss2: 0.0643 +Epoch [94/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0694, Loss2: 0.0655 +Epoch [94/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0605, Loss2: 0.0604 +Epoch [94/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0630, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 47.4159 % Model2 50.8013 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0756, Loss2: 0.0744 +Epoch [95/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0561, Loss2: 0.0571 +Epoch [95/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0517, Loss2: 0.0540 +Epoch [95/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0717, Loss2: 0.0705 +Epoch [95/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0741 +Epoch [95/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0796, Loss2: 0.0763 +Epoch [95/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0747, Loss2: 0.0769 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 48.8081 % Model2 50.2103 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0678, Loss2: 0.0598 +Epoch [96/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0529, Loss2: 0.0560 +Epoch [96/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0523, Loss2: 0.0485 +Epoch [96/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0557, Loss2: 0.0536 +Epoch [96/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 48.4375, Loss1: 0.0536, Loss2: 0.0585 +Epoch [96/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0701, Loss2: 0.0658 +Epoch [96/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0584, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 48.2973 % Model2 48.8482 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 69.5312, Loss1: 0.0860, Loss2: 0.0795 +Epoch [97/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0765, Loss2: 0.0764 +Epoch [97/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.8125, Loss1: 0.0590, Loss2: 0.0645 +Epoch [97/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0629, Loss2: 0.0625 +Epoch [97/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0475, Loss2: 0.0472 +Epoch [97/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0697, Loss2: 0.0643 +Epoch [97/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0600, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 49.2188 % Model2 50.9315 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0565, Loss2: 0.0518 +Epoch [98/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0621, Loss2: 0.0652 +Epoch [98/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0723, Loss2: 0.0699 +Epoch [98/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0570, Loss2: 0.0562 +Epoch [98/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0639, Loss2: 0.0613 +Epoch [98/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 45.3125, Loss1: 0.0532, Loss2: 0.0579 +Epoch [98/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0598, Loss2: 0.0621 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 48.5577 % Model2 50.5008 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0709, Loss2: 0.0651 +Epoch [99/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0569, Loss2: 0.0584 +Epoch [99/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0604, Loss2: 0.0604 +Epoch [99/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0623, Loss2: 0.0637 +Epoch [99/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0607, Loss2: 0.0571 +Epoch [99/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0750, Loss2: 0.0777 +Epoch [99/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0615, Loss2: 0.0582 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 49.0585 % Model2 50.2404 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0665, Loss2: 0.0669 +Epoch [100/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0647, Loss2: 0.0706 +Epoch [100/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0886, Loss2: 0.0811 +Epoch [100/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0678, Loss2: 0.0682 +Epoch [100/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0679, Loss2: 0.0624 +Epoch [100/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0579, Loss2: 0.0598 +Epoch [100/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0655, Loss2: 0.0696 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 48.9283 % Model2 49.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0783, Loss2: 0.0737 +Epoch [101/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0491, Loss2: 0.0473 +Epoch [101/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0673, Loss2: 0.0686 +Epoch [101/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0648, Loss2: 0.0651 +Epoch [101/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0567, Loss2: 0.0578 +Epoch [101/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0832, Loss2: 0.0739 +Epoch [101/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0509, Loss2: 0.0505 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 48.7179 % Model2 49.7997 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0546, Loss2: 0.0543 +Epoch [102/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0605, Loss2: 0.0566 +Epoch [102/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0755, Loss2: 0.0754 +Epoch [102/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0758, Loss2: 0.0740 +Epoch [102/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0641, Loss2: 0.0588 +Epoch [102/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0485, Loss2: 0.0487 +Epoch [102/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0733, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 48.5577 % Model2 50.1002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0694, Loss2: 0.0723 +Epoch [103/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0555, Loss2: 0.0522 +Epoch [103/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0634, Loss2: 0.0634 +Epoch [103/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0688, Loss2: 0.0672 +Epoch [103/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0664, Loss2: 0.0676 +Epoch [103/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0747, Loss2: 0.0682 +Epoch [103/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0592, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 47.8466 % Model2 50.1603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0652, Loss2: 0.0622 +Epoch [104/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0981, Loss2: 0.0943 +Epoch [104/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0548, Loss2: 0.0521 +Epoch [104/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 57.8125, Loss1: 0.0611, Loss2: 0.0534 +Epoch [104/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0759, Loss2: 0.0728 +Epoch [104/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0679, Loss2: 0.0634 +Epoch [104/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0560, Loss2: 0.0561 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 48.8682 % Model2 50.2003 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0785, Loss2: 0.0777 +Epoch [105/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.1046, Loss2: 0.1099 +Epoch [105/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0760, Loss2: 0.0753 +Epoch [105/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0543, Loss2: 0.0542 +Epoch [105/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0652, Loss2: 0.0622 +Epoch [105/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0681, Loss2: 0.0646 +Epoch [105/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0554, Loss2: 0.0600 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 48.6579 % Model2 50.4908 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0514, Loss2: 0.0513 +Epoch [106/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0626, Loss2: 0.0625 +Epoch [106/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0761, Loss2: 0.0747 +Epoch [106/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0674, Loss2: 0.0659 +Epoch [106/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0541, Loss2: 0.0562 +Epoch [106/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0587 +Epoch [106/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 52.3438, Loss1: 0.0601, Loss2: 0.0692 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 49.1987 % Model2 49.1587 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0620, Loss2: 0.0626 +Epoch [107/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0702, Loss2: 0.0666 +Epoch [107/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0908, Loss2: 0.0853 +Epoch [107/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0796, Loss2: 0.0746 +Epoch [107/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0705, Loss2: 0.0652 +Epoch [107/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0527, Loss2: 0.0534 +Epoch [107/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0834, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 49.0685 % Model2 50.4908 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1010, Loss2: 0.0973 +Epoch [108/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0659, Loss2: 0.0678 +Epoch [108/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0810, Loss2: 0.0738 +Epoch [108/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0540, Loss2: 0.0525 +Epoch [108/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0962, Loss2: 0.1013 +Epoch [108/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0819, Loss2: 0.0787 +Epoch [108/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0873, Loss2: 0.0824 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 48.8682 % Model2 49.5693 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0621, Loss2: 0.0603 +Epoch [109/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0622, Loss2: 0.0609 +Epoch [109/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0674, Loss2: 0.0662 +Epoch [109/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0742, Loss2: 0.0754 +Epoch [109/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0576, Loss2: 0.0597 +Epoch [109/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0620, Loss2: 0.0602 +Epoch [109/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0861, Loss2: 0.0892 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 48.2472 % Model2 49.6995 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0644, Loss2: 0.0705 +Epoch [110/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0616, Loss2: 0.0640 +Epoch [110/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0665, Loss2: 0.0662 +Epoch [110/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 56.2500, Loss1: 0.0600, Loss2: 0.0658 +Epoch [110/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0855, Loss2: 0.0841 +Epoch [110/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0792, Loss2: 0.0768 +Epoch [110/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0610, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 49.0585 % Model2 50.4407 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0545, Loss2: 0.0519 +Epoch [111/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0608, Loss2: 0.0579 +Epoch [111/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0638, Loss2: 0.0637 +Epoch [111/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0552, Loss2: 0.0579 +Epoch [111/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 59.3750, Loss1: 0.0683, Loss2: 0.0603 +Epoch [111/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0618, Loss2: 0.0645 +Epoch [111/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0778, Loss2: 0.0787 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 49.7396 % Model2 50.3606 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0832, Loss2: 0.0776 +Epoch [112/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0709, Loss2: 0.0680 +Epoch [112/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0802, Loss2: 0.0801 +Epoch [112/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0669, Loss2: 0.0647 +Epoch [112/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0863, Loss2: 0.0784 +Epoch [112/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0598, Loss2: 0.0623 +Epoch [112/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0601, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 48.5276 % Model2 49.7997 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0769, Loss2: 0.0766 +Epoch [113/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0750, Loss2: 0.0735 +Epoch [113/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0843, Loss2: 0.0848 +Epoch [113/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0874, Loss2: 0.0871 +Epoch [113/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 61.7188, Loss1: 0.0686, Loss2: 0.0791 +Epoch [113/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0766, Loss2: 0.0749 +Epoch [113/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0758, Loss2: 0.0728 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 47.9968 % Model2 49.1286 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0770, Loss2: 0.0737 +Epoch [114/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0574, Loss2: 0.0549 +Epoch [114/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0604, Loss2: 0.0655 +Epoch [114/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0647, Loss2: 0.0632 +Epoch [114/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0672, Loss2: 0.0619 +Epoch [114/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0674, Loss2: 0.0673 +Epoch [114/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0629, Loss2: 0.0618 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 48.5076 % Model2 49.6194 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0726, Loss2: 0.0761 +Epoch [115/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0810, Loss2: 0.0758 +Epoch [115/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 51.5625, Loss1: 0.0643, Loss2: 0.0719 +Epoch [115/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 55.4688, Loss1: 0.0523, Loss2: 0.0574 +Epoch [115/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0562, Loss2: 0.0599 +Epoch [115/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0657, Loss2: 0.0624 +Epoch [115/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0623, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 48.7780 % Model2 49.6695 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0828, Loss2: 0.0820 +Epoch [116/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0691, Loss2: 0.0716 +Epoch [116/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0745, Loss2: 0.0773 +Epoch [116/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0709, Loss2: 0.0648 +Epoch [116/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0697, Loss2: 0.0718 +Epoch [116/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 57.0312, Loss1: 0.0632, Loss2: 0.0736 +Epoch [116/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0637, Loss2: 0.0617 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 48.6779 % Model2 49.2188 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0871, Loss2: 0.0861 +Epoch [117/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0763, Loss2: 0.0673 +Epoch [117/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0738, Loss2: 0.0674 +Epoch [117/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0693, Loss2: 0.0698 +Epoch [117/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0595, Loss2: 0.0620 +Epoch [117/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0624, Loss2: 0.0583 +Epoch [117/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0678, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 48.3273 % Model2 49.5893 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0725, Loss2: 0.0823 +Epoch [118/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0757, Loss2: 0.0756 +Epoch [118/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0601, Loss2: 0.0571 +Epoch [118/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0582, Loss2: 0.0578 +Epoch [118/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0796, Loss2: 0.0749 +Epoch [118/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0624, Loss2: 0.0651 +Epoch [118/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0555, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 48.2071 % Model2 50.0501 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0679, Loss2: 0.0678 +Epoch [119/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0833, Loss2: 0.0867 +Epoch [119/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0787, Loss2: 0.0836 +Epoch [119/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0673, Loss2: 0.0620 +Epoch [119/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0597, Loss2: 0.0593 +Epoch [119/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0571 +Epoch [119/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0792, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 48.1671 % Model2 49.0385 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0739, Loss2: 0.0695 +Epoch [120/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0780, Loss2: 0.0765 +Epoch [120/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0860, Loss2: 0.0837 +Epoch [120/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0658, Loss2: 0.0653 +Epoch [120/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0571, Loss2: 0.0556 +Epoch [120/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0930, Loss2: 0.0940 +Epoch [120/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0810, Loss2: 0.0775 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 47.8365 % Model2 49.4291 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0736, Loss2: 0.0733 +Epoch [121/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0613, Loss2: 0.0625 +Epoch [121/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0579, Loss2: 0.0588 +Epoch [121/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0820, Loss2: 0.0867 +Epoch [121/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0777, Loss2: 0.0757 +Epoch [121/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0627, Loss2: 0.0616 +Epoch [121/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0655, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 47.7464 % Model2 49.5292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0757, Loss2: 0.0735 +Epoch [122/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0749, Loss2: 0.0783 +Epoch [122/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0686, Loss2: 0.0648 +Epoch [122/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0754, Loss2: 0.0792 +Epoch [122/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0573, Loss2: 0.0529 +Epoch [122/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0763, Loss2: 0.0763 +Epoch [122/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.0938, Loss1: 0.1000, Loss2: 0.0865 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 48.5076 % Model2 49.8498 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0586, Loss2: 0.0575 +Epoch [123/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 55.4688, Loss1: 0.0618, Loss2: 0.0693 +Epoch [123/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0764, Loss2: 0.0762 +Epoch [123/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0744, Loss2: 0.0676 +Epoch [123/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0700, Loss2: 0.0654 +Epoch [123/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0622, Loss2: 0.0641 +Epoch [123/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0695, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 48.3974 % Model2 49.1186 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0764, Loss2: 0.0783 +Epoch [124/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0607, Loss2: 0.0595 +Epoch [124/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0788, Loss2: 0.0810 +Epoch [124/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0778, Loss2: 0.0739 +Epoch [124/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0507, Loss2: 0.0488 +Epoch [124/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0716, Loss2: 0.0762 +Epoch [124/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0728, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 47.8666 % Model2 49.7496 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0711, Loss2: 0.0725 +Epoch [125/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0638, Loss2: 0.0675 +Epoch [125/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0588, Loss2: 0.0614 +Epoch [125/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0503, Loss2: 0.0533 +Epoch [125/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0687, Loss2: 0.0677 +Epoch [125/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0636, Loss2: 0.0604 +Epoch [125/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0757, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 48.6178 % Model2 49.7796 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0694, Loss2: 0.0652 +Epoch [126/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0816, Loss2: 0.0836 +Epoch [126/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 64.0625, Loss1: 0.0751, Loss2: 0.0661 +Epoch [126/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0791 +Epoch [126/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0575, Loss2: 0.0552 +Epoch [126/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0622 +Epoch [126/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0497, Loss2: 0.0482 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 48.7580 % Model2 49.5994 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0766, Loss2: 0.0755 +Epoch [127/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0665, Loss2: 0.0681 +Epoch [127/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0695, Loss2: 0.0621 +Epoch [127/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0781, Loss2: 0.0801 +Epoch [127/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1077, Loss2: 0.1051 +Epoch [127/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0835, Loss2: 0.0811 +Epoch [127/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0621, Loss2: 0.0632 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 48.0869 % Model2 49.0485 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0559, Loss2: 0.0547 +Epoch [128/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0736, Loss2: 0.0663 +Epoch [128/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0806, Loss2: 0.0869 +Epoch [128/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0667, Loss2: 0.0639 +Epoch [128/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0716, Loss2: 0.0718 +Epoch [128/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0908, Loss2: 0.0891 +Epoch [128/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0585, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 48.1771 % Model2 49.1587 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0714, Loss2: 0.0654 +Epoch [129/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0806, Loss2: 0.0790 +Epoch [129/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0732, Loss2: 0.0745 +Epoch [129/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1025, Loss2: 0.1055 +Epoch [129/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0926, Loss2: 0.0888 +Epoch [129/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0831, Loss2: 0.0861 +Epoch [129/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 55.4688, Loss1: 0.0612, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 48.2973 % Model2 49.3089 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0739, Loss2: 0.0694 +Epoch [130/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0692, Loss2: 0.0669 +Epoch [130/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0859, Loss2: 0.0930 +Epoch [130/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0767, Loss2: 0.0781 +Epoch [130/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0620, Loss2: 0.0589 +Epoch [130/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0769, Loss2: 0.0743 +Epoch [130/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0826, Loss2: 0.0863 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 48.1771 % Model2 50.6510 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0704, Loss2: 0.0721 +Epoch [131/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0622, Loss2: 0.0595 +Epoch [131/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0752, Loss2: 0.0747 +Epoch [131/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0797, Loss2: 0.0766 +Epoch [131/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0716, Loss2: 0.0770 +Epoch [131/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0635, Loss2: 0.0651 +Epoch [131/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.0902, Loss2: 0.0994 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 48.3874 % Model2 49.3289 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0934, Loss2: 0.0966 +Epoch [132/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0838, Loss2: 0.0790 +Epoch [132/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0814, Loss2: 0.0767 +Epoch [132/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0593, Loss2: 0.0629 +Epoch [132/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0868, Loss2: 0.0901 +Epoch [132/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0757, Loss2: 0.0786 +Epoch [132/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0814, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 48.6178 % Model2 49.6895 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 57.8125, Loss1: 0.0857, Loss2: 0.0965 +Epoch [133/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0891, Loss2: 0.0896 +Epoch [133/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0891, Loss2: 0.0867 +Epoch [133/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0676, Loss2: 0.0724 +Epoch [133/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0849, Loss2: 0.0819 +Epoch [133/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 56.2500, Loss1: 0.0748, Loss2: 0.0847 +Epoch [133/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0857, Loss2: 0.0817 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 47.8466 % Model2 49.4792 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0601, Loss2: 0.0589 +Epoch [134/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0990, Loss2: 0.0907 +Epoch [134/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0802, Loss2: 0.0788 +Epoch [134/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0690, Loss2: 0.0694 +Epoch [134/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0821, Loss2: 0.0748 +Epoch [134/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0916, Loss2: 0.0912 +Epoch [134/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0706, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 47.9367 % Model2 49.4491 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0607, Loss2: 0.0596 +Epoch [135/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 46.8750, Loss1: 0.0612, Loss2: 0.0693 +Epoch [135/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0780, Loss2: 0.0788 +Epoch [135/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 66.4062, Loss1: 0.0725, Loss2: 0.0637 +Epoch [135/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0875, Loss2: 0.0840 +Epoch [135/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.1098, Loss2: 0.1044 +Epoch [135/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0944, Loss2: 0.0896 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 47.3958 % Model2 49.5593 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0823, Loss2: 0.0756 +Epoch [136/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.0956, Loss2: 0.0910 +Epoch [136/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0502, Loss2: 0.0511 +Epoch [136/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0645, Loss2: 0.0627 +Epoch [136/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0648, Loss2: 0.0649 +Epoch [136/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.8750, Loss1: 0.0796, Loss2: 0.0708 +Epoch [136/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0655, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 47.2356 % Model2 49.5593 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0755, Loss2: 0.0750 +Epoch [137/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0949, Loss2: 0.0986 +Epoch [137/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0745, Loss2: 0.0724 +Epoch [137/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0706, Loss2: 0.0689 +Epoch [137/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0755, Loss2: 0.0739 +Epoch [137/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0791, Loss2: 0.0785 +Epoch [137/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0676, Loss2: 0.0639 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 48.2071 % Model2 49.1086 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.1059, Loss2: 0.0998 +Epoch [138/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0970, Loss2: 0.0993 +Epoch [138/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0837, Loss2: 0.0813 +Epoch [138/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0721, Loss2: 0.0805 +Epoch [138/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0825, Loss2: 0.0784 +Epoch [138/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0655, Loss2: 0.0655 +Epoch [138/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0790, Loss2: 0.0855 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 47.7063 % Model2 49.4992 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0637, Loss2: 0.0672 +Epoch [139/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0829, Loss2: 0.0832 +Epoch [139/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0475, Loss2: 0.0495 +Epoch [139/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0876, Loss2: 0.0841 +Epoch [139/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0949, Loss2: 0.0934 +Epoch [139/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0698, Loss2: 0.0721 +Epoch [139/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0583, Loss2: 0.0593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 48.0168 % Model2 49.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0618, Loss2: 0.0576 +Epoch [140/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0766, Loss2: 0.0772 +Epoch [140/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0694, Loss2: 0.0691 +Epoch [140/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0951, Loss2: 0.0944 +Epoch [140/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0782, Loss2: 0.0776 +Epoch [140/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0793, Loss2: 0.0846 +Epoch [140/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0752, Loss2: 0.0806 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 47.4659 % Model2 49.7696 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0715, Loss2: 0.0693 +Epoch [141/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0707, Loss2: 0.0713 +Epoch [141/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0689, Loss2: 0.0735 +Epoch [141/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0618, Loss2: 0.0633 +Epoch [141/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0748, Loss2: 0.0750 +Epoch [141/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0914, Loss2: 0.0898 +Epoch [141/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0574, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 47.3958 % Model2 49.1086 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0721, Loss2: 0.0739 +Epoch [142/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0913, Loss2: 0.0848 +Epoch [142/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0839, Loss2: 0.0930 +Epoch [142/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0920, Loss2: 0.1013 +Epoch [142/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0831, Loss2: 0.0861 +Epoch [142/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0760, Loss2: 0.0882 +Epoch [142/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0710, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 47.8365 % Model2 48.7480 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0927, Loss2: 0.0920 +Epoch [143/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0701, Loss2: 0.0733 +Epoch [143/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0914, Loss2: 0.0911 +Epoch [143/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0989, Loss2: 0.0950 +Epoch [143/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0768, Loss2: 0.0744 +Epoch [143/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0806, Loss2: 0.0831 +Epoch [143/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0781, Loss2: 0.0756 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 47.8966 % Model2 49.1486 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0733, Loss2: 0.0731 +Epoch [144/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0820 +Epoch [144/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0871, Loss2: 0.0805 +Epoch [144/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0618, Loss2: 0.0650 +Epoch [144/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0633, Loss2: 0.0594 +Epoch [144/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0635, Loss2: 0.0584 +Epoch [144/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.1021, Loss2: 0.0970 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 48.0469 % Model2 49.2288 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0628, Loss2: 0.0596 +Epoch [145/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0664, Loss2: 0.0713 +Epoch [145/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0936, Loss2: 0.0861 +Epoch [145/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0784, Loss2: 0.0845 +Epoch [145/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0804, Loss2: 0.0759 +Epoch [145/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0874, Loss2: 0.0860 +Epoch [145/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0695, Loss2: 0.0687 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 47.7464 % Model2 49.1186 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0986, Loss2: 0.0919 +Epoch [146/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0931, Loss2: 0.0910 +Epoch [146/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0647, Loss2: 0.0658 +Epoch [146/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0618, Loss2: 0.0591 +Epoch [146/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0930, Loss2: 0.0897 +Epoch [146/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0760, Loss2: 0.0759 +Epoch [146/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.9688, Loss1: 0.0623, Loss2: 0.0562 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 47.3758 % Model2 49.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0788, Loss2: 0.0839 +Epoch [147/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0859, Loss2: 0.0839 +Epoch [147/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0594, Loss2: 0.0618 +Epoch [147/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1104, Loss2: 0.1062 +Epoch [147/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1210, Loss2: 0.1154 +Epoch [147/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0753, Loss2: 0.0707 +Epoch [147/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0896, Loss2: 0.0903 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 47.4659 % Model2 49.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0716, Loss2: 0.0660 +Epoch [148/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0903, Loss2: 0.0868 +Epoch [148/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 68.7500, Loss1: 0.0751, Loss2: 0.0659 +Epoch [148/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.1097, Loss2: 0.1028 +Epoch [148/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1246, Loss2: 0.1165 +Epoch [148/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0713, Loss2: 0.0699 +Epoch [148/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0862, Loss2: 0.0874 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 46.4844 % Model2 49.2087 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0755, Loss2: 0.0773 +Epoch [149/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0772, Loss2: 0.0732 +Epoch [149/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0828, Loss2: 0.0775 +Epoch [149/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0785, Loss2: 0.0811 +Epoch [149/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0925, Loss2: 0.0869 +Epoch [149/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0710, Loss2: 0.0631 +Epoch [149/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0689 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 47.6763 % Model2 49.2388 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0810, Loss2: 0.0722 +Epoch [150/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0814, Loss2: 0.0736 +Epoch [150/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0848, Loss2: 0.0818 +Epoch [150/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0833, Loss2: 0.0851 +Epoch [150/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0756, Loss2: 0.0783 +Epoch [150/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.1432, Loss2: 0.1275 +Epoch [150/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0612, Loss2: 0.0632 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 48.1170 % Model2 49.2588 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0785, Loss2: 0.0795 +Epoch [151/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0777, Loss2: 0.0736 +Epoch [151/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0924, Loss2: 0.1010 +Epoch [151/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.1129, Loss2: 0.1191 +Epoch [151/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0634, Loss2: 0.0631 +Epoch [151/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0762, Loss2: 0.0796 +Epoch [151/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0599, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 47.8866 % Model2 49.0485 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1008, Loss2: 0.1093 +Epoch [152/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0962, Loss2: 0.1010 +Epoch [152/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1043, Loss2: 0.1031 +Epoch [152/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0660, Loss2: 0.0641 +Epoch [152/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0819, Loss2: 0.0792 +Epoch [152/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0611, Loss2: 0.0607 +Epoch [152/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0811, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 47.7264 % Model2 49.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1009, Loss2: 0.0979 +Epoch [153/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0788, Loss2: 0.0801 +Epoch [153/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0811, Loss2: 0.0897 +Epoch [153/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0726, Loss2: 0.0712 +Epoch [153/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0763, Loss2: 0.0734 +Epoch [153/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0639, Loss2: 0.0587 +Epoch [153/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0880, Loss2: 0.0940 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 47.5661 % Model2 48.6979 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0871, Loss2: 0.0814 +Epoch [154/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0793, Loss2: 0.0837 +Epoch [154/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0786, Loss2: 0.0782 +Epoch [154/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0936, Loss2: 0.0953 +Epoch [154/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0718, Loss2: 0.0765 +Epoch [154/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0750, Loss2: 0.0779 +Epoch [154/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0828, Loss2: 0.0784 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 47.4359 % Model2 48.4375 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.0000, Loss1: 0.1144, Loss2: 0.0955 +Epoch [155/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0907, Loss2: 0.0861 +Epoch [155/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0749, Loss2: 0.0679 +Epoch [155/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0716, Loss2: 0.0747 +Epoch [155/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0720, Loss2: 0.0640 +Epoch [155/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0762, Loss2: 0.0748 +Epoch [155/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0959, Loss2: 0.0890 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 47.8165 % Model2 48.6478 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0635, Loss2: 0.0611 +Epoch [156/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.1099, Loss2: 0.1113 +Epoch [156/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0778, Loss2: 0.0719 +Epoch [156/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0796, Loss2: 0.0762 +Epoch [156/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0630, Loss2: 0.0643 +Epoch [156/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0727, Loss2: 0.0745 +Epoch [156/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0919, Loss2: 0.0943 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 47.4359 % Model2 49.4391 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0759, Loss2: 0.0789 +Epoch [157/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0843, Loss2: 0.0779 +Epoch [157/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0899, Loss2: 0.0885 +Epoch [157/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.0625, Loss1: 0.0685, Loss2: 0.0708 +Epoch [157/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0823, Loss2: 0.0789 +Epoch [157/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.1044, Loss2: 0.0920 +Epoch [157/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0737, Loss2: 0.0702 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 47.9667 % Model2 49.2989 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0637, Loss2: 0.0656 +Epoch [158/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0850, Loss2: 0.0823 +Epoch [158/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0965, Loss2: 0.0986 +Epoch [158/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0647, Loss2: 0.0659 +Epoch [158/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.1209, Loss2: 0.1125 +Epoch [158/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1069, Loss2: 0.1073 +Epoch [158/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0982, Loss2: 0.0948 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 47.7664 % Model2 48.9583 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0842, Loss2: 0.0925 +Epoch [159/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0972, Loss2: 0.0905 +Epoch [159/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0797, Loss2: 0.0770 +Epoch [159/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0789, Loss2: 0.0731 +Epoch [159/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0918, Loss2: 0.0829 +Epoch [159/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0695, Loss2: 0.0672 +Epoch [159/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0749, Loss2: 0.0720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 47.4960 % Model2 49.0284 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0892, Loss2: 0.0934 +Epoch [160/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0989, Loss2: 0.1051 +Epoch [160/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0746, Loss2: 0.0766 +Epoch [160/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0756, Loss2: 0.0778 +Epoch [160/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.1224, Loss2: 0.1163 +Epoch [160/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0822, Loss2: 0.0774 +Epoch [160/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0793, Loss2: 0.0756 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 47.2456 % Model2 49.3089 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0835, Loss2: 0.0773 +Epoch [161/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0724, Loss2: 0.0686 +Epoch [161/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0964, Loss2: 0.0938 +Epoch [161/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0702, Loss2: 0.0729 +Epoch [161/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0794, Loss2: 0.0718 +Epoch [161/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0719, Loss2: 0.0756 +Epoch [161/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0822, Loss2: 0.0866 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 47.3758 % Model2 49.0986 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0776, Loss2: 0.0819 +Epoch [162/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0797, Loss2: 0.0748 +Epoch [162/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0685, Loss2: 0.0658 +Epoch [162/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0865, Loss2: 0.0924 +Epoch [162/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0837, Loss2: 0.0815 +Epoch [162/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0641, Loss2: 0.0678 +Epoch [162/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0783, Loss2: 0.0834 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 47.6162 % Model2 49.3089 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0746, Loss2: 0.0768 +Epoch [163/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 67.9688, Loss1: 0.1140, Loss2: 0.0902 +Epoch [163/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0740, Loss2: 0.0709 +Epoch [163/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0918, Loss2: 0.1029 +Epoch [163/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.1148, Loss2: 0.1015 +Epoch [163/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0847, Loss2: 0.0845 +Epoch [163/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0787, Loss2: 0.0847 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 47.2055 % Model2 49.2588 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0761, Loss2: 0.0789 +Epoch [164/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1205, Loss2: 0.1200 +Epoch [164/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0579, Loss2: 0.0600 +Epoch [164/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0772, Loss2: 0.0804 +Epoch [164/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0838, Loss2: 0.0840 +Epoch [164/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1103, Loss2: 0.1009 +Epoch [164/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0893, Loss2: 0.0942 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 47.4459 % Model2 49.3690 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0667, Loss2: 0.0697 +Epoch [165/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0825, Loss2: 0.0785 +Epoch [165/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0846, Loss2: 0.0840 +Epoch [165/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0673, Loss2: 0.0736 +Epoch [165/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0988, Loss2: 0.0994 +Epoch [165/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.0968, Loss2: 0.1036 +Epoch [165/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0810, Loss2: 0.0739 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 47.0853 % Model2 49.0685 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0983, Loss2: 0.0940 +Epoch [166/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0688, Loss2: 0.0695 +Epoch [166/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0761, Loss2: 0.0756 +Epoch [166/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0898, Loss2: 0.1018 +Epoch [166/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0991, Loss2: 0.0991 +Epoch [166/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0833, Loss2: 0.0820 +Epoch [166/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0842, Loss2: 0.0862 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 47.7163 % Model2 48.6779 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0722, Loss2: 0.0715 +Epoch [167/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0928, Loss2: 0.0892 +Epoch [167/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0726, Loss2: 0.0664 +Epoch [167/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0904, Loss2: 0.0890 +Epoch [167/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0831, Loss2: 0.0794 +Epoch [167/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0669, Loss2: 0.0619 +Epoch [167/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0624, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 47.1454 % Model2 49.1486 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0790, Loss2: 0.0721 +Epoch [168/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0936, Loss2: 0.0929 +Epoch [168/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0690, Loss2: 0.0666 +Epoch [168/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0686, Loss2: 0.0720 +Epoch [168/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 58.5938, Loss1: 0.0774, Loss2: 0.0858 +Epoch [168/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1046, Loss2: 0.0990 +Epoch [168/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0889, Loss2: 0.0864 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 47.4159 % Model2 49.1787 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0947, Loss2: 0.0876 +Epoch [169/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.1062, Loss2: 0.0973 +Epoch [169/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1030, Loss2: 0.0997 +Epoch [169/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0951, Loss2: 0.0913 +Epoch [169/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0928, Loss2: 0.0963 +Epoch [169/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0850, Loss2: 0.0861 +Epoch [169/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0825, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 46.9852 % Model2 48.4876 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.1303, Loss2: 0.1373 +Epoch [170/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0995, Loss2: 0.1077 +Epoch [170/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0717, Loss2: 0.0729 +Epoch [170/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0928, Loss2: 0.0920 +Epoch [170/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0848, Loss2: 0.0781 +Epoch [170/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0987, Loss2: 0.0947 +Epoch [170/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0645, Loss2: 0.0645 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 46.7348 % Model2 48.7780 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1000, Loss2: 0.1021 +Epoch [171/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0961, Loss2: 0.0914 +Epoch [171/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0755, Loss2: 0.0761 +Epoch [171/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0843, Loss2: 0.0895 +Epoch [171/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0891, Loss2: 0.0979 +Epoch [171/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0811, Loss2: 0.0853 +Epoch [171/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0936, Loss2: 0.0881 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 47.2656 % Model2 49.0385 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0928, Loss2: 0.0868 +Epoch [172/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1043, Loss2: 0.1068 +Epoch [172/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0756, Loss2: 0.0735 +Epoch [172/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0893, Loss2: 0.0838 +Epoch [172/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0744, Loss2: 0.0766 +Epoch [172/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 69.5312, Loss1: 0.0841, Loss2: 0.0715 +Epoch [172/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0675, Loss2: 0.0661 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 47.1154 % Model2 48.9183 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0885, Loss2: 0.0863 +Epoch [173/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1190, Loss2: 0.1125 +Epoch [173/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1047, Loss2: 0.1098 +Epoch [173/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1077, Loss2: 0.1097 +Epoch [173/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1269, Loss2: 0.1268 +Epoch [173/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1009, Loss2: 0.0995 +Epoch [173/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0942, Loss2: 0.0934 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 47.1755 % Model2 49.0284 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1125, Loss2: 0.1162 +Epoch [174/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0718, Loss2: 0.0705 +Epoch [174/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0869, Loss2: 0.0912 +Epoch [174/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0597, Loss2: 0.0658 +Epoch [174/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0816, Loss2: 0.0837 +Epoch [174/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1151, Loss2: 0.1190 +Epoch [174/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0788, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 47.2356 % Model2 49.2588 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.1024, Loss2: 0.0971 +Epoch [175/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0769, Loss2: 0.0819 +Epoch [175/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0944, Loss2: 0.0895 +Epoch [175/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0782, Loss2: 0.0753 +Epoch [175/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0943, Loss2: 0.0916 +Epoch [175/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0932, Loss2: 0.0968 +Epoch [175/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0791, Loss2: 0.0798 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 46.5845 % Model2 49.0885 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0932, Loss2: 0.1029 +Epoch [176/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0786, Loss2: 0.0773 +Epoch [176/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0944, Loss2: 0.0892 +Epoch [176/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0951, Loss2: 0.0938 +Epoch [176/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0854, Loss2: 0.0849 +Epoch [176/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0928, Loss2: 0.0898 +Epoch [176/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0746, Loss2: 0.0739 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 46.6647 % Model2 48.7780 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0783, Loss2: 0.0785 +Epoch [177/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0903, Loss2: 0.0954 +Epoch [177/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0812, Loss2: 0.0738 +Epoch [177/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1152, Loss2: 0.1064 +Epoch [177/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0908, Loss2: 0.0794 +Epoch [177/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1039, Loss2: 0.1125 +Epoch [177/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0760, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 47.0753 % Model2 48.7179 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0904, Loss2: 0.0826 +Epoch [178/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0903, Loss2: 0.0867 +Epoch [178/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0737, Loss2: 0.0714 +Epoch [178/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0812, Loss2: 0.0826 +Epoch [178/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0914, Loss2: 0.0881 +Epoch [178/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.1030, Loss2: 0.1191 +Epoch [178/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0930, Loss2: 0.0942 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 47.0954 % Model2 49.1386 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0739, Loss2: 0.0754 +Epoch [179/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.1010, Loss2: 0.1040 +Epoch [179/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0805, Loss2: 0.0764 +Epoch [179/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.1220, Loss2: 0.1049 +Epoch [179/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0998, Loss2: 0.1105 +Epoch [179/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1098, Loss2: 0.1102 +Epoch [179/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.1002, Loss2: 0.1036 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 46.8750 % Model2 48.9283 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0894, Loss2: 0.0968 +Epoch [180/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0810, Loss2: 0.0885 +Epoch [180/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0928, Loss2: 0.0949 +Epoch [180/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0750, Loss2: 0.0759 +Epoch [180/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0916, Loss2: 0.0842 +Epoch [180/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 65.6250, Loss1: 0.0725, Loss2: 0.0784 +Epoch [180/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1258, Loss2: 0.1214 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 47.0653 % Model2 48.8181 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0750, Loss2: 0.0765 +Epoch [181/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1098, Loss2: 0.1075 +Epoch [181/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0839, Loss2: 0.0785 +Epoch [181/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0934, Loss2: 0.1026 +Epoch [181/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.0906, Loss2: 0.0822 +Epoch [181/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0678, Loss2: 0.0692 +Epoch [181/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0680, Loss2: 0.0613 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 46.8149 % Model2 48.5076 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1070, Loss2: 0.1029 +Epoch [182/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 76.5625, Loss1: 0.1313, Loss2: 0.1077 +Epoch [182/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0934, Loss2: 0.1006 +Epoch [182/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0885, Loss2: 0.0862 +Epoch [182/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0713, Loss2: 0.0728 +Epoch [182/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0841, Loss2: 0.0833 +Epoch [182/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0733, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 46.7147 % Model2 48.9784 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1005, Loss2: 0.0952 +Epoch [183/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0731, Loss2: 0.0696 +Epoch [183/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0749, Loss2: 0.0815 +Epoch [183/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0993, Loss2: 0.1095 +Epoch [183/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0907, Loss2: 0.0791 +Epoch [183/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1135, Loss2: 0.1053 +Epoch [183/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0830, Loss2: 0.0884 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 46.7949 % Model2 48.6579 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0909, Loss2: 0.0952 +Epoch [184/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0925, Loss2: 0.0904 +Epoch [184/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0912, Loss2: 0.0946 +Epoch [184/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1124, Loss2: 0.1136 +Epoch [184/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0811, Loss2: 0.0812 +Epoch [184/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.1065, Loss2: 0.1116 +Epoch [184/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0775, Loss2: 0.0784 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 46.8450 % Model2 48.7179 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0925, Loss2: 0.0931 +Epoch [185/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0729, Loss2: 0.0658 +Epoch [185/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0916, Loss2: 0.0987 +Epoch [185/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0918, Loss2: 0.0947 +Epoch [185/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.1127, Loss2: 0.1050 +Epoch [185/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0926, Loss2: 0.0848 +Epoch [185/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0813, Loss2: 0.0745 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 46.7548 % Model2 48.6178 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1114, Loss2: 0.1050 +Epoch [186/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1542, Loss2: 0.1558 +Epoch [186/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1018, Loss2: 0.0948 +Epoch [186/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1085, Loss2: 0.1177 +Epoch [186/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0810, Loss2: 0.0850 +Epoch [186/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1017, Loss2: 0.0916 +Epoch [186/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1041, Loss2: 0.0991 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 46.4944 % Model2 48.4976 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0963, Loss2: 0.0969 +Epoch [187/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1009, Loss2: 0.0993 +Epoch [187/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0881, Loss2: 0.0946 +Epoch [187/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0967, Loss2: 0.0968 +Epoch [187/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.1434, Loss2: 0.1334 +Epoch [187/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0948, Loss2: 0.0882 +Epoch [187/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1019, Loss2: 0.1063 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 46.5244 % Model2 48.6478 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0872, Loss2: 0.0740 +Epoch [188/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1135, Loss2: 0.1029 +Epoch [188/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.1067, Loss2: 0.1004 +Epoch [188/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0810, Loss2: 0.0808 +Epoch [188/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0779, Loss2: 0.0725 +Epoch [188/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0737, Loss2: 0.0825 +Epoch [188/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0793, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 46.6046 % Model2 48.6979 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0836, Loss2: 0.0806 +Epoch [189/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0942, Loss2: 0.0957 +Epoch [189/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0912, Loss2: 0.0890 +Epoch [189/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0935, Loss2: 0.1005 +Epoch [189/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0921, Loss2: 0.0885 +Epoch [189/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0745, Loss2: 0.0780 +Epoch [189/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0766, Loss2: 0.0785 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 46.7248 % Model2 48.7179 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1255, Loss2: 0.1147 +Epoch [190/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1426, Loss2: 0.1410 +Epoch [190/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0988, Loss2: 0.0954 +Epoch [190/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0836, Loss2: 0.0798 +Epoch [190/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0982, Loss2: 0.0881 +Epoch [190/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0897, Loss2: 0.0915 +Epoch [190/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.1050, Loss2: 0.1123 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 46.6947 % Model2 48.6879 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0866, Loss2: 0.0981 +Epoch [191/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0685, Loss2: 0.0687 +Epoch [191/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1073, Loss2: 0.1075 +Epoch [191/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0898, Loss2: 0.0853 +Epoch [191/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0959, Loss2: 0.0865 +Epoch [191/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0762, Loss2: 0.0752 +Epoch [191/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0879, Loss2: 0.0838 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 46.4944 % Model2 48.8081 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1174, Loss2: 0.1151 +Epoch [192/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0784, Loss2: 0.0890 +Epoch [192/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1129, Loss2: 0.1101 +Epoch [192/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.1159, Loss2: 0.1030 +Epoch [192/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0875, Loss2: 0.0814 +Epoch [192/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0884, Loss2: 0.0854 +Epoch [192/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0854, Loss2: 0.0805 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 46.5946 % Model2 48.7780 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0973, Loss2: 0.0987 +Epoch [193/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0847, Loss2: 0.0862 +Epoch [193/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0802, Loss2: 0.0786 +Epoch [193/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0814, Loss2: 0.0822 +Epoch [193/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0836 +Epoch [193/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0929, Loss2: 0.0915 +Epoch [193/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.1126, Loss2: 0.1127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 46.2841 % Model2 48.7881 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1299, Loss2: 0.1175 +Epoch [194/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.0969, Loss2: 0.0969 +Epoch [194/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.1007, Loss2: 0.0968 +Epoch [194/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.1163, Loss2: 0.0967 +Epoch [194/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1665, Loss2: 0.1494 +Epoch [194/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.1061, Loss2: 0.1235 +Epoch [194/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0897, Loss2: 0.0857 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 46.4744 % Model2 48.7480 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0990, Loss2: 0.1028 +Epoch [195/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1005, Loss2: 0.1009 +Epoch [195/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1402, Loss2: 0.1305 +Epoch [195/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1334, Loss2: 0.1273 +Epoch [195/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0951, Loss2: 0.0912 +Epoch [195/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0768, Loss2: 0.0823 +Epoch [195/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0989, Loss2: 0.0879 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 46.4343 % Model2 48.5777 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0771, Loss2: 0.0772 +Epoch [196/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0888, Loss2: 0.0839 +Epoch [196/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.1266, Loss2: 0.1117 +Epoch [196/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.1014, Loss2: 0.0989 +Epoch [196/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0660, Loss2: 0.0609 +Epoch [196/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.1562, Loss1: 0.0824, Loss2: 0.0961 +Epoch [196/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1073, Loss2: 0.1125 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 46.3642 % Model2 48.3574 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1110, Loss2: 0.1074 +Epoch [197/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0897, Loss2: 0.0906 +Epoch [197/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0861, Loss2: 0.0819 +Epoch [197/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1088, Loss2: 0.1051 +Epoch [197/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0868, Loss2: 0.0881 +Epoch [197/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0947, Loss2: 0.1028 +Epoch [197/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0899, Loss2: 0.0967 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 46.4543 % Model2 48.7280 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0721, Loss2: 0.0747 +Epoch [198/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1393, Loss2: 0.1234 +Epoch [198/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.1011, Loss2: 0.1090 +Epoch [198/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0860, Loss2: 0.0797 +Epoch [198/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0943, Loss2: 0.1053 +Epoch [198/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1120, Loss2: 0.1168 +Epoch [198/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1117, Loss2: 0.1124 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 46.2941 % Model2 48.1370 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1404, Loss2: 0.1333 +Epoch [199/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0779, Loss2: 0.0833 +Epoch [199/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0726, Loss2: 0.0710 +Epoch [199/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1051, Loss2: 0.1073 +Epoch [199/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0770, Loss2: 0.0738 +Epoch [199/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0807, Loss2: 0.0738 +Epoch [199/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0852, Loss2: 0.0933 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 46.1739 % Model2 48.3073 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0867, Loss2: 0.0929 +Epoch [200/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0954, Loss2: 0.0862 +Epoch [200/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.0952, Loss2: 0.0910 +Epoch [200/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0803, Loss2: 0.0807 +Epoch [200/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0821, Loss2: 0.0873 +Epoch [200/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0883, Loss2: 0.0883 +Epoch [200/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0975, Loss2: 0.0987 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 46.1238 % Model2 48.3874 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_0_6.log b/other_methods/coteaching_plus/coteaching_plus_results/out_0_6.log new file mode 100644 index 0000000..1de65b3 --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_0_6.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 20.3125, Training Accuracy2: 20.3125, Loss1: 0.0175, Loss2: 0.0175 +Epoch [2/200], Iter [100/390] Training Accuracy1: 20.3125, Training Accuracy2: 20.3125, Loss1: 0.0176, Loss2: 0.0176 +Epoch [2/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 24.2188, Loss1: 0.0168, Loss2: 0.0169 +Epoch [2/200], Iter [200/390] Training Accuracy1: 23.4375, Training Accuracy2: 23.4375, Loss1: 0.0170, Loss2: 0.0170 +Epoch [2/200], Iter [250/390] Training Accuracy1: 23.4375, Training Accuracy2: 22.6562, Loss1: 0.0172, Loss2: 0.0171 +Epoch [2/200], Iter [300/390] Training Accuracy1: 21.0938, Training Accuracy2: 22.6562, Loss1: 0.0174, Loss2: 0.0174 +Epoch [2/200], Iter [350/390] Training Accuracy1: 23.4375, Training Accuracy2: 20.3125, Loss1: 0.0168, Loss2: 0.0170 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 15.0942 % Model2 14.8838 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 24.2188, Loss1: 0.0171, Loss2: 0.0172 +Epoch [3/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 24.2188, Loss1: 0.0175, Loss2: 0.0177 +Epoch [3/200], Iter [150/390] Training Accuracy1: 17.1875, Training Accuracy2: 19.5312, Loss1: 0.0172, Loss2: 0.0172 +Epoch [3/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 23.4375, Loss1: 0.0173, Loss2: 0.0177 +Epoch [3/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0163, Loss2: 0.0166 +Epoch [3/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 25.7812, Loss1: 0.0166, Loss2: 0.0165 +Epoch [3/200], Iter [350/390] Training Accuracy1: 20.3125, Training Accuracy2: 23.4375, Loss1: 0.0174, Loss2: 0.0176 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 18.2091 % Model2 16.5966 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 22.6562, Training Accuracy2: 22.6562, Loss1: 0.0172, Loss2: 0.0173 +Epoch [4/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 27.3438, Loss1: 0.0163, Loss2: 0.0162 +Epoch [4/200], Iter [150/390] Training Accuracy1: 23.4375, Training Accuracy2: 24.2188, Loss1: 0.0177, Loss2: 0.0181 +Epoch [4/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 24.2188, Loss1: 0.0172, Loss2: 0.0173 +Epoch [4/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0165, Loss2: 0.0165 +Epoch [4/200], Iter [300/390] Training Accuracy1: 22.6562, Training Accuracy2: 23.4375, Loss1: 0.0172, Loss2: 0.0173 +Epoch [4/200], Iter [350/390] Training Accuracy1: 21.8750, Training Accuracy2: 22.6562, Loss1: 0.0181, Loss2: 0.0182 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 18.1490 % Model2 18.0789 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0163, Loss2: 0.0164 +Epoch [5/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 28.9062, Loss1: 0.0169, Loss2: 0.0168 +Epoch [5/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0164, Loss2: 0.0163 +Epoch [5/200], Iter [200/390] Training Accuracy1: 23.4375, Training Accuracy2: 24.2188, Loss1: 0.0173, Loss2: 0.0174 +Epoch [5/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 22.6562, Loss1: 0.0177, Loss2: 0.0177 +Epoch [5/200], Iter [300/390] Training Accuracy1: 21.0938, Training Accuracy2: 25.0000, Loss1: 0.0174, Loss2: 0.0170 +Epoch [5/200], Iter [350/390] Training Accuracy1: 21.8750, Training Accuracy2: 25.7812, Loss1: 0.0172, Loss2: 0.0171 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 21.9351 % Model2 20.6631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 25.7812, Loss1: 0.0158, Loss2: 0.0158 +Epoch [6/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0182, Loss2: 0.0179 +Epoch [6/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0170, Loss2: 0.0165 +Epoch [6/200], Iter [200/390] Training Accuracy1: 21.0938, Training Accuracy2: 22.6562, Loss1: 0.0176, Loss2: 0.0176 +Epoch [6/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.7812, Loss1: 0.0163, Loss2: 0.0164 +Epoch [6/200], Iter [300/390] Training Accuracy1: 19.5312, Training Accuracy2: 21.0938, Loss1: 0.0178, Loss2: 0.0180 +Epoch [6/200], Iter [350/390] Training Accuracy1: 21.0938, Training Accuracy2: 20.3125, Loss1: 0.0202, Loss2: 0.0196 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 21.8550 % Model2 20.9936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0158, Loss2: 0.0159 +Epoch [7/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 24.2188, Loss1: 0.0172, Loss2: 0.0176 +Epoch [7/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 21.8750, Loss1: 0.0167, Loss2: 0.0168 +Epoch [7/200], Iter [200/390] Training Accuracy1: 21.0938, Training Accuracy2: 21.0938, Loss1: 0.0181, Loss2: 0.0180 +Epoch [7/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0168, Loss2: 0.0169 +Epoch [7/200], Iter [300/390] Training Accuracy1: 22.6562, Training Accuracy2: 23.4375, Loss1: 0.0176, Loss2: 0.0176 +Epoch [7/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 23.4375, Loss1: 0.0173, Loss2: 0.0173 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 21.8750 % Model2 21.4844 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 23.4375, Training Accuracy2: 23.4375, Loss1: 0.0176, Loss2: 0.0172 +Epoch [8/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 21.0938, Loss1: 0.0178, Loss2: 0.0176 +Epoch [8/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0171, Loss2: 0.0173 +Epoch [8/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0167, Loss2: 0.0172 +Epoch [8/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 30.4688, Loss1: 0.0162, Loss2: 0.0165 +Epoch [8/200], Iter [300/390] Training Accuracy1: 22.6562, Training Accuracy2: 26.5625, Loss1: 0.0164, Loss2: 0.0164 +Epoch [8/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0167, Loss2: 0.0165 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 21.7448 % Model2 23.6879 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 24.2188, Loss1: 0.0177, Loss2: 0.0175 +Epoch [9/200], Iter [100/390] Training Accuracy1: 21.0938, Training Accuracy2: 25.0000, Loss1: 0.0181, Loss2: 0.0175 +Epoch [9/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 23.4375, Loss1: 0.0166, Loss2: 0.0167 +Epoch [9/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.7812, Loss1: 0.0175, Loss2: 0.0169 +Epoch [9/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0168, Loss2: 0.0169 +Epoch [9/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0170, Loss2: 0.0167 +Epoch [9/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0151, Loss2: 0.0152 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 24.2087 % Model2 24.5793 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0163, Loss2: 0.0162 +Epoch [10/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0166, Loss2: 0.0163 +Epoch [10/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 21.8750, Loss1: 0.0175, Loss2: 0.0175 +Epoch [10/200], Iter [200/390] Training Accuracy1: 20.3125, Training Accuracy2: 19.5312, Loss1: 0.0187, Loss2: 0.0182 +Epoch [10/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0160, Loss2: 0.0161 +Epoch [10/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0160, Loss2: 0.0161 +Epoch [10/200], Iter [350/390] Training Accuracy1: 23.4375, Training Accuracy2: 26.5625, Loss1: 0.0174, Loss2: 0.0170 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 25.9816 % Model2 26.4824 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0163, Loss2: 0.0158 +Epoch [11/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0162, Loss2: 0.0156 +Epoch [11/200], Iter [150/390] Training Accuracy1: 20.3125, Training Accuracy2: 21.8750, Loss1: 0.0176, Loss2: 0.0173 +Epoch [11/200], Iter [200/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0167, Loss2: 0.0161 +Epoch [11/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 27.3438, Loss1: 0.0175, Loss2: 0.0174 +Epoch [11/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 24.2188, Loss1: 0.0173, Loss2: 0.0171 +Epoch [11/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0157, Loss2: 0.0153 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 23.0569 % Model2 25.0200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.1250, Loss1: 0.0165, Loss2: 0.0166 +Epoch [12/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0177, Loss2: 0.0175 +Epoch [12/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 25.0000, Loss1: 0.0174, Loss2: 0.0174 +Epoch [12/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0167, Loss2: 0.0171 +Epoch [12/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 27.3438, Loss1: 0.0156, Loss2: 0.0152 +Epoch [12/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0158, Loss2: 0.0156 +Epoch [12/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 27.3438, Loss1: 0.0167, Loss2: 0.0169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 22.7764 % Model2 22.8165 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 24.2188, Loss1: 0.0185, Loss2: 0.0180 +Epoch [13/200], Iter [100/390] Training Accuracy1: 25.0000, Training Accuracy2: 28.9062, Loss1: 0.0165, Loss2: 0.0164 +Epoch [13/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0173, Loss2: 0.0167 +Epoch [13/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.0000, Loss1: 0.0172, Loss2: 0.0173 +Epoch [13/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0173, Loss2: 0.0172 +Epoch [13/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0152, Loss2: 0.0146 +Epoch [13/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0154, Loss2: 0.0159 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 24.9700 % Model2 24.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 20.3125, Training Accuracy2: 16.4062, Loss1: 0.0192, Loss2: 0.0202 +Epoch [14/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.7812, Loss1: 0.0166, Loss2: 0.0157 +Epoch [14/200], Iter [150/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.8750, Loss1: 0.0183, Loss2: 0.0183 +Epoch [14/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0173, Loss2: 0.0179 +Epoch [14/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 27.3438, Loss1: 0.0167, Loss2: 0.0166 +Epoch [14/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0157 +Epoch [14/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0169, Loss2: 0.0169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 26.3121 % Model2 26.3522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0158, Loss2: 0.0166 +Epoch [15/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0151, Loss2: 0.0151 +Epoch [15/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 26.5625, Loss1: 0.0173, Loss2: 0.0170 +Epoch [15/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 27.3438, Loss1: 0.0170, Loss2: 0.0164 +Epoch [15/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 24.2188, Loss1: 0.0183, Loss2: 0.0175 +Epoch [15/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0164, Loss2: 0.0159 +Epoch [15/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0160, Loss2: 0.0152 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 26.0917 % Model2 25.6811 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0167, Loss2: 0.0164 +Epoch [16/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0156 +Epoch [16/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.9375, Loss1: 0.0159, Loss2: 0.0152 +Epoch [16/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 27.3438, Loss1: 0.0174, Loss2: 0.0165 +Epoch [16/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.1562, Loss1: 0.0168, Loss2: 0.0160 +Epoch [16/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0156 +Epoch [16/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.8125, Loss1: 0.0165, Loss2: 0.0156 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 27.1735 % Model2 27.5942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0145, Loss2: 0.0145 +Epoch [17/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0168, Loss2: 0.0165 +Epoch [17/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0164, Loss2: 0.0167 +Epoch [17/200], Iter [200/390] Training Accuracy1: 26.5625, Training Accuracy2: 31.2500, Loss1: 0.0167, Loss2: 0.0158 +Epoch [17/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.0312, Loss1: 0.0158, Loss2: 0.0151 +Epoch [17/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0154 +Epoch [17/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0160, Loss2: 0.0161 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 26.0817 % Model2 27.1735 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0161, Loss2: 0.0164 +Epoch [18/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0160 +Epoch [18/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0158, Loss2: 0.0160 +Epoch [18/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 26.5625, Loss1: 0.0149, Loss2: 0.0151 +Epoch [18/200], Iter [250/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.7812, Loss1: 0.0182, Loss2: 0.0170 +Epoch [18/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0157, Loss2: 0.0158 +Epoch [18/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 25.7812, Loss1: 0.0154, Loss2: 0.0158 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 23.9183 % Model2 24.5893 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0159, Loss2: 0.0161 +Epoch [19/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 31.2500, Loss1: 0.0167, Loss2: 0.0163 +Epoch [19/200], Iter [150/390] Training Accuracy1: 26.5625, Training Accuracy2: 31.2500, Loss1: 0.0164, Loss2: 0.0148 +Epoch [19/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.7812, Loss1: 0.0177, Loss2: 0.0174 +Epoch [19/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 31.2500, Loss1: 0.0161, Loss2: 0.0153 +Epoch [19/200], Iter [300/390] Training Accuracy1: 21.8750, Training Accuracy2: 19.5312, Loss1: 0.0162, Loss2: 0.0168 +Epoch [19/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0169, Loss2: 0.0173 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 25.2103 % Model2 26.7428 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0148, Loss2: 0.0142 +Epoch [20/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.1250, Loss1: 0.0163, Loss2: 0.0168 +Epoch [20/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 33.5938, Loss1: 0.0171, Loss2: 0.0161 +Epoch [20/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 20.3125, Loss1: 0.0181, Loss2: 0.0180 +Epoch [20/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0167, Loss2: 0.0159 +Epoch [20/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0157, Loss2: 0.0157 +Epoch [20/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0159, Loss2: 0.0157 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 26.8630 % Model2 27.3538 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0635, Loss2: 0.0631 +Epoch [21/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 24.2188, Loss1: 0.0525, Loss2: 0.0522 +Epoch [21/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.0000, Loss1: 0.0524, Loss2: 0.0523 +Epoch [21/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0603, Loss2: 0.0595 +Epoch [21/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.0000, Loss1: 0.0618, Loss2: 0.0632 +Epoch [21/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0708, Loss2: 0.0710 +Epoch [21/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 27.3438, Loss1: 0.0560, Loss2: 0.0568 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 17.8986 % Model2 21.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0702, Loss2: 0.0698 +Epoch [22/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.0312, Loss1: 0.0916, Loss2: 0.0925 +Epoch [22/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.9375, Loss1: 0.0531, Loss2: 0.0511 +Epoch [22/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 27.3438, Loss1: 0.0496, Loss2: 0.0490 +Epoch [22/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 27.3438, Loss1: 0.0436, Loss2: 0.0445 +Epoch [22/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 27.3438, Loss1: 0.0718, Loss2: 0.0740 +Epoch [22/200], Iter [350/390] Training Accuracy1: 19.5312, Training Accuracy2: 24.2188, Loss1: 0.0588, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 24.0785 % Model2 25.5008 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 32.0312, Loss1: 0.0622, Loss2: 0.0599 +Epoch [23/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0629, Loss2: 0.0616 +Epoch [23/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0613, Loss2: 0.0606 +Epoch [23/200], Iter [200/390] Training Accuracy1: 27.3438, Training Accuracy2: 25.7812, Loss1: 0.0652, Loss2: 0.0661 +Epoch [23/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 29.6875, Loss1: 0.0622, Loss2: 0.0608 +Epoch [23/200], Iter [300/390] Training Accuracy1: 20.3125, Training Accuracy2: 22.6562, Loss1: 0.0690, Loss2: 0.0683 +Epoch [23/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0683, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 23.5477 % Model2 24.6294 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 32.8125, Loss1: 0.0559, Loss2: 0.0582 +Epoch [24/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0595, Loss2: 0.0590 +Epoch [24/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 32.0312, Loss1: 0.0550, Loss2: 0.0533 +Epoch [24/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0644, Loss2: 0.0626 +Epoch [24/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 39.8438, Loss1: 0.0582, Loss2: 0.0548 +Epoch [24/200], Iter [300/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0509, Loss2: 0.0505 +Epoch [24/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0507, Loss2: 0.0509 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 26.4623 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0492, Loss2: 0.0488 +Epoch [25/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.9375, Loss1: 0.0500, Loss2: 0.0487 +Epoch [25/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 22.6562, Loss1: 0.0470, Loss2: 0.0482 +Epoch [25/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0536, Loss2: 0.0541 +Epoch [25/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0469, Loss2: 0.0472 +Epoch [25/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 34.3750, Loss1: 0.0524, Loss2: 0.0492 +Epoch [25/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 29.6875, Loss1: 0.0633, Loss2: 0.0631 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 26.9030 % Model2 27.3538 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 23.4375, Training Accuracy2: 28.1250, Loss1: 0.0551, Loss2: 0.0546 +Epoch [26/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.0312, Loss1: 0.0443, Loss2: 0.0429 +Epoch [26/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.0000, Loss1: 0.0458, Loss2: 0.0461 +Epoch [26/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0578, Loss2: 0.0562 +Epoch [26/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 36.7188, Loss1: 0.0585, Loss2: 0.0557 +Epoch [26/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0587, Loss2: 0.0576 +Epoch [26/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0471, Loss2: 0.0478 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 27.6843 % Model2 27.0132 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.1250, Loss1: 0.0494, Loss2: 0.0496 +Epoch [27/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0458, Loss2: 0.0462 +Epoch [27/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0388, Loss2: 0.0396 +Epoch [27/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 26.5625, Loss1: 0.0437, Loss2: 0.0439 +Epoch [27/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 26.5625, Loss1: 0.0574, Loss2: 0.0601 +Epoch [27/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 34.3750, Loss1: 0.0504, Loss2: 0.0483 +Epoch [27/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 29.6875, Loss1: 0.0548, Loss2: 0.0535 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 25.4407 % Model2 24.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 32.0312, Loss1: 0.0506, Loss2: 0.0497 +Epoch [28/200], Iter [100/390] Training Accuracy1: 22.6562, Training Accuracy2: 29.6875, Loss1: 0.0565, Loss2: 0.0538 +Epoch [28/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.0625, Loss1: 0.0470, Loss2: 0.0465 +Epoch [28/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0539, Loss2: 0.0521 +Epoch [28/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0473, Loss2: 0.0478 +Epoch [28/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0515, Loss2: 0.0529 +Epoch [28/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0656, Loss2: 0.0658 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 26.0817 % Model2 28.2752 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0457, Loss2: 0.0462 +Epoch [29/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0420, Loss2: 0.0409 +Epoch [29/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.8750, Loss1: 0.0516, Loss2: 0.0509 +Epoch [29/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.1562, Loss1: 0.0503, Loss2: 0.0514 +Epoch [29/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0637, Loss2: 0.0624 +Epoch [29/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0501, Loss2: 0.0500 +Epoch [29/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 32.8125, Loss1: 0.0657, Loss2: 0.0618 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 26.0317 % Model2 26.4123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0498, Loss2: 0.0495 +Epoch [30/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0575, Loss2: 0.0579 +Epoch [30/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0464, Loss2: 0.0466 +Epoch [30/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.8125, Loss1: 0.0466, Loss2: 0.0454 +Epoch [30/200], Iter [250/390] Training Accuracy1: 19.5312, Training Accuracy2: 25.0000, Loss1: 0.0529, Loss2: 0.0503 +Epoch [30/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0529, Loss2: 0.0523 +Epoch [30/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0471, Loss2: 0.0465 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 26.5725 % Model2 27.3938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0460, Loss2: 0.0461 +Epoch [31/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0678, Loss2: 0.0658 +Epoch [31/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0478, Loss2: 0.0472 +Epoch [31/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0498, Loss2: 0.0499 +Epoch [31/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0489, Loss2: 0.0498 +Epoch [31/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0467, Loss2: 0.0465 +Epoch [31/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0489, Loss2: 0.0474 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 24.3289 % Model2 26.6927 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0500, Loss2: 0.0500 +Epoch [32/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.9062, Loss1: 0.0438, Loss2: 0.0440 +Epoch [32/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0459, Loss2: 0.0463 +Epoch [32/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0502, Loss2: 0.0501 +Epoch [32/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0493, Loss2: 0.0479 +Epoch [32/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0434, Loss2: 0.0419 +Epoch [32/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 35.9375, Loss1: 0.0496, Loss2: 0.0469 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 23.5877 % Model2 24.6895 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0542, Loss2: 0.0530 +Epoch [33/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0534, Loss2: 0.0524 +Epoch [33/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 36.7188, Loss1: 0.0485, Loss2: 0.0493 +Epoch [33/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0504, Loss2: 0.0494 +Epoch [33/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.0625, Loss1: 0.0485, Loss2: 0.0468 +Epoch [33/200], Iter [300/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0520, Loss2: 0.0534 +Epoch [33/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0499, Loss2: 0.0480 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 26.2019 % Model2 26.6526 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0506, Loss2: 0.0510 +Epoch [34/200], Iter [100/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.7812, Loss1: 0.0514, Loss2: 0.0500 +Epoch [34/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 36.7188, Loss1: 0.0466, Loss2: 0.0452 +Epoch [34/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.1875, Loss1: 0.0524, Loss2: 0.0502 +Epoch [34/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0506, Loss2: 0.0495 +Epoch [34/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.1250, Loss1: 0.0565, Loss2: 0.0584 +Epoch [34/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0462, Loss2: 0.0450 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 26.6226 % Model2 25.3506 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0557, Loss2: 0.0539 +Epoch [35/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0410, Loss2: 0.0399 +Epoch [35/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0539, Loss2: 0.0528 +Epoch [35/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0411, Loss2: 0.0403 +Epoch [35/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0500, Loss2: 0.0491 +Epoch [35/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 40.6250, Loss1: 0.0579, Loss2: 0.0538 +Epoch [35/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0502, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 26.7528 % Model2 26.9431 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0501, Loss2: 0.0485 +Epoch [36/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 35.9375, Loss1: 0.0460, Loss2: 0.0480 +Epoch [36/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0462, Loss2: 0.0451 +Epoch [36/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0472, Loss2: 0.0466 +Epoch [36/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 39.8438, Loss1: 0.0472, Loss2: 0.0422 +Epoch [36/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0550, Loss2: 0.0553 +Epoch [36/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0577, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 26.8830 % Model2 26.1218 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0517, Loss2: 0.0490 +Epoch [37/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0553, Loss2: 0.0543 +Epoch [37/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0513, Loss2: 0.0508 +Epoch [37/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0502, Loss2: 0.0494 +Epoch [37/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 34.3750, Loss1: 0.0447, Loss2: 0.0427 +Epoch [37/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0448, Loss2: 0.0432 +Epoch [37/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 34.3750, Loss1: 0.0451, Loss2: 0.0432 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 26.6727 % Model2 25.7412 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0509, Loss2: 0.0501 +Epoch [38/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0459, Loss2: 0.0450 +Epoch [38/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0469, Loss2: 0.0462 +Epoch [38/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 35.1562, Loss1: 0.0482, Loss2: 0.0463 +Epoch [38/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0429, Loss2: 0.0407 +Epoch [38/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0500, Loss2: 0.0496 +Epoch [38/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0441, Loss2: 0.0433 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 26.6526 % Model2 26.4824 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0597, Loss2: 0.0605 +Epoch [39/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0421, Loss2: 0.0411 +Epoch [39/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.9688, Loss1: 0.0493, Loss2: 0.0469 +Epoch [39/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.8125, Loss1: 0.0534, Loss2: 0.0515 +Epoch [39/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0402, Loss2: 0.0400 +Epoch [39/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0486, Loss2: 0.0471 +Epoch [39/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0414, Loss2: 0.0403 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 25.9014 % Model2 26.7829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0425, Loss2: 0.0422 +Epoch [40/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0535, Loss2: 0.0519 +Epoch [40/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0407, Loss2: 0.0418 +Epoch [40/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 33.5938, Loss1: 0.0382, Loss2: 0.0360 +Epoch [40/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.1562, Loss1: 0.0456, Loss2: 0.0462 +Epoch [40/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0470, Loss2: 0.0459 +Epoch [40/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0473, Loss2: 0.0467 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 26.3121 % Model2 26.0116 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 36.7188, Loss1: 0.0471, Loss2: 0.0460 +Epoch [41/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0513, Loss2: 0.0534 +Epoch [41/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0611, Loss2: 0.0625 +Epoch [41/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0602, Loss2: 0.0577 +Epoch [41/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0510, Loss2: 0.0500 +Epoch [41/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0454, Loss2: 0.0449 +Epoch [41/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0390, Loss2: 0.0395 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 26.6226 % Model2 27.2736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 41.4062, Loss1: 0.0439, Loss2: 0.0408 +Epoch [42/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0531, Loss2: 0.0507 +Epoch [42/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0448, Loss2: 0.0441 +Epoch [42/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0596, Loss2: 0.0600 +Epoch [42/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 44.5312, Loss1: 0.0521, Loss2: 0.0483 +Epoch [42/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0452, Loss2: 0.0457 +Epoch [42/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 33.5938, Loss1: 0.0410, Loss2: 0.0392 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 26.1218 % Model2 26.6827 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 31.2500, Training Accuracy2: 39.8438, Loss1: 0.0465, Loss2: 0.0443 +Epoch [43/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 39.8438, Loss1: 0.0442, Loss2: 0.0414 +Epoch [43/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.7812, Loss1: 0.0489, Loss2: 0.0491 +Epoch [43/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 42.9688, Loss1: 0.0512, Loss2: 0.0475 +Epoch [43/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0470, Loss2: 0.0453 +Epoch [43/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 32.8125, Loss1: 0.0408, Loss2: 0.0384 +Epoch [43/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0441, Loss2: 0.0430 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 27.1635 % Model2 27.5341 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 41.4062, Loss1: 0.0435, Loss2: 0.0416 +Epoch [44/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0479, Loss2: 0.0477 +Epoch [44/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0423, Loss2: 0.0409 +Epoch [44/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 44.5312, Loss1: 0.0508, Loss2: 0.0478 +Epoch [44/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0556, Loss2: 0.0558 +Epoch [44/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0501, Loss2: 0.0494 +Epoch [44/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 34.3750, Loss1: 0.0436, Loss2: 0.0433 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 25.3305 % Model2 25.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0467 +Epoch [45/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0442, Loss2: 0.0436 +Epoch [45/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 44.5312, Loss1: 0.0545, Loss2: 0.0507 +Epoch [45/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0512, Loss2: 0.0500 +Epoch [45/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0436, Loss2: 0.0409 +Epoch [45/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0548, Loss2: 0.0519 +Epoch [45/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0463, Loss2: 0.0461 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 26.2821 % Model2 28.2452 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0513, Loss2: 0.0502 +Epoch [46/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0462, Loss2: 0.0465 +Epoch [46/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0430, Loss2: 0.0411 +Epoch [46/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0539, Loss2: 0.0532 +Epoch [46/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0604, Loss2: 0.0593 +Epoch [46/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 35.9375, Loss1: 0.0492, Loss2: 0.0505 +Epoch [46/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 33.5938, Loss1: 0.0415, Loss2: 0.0400 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 26.5525 % Model2 27.5040 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0536, Loss2: 0.0531 +Epoch [47/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0504, Loss2: 0.0472 +Epoch [47/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0545, Loss2: 0.0544 +Epoch [47/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0496, Loss2: 0.0476 +Epoch [47/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 42.9688, Loss1: 0.0464, Loss2: 0.0427 +Epoch [47/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.9375, Loss1: 0.0481, Loss2: 0.0467 +Epoch [47/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0487, Loss2: 0.0473 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 26.9431 % Model2 25.7512 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0456, Loss2: 0.0451 +Epoch [48/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 42.9688, Loss1: 0.0449, Loss2: 0.0398 +Epoch [48/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0470, Loss2: 0.0477 +Epoch [48/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.9688, Loss1: 0.0473, Loss2: 0.0445 +Epoch [48/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0527, Loss2: 0.0542 +Epoch [48/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 43.7500, Loss1: 0.0450, Loss2: 0.0408 +Epoch [48/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 37.5000, Loss1: 0.0470, Loss2: 0.0454 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 25.1402 % Model2 26.0216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0500, Loss2: 0.0476 +Epoch [49/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0523, Loss2: 0.0530 +Epoch [49/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0492, Loss2: 0.0501 +Epoch [49/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0564, Loss2: 0.0545 +Epoch [49/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0545, Loss2: 0.0513 +Epoch [49/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0538, Loss2: 0.0542 +Epoch [49/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0491, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 26.2320 % Model2 27.8245 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0442, Loss2: 0.0430 +Epoch [50/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.1562, Loss1: 0.0476, Loss2: 0.0482 +Epoch [50/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0480, Loss2: 0.0469 +Epoch [50/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0456, Loss2: 0.0454 +Epoch [50/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0447, Loss2: 0.0459 +Epoch [50/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0412, Loss2: 0.0409 +Epoch [50/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0437, Loss2: 0.0416 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 26.0417 % Model2 26.1719 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0499, Loss2: 0.0502 +Epoch [51/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0552, Loss2: 0.0534 +Epoch [51/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0495, Loss2: 0.0489 +Epoch [51/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0459, Loss2: 0.0460 +Epoch [51/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 40.6250, Loss1: 0.0497, Loss2: 0.0464 +Epoch [51/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 35.9375, Loss1: 0.0415, Loss2: 0.0430 +Epoch [51/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0567, Loss2: 0.0567 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 26.1018 % Model2 25.6711 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0451, Loss2: 0.0454 +Epoch [52/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0436, Loss2: 0.0435 +Epoch [52/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0598, Loss2: 0.0599 +Epoch [52/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0479, Loss2: 0.0504 +Epoch [52/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 43.7500, Loss1: 0.0448, Loss2: 0.0423 +Epoch [52/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0442, Loss2: 0.0420 +Epoch [52/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.0938, Loss1: 0.0509, Loss2: 0.0471 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 26.3421 % Model2 26.9431 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0491, Loss2: 0.0474 +Epoch [53/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0545, Loss2: 0.0522 +Epoch [53/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0410, Loss2: 0.0395 +Epoch [53/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 31.2500, Loss1: 0.0411, Loss2: 0.0427 +Epoch [53/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 40.6250, Loss1: 0.0418, Loss2: 0.0403 +Epoch [53/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0468, Loss2: 0.0458 +Epoch [53/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0434, Loss2: 0.0435 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 26.7528 % Model2 26.8830 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0468, Loss2: 0.0477 +Epoch [54/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 45.3125, Loss1: 0.0475, Loss2: 0.0442 +Epoch [54/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0477, Loss2: 0.0472 +Epoch [54/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0458, Loss2: 0.0463 +Epoch [54/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 27.3438, Loss1: 0.0447, Loss2: 0.0426 +Epoch [54/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0498, Loss2: 0.0474 +Epoch [54/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 47.6562, Loss1: 0.0482, Loss2: 0.0455 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 25.6611 % Model2 25.6210 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 46.8750, Loss1: 0.0554, Loss2: 0.0496 +Epoch [55/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0415, Loss2: 0.0416 +Epoch [55/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0453, Loss2: 0.0457 +Epoch [55/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0507, Loss2: 0.0506 +Epoch [55/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 41.4062, Loss1: 0.0399, Loss2: 0.0386 +Epoch [55/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0532, Loss2: 0.0525 +Epoch [55/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0464, Loss2: 0.0456 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 26.0317 % Model2 26.3522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0434, Loss2: 0.0457 +Epoch [56/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0526, Loss2: 0.0508 +Epoch [56/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.0625, Loss1: 0.0641, Loss2: 0.0593 +Epoch [56/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0407, Loss2: 0.0406 +Epoch [56/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 44.5312, Loss1: 0.0467, Loss2: 0.0425 +Epoch [56/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 47.6562, Loss1: 0.0465, Loss2: 0.0438 +Epoch [56/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0477, Loss2: 0.0481 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 26.3021 % Model2 27.1134 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0443, Loss2: 0.0437 +Epoch [57/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.7812, Loss1: 0.0606, Loss2: 0.0557 +Epoch [57/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0410, Loss2: 0.0406 +Epoch [57/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0391, Loss2: 0.0390 +Epoch [57/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.9375, Loss1: 0.0402, Loss2: 0.0417 +Epoch [57/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.8750, Loss1: 0.0471, Loss2: 0.0448 +Epoch [57/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0466, Loss2: 0.0451 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 27.0132 % Model2 27.4339 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0515, Loss2: 0.0514 +Epoch [58/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0533, Loss2: 0.0521 +Epoch [58/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0444, Loss2: 0.0451 +Epoch [58/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0494, Loss2: 0.0486 +Epoch [58/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0491, Loss2: 0.0490 +Epoch [58/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0539, Loss2: 0.0565 +Epoch [58/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0514, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 25.2804 % Model2 27.4639 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0397, Loss2: 0.0393 +Epoch [59/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0427, Loss2: 0.0420 +Epoch [59/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0426, Loss2: 0.0434 +Epoch [59/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0416, Loss2: 0.0414 +Epoch [59/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0457, Loss2: 0.0469 +Epoch [59/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0533, Loss2: 0.0526 +Epoch [59/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 41.4062, Loss1: 0.0463, Loss2: 0.0421 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 25.4107 % Model2 27.1935 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0488, Loss2: 0.0491 +Epoch [60/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0596, Loss2: 0.0589 +Epoch [60/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0454, Loss2: 0.0446 +Epoch [60/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0441, Loss2: 0.0441 +Epoch [60/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0466, Loss2: 0.0449 +Epoch [60/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0467, Loss2: 0.0479 +Epoch [60/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 40.6250, Loss1: 0.0506, Loss2: 0.0476 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 25.4908 % Model2 26.4123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0528, Loss2: 0.0553 +Epoch [61/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0490, Loss2: 0.0470 +Epoch [61/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0461, Loss2: 0.0439 +Epoch [61/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0450, Loss2: 0.0447 +Epoch [61/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 46.8750, Loss1: 0.0414, Loss2: 0.0381 +Epoch [61/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0461, Loss2: 0.0455 +Epoch [61/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0508, Loss2: 0.0498 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 25.8714 % Model2 26.5124 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0412, Loss2: 0.0400 +Epoch [62/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 49.2188, Loss1: 0.0495, Loss2: 0.0466 +Epoch [62/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 47.6562, Loss1: 0.0557, Loss2: 0.0530 +Epoch [62/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0446, Loss2: 0.0427 +Epoch [62/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0435, Loss2: 0.0431 +Epoch [62/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 43.7500, Loss1: 0.0497, Loss2: 0.0465 +Epoch [62/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0479, Loss2: 0.0472 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 26.7328 % Model2 26.2019 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0425, Loss2: 0.0419 +Epoch [63/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0612, Loss2: 0.0643 +Epoch [63/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0492, Loss2: 0.0496 +Epoch [63/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0446, Loss2: 0.0463 +Epoch [63/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.7812, Loss1: 0.0460, Loss2: 0.0422 +Epoch [63/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0424, Loss2: 0.0407 +Epoch [63/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0514, Loss2: 0.0504 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 24.8798 % Model2 26.2720 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0430, Loss2: 0.0426 +Epoch [64/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0502, Loss2: 0.0494 +Epoch [64/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0495, Loss2: 0.0496 +Epoch [64/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0442, Loss2: 0.0414 +Epoch [64/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 33.5938, Loss1: 0.0407, Loss2: 0.0445 +Epoch [64/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0493, Loss2: 0.0506 +Epoch [64/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 43.7500, Loss1: 0.0462, Loss2: 0.0434 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 25.7212 % Model2 25.0801 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0413, Loss2: 0.0414 +Epoch [65/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0517, Loss2: 0.0527 +Epoch [65/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0444, Loss2: 0.0440 +Epoch [65/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0543, Loss2: 0.0564 +Epoch [65/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0405, Loss2: 0.0397 +Epoch [65/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0505, Loss2: 0.0492 +Epoch [65/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0492, Loss2: 0.0479 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 26.2420 % Model2 25.2504 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0488, Loss2: 0.0488 +Epoch [66/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0517, Loss2: 0.0519 +Epoch [66/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0445, Loss2: 0.0466 +Epoch [66/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.7812, Loss1: 0.0517, Loss2: 0.0473 +Epoch [66/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0551, Loss2: 0.0549 +Epoch [66/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 37.5000, Loss1: 0.0491, Loss2: 0.0472 +Epoch [66/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0440, Loss2: 0.0432 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 27.3137 % Model2 27.0633 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0440, Loss2: 0.0439 +Epoch [67/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 40.6250, Loss1: 0.0408, Loss2: 0.0441 +Epoch [67/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.0625, Loss1: 0.0440, Loss2: 0.0460 +Epoch [67/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0447, Loss2: 0.0455 +Epoch [67/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0538, Loss2: 0.0521 +Epoch [67/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 42.1875, Loss1: 0.0470, Loss2: 0.0450 +Epoch [67/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0493, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 26.7829 % Model2 26.2019 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0444, Loss2: 0.0448 +Epoch [68/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0500, Loss2: 0.0477 +Epoch [68/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0427, Loss2: 0.0431 +Epoch [68/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0471, Loss2: 0.0455 +Epoch [68/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0552, Loss2: 0.0551 +Epoch [68/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0545, Loss2: 0.0564 +Epoch [68/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0499, Loss2: 0.0492 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 26.1018 % Model2 26.6126 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0515, Loss2: 0.0511 +Epoch [69/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0526, Loss2: 0.0526 +Epoch [69/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0588, Loss2: 0.0597 +Epoch [69/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0546, Loss2: 0.0529 +Epoch [69/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0479, Loss2: 0.0480 +Epoch [69/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0489, Loss2: 0.0485 +Epoch [69/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0494, Loss2: 0.0478 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 25.3506 % Model2 26.4724 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0433, Loss2: 0.0427 +Epoch [70/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.0000, Loss1: 0.0451, Loss2: 0.0436 +Epoch [70/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0489, Loss2: 0.0481 +Epoch [70/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0466, Loss2: 0.0454 +Epoch [70/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0497, Loss2: 0.0498 +Epoch [70/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0504, Loss2: 0.0510 +Epoch [70/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0444, Loss2: 0.0442 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 25.7812 % Model2 25.6010 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0467, Loss2: 0.0455 +Epoch [71/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0414, Loss2: 0.0403 +Epoch [71/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0393, Loss2: 0.0389 +Epoch [71/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0435, Loss2: 0.0444 +Epoch [71/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0469, Loss2: 0.0451 +Epoch [71/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0449, Loss2: 0.0457 +Epoch [71/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0498, Loss2: 0.0515 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 26.2821 % Model2 25.7712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0495, Loss2: 0.0483 +Epoch [72/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0491, Loss2: 0.0483 +Epoch [72/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0464, Loss2: 0.0439 +Epoch [72/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 34.3750, Loss1: 0.0486, Loss2: 0.0498 +Epoch [72/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 46.0938, Loss1: 0.0423, Loss2: 0.0388 +Epoch [72/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0493, Loss2: 0.0485 +Epoch [72/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0466, Loss2: 0.0476 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 26.0617 % Model2 26.2019 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0534, Loss2: 0.0518 +Epoch [73/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 48.4375, Loss1: 0.0446, Loss2: 0.0423 +Epoch [73/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0432, Loss2: 0.0439 +Epoch [73/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0443, Loss2: 0.0423 +Epoch [73/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0495, Loss2: 0.0467 +Epoch [73/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0495, Loss2: 0.0490 +Epoch [73/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0575, Loss2: 0.0583 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 26.1418 % Model2 26.8530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0543, Loss2: 0.0561 +Epoch [74/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0502, Loss2: 0.0498 +Epoch [74/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0480, Loss2: 0.0459 +Epoch [74/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.1250, Loss1: 0.0529, Loss2: 0.0511 +Epoch [74/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0490, Loss2: 0.0500 +Epoch [74/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 38.2812, Loss1: 0.0502, Loss2: 0.0482 +Epoch [74/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0462, Loss2: 0.0467 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 25.9415 % Model2 26.3922 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0488, Loss2: 0.0486 +Epoch [75/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0503, Loss2: 0.0492 +Epoch [75/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0411, Loss2: 0.0405 +Epoch [75/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0417, Loss2: 0.0435 +Epoch [75/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0461, Loss2: 0.0469 +Epoch [75/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0489, Loss2: 0.0512 +Epoch [75/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.0938, Loss1: 0.0508, Loss2: 0.0474 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 25.9115 % Model2 26.0717 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0549, Loss2: 0.0560 +Epoch [76/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0426, Loss2: 0.0417 +Epoch [76/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0436, Loss2: 0.0423 +Epoch [76/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0404, Loss2: 0.0396 +Epoch [76/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0506, Loss2: 0.0511 +Epoch [76/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0460, Loss2: 0.0457 +Epoch [76/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 30.4688, Loss1: 0.0388, Loss2: 0.0404 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 26.3622 % Model2 25.6711 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0526, Loss2: 0.0515 +Epoch [77/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 44.5312, Loss1: 0.0557, Loss2: 0.0530 +Epoch [77/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0463, Loss2: 0.0433 +Epoch [77/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0465, Loss2: 0.0457 +Epoch [77/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0425, Loss2: 0.0427 +Epoch [77/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0552, Loss2: 0.0550 +Epoch [77/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0430, Loss2: 0.0431 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 25.1302 % Model2 25.6110 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0459, Loss2: 0.0466 +Epoch [78/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0533, Loss2: 0.0523 +Epoch [78/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0496, Loss2: 0.0469 +Epoch [78/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0566, Loss2: 0.0594 +Epoch [78/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0522 +Epoch [78/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0451, Loss2: 0.0442 +Epoch [78/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0469, Loss2: 0.0476 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 26.2720 % Model2 27.2236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0474, Loss2: 0.0447 +Epoch [79/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0458, Loss2: 0.0465 +Epoch [79/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0476, Loss2: 0.0475 +Epoch [79/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0485, Loss2: 0.0480 +Epoch [79/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0484, Loss2: 0.0486 +Epoch [79/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0481, Loss2: 0.0472 +Epoch [79/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 34.3750, Loss1: 0.0455, Loss2: 0.0474 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 25.3906 % Model2 26.2019 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0428, Loss2: 0.0428 +Epoch [80/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0508, Loss2: 0.0546 +Epoch [80/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0490, Loss2: 0.0478 +Epoch [80/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0571, Loss2: 0.0568 +Epoch [80/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0434, Loss2: 0.0419 +Epoch [80/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 48.4375, Loss1: 0.0421, Loss2: 0.0400 +Epoch [80/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0415, Loss2: 0.0429 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 26.4924 % Model2 26.1819 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0548, Loss2: 0.0529 +Epoch [81/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0500, Loss2: 0.0478 +Epoch [81/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0445, Loss2: 0.0446 +Epoch [81/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0477, Loss2: 0.0479 +Epoch [81/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0486, Loss2: 0.0478 +Epoch [81/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0514, Loss2: 0.0519 +Epoch [81/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0467, Loss2: 0.0469 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 27.2135 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 35.9375, Loss1: 0.0511, Loss2: 0.0536 +Epoch [82/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 37.5000, Loss1: 0.0413, Loss2: 0.0454 +Epoch [82/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0523, Loss2: 0.0553 +Epoch [82/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0432, Loss2: 0.0429 +Epoch [82/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 35.1562, Loss1: 0.0461, Loss2: 0.0496 +Epoch [82/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0437, Loss2: 0.0435 +Epoch [82/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 47.6562, Loss1: 0.0460, Loss2: 0.0419 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 26.6627 % Model2 27.4239 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0469, Loss2: 0.0465 +Epoch [83/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 47.6562, Loss1: 0.0450, Loss2: 0.0409 +Epoch [83/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 51.5625, Loss1: 0.0496, Loss2: 0.0444 +Epoch [83/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0404, Loss2: 0.0391 +Epoch [83/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0440, Loss2: 0.0416 +Epoch [83/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0464, Loss2: 0.0453 +Epoch [83/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0444, Loss2: 0.0450 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 25.4607 % Model2 26.8129 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0521, Loss2: 0.0516 +Epoch [84/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0447, Loss2: 0.0428 +Epoch [84/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0473, Loss2: 0.0493 +Epoch [84/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0480, Loss2: 0.0462 +Epoch [84/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0481, Loss2: 0.0473 +Epoch [84/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0529, Loss2: 0.0517 +Epoch [84/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0453, Loss2: 0.0445 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 26.6727 % Model2 25.9415 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0417, Loss2: 0.0412 +Epoch [85/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 43.7500, Loss1: 0.0483, Loss2: 0.0471 +Epoch [85/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.0938, Loss1: 0.0510, Loss2: 0.0530 +Epoch [85/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0455, Loss2: 0.0440 +Epoch [85/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0413, Loss2: 0.0404 +Epoch [85/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0441, Loss2: 0.0446 +Epoch [85/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0462, Loss2: 0.0440 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 26.1919 % Model2 26.3822 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0467, Loss2: 0.0446 +Epoch [86/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0480, Loss2: 0.0455 +Epoch [86/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0498, Loss2: 0.0498 +Epoch [86/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0524, Loss2: 0.0517 +Epoch [86/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0529, Loss2: 0.0532 +Epoch [86/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0589, Loss2: 0.0577 +Epoch [86/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0476, Loss2: 0.0480 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 26.6627 % Model2 27.2336 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0574, Loss2: 0.0532 +Epoch [87/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0439, Loss2: 0.0451 +Epoch [87/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0496, Loss2: 0.0501 +Epoch [87/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0466, Loss2: 0.0455 +Epoch [87/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0371, Loss2: 0.0375 +Epoch [87/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0519, Loss2: 0.0532 +Epoch [87/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0486, Loss2: 0.0480 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 26.1819 % Model2 26.6326 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0477, Loss2: 0.0470 +Epoch [88/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0444, Loss2: 0.0432 +Epoch [88/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0452, Loss2: 0.0451 +Epoch [88/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0571, Loss2: 0.0559 +Epoch [88/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0637, Loss2: 0.0602 +Epoch [88/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0455, Loss2: 0.0450 +Epoch [88/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 44.5312, Loss1: 0.0426, Loss2: 0.0395 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 26.2821 % Model2 26.7628 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0509, Loss2: 0.0510 +Epoch [89/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0558, Loss2: 0.0526 +Epoch [89/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0462, Loss2: 0.0453 +Epoch [89/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0465, Loss2: 0.0453 +Epoch [89/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0447, Loss2: 0.0454 +Epoch [89/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0356, Loss2: 0.0367 +Epoch [89/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 49.2188, Loss1: 0.0463, Loss2: 0.0431 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 25.0701 % Model2 26.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0443, Loss2: 0.0443 +Epoch [90/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.0938, Loss1: 0.0481, Loss2: 0.0498 +Epoch [90/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 35.1562, Loss1: 0.0430, Loss2: 0.0456 +Epoch [90/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0501, Loss2: 0.0525 +Epoch [90/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0494, Loss2: 0.0491 +Epoch [90/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0560, Loss2: 0.0546 +Epoch [90/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0464, Loss2: 0.0456 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 25.6110 % Model2 26.1919 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0532, Loss2: 0.0554 +Epoch [91/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0424, Loss2: 0.0414 +Epoch [91/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0440, Loss2: 0.0435 +Epoch [91/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 49.2188, Loss1: 0.0513, Loss2: 0.0478 +Epoch [91/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0535, Loss2: 0.0524 +Epoch [91/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0441, Loss2: 0.0438 +Epoch [91/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0447, Loss2: 0.0436 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 26.7728 % Model2 25.5809 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0497, Loss2: 0.0483 +Epoch [92/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0427, Loss2: 0.0432 +Epoch [92/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 41.4062, Loss1: 0.0407, Loss2: 0.0437 +Epoch [92/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 57.0312, Loss1: 0.0596, Loss2: 0.0545 +Epoch [92/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.9062, Loss1: 0.0629, Loss2: 0.0588 +Epoch [92/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0440, Loss2: 0.0450 +Epoch [92/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0515, Loss2: 0.0507 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 26.4123 % Model2 26.9531 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0449, Loss2: 0.0456 +Epoch [93/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0487, Loss2: 0.0453 +Epoch [93/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0464, Loss2: 0.0451 +Epoch [93/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0488, Loss2: 0.0488 +Epoch [93/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0529, Loss2: 0.0533 +Epoch [93/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0491, Loss2: 0.0493 +Epoch [93/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0453, Loss2: 0.0469 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 25.2003 % Model2 26.2420 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0532, Loss2: 0.0540 +Epoch [94/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 46.0938, Loss1: 0.0462, Loss2: 0.0437 +Epoch [94/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0488, Loss2: 0.0516 +Epoch [94/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0445, Loss2: 0.0435 +Epoch [94/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 47.6562, Loss1: 0.0444, Loss2: 0.0426 +Epoch [94/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0511, Loss2: 0.0522 +Epoch [94/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0481, Loss2: 0.0465 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 25.4207 % Model2 25.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 47.6562, Loss1: 0.0442, Loss2: 0.0413 +Epoch [95/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0431, Loss2: 0.0418 +Epoch [95/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0524, Loss2: 0.0561 +Epoch [95/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.8750, Loss1: 0.0508, Loss2: 0.0484 +Epoch [95/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0472, Loss2: 0.0479 +Epoch [95/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.9062, Loss1: 0.0492, Loss2: 0.0468 +Epoch [95/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0538, Loss2: 0.0537 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 25.3506 % Model2 26.2821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0573, Loss2: 0.0529 +Epoch [96/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0427, Loss2: 0.0427 +Epoch [96/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0575, Loss2: 0.0599 +Epoch [96/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0500, Loss2: 0.0491 +Epoch [96/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 40.6250, Loss1: 0.0539, Loss2: 0.0578 +Epoch [96/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0508, Loss2: 0.0486 +Epoch [96/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0479, Loss2: 0.0490 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 25.6711 % Model2 26.0817 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0550, Loss2: 0.0542 +Epoch [97/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0485, Loss2: 0.0507 +Epoch [97/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0440, Loss2: 0.0448 +Epoch [97/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0517, Loss2: 0.0549 +Epoch [97/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0478, Loss2: 0.0459 +Epoch [97/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0525, Loss2: 0.0513 +Epoch [97/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0530, Loss2: 0.0517 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 25.4908 % Model2 25.9515 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0454, Loss2: 0.0450 +Epoch [98/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0452, Loss2: 0.0472 +Epoch [98/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0468, Loss2: 0.0467 +Epoch [98/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 49.2188, Loss1: 0.0471, Loss2: 0.0436 +Epoch [98/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0535, Loss2: 0.0563 +Epoch [98/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 49.2188, Loss1: 0.0482, Loss2: 0.0433 +Epoch [98/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0533, Loss2: 0.0502 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 26.8129 % Model2 26.4724 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0439, Loss2: 0.0425 +Epoch [99/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0532, Loss2: 0.0549 +Epoch [99/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0465, Loss2: 0.0468 +Epoch [99/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0579, Loss2: 0.0595 +Epoch [99/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0496, Loss2: 0.0478 +Epoch [99/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0469, Loss2: 0.0448 +Epoch [99/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0522, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 25.8614 % Model2 26.2720 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0515, Loss2: 0.0519 +Epoch [100/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.9062, Loss1: 0.0572, Loss2: 0.0515 +Epoch [100/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0572, Loss2: 0.0575 +Epoch [100/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 49.2188, Loss1: 0.0441, Loss2: 0.0478 +Epoch [100/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 43.7500, Loss1: 0.0495, Loss2: 0.0528 +Epoch [100/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0536, Loss2: 0.0498 +Epoch [100/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0482, Loss2: 0.0477 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 25.4808 % Model2 25.9315 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0507, Loss2: 0.0532 +Epoch [101/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0522, Loss2: 0.0505 +Epoch [101/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0516, Loss2: 0.0489 +Epoch [101/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0539 +Epoch [101/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0466, Loss2: 0.0449 +Epoch [101/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0520, Loss2: 0.0511 +Epoch [101/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 48.4375, Loss1: 0.0583, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 26.2520 % Model2 26.1318 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0568, Loss2: 0.0533 +Epoch [102/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0421, Loss2: 0.0419 +Epoch [102/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0525, Loss2: 0.0511 +Epoch [102/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0523, Loss2: 0.0510 +Epoch [102/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0486, Loss2: 0.0504 +Epoch [102/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 46.8750, Loss1: 0.0376, Loss2: 0.0354 +Epoch [102/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 48.4375, Loss1: 0.0438, Loss2: 0.0404 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 26.0617 % Model2 25.7812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0493, Loss2: 0.0477 +Epoch [103/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0466, Loss2: 0.0484 +Epoch [103/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0440, Loss2: 0.0456 +Epoch [103/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0626, Loss2: 0.0574 +Epoch [103/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0512, Loss2: 0.0526 +Epoch [103/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0467, Loss2: 0.0477 +Epoch [103/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0460, Loss2: 0.0452 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 26.2520 % Model2 25.9515 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0601, Loss2: 0.0602 +Epoch [104/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 46.0938, Loss1: 0.0527, Loss2: 0.0483 +Epoch [104/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.0938, Loss1: 0.0535, Loss2: 0.0498 +Epoch [104/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0470, Loss2: 0.0456 +Epoch [104/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0452, Loss2: 0.0455 +Epoch [104/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0479, Loss2: 0.0480 +Epoch [104/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 42.9688, Loss1: 0.0418, Loss2: 0.0447 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 25.4407 % Model2 25.4006 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0576, Loss2: 0.0586 +Epoch [105/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0461, Loss2: 0.0464 +Epoch [105/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0493, Loss2: 0.0481 +Epoch [105/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 55.4688, Loss1: 0.0501, Loss2: 0.0471 +Epoch [105/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0371, Loss2: 0.0379 +Epoch [105/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.1562, Loss1: 0.0472, Loss2: 0.0500 +Epoch [105/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0499, Loss2: 0.0511 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 25.2204 % Model2 26.4523 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 48.4375, Loss1: 0.0469, Loss2: 0.0434 +Epoch [106/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0590, Loss2: 0.0589 +Epoch [106/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0429, Loss2: 0.0417 +Epoch [106/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0441, Loss2: 0.0438 +Epoch [106/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0498, Loss2: 0.0503 +Epoch [106/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0501, Loss2: 0.0539 +Epoch [106/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0580, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 25.7011 % Model2 26.5224 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0464, Loss2: 0.0457 +Epoch [107/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0486, Loss2: 0.0494 +Epoch [107/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0546, Loss2: 0.0548 +Epoch [107/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0530, Loss2: 0.0519 +Epoch [107/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0626, Loss2: 0.0612 +Epoch [107/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0464, Loss2: 0.0440 +Epoch [107/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0525, Loss2: 0.0482 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 26.0116 % Model2 25.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0503, Loss2: 0.0514 +Epoch [108/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0578, Loss2: 0.0546 +Epoch [108/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0659, Loss2: 0.0667 +Epoch [108/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0603, Loss2: 0.0594 +Epoch [108/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0508, Loss2: 0.0481 +Epoch [108/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0468, Loss2: 0.0490 +Epoch [108/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0495, Loss2: 0.0496 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 26.3722 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0503, Loss2: 0.0490 +Epoch [109/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0664, Loss2: 0.0629 +Epoch [109/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0436, Loss2: 0.0418 +Epoch [109/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 53.9062, Loss1: 0.0603, Loss2: 0.0634 +Epoch [109/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0594, Loss2: 0.0607 +Epoch [109/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0524, Loss2: 0.0487 +Epoch [109/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 54.6875, Loss1: 0.0567, Loss2: 0.0530 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 25.4207 % Model2 26.4323 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0591, Loss2: 0.0602 +Epoch [110/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0596, Loss2: 0.0620 +Epoch [110/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0510, Loss2: 0.0486 +Epoch [110/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0560, Loss2: 0.0556 +Epoch [110/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0489, Loss2: 0.0465 +Epoch [110/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0480, Loss2: 0.0480 +Epoch [110/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0491 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 25.9315 % Model2 26.3421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0549, Loss2: 0.0555 +Epoch [111/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0417, Loss2: 0.0411 +Epoch [111/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0496, Loss2: 0.0500 +Epoch [111/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0454, Loss2: 0.0446 +Epoch [111/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0445, Loss2: 0.0426 +Epoch [111/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0645, Loss2: 0.0597 +Epoch [111/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0514, Loss2: 0.0513 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 25.9315 % Model2 25.5409 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0540, Loss2: 0.0541 +Epoch [112/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0638, Loss2: 0.0601 +Epoch [112/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0527, Loss2: 0.0511 +Epoch [112/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0468, Loss2: 0.0457 +Epoch [112/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0594, Loss2: 0.0582 +Epoch [112/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0469, Loss2: 0.0464 +Epoch [112/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 41.4062, Loss1: 0.0547, Loss2: 0.0562 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 25.9115 % Model2 25.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0575, Loss2: 0.0592 +Epoch [113/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0534, Loss2: 0.0533 +Epoch [113/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0523, Loss2: 0.0516 +Epoch [113/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0505, Loss2: 0.0473 +Epoch [113/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0501 +Epoch [113/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0637, Loss2: 0.0626 +Epoch [113/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 47.6562, Loss1: 0.0588, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 25.9716 % Model2 25.5308 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0494, Loss2: 0.0505 +Epoch [114/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0585, Loss2: 0.0561 +Epoch [114/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0456, Loss2: 0.0449 +Epoch [114/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 46.0938, Loss1: 0.0486, Loss2: 0.0447 +Epoch [114/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0502, Loss2: 0.0503 +Epoch [114/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0470, Loss2: 0.0494 +Epoch [114/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0614, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 25.7712 % Model2 26.3522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0666, Loss2: 0.0660 +Epoch [115/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0511, Loss2: 0.0490 +Epoch [115/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0599, Loss2: 0.0597 +Epoch [115/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0509, Loss2: 0.0497 +Epoch [115/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0523, Loss2: 0.0512 +Epoch [115/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0457, Loss2: 0.0447 +Epoch [115/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0499, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 26.0717 % Model2 25.7712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0509, Loss2: 0.0487 +Epoch [116/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0649, Loss2: 0.0639 +Epoch [116/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0491, Loss2: 0.0504 +Epoch [116/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0468, Loss2: 0.0462 +Epoch [116/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 42.9688, Loss1: 0.0464, Loss2: 0.0510 +Epoch [116/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0469, Loss2: 0.0493 +Epoch [116/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0447, Loss2: 0.0446 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 25.9716 % Model2 26.0817 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0620, Loss2: 0.0621 +Epoch [117/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0489, Loss2: 0.0519 +Epoch [117/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0534, Loss2: 0.0523 +Epoch [117/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0511, Loss2: 0.0493 +Epoch [117/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.0000, Loss1: 0.0591, Loss2: 0.0569 +Epoch [117/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0453, Loss2: 0.0425 +Epoch [117/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0508, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 25.9115 % Model2 26.0417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0479, Loss2: 0.0498 +Epoch [118/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0477, Loss2: 0.0468 +Epoch [118/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0442, Loss2: 0.0455 +Epoch [118/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0640, Loss2: 0.0639 +Epoch [118/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0484, Loss2: 0.0504 +Epoch [118/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0544, Loss2: 0.0562 +Epoch [118/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0520, Loss2: 0.0504 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 25.5409 % Model2 26.6326 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 62.5000, Loss1: 0.0558, Loss2: 0.0503 +Epoch [119/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0522, Loss2: 0.0505 +Epoch [119/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0631, Loss2: 0.0611 +Epoch [119/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0545, Loss2: 0.0531 +Epoch [119/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0518, Loss2: 0.0507 +Epoch [119/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0501, Loss2: 0.0500 +Epoch [119/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0565, Loss2: 0.0527 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 25.7913 % Model2 25.9716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0551, Loss2: 0.0553 +Epoch [120/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0571, Loss2: 0.0561 +Epoch [120/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0487, Loss2: 0.0506 +Epoch [120/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0482, Loss2: 0.0477 +Epoch [120/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0494, Loss2: 0.0487 +Epoch [120/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0546, Loss2: 0.0545 +Epoch [120/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0450, Loss2: 0.0469 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 25.5709 % Model2 25.5709 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0484, Loss2: 0.0501 +Epoch [121/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0522, Loss2: 0.0518 +Epoch [121/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0512, Loss2: 0.0504 +Epoch [121/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0525, Loss2: 0.0528 +Epoch [121/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0634, Loss2: 0.0594 +Epoch [121/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0523, Loss2: 0.0510 +Epoch [121/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 42.1875, Loss1: 0.0425, Loss2: 0.0452 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 25.5308 % Model2 25.4307 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0712, Loss2: 0.0653 +Epoch [122/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0560, Loss2: 0.0541 +Epoch [122/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0501, Loss2: 0.0505 +Epoch [122/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0673, Loss2: 0.0661 +Epoch [122/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0495, Loss2: 0.0488 +Epoch [122/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0591, Loss2: 0.0588 +Epoch [122/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0488, Loss2: 0.0499 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 26.1118 % Model2 26.2320 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0643, Loss2: 0.0565 +Epoch [123/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0434, Loss2: 0.0433 +Epoch [123/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0469, Loss2: 0.0464 +Epoch [123/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0411, Loss2: 0.0426 +Epoch [123/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0537, Loss2: 0.0529 +Epoch [123/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0470, Loss2: 0.0497 +Epoch [123/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0500, Loss2: 0.0501 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 26.2420 % Model2 26.5725 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0765, Loss2: 0.0681 +Epoch [124/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0501, Loss2: 0.0505 +Epoch [124/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0547, Loss2: 0.0566 +Epoch [124/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0627, Loss2: 0.0626 +Epoch [124/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0566, Loss2: 0.0538 +Epoch [124/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0547, Loss2: 0.0535 +Epoch [124/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0452, Loss2: 0.0475 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 25.7011 % Model2 25.9115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0505, Loss2: 0.0514 +Epoch [125/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0428, Loss2: 0.0424 +Epoch [125/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0464, Loss2: 0.0456 +Epoch [125/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0547, Loss2: 0.0558 +Epoch [125/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0667, Loss2: 0.0682 +Epoch [125/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0530, Loss2: 0.0521 +Epoch [125/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0540, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 25.6711 % Model2 25.8413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0549, Loss2: 0.0541 +Epoch [126/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0551, Loss2: 0.0539 +Epoch [126/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0577, Loss2: 0.0592 +Epoch [126/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0495, Loss2: 0.0486 +Epoch [126/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0455, Loss2: 0.0446 +Epoch [126/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0534, Loss2: 0.0520 +Epoch [126/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0611, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 26.5124 % Model2 26.2320 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0501, Loss2: 0.0495 +Epoch [127/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0519, Loss2: 0.0531 +Epoch [127/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0584, Loss2: 0.0637 +Epoch [127/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0557, Loss2: 0.0544 +Epoch [127/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0411, Loss2: 0.0400 +Epoch [127/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0501, Loss2: 0.0470 +Epoch [127/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0531, Loss2: 0.0514 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 26.0717 % Model2 26.3822 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0462, Loss2: 0.0488 +Epoch [128/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0537, Loss2: 0.0559 +Epoch [128/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0609, Loss2: 0.0618 +Epoch [128/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0481, Loss2: 0.0505 +Epoch [128/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0557, Loss2: 0.0553 +Epoch [128/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 49.2188, Loss1: 0.0485, Loss2: 0.0455 +Epoch [128/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.7812, Loss1: 0.0602, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 25.5809 % Model2 25.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0503, Loss2: 0.0509 +Epoch [129/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0625, Loss2: 0.0587 +Epoch [129/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0574, Loss2: 0.0518 +Epoch [129/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0781, Loss2: 0.0775 +Epoch [129/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0465, Loss2: 0.0470 +Epoch [129/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0500, Loss2: 0.0504 +Epoch [129/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0573, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 25.8514 % Model2 26.4623 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0555, Loss2: 0.0547 +Epoch [130/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0573, Loss2: 0.0571 +Epoch [130/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0575, Loss2: 0.0583 +Epoch [130/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0478, Loss2: 0.0459 +Epoch [130/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0463, Loss2: 0.0453 +Epoch [130/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0583, Loss2: 0.0551 +Epoch [130/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0553, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 26.3522 % Model2 26.2821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0513, Loss2: 0.0502 +Epoch [131/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0589, Loss2: 0.0562 +Epoch [131/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0496, Loss2: 0.0476 +Epoch [131/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0453, Loss2: 0.0477 +Epoch [131/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0545, Loss2: 0.0546 +Epoch [131/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0594, Loss2: 0.0629 +Epoch [131/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0483, Loss2: 0.0458 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 25.6010 % Model2 25.8814 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0733, Loss2: 0.0686 +Epoch [132/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0496, Loss2: 0.0470 +Epoch [132/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0548, Loss2: 0.0539 +Epoch [132/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0620, Loss2: 0.0589 +Epoch [132/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0572, Loss2: 0.0599 +Epoch [132/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0569, Loss2: 0.0563 +Epoch [132/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0570, Loss2: 0.0559 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 25.5909 % Model2 26.2220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0530, Loss2: 0.0512 +Epoch [133/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0512, Loss2: 0.0538 +Epoch [133/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0544, Loss2: 0.0525 +Epoch [133/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0517, Loss2: 0.0511 +Epoch [133/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0496, Loss2: 0.0461 +Epoch [133/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0553, Loss2: 0.0544 +Epoch [133/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0523, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 25.9816 % Model2 26.2821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0533, Loss2: 0.0556 +Epoch [134/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0524, Loss2: 0.0523 +Epoch [134/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0551, Loss2: 0.0534 +Epoch [134/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0539, Loss2: 0.0511 +Epoch [134/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0493, Loss2: 0.0497 +Epoch [134/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0596, Loss2: 0.0588 +Epoch [134/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0627, Loss2: 0.0597 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 25.8413 % Model2 26.1819 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0561, Loss2: 0.0575 +Epoch [135/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0449, Loss2: 0.0462 +Epoch [135/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0531, Loss2: 0.0541 +Epoch [135/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0510, Loss2: 0.0518 +Epoch [135/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0608, Loss2: 0.0602 +Epoch [135/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0483, Loss2: 0.0508 +Epoch [135/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0666, Loss2: 0.0686 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 25.8213 % Model2 26.2520 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0581, Loss2: 0.0544 +Epoch [136/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0632, Loss2: 0.0608 +Epoch [136/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0559, Loss2: 0.0523 +Epoch [136/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0489, Loss2: 0.0483 +Epoch [136/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0587, Loss2: 0.0598 +Epoch [136/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0552, Loss2: 0.0556 +Epoch [136/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0515, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 26.2220 % Model2 26.0016 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0591, Loss2: 0.0531 +Epoch [137/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0644, Loss2: 0.0631 +Epoch [137/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0583, Loss2: 0.0576 +Epoch [137/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0545, Loss2: 0.0531 +Epoch [137/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0667, Loss2: 0.0675 +Epoch [137/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0476, Loss2: 0.0450 +Epoch [137/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0599, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 25.9315 % Model2 26.1418 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 51.5625, Loss1: 0.0524, Loss2: 0.0486 +Epoch [138/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0464, Loss2: 0.0459 +Epoch [138/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0565, Loss2: 0.0533 +Epoch [138/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0558, Loss2: 0.0546 +Epoch [138/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0477, Loss2: 0.0478 +Epoch [138/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0558, Loss2: 0.0531 +Epoch [138/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0514, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 26.4523 % Model2 25.6310 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0482, Loss2: 0.0497 +Epoch [139/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0522, Loss2: 0.0514 +Epoch [139/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0493, Loss2: 0.0471 +Epoch [139/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0529, Loss2: 0.0527 +Epoch [139/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0527, Loss2: 0.0557 +Epoch [139/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0509, Loss2: 0.0490 +Epoch [139/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0612, Loss2: 0.0547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 25.1202 % Model2 25.8614 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0679, Loss2: 0.0676 +Epoch [140/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0679, Loss2: 0.0718 +Epoch [140/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0468, Loss2: 0.0450 +Epoch [140/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0616, Loss2: 0.0577 +Epoch [140/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0670, Loss2: 0.0622 +Epoch [140/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0575, Loss2: 0.0607 +Epoch [140/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0491, Loss2: 0.0499 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 25.6611 % Model2 26.1018 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0600, Loss2: 0.0586 +Epoch [141/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0584, Loss2: 0.0583 +Epoch [141/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0576, Loss2: 0.0542 +Epoch [141/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0585, Loss2: 0.0589 +Epoch [141/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0660, Loss2: 0.0635 +Epoch [141/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0553, Loss2: 0.0594 +Epoch [141/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0497, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 25.9215 % Model2 26.1619 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0595, Loss2: 0.0546 +Epoch [142/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0654, Loss2: 0.0699 +Epoch [142/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0552, Loss2: 0.0583 +Epoch [142/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0574, Loss2: 0.0558 +Epoch [142/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0591, Loss2: 0.0557 +Epoch [142/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 47.6562, Loss1: 0.0446, Loss2: 0.0424 +Epoch [142/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0423, Loss2: 0.0428 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 26.0817 % Model2 25.9115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0584, Loss2: 0.0609 +Epoch [143/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0482, Loss2: 0.0494 +Epoch [143/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0560, Loss2: 0.0519 +Epoch [143/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0658, Loss2: 0.0618 +Epoch [143/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0511, Loss2: 0.0510 +Epoch [143/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0588, Loss2: 0.0574 +Epoch [143/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0538, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 25.9515 % Model2 26.3922 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0527, Loss2: 0.0522 +Epoch [144/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.9062, Loss1: 0.0571, Loss2: 0.0533 +Epoch [144/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0535, Loss2: 0.0538 +Epoch [144/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0518, Loss2: 0.0525 +Epoch [144/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0637, Loss2: 0.0652 +Epoch [144/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0751, Loss2: 0.0752 +Epoch [144/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0541, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 25.8814 % Model2 26.1518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0545, Loss2: 0.0579 +Epoch [145/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0644, Loss2: 0.0599 +Epoch [145/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0674, Loss2: 0.0636 +Epoch [145/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0574, Loss2: 0.0551 +Epoch [145/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0507, Loss2: 0.0516 +Epoch [145/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0596, Loss2: 0.0579 +Epoch [145/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0550, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 26.3221 % Model2 26.4123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0582, Loss2: 0.0591 +Epoch [146/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0672, Loss2: 0.0665 +Epoch [146/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0486, Loss2: 0.0495 +Epoch [146/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0588, Loss2: 0.0633 +Epoch [146/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0480, Loss2: 0.0519 +Epoch [146/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0559, Loss2: 0.0527 +Epoch [146/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0630, Loss2: 0.0637 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 25.7712 % Model2 26.2019 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0469, Loss2: 0.0457 +Epoch [147/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0616, Loss2: 0.0605 +Epoch [147/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0637, Loss2: 0.0599 +Epoch [147/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0556, Loss2: 0.0570 +Epoch [147/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0498, Loss2: 0.0500 +Epoch [147/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0711, Loss2: 0.0684 +Epoch [147/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0586, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 25.9415 % Model2 26.0116 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0581, Loss2: 0.0575 +Epoch [148/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0609, Loss2: 0.0605 +Epoch [148/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0718, Loss2: 0.0692 +Epoch [148/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0590, Loss2: 0.0562 +Epoch [148/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 59.3750, Loss1: 0.0531, Loss2: 0.0483 +Epoch [148/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0558, Loss2: 0.0524 +Epoch [148/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 25.6010 % Model2 25.8313 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0640, Loss2: 0.0615 +Epoch [149/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0573, Loss2: 0.0565 +Epoch [149/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0520, Loss2: 0.0483 +Epoch [149/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0627, Loss2: 0.0565 +Epoch [149/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0580, Loss2: 0.0592 +Epoch [149/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0690, Loss2: 0.0734 +Epoch [149/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0526, Loss2: 0.0552 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 26.0216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0679, Loss2: 0.0710 +Epoch [150/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0512, Loss2: 0.0504 +Epoch [150/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0483, Loss2: 0.0516 +Epoch [150/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0615, Loss2: 0.0638 +Epoch [150/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0558, Loss2: 0.0542 +Epoch [150/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0722, Loss2: 0.0717 +Epoch [150/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0622, Loss2: 0.0582 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 25.7011 % Model2 25.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0553, Loss2: 0.0559 +Epoch [151/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0524, Loss2: 0.0534 +Epoch [151/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0709, Loss2: 0.0697 +Epoch [151/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0551, Loss2: 0.0554 +Epoch [151/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0581, Loss2: 0.0572 +Epoch [151/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0450, Loss2: 0.0479 +Epoch [151/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0637, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 26.0317 % Model2 26.1318 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0622, Loss2: 0.0606 +Epoch [152/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0476, Loss2: 0.0472 +Epoch [152/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0661, Loss2: 0.0641 +Epoch [152/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0575, Loss2: 0.0564 +Epoch [152/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0609, Loss2: 0.0611 +Epoch [152/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0665, Loss2: 0.0651 +Epoch [152/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0594, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 26.0116 % Model2 26.2821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0621, Loss2: 0.0639 +Epoch [153/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0573, Loss2: 0.0583 +Epoch [153/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0694, Loss2: 0.0689 +Epoch [153/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0548, Loss2: 0.0544 +Epoch [153/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0537, Loss2: 0.0564 +Epoch [153/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0452, Loss2: 0.0462 +Epoch [153/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.9062, Loss1: 0.0554, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 25.9315 % Model2 26.1919 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0612, Loss2: 0.0604 +Epoch [154/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0672, Loss2: 0.0636 +Epoch [154/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0586, Loss2: 0.0638 +Epoch [154/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0616, Loss2: 0.0616 +Epoch [154/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0540, Loss2: 0.0551 +Epoch [154/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0557, Loss2: 0.0602 +Epoch [154/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0537, Loss2: 0.0519 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 26.0016 % Model2 26.0016 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0618, Loss2: 0.0617 +Epoch [155/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0544 +Epoch [155/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0607, Loss2: 0.0664 +Epoch [155/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0624, Loss2: 0.0611 +Epoch [155/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0493, Loss2: 0.0492 +Epoch [155/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0603, Loss2: 0.0598 +Epoch [155/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0513, Loss2: 0.0514 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 25.8013 % Model2 26.0517 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0789, Loss2: 0.0741 +Epoch [156/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0624, Loss2: 0.0610 +Epoch [156/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0629 +Epoch [156/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0552, Loss2: 0.0548 +Epoch [156/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0491, Loss2: 0.0514 +Epoch [156/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0586, Loss2: 0.0571 +Epoch [156/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0658, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 25.6911 % Model2 26.1418 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0587, Loss2: 0.0570 +Epoch [157/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0551, Loss2: 0.0547 +Epoch [157/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0728, Loss2: 0.0730 +Epoch [157/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0597, Loss2: 0.0601 +Epoch [157/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0462, Loss2: 0.0476 +Epoch [157/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0543, Loss2: 0.0529 +Epoch [157/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0620, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 25.7412 % Model2 26.2520 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0687, Loss2: 0.0639 +Epoch [158/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0670, Loss2: 0.0688 +Epoch [158/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0630, Loss2: 0.0645 +Epoch [158/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0554, Loss2: 0.0511 +Epoch [158/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0589, Loss2: 0.0586 +Epoch [158/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0534, Loss2: 0.0566 +Epoch [158/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0643, Loss2: 0.0641 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 25.5308 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0599, Loss2: 0.0558 +Epoch [159/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0562, Loss2: 0.0575 +Epoch [159/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0550, Loss2: 0.0546 +Epoch [159/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0750, Loss2: 0.0753 +Epoch [159/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0548, Loss2: 0.0542 +Epoch [159/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0586, Loss2: 0.0624 +Epoch [159/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0769, Loss2: 0.0684 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 26.1218 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0517, Loss2: 0.0507 +Epoch [160/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0781, Loss2: 0.0836 +Epoch [160/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0479, Loss2: 0.0493 +Epoch [160/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0619, Loss2: 0.0621 +Epoch [160/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0562, Loss2: 0.0531 +Epoch [160/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0548, Loss2: 0.0547 +Epoch [160/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0607, Loss2: 0.0551 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 26.0917 % Model2 25.6210 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0655, Loss2: 0.0666 +Epoch [161/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0535, Loss2: 0.0516 +Epoch [161/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0683, Loss2: 0.0665 +Epoch [161/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0562, Loss2: 0.0535 +Epoch [161/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0570, Loss2: 0.0551 +Epoch [161/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0511, Loss2: 0.0517 +Epoch [161/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0483, Loss2: 0.0479 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 25.5409 % Model2 25.6210 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0604, Loss2: 0.0632 +Epoch [162/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0518, Loss2: 0.0509 +Epoch [162/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0509, Loss2: 0.0528 +Epoch [162/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0605, Loss2: 0.0623 +Epoch [162/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0623, Loss2: 0.0562 +Epoch [162/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0540, Loss2: 0.0563 +Epoch [162/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0576, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 25.6510 % Model2 25.9615 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0684, Loss2: 0.0643 +Epoch [163/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0541, Loss2: 0.0570 +Epoch [163/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0648, Loss2: 0.0667 +Epoch [163/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0592, Loss2: 0.0608 +Epoch [163/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0683, Loss2: 0.0678 +Epoch [163/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0621, Loss2: 0.0616 +Epoch [163/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0584, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 25.9215 % Model2 26.2119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0642, Loss2: 0.0632 +Epoch [164/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0629, Loss2: 0.0633 +Epoch [164/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0614, Loss2: 0.0588 +Epoch [164/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0589, Loss2: 0.0619 +Epoch [164/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0489, Loss2: 0.0499 +Epoch [164/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0660, Loss2: 0.0652 +Epoch [164/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0573, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 25.8113 % Model2 26.1518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0475, Loss2: 0.0458 +Epoch [165/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0776, Loss2: 0.0772 +Epoch [165/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0600, Loss2: 0.0559 +Epoch [165/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0572, Loss2: 0.0566 +Epoch [165/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0596, Loss2: 0.0603 +Epoch [165/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0713, Loss2: 0.0740 +Epoch [165/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0684, Loss2: 0.0738 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 25.5509 % Model2 26.0216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0618, Loss2: 0.0582 +Epoch [166/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0618, Loss2: 0.0613 +Epoch [166/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0589, Loss2: 0.0600 +Epoch [166/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0561, Loss2: 0.0527 +Epoch [166/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0616, Loss2: 0.0579 +Epoch [166/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0671, Loss2: 0.0634 +Epoch [166/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0571, Loss2: 0.0557 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 25.6510 % Model2 26.1919 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0530, Loss2: 0.0560 +Epoch [167/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0693, Loss2: 0.0666 +Epoch [167/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0590, Loss2: 0.0622 +Epoch [167/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0597, Loss2: 0.0610 +Epoch [167/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0566, Loss2: 0.0556 +Epoch [167/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0560, Loss2: 0.0527 +Epoch [167/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0623, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 25.9415 % Model2 26.1118 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0563, Loss2: 0.0519 +Epoch [168/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0577, Loss2: 0.0562 +Epoch [168/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0527, Loss2: 0.0532 +Epoch [168/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0553, Loss2: 0.0533 +Epoch [168/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0601, Loss2: 0.0573 +Epoch [168/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0731, Loss2: 0.0756 +Epoch [168/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0667, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 25.6410 % Model2 26.0216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0650, Loss2: 0.0623 +Epoch [169/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0659, Loss2: 0.0658 +Epoch [169/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0695, Loss2: 0.0703 +Epoch [169/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0690, Loss2: 0.0704 +Epoch [169/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.8438, Loss1: 0.0500, Loss2: 0.0544 +Epoch [169/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 51.5625, Loss1: 0.0672, Loss2: 0.0603 +Epoch [169/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0752, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 25.9415 % Model2 25.7913 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0582, Loss2: 0.0575 +Epoch [170/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0505, Loss2: 0.0505 +Epoch [170/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0684, Loss2: 0.0669 +Epoch [170/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0608, Loss2: 0.0589 +Epoch [170/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0646, Loss2: 0.0601 +Epoch [170/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0722, Loss2: 0.0761 +Epoch [170/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0618, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 25.7111 % Model2 26.3722 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0624, Loss2: 0.0630 +Epoch [171/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0498, Loss2: 0.0489 +Epoch [171/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0619, Loss2: 0.0620 +Epoch [171/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0569, Loss2: 0.0537 +Epoch [171/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0551, Loss2: 0.0558 +Epoch [171/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0593, Loss2: 0.0612 +Epoch [171/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0669, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 25.6811 % Model2 25.8013 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0589, Loss2: 0.0608 +Epoch [172/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0643, Loss2: 0.0633 +Epoch [172/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0484, Loss2: 0.0496 +Epoch [172/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0799, Loss2: 0.0793 +Epoch [172/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0669, Loss2: 0.0660 +Epoch [172/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0641, Loss2: 0.0634 +Epoch [172/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0491, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 25.6611 % Model2 25.9716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0654, Loss2: 0.0620 +Epoch [173/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0671, Loss2: 0.0647 +Epoch [173/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0576, Loss2: 0.0572 +Epoch [173/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0720, Loss2: 0.0707 +Epoch [173/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0613, Loss2: 0.0611 +Epoch [173/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0536, Loss2: 0.0562 +Epoch [173/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0550, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 25.6611 % Model2 26.1318 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0704, Loss2: 0.0679 +Epoch [174/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0757, Loss2: 0.0740 +Epoch [174/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0721, Loss2: 0.0700 +Epoch [174/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 44.5312, Loss1: 0.0529, Loss2: 0.0586 +Epoch [174/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0550, Loss2: 0.0532 +Epoch [174/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0600 +Epoch [174/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0596, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 25.5409 % Model2 25.9615 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0537, Loss2: 0.0562 +Epoch [175/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0547, Loss2: 0.0544 +Epoch [175/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0693 +Epoch [175/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0615 +Epoch [175/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.0938, Loss1: 0.0543, Loss2: 0.0600 +Epoch [175/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0602, Loss2: 0.0608 +Epoch [175/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0538, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 25.7512 % Model2 25.8614 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0672, Loss2: 0.0686 +Epoch [176/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0572, Loss2: 0.0538 +Epoch [176/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0565, Loss2: 0.0589 +Epoch [176/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0682, Loss2: 0.0717 +Epoch [176/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0573, Loss2: 0.0603 +Epoch [176/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0618, Loss2: 0.0611 +Epoch [176/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0567, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 25.8614 % Model2 26.0917 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0533, Loss2: 0.0496 +Epoch [177/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0691, Loss2: 0.0638 +Epoch [177/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0592, Loss2: 0.0596 +Epoch [177/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0595, Loss2: 0.0613 +Epoch [177/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0618, Loss2: 0.0621 +Epoch [177/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0645, Loss2: 0.0656 +Epoch [177/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0657, Loss2: 0.0623 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 25.7212 % Model2 25.7011 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0617, Loss2: 0.0642 +Epoch [178/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0674, Loss2: 0.0627 +Epoch [178/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0562, Loss2: 0.0568 +Epoch [178/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0666, Loss2: 0.0622 +Epoch [178/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0630, Loss2: 0.0588 +Epoch [178/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0564, Loss2: 0.0579 +Epoch [178/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0581, Loss2: 0.0568 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 25.6811 % Model2 26.0917 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0608, Loss2: 0.0651 +Epoch [179/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0659, Loss2: 0.0632 +Epoch [179/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0642, Loss2: 0.0603 +Epoch [179/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0704, Loss2: 0.0687 +Epoch [179/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 42.9688, Loss1: 0.0497, Loss2: 0.0519 +Epoch [179/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0808, Loss2: 0.0853 +Epoch [179/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0553, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 25.6711 % Model2 25.8313 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0649, Loss2: 0.0626 +Epoch [180/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0695, Loss2: 0.0691 +Epoch [180/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0667, Loss2: 0.0618 +Epoch [180/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0564, Loss2: 0.0517 +Epoch [180/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0513, Loss2: 0.0505 +Epoch [180/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0556, Loss2: 0.0546 +Epoch [180/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0755, Loss2: 0.0751 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 25.7812 % Model2 25.9415 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0589, Loss2: 0.0590 +Epoch [181/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0727, Loss2: 0.0760 +Epoch [181/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0682, Loss2: 0.0691 +Epoch [181/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0712, Loss2: 0.0716 +Epoch [181/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0575, Loss2: 0.0592 +Epoch [181/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0617, Loss2: 0.0590 +Epoch [181/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0638, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 25.7412 % Model2 25.7512 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0582, Loss2: 0.0628 +Epoch [182/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0615, Loss2: 0.0599 +Epoch [182/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0681, Loss2: 0.0726 +Epoch [182/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0672, Loss2: 0.0586 +Epoch [182/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0568, Loss2: 0.0590 +Epoch [182/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0669, Loss2: 0.0708 +Epoch [182/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0519, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 25.9014 % Model2 25.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0693, Loss2: 0.0696 +Epoch [183/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0624, Loss2: 0.0656 +Epoch [183/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0684, Loss2: 0.0667 +Epoch [183/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0608, Loss2: 0.0626 +Epoch [183/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0676, Loss2: 0.0662 +Epoch [183/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0642, Loss2: 0.0662 +Epoch [183/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0617, Loss2: 0.0577 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 25.6811 % Model2 26.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0582, Loss2: 0.0581 +Epoch [184/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0544, Loss2: 0.0526 +Epoch [184/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0713, Loss2: 0.0728 +Epoch [184/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0660, Loss2: 0.0656 +Epoch [184/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0682, Loss2: 0.0706 +Epoch [184/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0706, Loss2: 0.0705 +Epoch [184/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0633, Loss2: 0.0620 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 25.7312 % Model2 25.8213 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0607, Loss2: 0.0594 +Epoch [185/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0611, Loss2: 0.0630 +Epoch [185/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0644, Loss2: 0.0685 +Epoch [185/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 42.9688, Loss1: 0.0597, Loss2: 0.0654 +Epoch [185/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0628, Loss2: 0.0575 +Epoch [185/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0696, Loss2: 0.0707 +Epoch [185/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0588, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 26.1218 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0681, Loss2: 0.0697 +Epoch [186/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.7812, Loss1: 0.0636, Loss2: 0.0688 +Epoch [186/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.8750, Loss1: 0.0525, Loss2: 0.0575 +Epoch [186/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0548, Loss2: 0.0557 +Epoch [186/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0677, Loss2: 0.0714 +Epoch [186/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0763, Loss2: 0.0717 +Epoch [186/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0536, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 25.6510 % Model2 25.9916 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0629, Loss2: 0.0605 +Epoch [187/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0681, Loss2: 0.0661 +Epoch [187/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0753, Loss2: 0.0750 +Epoch [187/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0685, Loss2: 0.0727 +Epoch [187/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0683, Loss2: 0.0617 +Epoch [187/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0682, Loss2: 0.0615 +Epoch [187/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0914, Loss2: 0.0933 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 25.5709 % Model2 25.8814 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0555, Loss2: 0.0574 +Epoch [188/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0614, Loss2: 0.0565 +Epoch [188/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 44.5312, Loss1: 0.0530, Loss2: 0.0555 +Epoch [188/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0472, Loss2: 0.0472 +Epoch [188/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0553, Loss2: 0.0569 +Epoch [188/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0500, Loss2: 0.0480 +Epoch [188/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0605, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 25.5008 % Model2 25.9515 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0593, Loss2: 0.0600 +Epoch [189/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0595, Loss2: 0.0649 +Epoch [189/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0751, Loss2: 0.0739 +Epoch [189/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0623, Loss2: 0.0663 +Epoch [189/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 46.0938, Loss1: 0.0536, Loss2: 0.0583 +Epoch [189/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0708, Loss2: 0.0678 +Epoch [189/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0749, Loss2: 0.0760 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 25.6010 % Model2 25.9916 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0714, Loss2: 0.0708 +Epoch [190/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0574, Loss2: 0.0602 +Epoch [190/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0672, Loss2: 0.0696 +Epoch [190/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0653, Loss2: 0.0713 +Epoch [190/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0749, Loss2: 0.0771 +Epoch [190/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0617, Loss2: 0.0675 +Epoch [190/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0677, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 25.5709 % Model2 25.9515 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0601 +Epoch [191/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0655, Loss2: 0.0662 +Epoch [191/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0584, Loss2: 0.0544 +Epoch [191/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 55.4688, Loss1: 0.0584, Loss2: 0.0625 +Epoch [191/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0651, Loss2: 0.0664 +Epoch [191/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.0938, Loss1: 0.0573, Loss2: 0.0627 +Epoch [191/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0707, Loss2: 0.0698 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 25.9215 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0733, Loss2: 0.0668 +Epoch [192/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0659, Loss2: 0.0597 +Epoch [192/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0776, Loss2: 0.0764 +Epoch [192/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0759, Loss2: 0.0771 +Epoch [192/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0668, Loss2: 0.0679 +Epoch [192/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0631, Loss2: 0.0639 +Epoch [192/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0556, Loss2: 0.0568 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 25.4607 % Model2 25.9816 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0841, Loss2: 0.0822 +Epoch [193/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0588, Loss2: 0.0595 +Epoch [193/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 41.4062, Loss1: 0.0511, Loss2: 0.0540 +Epoch [193/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0780, Loss2: 0.0783 +Epoch [193/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0494, Loss2: 0.0488 +Epoch [193/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0525, Loss2: 0.0489 +Epoch [193/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0863, Loss2: 0.0831 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 25.4607 % Model2 25.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0722, Loss2: 0.0687 +Epoch [194/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0563, Loss2: 0.0586 +Epoch [194/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 48.4375, Loss1: 0.0678, Loss2: 0.0747 +Epoch [194/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0708, Loss2: 0.0715 +Epoch [194/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0691, Loss2: 0.0688 +Epoch [194/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0605, Loss2: 0.0574 +Epoch [194/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0869, Loss2: 0.0853 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 25.4708 % Model2 25.8614 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0610, Loss2: 0.0591 +Epoch [195/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0662, Loss2: 0.0668 +Epoch [195/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0589 +Epoch [195/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0630, Loss2: 0.0658 +Epoch [195/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0549, Loss2: 0.0551 +Epoch [195/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0603, Loss2: 0.0610 +Epoch [195/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0604, Loss2: 0.0639 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 25.4107 % Model2 26.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0792, Loss2: 0.0724 +Epoch [196/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0641, Loss2: 0.0604 +Epoch [196/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0618, Loss2: 0.0575 +Epoch [196/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0524, Loss2: 0.0567 +Epoch [196/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0771, Loss2: 0.0824 +Epoch [196/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0597, Loss2: 0.0620 +Epoch [196/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0662, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 25.5108 % Model2 25.9415 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0725, Loss2: 0.0685 +Epoch [197/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0754, Loss2: 0.0690 +Epoch [197/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0470, Loss2: 0.0490 +Epoch [197/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0973, Loss2: 0.0949 +Epoch [197/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0614, Loss2: 0.0569 +Epoch [197/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0677, Loss2: 0.0658 +Epoch [197/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0547, Loss2: 0.0534 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 25.3706 % Model2 25.8514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0581, Loss2: 0.0622 +Epoch [198/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0760, Loss2: 0.0739 +Epoch [198/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0648, Loss2: 0.0674 +Epoch [198/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0705, Loss2: 0.0658 +Epoch [198/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0588, Loss2: 0.0633 +Epoch [198/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0764, Loss2: 0.0747 +Epoch [198/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0718, Loss2: 0.0721 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 25.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0647, Loss2: 0.0618 +Epoch [199/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0519, Loss2: 0.0524 +Epoch [199/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0568, Loss2: 0.0607 +Epoch [199/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0618, Loss2: 0.0612 +Epoch [199/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0625, Loss2: 0.0656 +Epoch [199/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0577, Loss2: 0.0618 +Epoch [199/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0657, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 25.4207 % Model2 25.8914 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0643, Loss2: 0.0674 +Epoch [200/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0658, Loss2: 0.0676 +Epoch [200/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0805, Loss2: 0.0712 +Epoch [200/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0478, Loss2: 0.0489 +Epoch [200/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0489, Loss2: 0.0494 +Epoch [200/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.1025, Loss2: 0.0916 +Epoch [200/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0542, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 25.4708 % Model2 25.8514 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_2_2.log b/other_methods/coteaching_plus/coteaching_plus_results/out_2_2.log new file mode 100644 index 0000000..dde9f7a --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_2_2.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 18.7500, Training Accuracy2: 18.7500, Loss1: 0.0181, Loss2: 0.0177 +Epoch [2/200], Iter [100/390] Training Accuracy1: 17.1875, Training Accuracy2: 19.5312, Loss1: 0.0170, Loss2: 0.0166 +Epoch [2/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0159, Loss2: 0.0162 +Epoch [2/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 36.7188, Loss1: 0.0164, Loss2: 0.0162 +Epoch [2/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0152, Loss2: 0.0148 +Epoch [2/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0139, Loss2: 0.0138 +Epoch [2/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0154, Loss2: 0.0153 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 35.8373 % Model2 35.9976 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0157, Loss2: 0.0160 +Epoch [3/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 32.0312, Loss1: 0.0147, Loss2: 0.0159 +Epoch [3/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0161, Loss2: 0.0150 +Epoch [3/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0145, Loss2: 0.0153 +Epoch [3/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0138, Loss2: 0.0136 +Epoch [3/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0140, Loss2: 0.0138 +Epoch [3/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0134, Loss2: 0.0133 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 39.9740 % Model2 40.3946 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0143, Loss2: 0.0138 +Epoch [4/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 42.9688, Loss1: 0.0136, Loss2: 0.0138 +Epoch [4/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0134, Loss2: 0.0134 +Epoch [4/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0131, Loss2: 0.0128 +Epoch [4/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0122 +Epoch [4/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0133, Loss2: 0.0132 +Epoch [4/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.8125, Loss1: 0.0148, Loss2: 0.0141 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 45.6430 % Model2 45.8233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.0938, Loss1: 0.0130, Loss2: 0.0124 +Epoch [5/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0140, Loss2: 0.0131 +Epoch [5/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0119, Loss2: 0.0116 +Epoch [5/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0137, Loss2: 0.0133 +Epoch [5/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0127, Loss2: 0.0129 +Epoch [5/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0139, Loss2: 0.0141 +Epoch [5/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0141, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 48.1671 % Model2 48.4776 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0114, Loss2: 0.0113 +Epoch [6/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0117, Loss2: 0.0119 +Epoch [6/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0116 +Epoch [6/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0131 +Epoch [6/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0129, Loss2: 0.0124 +Epoch [6/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0116, Loss2: 0.0111 +Epoch [6/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0132, Loss2: 0.0130 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 49.2288 % Model2 50.0901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0103, Loss2: 0.0104 +Epoch [7/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0105, Loss2: 0.0106 +Epoch [7/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0130, Loss2: 0.0129 +Epoch [7/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0128, Loss2: 0.0124 +Epoch [7/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0117, Loss2: 0.0118 +Epoch [7/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0111, Loss2: 0.0114 +Epoch [7/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0114, Loss2: 0.0118 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 51.4123 % Model2 51.7528 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0105, Loss2: 0.0104 +Epoch [8/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0119, Loss2: 0.0121 +Epoch [8/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 43.7500, Loss1: 0.0098, Loss2: 0.0101 +Epoch [8/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 51.5625, Loss1: 0.0103, Loss2: 0.0106 +Epoch [8/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0098, Loss2: 0.0095 +Epoch [8/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0134, Loss2: 0.0134 +Epoch [8/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0119, Loss2: 0.0117 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 52.9247 % Model2 53.3053 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 42.9688, Loss1: 0.0104, Loss2: 0.0111 +Epoch [9/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0119, Loss2: 0.0118 +Epoch [9/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0120, Loss2: 0.0107 +Epoch [9/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0107, Loss2: 0.0101 +Epoch [9/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0113, Loss2: 0.0113 +Epoch [9/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0116, Loss2: 0.0107 +Epoch [9/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0116, Loss2: 0.0109 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 52.7845 % Model2 54.7576 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0116, Loss2: 0.0117 +Epoch [10/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0096, Loss2: 0.0091 +Epoch [10/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 50.0000, Loss1: 0.0122, Loss2: 0.0106 +Epoch [10/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0112, Loss2: 0.0115 +Epoch [10/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0099, Loss2: 0.0091 +Epoch [10/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0119, Loss2: 0.0116 +Epoch [10/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.8438, Loss1: 0.0103, Loss2: 0.0118 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 52.8946 % Model2 54.4772 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0108, Loss2: 0.0104 +Epoch [11/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0079, Loss2: 0.0084 +Epoch [11/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0137, Loss2: 0.0129 +Epoch [11/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0078, Loss2: 0.0086 +Epoch [11/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0099, Loss2: 0.0099 +Epoch [11/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0094, Loss2: 0.0085 +Epoch [11/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0113, Loss2: 0.0105 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 54.2468 % Model2 55.4688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0090, Loss2: 0.0089 +Epoch [12/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0111, Loss2: 0.0107 +Epoch [12/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0099, Loss2: 0.0107 +Epoch [12/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0098, Loss2: 0.0099 +Epoch [12/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0126, Loss2: 0.0130 +Epoch [12/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0093, Loss2: 0.0085 +Epoch [12/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0095, Loss2: 0.0101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 55.2784 % Model2 55.7893 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0098, Loss2: 0.0094 +Epoch [13/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0111, Loss2: 0.0106 +Epoch [13/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0102, Loss2: 0.0103 +Epoch [13/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0099, Loss2: 0.0099 +Epoch [13/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0084, Loss2: 0.0087 +Epoch [13/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0089, Loss2: 0.0080 +Epoch [13/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0104, Loss2: 0.0099 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 55.4587 % Model2 56.3301 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0087, Loss2: 0.0081 +Epoch [14/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0113, Loss2: 0.0105 +Epoch [14/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0096, Loss2: 0.0089 +Epoch [14/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0103, Loss2: 0.0109 +Epoch [14/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0092, Loss2: 0.0101 +Epoch [14/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0090, Loss2: 0.0097 +Epoch [14/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0101, Loss2: 0.0105 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 55.9095 % Model2 57.4619 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0094, Loss2: 0.0084 +Epoch [15/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0097, Loss2: 0.0094 +Epoch [15/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0098, Loss2: 0.0111 +Epoch [15/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0101, Loss2: 0.0102 +Epoch [15/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0093, Loss2: 0.0094 +Epoch [15/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0101, Loss2: 0.0101 +Epoch [15/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0100, Loss2: 0.0099 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 55.1783 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0108, Loss2: 0.0106 +Epoch [16/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0084, Loss2: 0.0076 +Epoch [16/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0084, Loss2: 0.0086 +Epoch [16/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0083, Loss2: 0.0081 +Epoch [16/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0105, Loss2: 0.0093 +Epoch [16/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0113, Loss2: 0.0111 +Epoch [16/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0092, Loss2: 0.0101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 56.4002 % Model2 57.2716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0086, Loss2: 0.0089 +Epoch [17/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0081, Loss2: 0.0084 +Epoch [17/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0096, Loss2: 0.0094 +Epoch [17/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0112, Loss2: 0.0100 +Epoch [17/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0098, Loss2: 0.0100 +Epoch [17/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0098, Loss2: 0.0100 +Epoch [17/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0088, Loss2: 0.0086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 57.3718 % Model2 57.5621 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0077, Loss2: 0.0085 +Epoch [18/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0104, Loss2: 0.0107 +Epoch [18/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0096, Loss2: 0.0091 +Epoch [18/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0093, Loss2: 0.0092 +Epoch [18/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0099, Loss2: 0.0101 +Epoch [18/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0087, Loss2: 0.0091 +Epoch [18/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0080, Loss2: 0.0080 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 56.5204 % Model2 56.8810 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0117, Loss2: 0.0112 +Epoch [19/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.9062, Loss1: 0.0103, Loss2: 0.0090 +Epoch [19/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0078, Loss2: 0.0079 +Epoch [19/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0085, Loss2: 0.0083 +Epoch [19/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 61.7188, Loss1: 0.0082, Loss2: 0.0074 +Epoch [19/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0118, Loss2: 0.0115 +Epoch [19/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0083, Loss2: 0.0090 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 56.7909 % Model2 57.8926 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0081, Loss2: 0.0080 +Epoch [20/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0106, Loss2: 0.0101 +Epoch [20/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0055, Loss2: 0.0060 +Epoch [20/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0090, Loss2: 0.0087 +Epoch [20/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0089, Loss2: 0.0077 +Epoch [20/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0071, Loss2: 0.0078 +Epoch [20/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0093, Loss2: 0.0091 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 57.6122 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0654, Loss2: 0.0648 +Epoch [21/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0537, Loss2: 0.0532 +Epoch [21/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0713, Loss2: 0.0731 +Epoch [21/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0482, Loss2: 0.0494 +Epoch [21/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0453, Loss2: 0.0462 +Epoch [21/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0588, Loss2: 0.0595 +Epoch [21/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0645, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 52.2135 % Model2 54.5974 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0546, Loss2: 0.0549 +Epoch [22/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0575, Loss2: 0.0574 +Epoch [22/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0598, Loss2: 0.0615 +Epoch [22/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0823, Loss2: 0.0816 +Epoch [22/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0795, Loss2: 0.0784 +Epoch [22/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 49.2188, Loss1: 0.0514, Loss2: 0.0485 +Epoch [22/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0525, Loss2: 0.0522 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 51.8530 % Model2 56.0497 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0832, Loss2: 0.0793 +Epoch [23/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0876, Loss2: 0.0914 +Epoch [23/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0653, Loss2: 0.0664 +Epoch [23/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0650, Loss2: 0.0624 +Epoch [23/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0635, Loss2: 0.0627 +Epoch [23/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0583, Loss2: 0.0582 +Epoch [23/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0687, Loss2: 0.0693 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 53.0649 % Model2 56.5304 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0752, Loss2: 0.0729 +Epoch [24/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0541, Loss2: 0.0535 +Epoch [24/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 45.3125, Loss1: 0.0620, Loss2: 0.0656 +Epoch [24/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0765, Loss2: 0.0799 +Epoch [24/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0558, Loss2: 0.0565 +Epoch [24/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0781, Loss2: 0.0748 +Epoch [24/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0752, Loss2: 0.0752 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 56.1899 % Model2 57.2216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0648, Loss2: 0.0618 +Epoch [25/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0492, Loss2: 0.0474 +Epoch [25/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0695, Loss2: 0.0754 +Epoch [25/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0833, Loss2: 0.0834 +Epoch [25/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0551, Loss2: 0.0523 +Epoch [25/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0892, Loss2: 0.0848 +Epoch [25/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0730, Loss2: 0.0735 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 56.7808 % Model2 54.7776 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0795, Loss2: 0.0828 +Epoch [26/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0677, Loss2: 0.0658 +Epoch [26/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0588, Loss2: 0.0621 +Epoch [26/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0563, Loss2: 0.0544 +Epoch [26/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0661, Loss2: 0.0659 +Epoch [26/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0687, Loss2: 0.0726 +Epoch [26/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0605, Loss2: 0.0572 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 56.1899 % Model2 56.9812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0641, Loss2: 0.0652 +Epoch [27/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0673, Loss2: 0.0672 +Epoch [27/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 56.2500, Loss1: 0.0596, Loss2: 0.0534 +Epoch [27/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0550, Loss2: 0.0555 +Epoch [27/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0713, Loss2: 0.0722 +Epoch [27/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0754, Loss2: 0.0745 +Epoch [27/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0594, Loss2: 0.0585 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 56.0397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0604, Loss2: 0.0608 +Epoch [28/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0678, Loss2: 0.0653 +Epoch [28/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0672, Loss2: 0.0606 +Epoch [28/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 59.3750, Loss1: 0.0707, Loss2: 0.0637 +Epoch [28/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0729, Loss2: 0.0701 +Epoch [28/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 51.5625, Loss1: 0.0699, Loss2: 0.0642 +Epoch [28/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0751, Loss2: 0.0708 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 55.9996 % Model2 57.4920 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0535, Loss2: 0.0560 +Epoch [29/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0710, Loss2: 0.0716 +Epoch [29/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0636, Loss2: 0.0628 +Epoch [29/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0620, Loss2: 0.0578 +Epoch [29/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0640, Loss2: 0.0636 +Epoch [29/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0589, Loss2: 0.0600 +Epoch [29/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0654, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 56.5405 % Model2 56.4603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0722, Loss2: 0.0729 +Epoch [30/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0653, Loss2: 0.0659 +Epoch [30/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0468, Loss2: 0.0475 +Epoch [30/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0595, Loss2: 0.0594 +Epoch [30/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0647, Loss2: 0.0600 +Epoch [30/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0587, Loss2: 0.0592 +Epoch [30/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0640, Loss2: 0.0578 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 55.4387 % Model2 57.3317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0527, Loss2: 0.0564 +Epoch [31/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0710, Loss2: 0.0700 +Epoch [31/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0717, Loss2: 0.0703 +Epoch [31/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0765, Loss2: 0.0725 +Epoch [31/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0494, Loss2: 0.0480 +Epoch [31/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0621, Loss2: 0.0623 +Epoch [31/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0732, Loss2: 0.0714 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 56.6607 % Model2 58.3133 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0623, Loss2: 0.0604 +Epoch [32/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0765, Loss2: 0.0751 +Epoch [32/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 61.7188, Loss1: 0.0637, Loss2: 0.0584 +Epoch [32/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0623, Loss2: 0.0613 +Epoch [32/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0821, Loss2: 0.0792 +Epoch [32/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0689, Loss2: 0.0706 +Epoch [32/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0602, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 56.1098 % Model2 58.4034 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0696, Loss2: 0.0725 +Epoch [33/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0605, Loss2: 0.0626 +Epoch [33/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0789, Loss2: 0.0745 +Epoch [33/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0777, Loss2: 0.0818 +Epoch [33/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0604, Loss2: 0.0606 +Epoch [33/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0652, Loss2: 0.0633 +Epoch [33/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0583, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 53.5457 % Model2 58.7139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0502, Loss2: 0.0531 +Epoch [34/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0794, Loss2: 0.0752 +Epoch [34/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0788, Loss2: 0.0734 +Epoch [34/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0720, Loss2: 0.0688 +Epoch [34/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0756, Loss2: 0.0728 +Epoch [34/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0514, Loss2: 0.0526 +Epoch [34/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0564, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 55.7492 % Model2 56.7508 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0915, Loss2: 0.0945 +Epoch [35/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0831, Loss2: 0.0860 +Epoch [35/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0629, Loss2: 0.0632 +Epoch [35/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0576, Loss2: 0.0600 +Epoch [35/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0720, Loss2: 0.0767 +Epoch [35/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0562, Loss2: 0.0584 +Epoch [35/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0848, Loss2: 0.0929 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 57.2316 % Model2 58.2232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0606, Loss2: 0.0601 +Epoch [36/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0675, Loss2: 0.0718 +Epoch [36/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0701, Loss2: 0.0648 +Epoch [36/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0681, Loss2: 0.0655 +Epoch [36/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0607, Loss2: 0.0586 +Epoch [36/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0701, Loss2: 0.0721 +Epoch [36/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0781, Loss2: 0.0728 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 56.5405 % Model2 58.8742 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0763, Loss2: 0.0766 +Epoch [37/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0713, Loss2: 0.0743 +Epoch [37/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0710, Loss2: 0.0697 +Epoch [37/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0860, Loss2: 0.0810 +Epoch [37/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0655 +Epoch [37/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0747, Loss2: 0.0728 +Epoch [37/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0610, Loss2: 0.0628 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 57.1514 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0654, Loss2: 0.0634 +Epoch [38/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0757, Loss2: 0.0738 +Epoch [38/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0551, Loss2: 0.0519 +Epoch [38/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0699, Loss2: 0.0734 +Epoch [38/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0715, Loss2: 0.0672 +Epoch [38/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0853, Loss2: 0.0886 +Epoch [38/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0638, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 56.1999 % Model2 59.4251 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0686, Loss2: 0.0636 +Epoch [39/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1032, Loss2: 0.1048 +Epoch [39/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0714, Loss2: 0.0694 +Epoch [39/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0761, Loss2: 0.0742 +Epoch [39/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0637, Loss2: 0.0634 +Epoch [39/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0642, Loss2: 0.0632 +Epoch [39/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0753, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 56.7007 % Model2 58.0629 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0649, Loss2: 0.0668 +Epoch [40/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0668, Loss2: 0.0701 +Epoch [40/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0571, Loss2: 0.0598 +Epoch [40/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0534, Loss2: 0.0541 +Epoch [40/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0667, Loss2: 0.0715 +Epoch [40/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0812, Loss2: 0.0785 +Epoch [40/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0562, Loss2: 0.0570 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 57.4820 % Model2 57.8626 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0746, Loss2: 0.0740 +Epoch [41/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0771, Loss2: 0.0710 +Epoch [41/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0602, Loss2: 0.0608 +Epoch [41/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0536, Loss2: 0.0523 +Epoch [41/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0558, Loss2: 0.0580 +Epoch [41/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0721, Loss2: 0.0737 +Epoch [41/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0521, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 55.7292 % Model2 58.3634 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0626, Loss2: 0.0592 +Epoch [42/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0687, Loss2: 0.0703 +Epoch [42/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0573, Loss2: 0.0568 +Epoch [42/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0684, Loss2: 0.0715 +Epoch [42/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0909, Loss2: 0.0926 +Epoch [42/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0619, Loss2: 0.0651 +Epoch [42/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0842, Loss2: 0.0874 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 56.6807 % Model2 58.8942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.9375, Loss1: 0.0530, Loss2: 0.0491 +Epoch [43/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0804, Loss2: 0.0778 +Epoch [43/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0596, Loss2: 0.0587 +Epoch [43/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0686, Loss2: 0.0689 +Epoch [43/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0628, Loss2: 0.0601 +Epoch [43/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.1021, Loss2: 0.0980 +Epoch [43/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0495, Loss2: 0.0522 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 57.3117 % Model2 58.7841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0654, Loss2: 0.0639 +Epoch [44/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0792, Loss2: 0.0794 +Epoch [44/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0624, Loss2: 0.0580 +Epoch [44/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 57.8125, Loss1: 0.0759, Loss2: 0.0660 +Epoch [44/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0659, Loss2: 0.0601 +Epoch [44/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0818, Loss2: 0.0844 +Epoch [44/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 61.7188, Loss1: 0.0682, Loss2: 0.0607 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 56.8910 % Model2 57.7324 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0804, Loss2: 0.0797 +Epoch [45/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0602, Loss2: 0.0582 +Epoch [45/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0594, Loss2: 0.0579 +Epoch [45/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0582, Loss2: 0.0582 +Epoch [45/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0757, Loss2: 0.0767 +Epoch [45/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0631, Loss2: 0.0625 +Epoch [45/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 53.1250, Loss1: 0.0830, Loss2: 0.0918 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 57.5621 % Model2 58.8642 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0726, Loss2: 0.0728 +Epoch [46/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0666, Loss2: 0.0678 +Epoch [46/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0683, Loss2: 0.0634 +Epoch [46/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0749, Loss2: 0.0727 +Epoch [46/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0664, Loss2: 0.0641 +Epoch [46/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0793, Loss2: 0.0839 +Epoch [46/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0643, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 57.0212 % Model2 59.0645 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0662, Loss2: 0.0671 +Epoch [47/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0609, Loss2: 0.0660 +Epoch [47/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0861, Loss2: 0.0833 +Epoch [47/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0666, Loss2: 0.0721 +Epoch [47/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0596, Loss2: 0.0581 +Epoch [47/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0946, Loss2: 0.0943 +Epoch [47/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0594, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 56.2600 % Model2 58.0228 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0804, Loss2: 0.0760 +Epoch [48/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0851, Loss2: 0.0827 +Epoch [48/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0782, Loss2: 0.0766 +Epoch [48/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0663, Loss2: 0.0715 +Epoch [48/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0547, Loss2: 0.0525 +Epoch [48/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0558, Loss2: 0.0543 +Epoch [48/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0634, Loss2: 0.0632 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 55.8093 % Model2 57.9928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0744, Loss2: 0.0690 +Epoch [49/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.0312, Loss1: 0.0553, Loss2: 0.0500 +Epoch [49/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0626, Loss2: 0.0606 +Epoch [49/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0835, Loss2: 0.0841 +Epoch [49/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0750, Loss2: 0.0744 +Epoch [49/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0605, Loss2: 0.0608 +Epoch [49/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0693, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 56.5204 % Model2 58.7440 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0895, Loss2: 0.0774 +Epoch [50/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0642, Loss2: 0.0657 +Epoch [50/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0759, Loss2: 0.0754 +Epoch [50/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0808, Loss2: 0.0812 +Epoch [50/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0723, Loss2: 0.0656 +Epoch [50/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0831, Loss2: 0.0826 +Epoch [50/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0638, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 57.9227 % Model2 58.4335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0750, Loss2: 0.0755 +Epoch [51/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0592, Loss2: 0.0635 +Epoch [51/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0724, Loss2: 0.0666 +Epoch [51/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0609, Loss2: 0.0647 +Epoch [51/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0693, Loss2: 0.0633 +Epoch [51/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0616, Loss2: 0.0631 +Epoch [51/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0669, Loss2: 0.0674 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 55.4788 % Model2 56.2500 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0809, Loss2: 0.0804 +Epoch [52/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0700, Loss2: 0.0696 +Epoch [52/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0558, Loss2: 0.0579 +Epoch [52/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 54.6875, Loss1: 0.0585, Loss2: 0.0680 +Epoch [52/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0751, Loss2: 0.0731 +Epoch [52/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0639, Loss2: 0.0661 +Epoch [52/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0671, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 57.0813 % Model2 57.5220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0777, Loss2: 0.0748 +Epoch [53/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0666, Loss2: 0.0630 +Epoch [53/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0497, Loss2: 0.0507 +Epoch [53/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0758, Loss2: 0.0762 +Epoch [53/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0804, Loss2: 0.0789 +Epoch [53/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0572, Loss2: 0.0584 +Epoch [53/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0640, Loss2: 0.0635 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 57.4920 % Model2 57.7524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0618, Loss2: 0.0626 +Epoch [54/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0808, Loss2: 0.0726 +Epoch [54/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0749, Loss2: 0.0819 +Epoch [54/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0688, Loss2: 0.0723 +Epoch [54/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0882, Loss2: 0.0834 +Epoch [54/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0782, Loss2: 0.0694 +Epoch [54/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0744, Loss2: 0.0720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 56.9611 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0595, Loss2: 0.0587 +Epoch [55/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0724, Loss2: 0.0675 +Epoch [55/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0607, Loss2: 0.0589 +Epoch [55/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0724, Loss2: 0.0695 +Epoch [55/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0658, Loss2: 0.0646 +Epoch [55/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0659 +Epoch [55/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0644, Loss2: 0.0637 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 56.4203 % Model2 57.4219 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0666, Loss2: 0.0682 +Epoch [56/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0875, Loss2: 0.0882 +Epoch [56/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0645, Loss2: 0.0663 +Epoch [56/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0625, Loss2: 0.0632 +Epoch [56/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0675, Loss2: 0.0651 +Epoch [56/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0792, Loss2: 0.0803 +Epoch [56/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0630, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 56.7508 % Model2 57.5020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.8125, Loss1: 0.0692, Loss2: 0.0750 +Epoch [57/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0741, Loss2: 0.0771 +Epoch [57/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 54.6875, Loss1: 0.0605, Loss2: 0.0695 +Epoch [57/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0757, Loss2: 0.0705 +Epoch [57/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0576, Loss2: 0.0590 +Epoch [57/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0639, Loss2: 0.0645 +Epoch [57/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0698, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 57.9026 % Model2 58.8642 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0671, Loss2: 0.0667 +Epoch [58/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 66.4062, Loss1: 0.0658, Loss2: 0.0605 +Epoch [58/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0568, Loss2: 0.0574 +Epoch [58/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.9375, Loss1: 0.0846, Loss2: 0.0727 +Epoch [58/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0655, Loss2: 0.0729 +Epoch [58/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0491, Loss2: 0.0477 +Epoch [58/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0683, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 56.7107 % Model2 58.0228 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0704, Loss2: 0.0650 +Epoch [59/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0771, Loss2: 0.0800 +Epoch [59/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 64.0625, Loss1: 0.0586, Loss2: 0.0559 +Epoch [59/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0704, Loss2: 0.0707 +Epoch [59/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0691, Loss2: 0.0638 +Epoch [59/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0531, Loss2: 0.0498 +Epoch [59/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0636, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 56.9311 % Model2 58.0729 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0841, Loss2: 0.0878 +Epoch [60/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0671, Loss2: 0.0670 +Epoch [60/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0977, Loss2: 0.0994 +Epoch [60/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0638, Loss2: 0.0667 +Epoch [60/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0706, Loss2: 0.0656 +Epoch [60/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0758, Loss2: 0.0787 +Epoch [60/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0834, Loss2: 0.0907 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 56.4503 % Model2 57.0312 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0622, Loss2: 0.0609 +Epoch [61/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0654, Loss2: 0.0697 +Epoch [61/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0749, Loss2: 0.0724 +Epoch [61/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 65.6250, Loss1: 0.0628, Loss2: 0.0542 +Epoch [61/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0709, Loss2: 0.0673 +Epoch [61/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0773, Loss2: 0.0726 +Epoch [61/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0738, Loss2: 0.0764 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 56.9611 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0810, Loss2: 0.0765 +Epoch [62/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0579 +Epoch [62/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.0312, Loss1: 0.0470, Loss2: 0.0504 +Epoch [62/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0768, Loss2: 0.0748 +Epoch [62/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0833, Loss2: 0.0818 +Epoch [62/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0697, Loss2: 0.0669 +Epoch [62/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0572, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 56.1999 % Model2 57.6422 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0705, Loss2: 0.0762 +Epoch [63/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0582, Loss2: 0.0568 +Epoch [63/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0834, Loss2: 0.0783 +Epoch [63/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0935, Loss2: 0.0934 +Epoch [63/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0785, Loss2: 0.0735 +Epoch [63/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0777, Loss2: 0.0686 +Epoch [63/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0967, Loss2: 0.0918 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 56.9511 % Model2 58.2232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0840, Loss2: 0.0825 +Epoch [64/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 69.5312, Loss1: 0.0724, Loss2: 0.0634 +Epoch [64/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0755, Loss2: 0.0796 +Epoch [64/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0615, Loss2: 0.0592 +Epoch [64/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0677, Loss2: 0.0642 +Epoch [64/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0556, Loss2: 0.0560 +Epoch [64/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0803, Loss2: 0.0742 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 57.0413 % Model2 57.7624 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0861, Loss2: 0.0871 +Epoch [65/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0612, Loss2: 0.0581 +Epoch [65/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0697, Loss2: 0.0651 +Epoch [65/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 58.5938, Loss1: 0.0647, Loss2: 0.0755 +Epoch [65/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0632, Loss2: 0.0620 +Epoch [65/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0805, Loss2: 0.0768 +Epoch [65/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0741, Loss2: 0.0779 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 56.7408 % Model2 58.6438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0630, Loss2: 0.0609 +Epoch [66/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0652, Loss2: 0.0639 +Epoch [66/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0737, Loss2: 0.0755 +Epoch [66/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0566, Loss2: 0.0608 +Epoch [66/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0665, Loss2: 0.0670 +Epoch [66/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0527, Loss2: 0.0517 +Epoch [66/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0675, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 56.3201 % Model2 58.3233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0632, Loss2: 0.0634 +Epoch [67/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0729, Loss2: 0.0715 +Epoch [67/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0622, Loss2: 0.0624 +Epoch [67/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0744, Loss2: 0.0731 +Epoch [67/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0607, Loss2: 0.0594 +Epoch [67/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0670, Loss2: 0.0728 +Epoch [67/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0673, Loss2: 0.0707 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 57.1114 % Model2 57.9627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0741, Loss2: 0.0766 +Epoch [68/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 69.5312, Loss1: 0.0614, Loss2: 0.0538 +Epoch [68/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0746, Loss2: 0.0714 +Epoch [68/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0862, Loss2: 0.0881 +Epoch [68/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0764, Loss2: 0.0707 +Epoch [68/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 73.4375, Loss1: 0.0767, Loss2: 0.0686 +Epoch [68/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.0901, Loss2: 0.0941 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 56.6406 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 67.1875, Loss1: 0.0691, Loss2: 0.0602 +Epoch [69/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0896, Loss2: 0.0818 +Epoch [69/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0789, Loss2: 0.0771 +Epoch [69/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0716, Loss2: 0.0669 +Epoch [69/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0674, Loss2: 0.0691 +Epoch [69/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0807, Loss2: 0.0775 +Epoch [69/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 54.6875, Loss1: 0.0541, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 55.0681 % Model2 57.1915 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0711, Loss2: 0.0719 +Epoch [70/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0763, Loss2: 0.0718 +Epoch [70/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0812, Loss2: 0.0774 +Epoch [70/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0786, Loss2: 0.0741 +Epoch [70/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0661, Loss2: 0.0658 +Epoch [70/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0762, Loss2: 0.0757 +Epoch [70/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0687, Loss2: 0.0731 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 56.4704 % Model2 58.1530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0793, Loss2: 0.0814 +Epoch [71/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0831, Loss2: 0.0911 +Epoch [71/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0822, Loss2: 0.0821 +Epoch [71/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 64.8438, Loss1: 0.0565, Loss2: 0.0515 +Epoch [71/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0977, Loss2: 0.0990 +Epoch [71/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0722, Loss2: 0.0748 +Epoch [71/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 68.7500, Loss1: 0.0614, Loss2: 0.0552 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 56.5605 % Model2 58.3534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0751, Loss2: 0.0829 +Epoch [72/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 67.1875, Loss1: 0.0717, Loss2: 0.0670 +Epoch [72/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0747, Loss2: 0.0775 +Epoch [72/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.1144, Loss2: 0.1082 +Epoch [72/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0624, Loss2: 0.0593 +Epoch [72/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0982, Loss2: 0.1014 +Epoch [72/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0822, Loss2: 0.0857 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 58.1330 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0671, Loss2: 0.0607 +Epoch [73/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0892, Loss2: 0.0947 +Epoch [73/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0642, Loss2: 0.0639 +Epoch [73/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0750, Loss2: 0.0687 +Epoch [73/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.1354, Loss2: 0.1273 +Epoch [73/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0730, Loss2: 0.0722 +Epoch [73/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0828, Loss2: 0.0848 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 56.5405 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0961, Loss2: 0.0935 +Epoch [74/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0818, Loss2: 0.0771 +Epoch [74/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.9688, Loss1: 0.0836, Loss2: 0.0731 +Epoch [74/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0697, Loss2: 0.0667 +Epoch [74/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0689 +Epoch [74/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0743, Loss2: 0.0740 +Epoch [74/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0618, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 56.4804 % Model2 58.0128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0742, Loss2: 0.0732 +Epoch [75/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0616, Loss2: 0.0647 +Epoch [75/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 57.8125, Loss1: 0.0697, Loss2: 0.0813 +Epoch [75/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0829, Loss2: 0.0848 +Epoch [75/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1190, Loss2: 0.1145 +Epoch [75/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0730, Loss2: 0.0698 +Epoch [75/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0699, Loss2: 0.0765 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 56.5805 % Model2 58.4034 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0695, Loss2: 0.0657 +Epoch [76/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0620, Loss2: 0.0614 +Epoch [76/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0648, Loss2: 0.0648 +Epoch [76/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0684, Loss2: 0.0697 +Epoch [76/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0664, Loss2: 0.0668 +Epoch [76/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0913, Loss2: 0.0918 +Epoch [76/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0658, Loss2: 0.0620 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 56.3902 % Model2 57.0312 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0691, Loss2: 0.0724 +Epoch [77/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0781, Loss2: 0.0818 +Epoch [77/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.0312, Loss1: 0.0694, Loss2: 0.0742 +Epoch [77/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0632, Loss2: 0.0667 +Epoch [77/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0832, Loss2: 0.0846 +Epoch [77/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0615, Loss2: 0.0587 +Epoch [77/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0769, Loss2: 0.0764 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 56.6306 % Model2 58.0329 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0651, Loss2: 0.0672 +Epoch [78/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0786, Loss2: 0.0750 +Epoch [78/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0760, Loss2: 0.0757 +Epoch [78/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0900, Loss2: 0.0845 +Epoch [78/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0668, Loss2: 0.0651 +Epoch [78/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0808, Loss2: 0.0851 +Epoch [78/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0755, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 55.8093 % Model2 57.5821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0808, Loss2: 0.0750 +Epoch [79/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0736, Loss2: 0.0700 +Epoch [79/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0547, Loss2: 0.0537 +Epoch [79/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.1046, Loss2: 0.0996 +Epoch [79/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.9375, Loss1: 0.0663, Loss2: 0.0724 +Epoch [79/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0826, Loss2: 0.0772 +Epoch [79/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0620, Loss2: 0.0612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 56.4203 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0639, Loss2: 0.0617 +Epoch [80/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0571, Loss2: 0.0536 +Epoch [80/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0709, Loss2: 0.0699 +Epoch [80/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0765, Loss2: 0.0714 +Epoch [80/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0846, Loss2: 0.0824 +Epoch [80/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0876, Loss2: 0.0877 +Epoch [80/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0668, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 56.4704 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 56.2500, Loss1: 0.0689, Loss2: 0.0735 +Epoch [81/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0768, Loss2: 0.0792 +Epoch [81/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0651, Loss2: 0.0629 +Epoch [81/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0705, Loss2: 0.0651 +Epoch [81/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0592, Loss2: 0.0533 +Epoch [81/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0855, Loss2: 0.0800 +Epoch [81/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0699, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 55.1683 % Model2 57.1815 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0706, Loss2: 0.0757 +Epoch [82/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.0759, Loss2: 0.0676 +Epoch [82/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0826, Loss2: 0.0748 +Epoch [82/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0704, Loss2: 0.0709 +Epoch [82/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0758, Loss2: 0.0750 +Epoch [82/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0549, Loss2: 0.0518 +Epoch [82/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0735, Loss2: 0.0745 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 56.0296 % Model2 57.2817 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0656, Loss2: 0.0644 +Epoch [83/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0648, Loss2: 0.0649 +Epoch [83/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0694, Loss2: 0.0690 +Epoch [83/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0640, Loss2: 0.0663 +Epoch [83/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0638, Loss2: 0.0656 +Epoch [83/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0622, Loss2: 0.0615 +Epoch [83/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0694, Loss2: 0.0742 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 56.5805 % Model2 57.2917 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0825, Loss2: 0.0843 +Epoch [84/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0667, Loss2: 0.0640 +Epoch [84/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0582, Loss2: 0.0565 +Epoch [84/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0571, Loss2: 0.0529 +Epoch [84/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0712, Loss2: 0.0695 +Epoch [84/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0771, Loss2: 0.0731 +Epoch [84/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0721, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 53.3854 % Model2 56.6807 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0610, Loss2: 0.0591 +Epoch [85/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0705, Loss2: 0.0683 +Epoch [85/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0675, Loss2: 0.0625 +Epoch [85/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0930, Loss2: 0.0955 +Epoch [85/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0797, Loss2: 0.0836 +Epoch [85/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0870, Loss2: 0.0771 +Epoch [85/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0898, Loss2: 0.0849 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 55.9896 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0806, Loss2: 0.0792 +Epoch [86/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.0625, Loss1: 0.0826, Loss2: 0.0901 +Epoch [86/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0704, Loss2: 0.0715 +Epoch [86/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1103, Loss2: 0.1098 +Epoch [86/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0772, Loss2: 0.0771 +Epoch [86/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0637, Loss2: 0.0614 +Epoch [86/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0736, Loss2: 0.0754 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 55.9195 % Model2 57.7424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0833, Loss2: 0.0862 +Epoch [87/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0779, Loss2: 0.0776 +Epoch [87/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0891, Loss2: 0.0915 +Epoch [87/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0780, Loss2: 0.0750 +Epoch [87/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0713, Loss2: 0.0734 +Epoch [87/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0603, Loss2: 0.0576 +Epoch [87/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 55.8794 % Model2 57.7424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 60.9375, Loss1: 0.0704, Loss2: 0.0795 +Epoch [88/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.7812, Loss1: 0.0829, Loss2: 0.0768 +Epoch [88/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0659, Loss2: 0.0587 +Epoch [88/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0682, Loss2: 0.0680 +Epoch [88/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0698, Loss2: 0.0713 +Epoch [88/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0632, Loss2: 0.0670 +Epoch [88/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0675, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 55.8894 % Model2 56.6807 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0730, Loss2: 0.0712 +Epoch [89/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0909, Loss2: 0.0867 +Epoch [89/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0757, Loss2: 0.0732 +Epoch [89/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0754, Loss2: 0.0679 +Epoch [89/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0752, Loss2: 0.0731 +Epoch [89/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0564, Loss2: 0.0579 +Epoch [89/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0802, Loss2: 0.0770 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 55.4087 % Model2 57.3518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0925, Loss2: 0.0839 +Epoch [90/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0864, Loss2: 0.0855 +Epoch [90/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0722, Loss2: 0.0727 +Epoch [90/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1023, Loss2: 0.1021 +Epoch [90/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0701, Loss2: 0.0693 +Epoch [90/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0941, Loss2: 0.0948 +Epoch [90/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0579, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 55.8193 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0643, Loss2: 0.0633 +Epoch [91/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0832, Loss2: 0.0802 +Epoch [91/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0745, Loss2: 0.0835 +Epoch [91/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0646, Loss2: 0.0656 +Epoch [91/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0853, Loss2: 0.0837 +Epoch [91/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0847, Loss2: 0.0943 +Epoch [91/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1019, Loss2: 0.1061 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 56.2500 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 65.6250, Loss1: 0.0777, Loss2: 0.0816 +Epoch [92/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 69.5312, Loss1: 0.0769, Loss2: 0.0661 +Epoch [92/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0701, Loss2: 0.0725 +Epoch [92/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0761, Loss2: 0.0741 +Epoch [92/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 64.8438, Loss1: 0.0620, Loss2: 0.0540 +Epoch [92/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0551, Loss2: 0.0558 +Epoch [92/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0997, Loss2: 0.1062 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 55.5990 % Model2 58.3233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0881, Loss2: 0.0858 +Epoch [93/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0999, Loss2: 0.0977 +Epoch [93/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0685, Loss2: 0.0666 +Epoch [93/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0909, Loss2: 0.0852 +Epoch [93/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0868, Loss2: 0.0848 +Epoch [93/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0741, Loss2: 0.0756 +Epoch [93/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 54.6875, Loss1: 0.0633, Loss2: 0.0684 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 55.8994 % Model2 57.3918 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1082, Loss2: 0.1005 +Epoch [94/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0887, Loss2: 0.1016 +Epoch [94/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1007, Loss2: 0.1049 +Epoch [94/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0699, Loss2: 0.0715 +Epoch [94/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0585, Loss2: 0.0599 +Epoch [94/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 51.5625, Loss1: 0.0488, Loss2: 0.0508 +Epoch [94/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0702, Loss2: 0.0717 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 55.6891 % Model2 57.6422 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0677, Loss2: 0.0755 +Epoch [95/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0580, Loss2: 0.0592 +Epoch [95/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0820, Loss2: 0.0731 +Epoch [95/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0738, Loss2: 0.0735 +Epoch [95/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0695, Loss2: 0.0715 +Epoch [95/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0894, Loss2: 0.0880 +Epoch [95/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1248, Loss2: 0.1248 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 57.1514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0693, Loss2: 0.0619 +Epoch [96/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0903, Loss2: 0.0927 +Epoch [96/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0698, Loss2: 0.0633 +Epoch [96/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0842, Loss2: 0.0758 +Epoch [96/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0922, Loss2: 0.0837 +Epoch [96/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0602, Loss2: 0.0655 +Epoch [96/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0744, Loss2: 0.0783 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 55.1082 % Model2 57.4219 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.1043, Loss2: 0.0981 +Epoch [97/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0676, Loss2: 0.0652 +Epoch [97/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0653, Loss2: 0.0623 +Epoch [97/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0705, Loss2: 0.0733 +Epoch [97/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0638, Loss2: 0.0650 +Epoch [97/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0819, Loss2: 0.0764 +Epoch [97/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0869, Loss2: 0.0805 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 55.4287 % Model2 57.4820 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.1121, Loss2: 0.1058 +Epoch [98/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0712, Loss2: 0.0666 +Epoch [98/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0585, Loss2: 0.0561 +Epoch [98/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0977, Loss2: 0.0952 +Epoch [98/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0914, Loss2: 0.0865 +Epoch [98/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0741, Loss2: 0.0781 +Epoch [98/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0853, Loss2: 0.0734 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 55.4688 % Model2 58.1030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0721, Loss2: 0.0758 +Epoch [99/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0902, Loss2: 0.0842 +Epoch [99/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0736, Loss2: 0.0734 +Epoch [99/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1019, Loss2: 0.1056 +Epoch [99/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0869, Loss2: 0.0890 +Epoch [99/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0620, Loss2: 0.0575 +Epoch [99/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0675, Loss2: 0.0686 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 55.4688 % Model2 57.6823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0569, Loss2: 0.0583 +Epoch [100/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1097, Loss2: 0.1091 +Epoch [100/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 65.6250, Loss1: 0.1002, Loss2: 0.1129 +Epoch [100/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0895, Loss2: 0.0849 +Epoch [100/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0765, Loss2: 0.0698 +Epoch [100/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0725, Loss2: 0.0784 +Epoch [100/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0947, Loss2: 0.0980 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 57.4619 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1152, Loss2: 0.1083 +Epoch [101/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0722, Loss2: 0.0756 +Epoch [101/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.1131, Loss2: 0.1066 +Epoch [101/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0990, Loss2: 0.0940 +Epoch [101/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0754, Loss2: 0.0672 +Epoch [101/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1027, Loss2: 0.1049 +Epoch [101/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0907, Loss2: 0.0861 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 56.1498 % Model2 57.8025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0736, Loss2: 0.0761 +Epoch [102/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0719, Loss2: 0.0715 +Epoch [102/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0868, Loss2: 0.0914 +Epoch [102/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0882, Loss2: 0.0889 +Epoch [102/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0963, Loss2: 0.0922 +Epoch [102/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0708, Loss2: 0.0721 +Epoch [102/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0628, Loss2: 0.0570 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 54.6575 % Model2 57.6623 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0719, Loss2: 0.0727 +Epoch [103/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0901, Loss2: 0.0878 +Epoch [103/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0807, Loss2: 0.0767 +Epoch [103/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0683, Loss2: 0.0695 +Epoch [103/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1169, Loss2: 0.1120 +Epoch [103/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0900, Loss2: 0.0932 +Epoch [103/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0777, Loss2: 0.0778 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 55.2083 % Model2 57.4119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0707, Loss2: 0.0728 +Epoch [104/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1089, Loss2: 0.1051 +Epoch [104/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 66.4062, Loss1: 0.0810, Loss2: 0.0714 +Epoch [104/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0647, Loss2: 0.0655 +Epoch [104/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0784, Loss2: 0.0793 +Epoch [104/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0792, Loss2: 0.0710 +Epoch [104/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0652, Loss2: 0.0677 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 55.3185 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1142, Loss2: 0.1271 +Epoch [105/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1076, Loss2: 0.1087 +Epoch [105/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0691, Loss2: 0.0679 +Epoch [105/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0733, Loss2: 0.0773 +Epoch [105/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0762, Loss2: 0.0766 +Epoch [105/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0631, Loss2: 0.0626 +Epoch [105/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0966, Loss2: 0.0883 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 57.0513 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0584, Loss2: 0.0579 +Epoch [106/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1026, Loss2: 0.0969 +Epoch [106/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0948, Loss2: 0.0897 +Epoch [106/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0901, Loss2: 0.0940 +Epoch [106/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0756, Loss2: 0.0709 +Epoch [106/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0814, Loss2: 0.0807 +Epoch [106/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0880, Loss2: 0.0801 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 55.7692 % Model2 57.3818 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0875, Loss2: 0.0797 +Epoch [107/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0678, Loss2: 0.0686 +Epoch [107/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0765, Loss2: 0.0772 +Epoch [107/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1139, Loss2: 0.1085 +Epoch [107/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.0790, Loss2: 0.0713 +Epoch [107/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0589, Loss2: 0.0566 +Epoch [107/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0853, Loss2: 0.0839 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 56.0397 % Model2 57.2616 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0989, Loss2: 0.0964 +Epoch [108/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0841, Loss2: 0.0876 +Epoch [108/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0757, Loss2: 0.0692 +Epoch [108/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0839, Loss2: 0.0799 +Epoch [108/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.0896, Loss2: 0.0951 +Epoch [108/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0986, Loss2: 0.1006 +Epoch [108/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0806, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 54.8778 % Model2 56.2800 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0940, Loss2: 0.0933 +Epoch [109/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1097, Loss2: 0.1010 +Epoch [109/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0752, Loss2: 0.0713 +Epoch [109/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0906, Loss2: 0.0882 +Epoch [109/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0801, Loss2: 0.0829 +Epoch [109/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0864, Loss2: 0.0856 +Epoch [109/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0886, Loss2: 0.0842 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 55.1182 % Model2 57.0312 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0713, Loss2: 0.0713 +Epoch [110/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0725, Loss2: 0.0659 +Epoch [110/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0842, Loss2: 0.0880 +Epoch [110/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0982, Loss2: 0.0942 +Epoch [110/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0841, Loss2: 0.0855 +Epoch [110/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1091, Loss2: 0.1157 +Epoch [110/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0900, Loss2: 0.0826 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 55.4587 % Model2 57.5321 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0835, Loss2: 0.0829 +Epoch [111/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0818, Loss2: 0.0813 +Epoch [111/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0735, Loss2: 0.0735 +Epoch [111/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0967, Loss2: 0.0984 +Epoch [111/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0834, Loss2: 0.0777 +Epoch [111/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.0938, Loss1: 0.0787, Loss2: 0.0886 +Epoch [111/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1018, Loss2: 0.1094 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 55.1583 % Model2 57.6923 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0659, Loss2: 0.0651 +Epoch [112/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0753, Loss2: 0.0701 +Epoch [112/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0700, Loss2: 0.0703 +Epoch [112/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0746, Loss2: 0.0696 +Epoch [112/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 68.7500, Loss1: 0.0834, Loss2: 0.0731 +Epoch [112/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0721, Loss2: 0.0755 +Epoch [112/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1068, Loss2: 0.1048 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 55.6991 % Model2 57.3618 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1106, Loss2: 0.1027 +Epoch [113/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1048, Loss2: 0.0987 +Epoch [113/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.1155, Loss2: 0.1008 +Epoch [113/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0849, Loss2: 0.0842 +Epoch [113/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0902, Loss2: 0.0822 +Epoch [113/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 76.5625, Loss1: 0.0797, Loss2: 0.0708 +Epoch [113/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0882, Loss2: 0.0874 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 54.6274 % Model2 57.5421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0756, Loss2: 0.0715 +Epoch [114/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0928, Loss2: 0.0898 +Epoch [114/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0834, Loss2: 0.0789 +Epoch [114/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0710, Loss2: 0.0670 +Epoch [114/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0759, Loss2: 0.0785 +Epoch [114/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0814, Loss2: 0.0868 +Epoch [114/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0836, Loss2: 0.0725 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 55.1583 % Model2 57.3818 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0682, Loss2: 0.0653 +Epoch [115/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0861, Loss2: 0.0790 +Epoch [115/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0951, Loss2: 0.0920 +Epoch [115/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0757, Loss2: 0.0790 +Epoch [115/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0810, Loss2: 0.0858 +Epoch [115/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.1040, Loss2: 0.0978 +Epoch [115/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0909, Loss2: 0.0899 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 55.1082 % Model2 57.2115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1038, Loss2: 0.1064 +Epoch [116/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.1165, Loss2: 0.1061 +Epoch [116/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.1004, Loss2: 0.1105 +Epoch [116/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0764, Loss2: 0.0785 +Epoch [116/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0689, Loss2: 0.0653 +Epoch [116/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0801, Loss2: 0.0770 +Epoch [116/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.9688, Loss1: 0.0778, Loss2: 0.0725 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 57.0713 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0928, Loss2: 0.0891 +Epoch [117/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0954, Loss2: 0.1018 +Epoch [117/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0855, Loss2: 0.0827 +Epoch [117/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0648, Loss2: 0.0637 +Epoch [117/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0831, Loss2: 0.0829 +Epoch [117/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0843, Loss2: 0.0883 +Epoch [117/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0776, Loss2: 0.0756 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 55.7091 % Model2 57.6823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 66.4062, Loss1: 0.0748, Loss2: 0.0825 +Epoch [118/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.0869, Loss2: 0.0849 +Epoch [118/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0866, Loss2: 0.0855 +Epoch [118/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1101, Loss2: 0.1162 +Epoch [118/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0760, Loss2: 0.0800 +Epoch [118/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1219, Loss2: 0.1165 +Epoch [118/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0759, Loss2: 0.0731 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 55.6991 % Model2 57.3518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0834, Loss2: 0.0783 +Epoch [119/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0900, Loss2: 0.0837 +Epoch [119/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1074, Loss2: 0.1076 +Epoch [119/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0877, Loss2: 0.0817 +Epoch [119/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0763, Loss2: 0.0736 +Epoch [119/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0764, Loss2: 0.0738 +Epoch [119/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0941, Loss2: 0.0805 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 56.0897 % Model2 57.4920 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0895, Loss2: 0.0777 +Epoch [120/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0904, Loss2: 0.0920 +Epoch [120/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0648, Loss2: 0.0667 +Epoch [120/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0549, Loss2: 0.0539 +Epoch [120/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0997, Loss2: 0.0883 +Epoch [120/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 65.6250, Loss1: 0.0923, Loss2: 0.1027 +Epoch [120/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0951, Loss2: 0.0908 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 54.6775 % Model2 57.2516 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0879, Loss2: 0.0863 +Epoch [121/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0803, Loss2: 0.0768 +Epoch [121/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0898, Loss2: 0.0927 +Epoch [121/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0833, Loss2: 0.0823 +Epoch [121/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0894, Loss2: 0.0857 +Epoch [121/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.1019, Loss2: 0.1025 +Epoch [121/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0629, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 55.0280 % Model2 57.5120 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0733, Loss2: 0.0780 +Epoch [122/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1214, Loss2: 0.1146 +Epoch [122/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1065, Loss2: 0.0997 +Epoch [122/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0980, Loss2: 0.0989 +Epoch [122/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1158, Loss2: 0.1145 +Epoch [122/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1154, Loss2: 0.1019 +Epoch [122/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1050, Loss2: 0.1056 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 57.4519 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0849, Loss2: 0.0808 +Epoch [123/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.1091, Loss2: 0.0918 +Epoch [123/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0769, Loss2: 0.0796 +Epoch [123/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0811, Loss2: 0.0781 +Epoch [123/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0789, Loss2: 0.0755 +Epoch [123/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0828, Loss2: 0.0799 +Epoch [123/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0823, Loss2: 0.0760 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 55.9395 % Model2 57.9527 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1171, Loss2: 0.1088 +Epoch [124/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0992, Loss2: 0.1019 +Epoch [124/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0847, Loss2: 0.0861 +Epoch [124/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1286, Loss2: 0.1348 +Epoch [124/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0831, Loss2: 0.0897 +Epoch [124/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.1005, Loss2: 0.1043 +Epoch [124/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0701, Loss2: 0.0703 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 54.9679 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0821, Loss2: 0.0828 +Epoch [125/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0946, Loss2: 0.0903 +Epoch [125/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1101, Loss2: 0.1083 +Epoch [125/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0894, Loss2: 0.0830 +Epoch [125/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0865, Loss2: 0.0842 +Epoch [125/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.9375, Loss1: 0.0751, Loss2: 0.0656 +Epoch [125/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0949, Loss2: 0.0862 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 55.3486 % Model2 57.4720 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0858, Loss2: 0.0887 +Epoch [126/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1112, Loss2: 0.1098 +Epoch [126/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0714, Loss2: 0.0716 +Epoch [126/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0928, Loss2: 0.0869 +Epoch [126/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0887, Loss2: 0.0870 +Epoch [126/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1115, Loss2: 0.1015 +Epoch [126/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1087, Loss2: 0.1051 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 57.4419 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0953, Loss2: 0.0962 +Epoch [127/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.0859, Loss2: 0.0935 +Epoch [127/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0956, Loss2: 0.0930 +Epoch [127/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0772, Loss2: 0.0729 +Epoch [127/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1440, Loss2: 0.1407 +Epoch [127/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1084, Loss2: 0.1041 +Epoch [127/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.1142, Loss2: 0.0962 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 55.0681 % Model2 57.3818 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1102, Loss2: 0.1078 +Epoch [128/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0778, Loss2: 0.0758 +Epoch [128/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.1145, Loss2: 0.1009 +Epoch [128/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0829, Loss2: 0.0802 +Epoch [128/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1187, Loss2: 0.1110 +Epoch [128/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.1061, Loss2: 0.1060 +Epoch [128/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0736, Loss2: 0.0718 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 55.1783 % Model2 57.5120 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0773, Loss2: 0.0700 +Epoch [129/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0857, Loss2: 0.0909 +Epoch [129/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0826, Loss2: 0.0817 +Epoch [129/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.1044, Loss2: 0.0990 +Epoch [129/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1155, Loss2: 0.1238 +Epoch [129/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0856, Loss2: 0.0895 +Epoch [129/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0939, Loss2: 0.0847 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 54.9179 % Model2 57.2015 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0836, Loss2: 0.0740 +Epoch [130/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0766, Loss2: 0.0713 +Epoch [130/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0821, Loss2: 0.0884 +Epoch [130/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1223, Loss2: 0.1298 +Epoch [130/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0751, Loss2: 0.0733 +Epoch [130/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0949, Loss2: 0.0881 +Epoch [130/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1235, Loss2: 0.1181 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 55.2985 % Model2 57.4018 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1139, Loss2: 0.1124 +Epoch [131/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0781, Loss2: 0.0752 +Epoch [131/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.1270, Loss2: 0.1385 +Epoch [131/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0908, Loss2: 0.0878 +Epoch [131/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 75.7812, Loss1: 0.1102, Loss2: 0.0969 +Epoch [131/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.1036, Loss2: 0.0918 +Epoch [131/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0923, Loss2: 0.0889 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 54.9780 % Model2 56.5705 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0959, Loss2: 0.0893 +Epoch [132/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0843, Loss2: 0.0872 +Epoch [132/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0783, Loss2: 0.0791 +Epoch [132/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0676, Loss2: 0.0690 +Epoch [132/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.0973, Loss2: 0.0931 +Epoch [132/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1143, Loss2: 0.1084 +Epoch [132/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1052, Loss2: 0.1062 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 54.7676 % Model2 57.5421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0845, Loss2: 0.0766 +Epoch [133/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.1631, Loss2: 0.1719 +Epoch [133/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0863, Loss2: 0.0884 +Epoch [133/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0782, Loss2: 0.0861 +Epoch [133/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1165, Loss2: 0.1110 +Epoch [133/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1053, Loss2: 0.1022 +Epoch [133/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1138, Loss2: 0.1147 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 54.6074 % Model2 57.4119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1074, Loss2: 0.1109 +Epoch [134/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.1268, Loss2: 0.1346 +Epoch [134/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0754, Loss2: 0.0730 +Epoch [134/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0875, Loss2: 0.0841 +Epoch [134/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 68.7500, Loss1: 0.0785, Loss2: 0.0682 +Epoch [134/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0913, Loss2: 0.0966 +Epoch [134/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0880, Loss2: 0.0911 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 55.1783 % Model2 57.1514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0759, Loss2: 0.0730 +Epoch [135/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.0825, Loss2: 0.0772 +Epoch [135/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0932, Loss2: 0.0857 +Epoch [135/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1145, Loss2: 0.1175 +Epoch [135/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0838, Loss2: 0.0854 +Epoch [135/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.1005, Loss2: 0.1101 +Epoch [135/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.1062, Loss2: 0.1021 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 55.4888 % Model2 57.4419 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0859, Loss2: 0.0924 +Epoch [136/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1184, Loss2: 0.1213 +Epoch [136/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1393, Loss2: 0.1276 +Epoch [136/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0637, Loss2: 0.0641 +Epoch [136/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1159, Loss2: 0.1155 +Epoch [136/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0966, Loss2: 0.0926 +Epoch [136/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0853, Loss2: 0.0840 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 54.7376 % Model2 57.2516 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1088, Loss2: 0.0972 +Epoch [137/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1209, Loss2: 0.1129 +Epoch [137/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0813, Loss2: 0.0855 +Epoch [137/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0883, Loss2: 0.0865 +Epoch [137/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1197, Loss2: 0.1231 +Epoch [137/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0913, Loss2: 0.0856 +Epoch [137/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0984, Loss2: 0.0979 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 55.0280 % Model2 57.5020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 72.6562, Loss1: 0.0706, Loss2: 0.0626 +Epoch [138/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0941, Loss2: 0.0860 +Epoch [138/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1033, Loss2: 0.1106 +Epoch [138/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1135, Loss2: 0.1100 +Epoch [138/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1407, Loss2: 0.1496 +Epoch [138/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0824, Loss2: 0.0776 +Epoch [138/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1134, Loss2: 0.1157 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 57.5421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0743, Loss2: 0.0754 +Epoch [139/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1040, Loss2: 0.0940 +Epoch [139/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.7812, Loss1: 0.1068, Loss2: 0.1003 +Epoch [139/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0756, Loss2: 0.0760 +Epoch [139/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0965, Loss2: 0.0857 +Epoch [139/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.1059, Loss2: 0.0985 +Epoch [139/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1146, Loss2: 0.1196 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 56.3902 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 79.6875, Loss1: 0.0942, Loss2: 0.0829 +Epoch [140/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0998, Loss2: 0.0901 +Epoch [140/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0803, Loss2: 0.0794 +Epoch [140/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1286, Loss2: 0.1307 +Epoch [140/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1297, Loss2: 0.1326 +Epoch [140/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.0939, Loss2: 0.0865 +Epoch [140/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0756, Loss2: 0.0738 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 54.3269 % Model2 57.0012 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1348, Loss2: 0.1210 +Epoch [141/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.1194, Loss2: 0.1183 +Epoch [141/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.1113, Loss2: 0.0992 +Epoch [141/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1285, Loss2: 0.1352 +Epoch [141/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0738, Loss2: 0.0769 +Epoch [141/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0940, Loss2: 0.0984 +Epoch [141/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1116, Loss2: 0.1082 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 57.0413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1095, Loss2: 0.0973 +Epoch [142/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1323, Loss2: 0.1310 +Epoch [142/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1091, Loss2: 0.1107 +Epoch [142/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1138, Loss2: 0.1068 +Epoch [142/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0869, Loss2: 0.0853 +Epoch [142/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1104, Loss2: 0.1075 +Epoch [142/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.0945, Loss2: 0.0841 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 54.6074 % Model2 56.9812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1015, Loss2: 0.1031 +Epoch [143/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1074, Loss2: 0.0978 +Epoch [143/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0910, Loss2: 0.0883 +Epoch [143/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.1188, Loss2: 0.1106 +Epoch [143/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1055, Loss2: 0.1112 +Epoch [143/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0948, Loss2: 0.0878 +Epoch [143/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1026, Loss2: 0.0980 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 55.2684 % Model2 56.7708 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.0926, Loss2: 0.0885 +Epoch [144/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0900, Loss2: 0.0885 +Epoch [144/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0871, Loss2: 0.0882 +Epoch [144/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0934, Loss2: 0.0979 +Epoch [144/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1199, Loss2: 0.1175 +Epoch [144/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1114, Loss2: 0.1129 +Epoch [144/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1035, Loss2: 0.1028 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 54.8277 % Model2 56.9712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1257, Loss2: 0.1318 +Epoch [145/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1080, Loss2: 0.1092 +Epoch [145/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1258, Loss2: 0.1291 +Epoch [145/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0925, Loss2: 0.0887 +Epoch [145/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1270, Loss2: 0.1180 +Epoch [145/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1110, Loss2: 0.1103 +Epoch [145/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1207, Loss2: 0.1224 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 55.2284 % Model2 57.2216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 77.3438, Loss1: 0.1097, Loss2: 0.1260 +Epoch [146/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1022, Loss2: 0.1020 +Epoch [146/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0873, Loss2: 0.0925 +Epoch [146/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1370, Loss2: 0.1348 +Epoch [146/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1045, Loss2: 0.1041 +Epoch [146/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1030, Loss2: 0.1059 +Epoch [146/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0979, Loss2: 0.0915 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 54.5473 % Model2 56.9211 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.1082, Loss2: 0.1052 +Epoch [147/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 79.6875, Loss1: 0.1331, Loss2: 0.1216 +Epoch [147/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1303, Loss2: 0.1327 +Epoch [147/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 68.7500, Loss1: 0.0787, Loss2: 0.0854 +Epoch [147/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0909, Loss2: 0.0892 +Epoch [147/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1026, Loss2: 0.1013 +Epoch [147/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1157, Loss2: 0.1127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 54.7676 % Model2 57.3518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0856, Loss2: 0.0913 +Epoch [148/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1659, Loss2: 0.1692 +Epoch [148/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0847, Loss2: 0.0859 +Epoch [148/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1100, Loss2: 0.1136 +Epoch [148/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1157, Loss2: 0.1309 +Epoch [148/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.1875, Loss1: 0.1100, Loss2: 0.1146 +Epoch [148/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0982, Loss2: 0.0955 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 54.5873 % Model2 57.2015 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1099, Loss2: 0.0996 +Epoch [149/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0798, Loss2: 0.0803 +Epoch [149/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1045, Loss2: 0.1037 +Epoch [149/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1219, Loss2: 0.1123 +Epoch [149/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 72.6562, Loss1: 0.0851, Loss2: 0.0732 +Epoch [149/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1096, Loss2: 0.1104 +Epoch [149/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0879, Loss2: 0.0889 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 54.5773 % Model2 56.9611 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0930, Loss2: 0.0919 +Epoch [150/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 77.3438, Loss1: 0.1245, Loss2: 0.1071 +Epoch [150/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1044, Loss2: 0.1055 +Epoch [150/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1003, Loss2: 0.1054 +Epoch [150/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0806, Loss2: 0.0827 +Epoch [150/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1002, Loss2: 0.1054 +Epoch [150/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.1003, Loss2: 0.1056 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 54.7676 % Model2 57.3317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.1303, Loss2: 0.1120 +Epoch [151/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0891, Loss2: 0.0907 +Epoch [151/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.1232, Loss2: 0.1148 +Epoch [151/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1363, Loss2: 0.1227 +Epoch [151/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1172, Loss2: 0.1165 +Epoch [151/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0858, Loss2: 0.0790 +Epoch [151/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1332, Loss2: 0.1227 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 54.4171 % Model2 57.3918 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1363, Loss2: 0.1367 +Epoch [152/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 69.5312, Loss1: 0.0795, Loss2: 0.0682 +Epoch [152/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0943, Loss2: 0.0973 +Epoch [152/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1112, Loss2: 0.1166 +Epoch [152/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1037, Loss2: 0.1038 +Epoch [152/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1057, Loss2: 0.0990 +Epoch [152/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 83.5938, Loss1: 0.1985, Loss2: 0.1612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 54.8578 % Model2 57.3618 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1015, Loss2: 0.0916 +Epoch [153/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1842, Loss2: 0.1970 +Epoch [153/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0952, Loss2: 0.0950 +Epoch [153/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1187, Loss2: 0.1151 +Epoch [153/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0959, Loss2: 0.0929 +Epoch [153/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0914, Loss2: 0.0886 +Epoch [153/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1767, Loss2: 0.1728 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 54.8978 % Model2 57.0613 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0984, Loss2: 0.0987 +Epoch [154/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1411, Loss2: 0.1368 +Epoch [154/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1429, Loss2: 0.1311 +Epoch [154/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1320, Loss2: 0.1271 +Epoch [154/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.1084, Loss2: 0.1122 +Epoch [154/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0948, Loss2: 0.0881 +Epoch [154/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0937, Loss2: 0.0913 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 54.4071 % Model2 57.0513 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1497, Loss2: 0.1355 +Epoch [155/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1434, Loss2: 0.1399 +Epoch [155/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0787, Loss2: 0.0819 +Epoch [155/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1159, Loss2: 0.1212 +Epoch [155/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1452, Loss2: 0.1453 +Epoch [155/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1063, Loss2: 0.1021 +Epoch [155/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1171, Loss2: 0.1127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 54.7476 % Model2 56.6907 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.1324, Loss2: 0.1377 +Epoch [156/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.1458, Loss2: 0.1383 +Epoch [156/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.1575, Loss2: 0.1682 +Epoch [156/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1025, Loss2: 0.1010 +Epoch [156/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0992, Loss2: 0.0971 +Epoch [156/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0989, Loss2: 0.0916 +Epoch [156/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1262, Loss2: 0.1279 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 54.9379 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1006, Loss2: 0.1090 +Epoch [157/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1178, Loss2: 0.1221 +Epoch [157/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 75.7812, Loss1: 0.0936, Loss2: 0.0834 +Epoch [157/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1826, Loss2: 0.1766 +Epoch [157/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0883, Loss2: 0.0911 +Epoch [157/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1032, Loss2: 0.1090 +Epoch [157/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1704, Loss2: 0.1720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 54.7075 % Model2 57.0513 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1027, Loss2: 0.1131 +Epoch [158/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0974, Loss2: 0.0988 +Epoch [158/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1130, Loss2: 0.0996 +Epoch [158/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1090, Loss2: 0.1176 +Epoch [158/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.1106, Loss2: 0.1152 +Epoch [158/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1116, Loss2: 0.1164 +Epoch [158/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1323, Loss2: 0.1297 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 54.1867 % Model2 56.6406 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0939, Loss2: 0.0957 +Epoch [159/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1048, Loss2: 0.1132 +Epoch [159/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.1144, Loss2: 0.1211 +Epoch [159/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0988, Loss2: 0.0960 +Epoch [159/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1176, Loss2: 0.1080 +Epoch [159/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0887, Loss2: 0.0928 +Epoch [159/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.1445, Loss2: 0.1649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 54.2368 % Model2 57.0012 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 78.1250, Loss1: 0.0878, Loss2: 0.0785 +Epoch [160/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1048, Loss2: 0.0995 +Epoch [160/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.1481, Loss2: 0.1275 +Epoch [160/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0976, Loss2: 0.0966 +Epoch [160/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1342, Loss2: 0.1303 +Epoch [160/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1097, Loss2: 0.1047 +Epoch [160/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1064, Loss2: 0.0995 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 54.9079 % Model2 56.8209 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1229, Loss2: 0.1208 +Epoch [161/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0959, Loss2: 0.0871 +Epoch [161/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.1212, Loss2: 0.1121 +Epoch [161/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1003, Loss2: 0.1092 +Epoch [161/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.7812, Loss1: 0.1097, Loss2: 0.1144 +Epoch [161/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0954, Loss2: 0.0997 +Epoch [161/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.1074, Loss2: 0.0981 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 55.0080 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.1208, Loss2: 0.1029 +Epoch [162/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1279, Loss2: 0.1239 +Epoch [162/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1542, Loss2: 0.1543 +Epoch [162/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.1110, Loss2: 0.1036 +Epoch [162/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0868, Loss2: 0.0786 +Epoch [162/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1130, Loss2: 0.1192 +Epoch [162/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0849, Loss2: 0.0873 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 54.5673 % Model2 57.2115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.1120, Loss2: 0.0977 +Epoch [163/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1302, Loss2: 0.1356 +Epoch [163/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1129, Loss2: 0.1101 +Epoch [163/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1120, Loss2: 0.1089 +Epoch [163/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.0927, Loss2: 0.0902 +Epoch [163/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0862, Loss2: 0.0902 +Epoch [163/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1491, Loss2: 0.1418 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 54.7476 % Model2 56.7909 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1080, Loss2: 0.1079 +Epoch [164/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1597, Loss2: 0.1631 +Epoch [164/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1252, Loss2: 0.1280 +Epoch [164/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0948, Loss2: 0.0971 +Epoch [164/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1517, Loss2: 0.1620 +Epoch [164/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1015, Loss2: 0.1013 +Epoch [164/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1231, Loss2: 0.1228 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 54.6374 % Model2 56.6707 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1072, Loss2: 0.1194 +Epoch [165/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.1305, Loss2: 0.1346 +Epoch [165/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1244, Loss2: 0.1176 +Epoch [165/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.1026, Loss2: 0.1136 +Epoch [165/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.1071, Loss2: 0.0904 +Epoch [165/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1219, Loss2: 0.1224 +Epoch [165/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1322, Loss2: 0.1340 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 54.6374 % Model2 56.5805 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1120, Loss2: 0.1021 +Epoch [166/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0843, Loss2: 0.0930 +Epoch [166/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1549, Loss2: 0.1411 +Epoch [166/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1336, Loss2: 0.1430 +Epoch [166/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.1408, Loss2: 0.1299 +Epoch [166/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0846, Loss2: 0.0817 +Epoch [166/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 67.9688, Loss1: 0.1046, Loss2: 0.1145 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 54.3870 % Model2 56.7208 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0985, Loss2: 0.1019 +Epoch [167/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1435, Loss2: 0.1416 +Epoch [167/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1328, Loss2: 0.1271 +Epoch [167/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1355, Loss2: 0.1240 +Epoch [167/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1107, Loss2: 0.1054 +Epoch [167/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1405, Loss2: 0.1438 +Epoch [167/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1019, Loss2: 0.1021 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 54.7376 % Model2 56.8309 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1318, Loss2: 0.1341 +Epoch [168/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1667, Loss2: 0.1403 +Epoch [168/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1032, Loss2: 0.1104 +Epoch [168/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1169, Loss2: 0.1130 +Epoch [168/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1490, Loss2: 0.1620 +Epoch [168/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1011, Loss2: 0.1012 +Epoch [168/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1217, Loss2: 0.1333 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 54.5172 % Model2 57.1815 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1290, Loss2: 0.1431 +Epoch [169/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0846, Loss2: 0.0857 +Epoch [169/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1181, Loss2: 0.1164 +Epoch [169/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1170, Loss2: 0.1157 +Epoch [169/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0894, Loss2: 0.0913 +Epoch [169/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0966, Loss2: 0.1009 +Epoch [169/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1347, Loss2: 0.1457 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 54.4972 % Model2 56.9511 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1886, Loss2: 0.1941 +Epoch [170/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1333, Loss2: 0.1400 +Epoch [170/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1205, Loss2: 0.1222 +Epoch [170/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0814, Loss2: 0.0797 +Epoch [170/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1396, Loss2: 0.1339 +Epoch [170/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1791, Loss2: 0.1537 +Epoch [170/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1400, Loss2: 0.1323 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 54.5272 % Model2 56.7508 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0868, Loss2: 0.0902 +Epoch [171/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1041, Loss2: 0.0943 +Epoch [171/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 65.6250, Loss1: 0.0755, Loss2: 0.0884 +Epoch [171/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1675, Loss2: 0.1569 +Epoch [171/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1379, Loss2: 0.1343 +Epoch [171/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1421, Loss2: 0.1364 +Epoch [171/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1067, Loss2: 0.1051 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 53.8662 % Model2 57.0513 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.0727, Loss2: 0.0810 +Epoch [172/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1303, Loss2: 0.1240 +Epoch [172/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1413, Loss2: 0.1288 +Epoch [172/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1525, Loss2: 0.1535 +Epoch [172/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1264, Loss2: 0.1328 +Epoch [172/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1139, Loss2: 0.1018 +Epoch [172/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1219, Loss2: 0.1257 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 54.4772 % Model2 56.8309 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1254, Loss2: 0.1187 +Epoch [173/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1003, Loss2: 0.1060 +Epoch [173/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1055, Loss2: 0.1104 +Epoch [173/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1332, Loss2: 0.1306 +Epoch [173/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1262, Loss2: 0.1130 +Epoch [173/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1531, Loss2: 0.1598 +Epoch [173/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1028, Loss2: 0.1058 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 54.6074 % Model2 56.8610 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.1875, Loss1: 0.0910, Loss2: 0.1044 +Epoch [174/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1391, Loss2: 0.1421 +Epoch [174/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1160, Loss2: 0.1272 +Epoch [174/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1115, Loss2: 0.1117 +Epoch [174/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1164, Loss2: 0.1146 +Epoch [174/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1868, Loss2: 0.1768 +Epoch [174/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0936, Loss2: 0.0895 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 54.4571 % Model2 57.1014 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0893, Loss2: 0.0934 +Epoch [175/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1417, Loss2: 0.1419 +Epoch [175/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 78.1250, Loss1: 0.1001, Loss2: 0.0904 +Epoch [175/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1078, Loss2: 0.1081 +Epoch [175/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.1103, Loss2: 0.1015 +Epoch [175/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1043, Loss2: 0.1072 +Epoch [175/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1561, Loss2: 0.1590 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 54.4972 % Model2 56.8309 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0997, Loss2: 0.0955 +Epoch [176/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0880, Loss2: 0.0901 +Epoch [176/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1238, Loss2: 0.1305 +Epoch [176/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1641, Loss2: 0.1687 +Epoch [176/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.1504, Loss2: 0.1698 +Epoch [176/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1551, Loss2: 0.1820 +Epoch [176/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1622, Loss2: 0.1650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 54.2167 % Model2 56.7608 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0861, Loss2: 0.0886 +Epoch [177/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 80.4688, Loss1: 0.1867, Loss2: 0.1606 +Epoch [177/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0812, Loss2: 0.0757 +Epoch [177/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1846, Loss2: 0.1774 +Epoch [177/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.1202, Loss2: 0.1366 +Epoch [177/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 73.4375, Loss1: 0.1009, Loss2: 0.1138 +Epoch [177/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1161, Loss2: 0.1091 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 54.1567 % Model2 56.6106 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0928, Loss2: 0.1028 +Epoch [178/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.1544, Loss2: 0.1419 +Epoch [178/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.1235, Loss2: 0.1273 +Epoch [178/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1090, Loss2: 0.1089 +Epoch [178/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1051, Loss2: 0.0981 +Epoch [178/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.1454, Loss2: 0.1448 +Epoch [178/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1902, Loss2: 0.1835 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 54.5172 % Model2 56.7608 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1408, Loss2: 0.1301 +Epoch [179/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0909, Loss2: 0.0892 +Epoch [179/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1127, Loss2: 0.1195 +Epoch [179/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.2427, Loss2: 0.2425 +Epoch [179/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1141, Loss2: 0.1123 +Epoch [179/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1860, Loss2: 0.1898 +Epoch [179/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1126, Loss2: 0.1100 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 54.5272 % Model2 56.8109 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1298, Loss2: 0.1379 +Epoch [180/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 73.4375, Loss1: 0.1204, Loss2: 0.1438 +Epoch [180/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 74.2188, Loss1: 0.1455, Loss2: 0.1265 +Epoch [180/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1378, Loss2: 0.1417 +Epoch [180/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.1480, Loss2: 0.1289 +Epoch [180/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1336, Loss2: 0.1425 +Epoch [180/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1167, Loss2: 0.1156 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 54.5473 % Model2 56.8309 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0925, Loss2: 0.0897 +Epoch [181/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.1637, Loss2: 0.1907 +Epoch [181/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1539, Loss2: 0.1679 +Epoch [181/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1188, Loss2: 0.1177 +Epoch [181/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1827, Loss2: 0.1737 +Epoch [181/200], Iter [300/390] Training Accuracy1: 81.2500, Training Accuracy2: 77.3438, Loss1: 0.1163, Loss2: 0.1324 +Epoch [181/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1019, Loss2: 0.1048 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 54.3670 % Model2 56.7408 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1295, Loss2: 0.1337 +Epoch [182/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1289, Loss2: 0.1277 +Epoch [182/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.2158, Loss2: 0.2202 +Epoch [182/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1642, Loss2: 0.1681 +Epoch [182/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 73.4375, Loss1: 0.1225, Loss2: 0.1445 +Epoch [182/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1534, Loss2: 0.1559 +Epoch [182/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1190, Loss2: 0.1107 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 54.3369 % Model2 56.4804 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1351, Loss2: 0.1271 +Epoch [183/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1089, Loss2: 0.1070 +Epoch [183/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1193, Loss2: 0.1262 +Epoch [183/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1137, Loss2: 0.1116 +Epoch [183/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1433, Loss2: 0.1287 +Epoch [183/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1451, Loss2: 0.1421 +Epoch [183/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1100, Loss2: 0.1063 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 54.2568 % Model2 56.4804 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1532, Loss2: 0.1485 +Epoch [184/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1448, Loss2: 0.1337 +Epoch [184/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1497, Loss2: 0.1466 +Epoch [184/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 77.3438, Loss1: 0.1739, Loss2: 0.2095 +Epoch [184/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.1311, Loss2: 0.1454 +Epoch [184/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1063, Loss2: 0.1014 +Epoch [184/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.1359, Loss2: 0.1132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 54.5974 % Model2 56.5705 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1180, Loss2: 0.1246 +Epoch [185/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.1650, Loss2: 0.1656 +Epoch [185/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1375, Loss2: 0.1339 +Epoch [185/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1098, Loss2: 0.1205 +Epoch [185/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1183, Loss2: 0.1339 +Epoch [185/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1377, Loss2: 0.1399 +Epoch [185/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.1672, Loss2: 0.2003 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 54.4371 % Model2 56.3702 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1922, Loss2: 0.1898 +Epoch [186/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1171, Loss2: 0.1059 +Epoch [186/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1342, Loss2: 0.1433 +Epoch [186/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.1288, Loss2: 0.1341 +Epoch [186/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.1405, Loss2: 0.1223 +Epoch [186/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1443, Loss2: 0.1459 +Epoch [186/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.2124, Loss2: 0.2058 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 54.1767 % Model2 56.4904 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1307, Loss2: 0.1336 +Epoch [187/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0938, Loss2: 0.0945 +Epoch [187/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.1502, Loss2: 0.1680 +Epoch [187/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1775, Loss2: 0.1907 +Epoch [187/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1455, Loss2: 0.1636 +Epoch [187/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1902, Loss2: 0.1936 +Epoch [187/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1209, Loss2: 0.1219 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 54.3369 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1072, Loss2: 0.1031 +Epoch [188/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1582, Loss2: 0.1709 +Epoch [188/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.1118, Loss2: 0.0974 +Epoch [188/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1250, Loss2: 0.1266 +Epoch [188/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.1419, Loss2: 0.1629 +Epoch [188/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.1084, Loss2: 0.1139 +Epoch [188/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.1113, Loss2: 0.1222 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 54.4271 % Model2 56.3401 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1425, Loss2: 0.1501 +Epoch [189/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1909, Loss2: 0.2051 +Epoch [189/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1213, Loss2: 0.1241 +Epoch [189/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1403, Loss2: 0.1417 +Epoch [189/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.1153, Loss2: 0.1178 +Epoch [189/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1282, Loss2: 0.1408 +Epoch [189/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.0945, Loss2: 0.0921 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 54.2067 % Model2 56.3802 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1043, Loss2: 0.0991 +Epoch [190/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1540, Loss2: 0.1362 +Epoch [190/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1087, Loss2: 0.1150 +Epoch [190/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1172, Loss2: 0.1088 +Epoch [190/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1530, Loss2: 0.1446 +Epoch [190/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1160, Loss2: 0.1206 +Epoch [190/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1326, Loss2: 0.1340 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 54.3269 % Model2 56.3802 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1663, Loss2: 0.1629 +Epoch [191/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.1526, Loss2: 0.1307 +Epoch [191/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.1441, Loss2: 0.1303 +Epoch [191/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1712, Loss2: 0.1762 +Epoch [191/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1245, Loss2: 0.1335 +Epoch [191/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1881, Loss2: 0.2146 +Epoch [191/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.1147, Loss2: 0.1068 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 54.1366 % Model2 56.2700 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1371, Loss2: 0.1362 +Epoch [192/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1471, Loss2: 0.1432 +Epoch [192/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1451, Loss2: 0.1379 +Epoch [192/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1309, Loss2: 0.1451 +Epoch [192/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1234, Loss2: 0.1267 +Epoch [192/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1364, Loss2: 0.1293 +Epoch [192/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.2038, Loss2: 0.1827 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 54.1967 % Model2 56.2099 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.1076, Loss2: 0.1205 +Epoch [193/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1397, Loss2: 0.1339 +Epoch [193/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1922, Loss2: 0.1770 +Epoch [193/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1266, Loss2: 0.1297 +Epoch [193/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.2067, Loss2: 0.2143 +Epoch [193/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1641, Loss2: 0.1655 +Epoch [193/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1279, Loss2: 0.1339 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 54.0765 % Model2 56.1799 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1690, Loss2: 0.1451 +Epoch [194/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1309, Loss2: 0.1301 +Epoch [194/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1126, Loss2: 0.1106 +Epoch [194/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1429, Loss2: 0.1417 +Epoch [194/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1339, Loss2: 0.1236 +Epoch [194/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1656, Loss2: 0.1741 +Epoch [194/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1901, Loss2: 0.1893 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 54.0465 % Model2 56.3101 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1742, Loss2: 0.1716 +Epoch [195/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.1545, Loss2: 0.1553 +Epoch [195/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.1933, Loss2: 0.2040 +Epoch [195/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 75.7812, Loss1: 0.1172, Loss2: 0.1333 +Epoch [195/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1171, Loss2: 0.1102 +Epoch [195/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1195, Loss2: 0.1235 +Epoch [195/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0995, Loss2: 0.1091 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 53.9864 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1201, Loss2: 0.1189 +Epoch [196/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1353, Loss2: 0.1370 +Epoch [196/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1155, Loss2: 0.1100 +Epoch [196/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1394, Loss2: 0.1412 +Epoch [196/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1223, Loss2: 0.1207 +Epoch [196/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1447, Loss2: 0.1341 +Epoch [196/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1337, Loss2: 0.1373 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 53.9964 % Model2 56.1699 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1495, Loss2: 0.1546 +Epoch [197/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1401, Loss2: 0.1302 +Epoch [197/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1415, Loss2: 0.1395 +Epoch [197/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 80.4688, Loss1: 0.1407, Loss2: 0.1483 +Epoch [197/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1692, Loss2: 0.1615 +Epoch [197/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1569, Loss2: 0.1530 +Epoch [197/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0868, Loss2: 0.0839 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 53.8862 % Model2 56.1899 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1142, Loss2: 0.1014 +Epoch [198/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1836, Loss2: 0.1831 +Epoch [198/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0935, Loss2: 0.1023 +Epoch [198/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1140, Loss2: 0.1053 +Epoch [198/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 80.4688, Loss1: 0.1541, Loss2: 0.1375 +Epoch [198/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1292, Loss2: 0.1269 +Epoch [198/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1304, Loss2: 0.1301 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 54.0565 % Model2 56.2901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1114, Loss2: 0.1128 +Epoch [199/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1302, Loss2: 0.1337 +Epoch [199/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.1335, Loss2: 0.1327 +Epoch [199/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1054, Loss2: 0.1135 +Epoch [199/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1706, Loss2: 0.2041 +Epoch [199/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.1175, Loss2: 0.1337 +Epoch [199/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.1445, Loss2: 0.1265 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 53.9062 % Model2 56.0897 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1447, Loss2: 0.1298 +Epoch [200/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1261, Loss2: 0.1164 +Epoch [200/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1458, Loss2: 0.1595 +Epoch [200/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1402, Loss2: 0.1275 +Epoch [200/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1423, Loss2: 0.1374 +Epoch [200/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1785, Loss2: 0.1764 +Epoch [200/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.1626, Loss2: 0.1970 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 53.9163 % Model2 56.1599 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_2_4.log b/other_methods/coteaching_plus/coteaching_plus_results/out_2_4.log new file mode 100644 index 0000000..24bc7d6 --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_2_4.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 14.0625, Training Accuracy2: 12.5000, Loss1: 0.0179, Loss2: 0.0179 +Epoch [2/200], Iter [100/390] Training Accuracy1: 21.0938, Training Accuracy2: 21.0938, Loss1: 0.0174, Loss2: 0.0176 +Epoch [2/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 35.1562, Loss1: 0.0156, Loss2: 0.0156 +Epoch [2/200], Iter [200/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0162, Loss2: 0.0164 +Epoch [2/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 24.2188, Loss1: 0.0151, Loss2: 0.0153 +Epoch [2/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0151 +Epoch [2/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 21.8750, Loss1: 0.0160, Loss2: 0.0168 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 30.1482 % Model2 29.6074 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0150, Loss2: 0.0152 +Epoch [3/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.7812, Loss1: 0.0150, Loss2: 0.0157 +Epoch [3/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 27.3438, Loss1: 0.0151, Loss2: 0.0153 +Epoch [3/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0161, Loss2: 0.0164 +Epoch [3/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 30.4688, Loss1: 0.0148, Loss2: 0.0147 +Epoch [3/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0149, Loss2: 0.0149 +Epoch [3/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0144, Loss2: 0.0146 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 37.9307 % Model2 36.1478 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 28.9062, Loss1: 0.0153, Loss2: 0.0155 +Epoch [4/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0137, Loss2: 0.0135 +Epoch [4/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 26.5625, Loss1: 0.0152, Loss2: 0.0156 +Epoch [4/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0150, Loss2: 0.0148 +Epoch [4/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 29.6875, Loss1: 0.0148, Loss2: 0.0148 +Epoch [4/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0147, Loss2: 0.0144 +Epoch [4/200], Iter [350/390] Training Accuracy1: 25.0000, Training Accuracy2: 22.6562, Loss1: 0.0160, Loss2: 0.0163 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 40.3746 % Model2 41.8870 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 31.2500, Loss1: 0.0152, Loss2: 0.0149 +Epoch [5/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0149, Loss2: 0.0147 +Epoch [5/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0143, Loss2: 0.0142 +Epoch [5/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0160, Loss2: 0.0157 +Epoch [5/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0151, Loss2: 0.0147 +Epoch [5/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 29.6875, Loss1: 0.0146, Loss2: 0.0148 +Epoch [5/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0152, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 38.6619 % Model2 38.6418 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0130, Loss2: 0.0135 +Epoch [6/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0128, Loss2: 0.0127 +Epoch [6/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0131 +Epoch [6/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0146, Loss2: 0.0144 +Epoch [6/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0147, Loss2: 0.0144 +Epoch [6/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 34.3750, Loss1: 0.0139, Loss2: 0.0140 +Epoch [6/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0142, Loss2: 0.0141 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 44.8017 % Model2 45.8534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0137, Loss2: 0.0136 +Epoch [7/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0146, Loss2: 0.0143 +Epoch [7/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0133, Loss2: 0.0139 +Epoch [7/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0149, Loss2: 0.0138 +Epoch [7/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0152, Loss2: 0.0146 +Epoch [7/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.1562, Loss1: 0.0136, Loss2: 0.0136 +Epoch [7/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0130, Loss2: 0.0132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 47.6663 % Model2 48.4675 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0133, Loss2: 0.0136 +Epoch [8/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 31.2500, Loss1: 0.0148, Loss2: 0.0147 +Epoch [8/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0129, Loss2: 0.0121 +Epoch [8/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 41.4062, Loss1: 0.0141, Loss2: 0.0134 +Epoch [8/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0130 +Epoch [8/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0151, Loss2: 0.0150 +Epoch [8/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0141, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 47.6462 % Model2 46.0236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0150 +Epoch [9/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0145, Loss2: 0.0144 +Epoch [9/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.0312, Loss1: 0.0155, Loss2: 0.0154 +Epoch [9/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0131, Loss2: 0.0131 +Epoch [9/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0126, Loss2: 0.0118 +Epoch [9/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 44.5312, Loss1: 0.0129, Loss2: 0.0121 +Epoch [9/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0135, Loss2: 0.0136 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 44.9219 % Model2 46.7748 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0126, Loss2: 0.0128 +Epoch [10/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0130, Loss2: 0.0131 +Epoch [10/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0137, Loss2: 0.0132 +Epoch [10/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0134 +Epoch [10/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0140, Loss2: 0.0144 +Epoch [10/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0145, Loss2: 0.0136 +Epoch [10/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0126, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 47.1054 % Model2 48.1771 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0114, Loss2: 0.0120 +Epoch [11/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 40.6250, Loss1: 0.0127, Loss2: 0.0131 +Epoch [11/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0150, Loss2: 0.0138 +Epoch [11/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0124, Loss2: 0.0126 +Epoch [11/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0138, Loss2: 0.0151 +Epoch [11/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.1250, Loss1: 0.0166, Loss2: 0.0162 +Epoch [11/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0136, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 47.6062 % Model2 49.4892 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.8438, Loss1: 0.0117, Loss2: 0.0119 +Epoch [12/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 32.8125, Loss1: 0.0138, Loss2: 0.0146 +Epoch [12/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0125, Loss2: 0.0131 +Epoch [12/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0141, Loss2: 0.0144 +Epoch [12/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0123, Loss2: 0.0130 +Epoch [12/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0110, Loss2: 0.0109 +Epoch [12/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0121, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 48.8682 % Model2 49.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0122, Loss2: 0.0114 +Epoch [13/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0137, Loss2: 0.0137 +Epoch [13/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0137, Loss2: 0.0135 +Epoch [13/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0134 +Epoch [13/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0111, Loss2: 0.0113 +Epoch [13/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0123, Loss2: 0.0117 +Epoch [13/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0135, Loss2: 0.0130 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 49.2188 % Model2 50.7011 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0119 +Epoch [14/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0129 +Epoch [14/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0121, Loss2: 0.0123 +Epoch [14/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0138, Loss2: 0.0133 +Epoch [14/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0146, Loss2: 0.0136 +Epoch [14/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0126 +Epoch [14/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0129, Loss2: 0.0131 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 45.8834 % Model2 48.8381 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 32.0312, Loss1: 0.0123, Loss2: 0.0136 +Epoch [15/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0120, Loss2: 0.0122 +Epoch [15/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0131, Loss2: 0.0128 +Epoch [15/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0119, Loss2: 0.0122 +Epoch [15/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0123, Loss2: 0.0124 +Epoch [15/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0135, Loss2: 0.0135 +Epoch [15/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0119, Loss2: 0.0119 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 50.1703 % Model2 50.6310 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 38.2812, Loss1: 0.0133, Loss2: 0.0132 +Epoch [16/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0118, Loss2: 0.0118 +Epoch [16/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0110, Loss2: 0.0120 +Epoch [16/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0120, Loss2: 0.0130 +Epoch [16/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0120, Loss2: 0.0119 +Epoch [16/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0130, Loss2: 0.0121 +Epoch [16/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0118, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 49.7095 % Model2 49.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0097, Loss2: 0.0093 +Epoch [17/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0129, Loss2: 0.0130 +Epoch [17/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0139, Loss2: 0.0138 +Epoch [17/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0128, Loss2: 0.0126 +Epoch [17/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.8438, Loss1: 0.0119, Loss2: 0.0126 +Epoch [17/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0125, Loss2: 0.0116 +Epoch [17/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 40.6250, Loss1: 0.0120, Loss2: 0.0124 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 51.0317 % Model2 51.1719 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0128, Loss2: 0.0132 +Epoch [18/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0129 +Epoch [18/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 36.7188, Loss1: 0.0121, Loss2: 0.0127 +Epoch [18/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.0625, Loss1: 0.0127, Loss2: 0.0132 +Epoch [18/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 49.2188, Loss1: 0.0114, Loss2: 0.0103 +Epoch [18/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0092, Loss2: 0.0097 +Epoch [18/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0116, Loss2: 0.0111 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 50.0000 % Model2 50.8714 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0114, Loss2: 0.0112 +Epoch [19/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0115, Loss2: 0.0114 +Epoch [19/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0114, Loss2: 0.0103 +Epoch [19/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.8438, Loss1: 0.0119, Loss2: 0.0124 +Epoch [19/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0121, Loss2: 0.0119 +Epoch [19/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0111, Loss2: 0.0108 +Epoch [19/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0142, Loss2: 0.0136 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 50.5308 % Model2 52.1835 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 36.7188, Loss1: 0.0115, Loss2: 0.0116 +Epoch [20/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0148, Loss2: 0.0144 +Epoch [20/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0121, Loss2: 0.0119 +Epoch [20/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0131, Loss2: 0.0132 +Epoch [20/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0105, Loss2: 0.0114 +Epoch [20/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0133, Loss2: 0.0130 +Epoch [20/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0130, Loss2: 0.0127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 53.1350 % Model2 53.4455 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0566, Loss2: 0.0553 +Epoch [21/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0713, Loss2: 0.0700 +Epoch [21/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0491, Loss2: 0.0482 +Epoch [21/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0574, Loss2: 0.0565 +Epoch [21/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0485, Loss2: 0.0483 +Epoch [21/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.1562, Loss1: 0.0532, Loss2: 0.0537 +Epoch [21/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0519, Loss2: 0.0514 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 49.0184 % Model2 49.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0512, Loss2: 0.0521 +Epoch [22/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0582, Loss2: 0.0575 +Epoch [22/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0524, Loss2: 0.0511 +Epoch [22/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0643, Loss2: 0.0636 +Epoch [22/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0558, Loss2: 0.0535 +Epoch [22/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0539, Loss2: 0.0538 +Epoch [22/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0622, Loss2: 0.0637 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 50.2304 % Model2 51.0417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0526, Loss2: 0.0539 +Epoch [23/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0488, Loss2: 0.0475 +Epoch [23/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0605, Loss2: 0.0616 +Epoch [23/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0580, Loss2: 0.0580 +Epoch [23/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0675, Loss2: 0.0667 +Epoch [23/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0611, Loss2: 0.0609 +Epoch [23/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0501, Loss2: 0.0507 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 49.0284 % Model2 51.0016 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0543, Loss2: 0.0550 +Epoch [24/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 48.4375, Loss1: 0.0632, Loss2: 0.0593 +Epoch [24/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0587, Loss2: 0.0579 +Epoch [24/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0553, Loss2: 0.0552 +Epoch [24/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0599, Loss2: 0.0604 +Epoch [24/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0535, Loss2: 0.0532 +Epoch [24/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0519, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 52.6743 % Model2 52.2736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0491, Loss2: 0.0485 +Epoch [25/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0547, Loss2: 0.0545 +Epoch [25/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0664, Loss2: 0.0672 +Epoch [25/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0574, Loss2: 0.0579 +Epoch [25/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0541, Loss2: 0.0550 +Epoch [25/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 35.9375, Loss1: 0.0566, Loss2: 0.0610 +Epoch [25/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0594, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 47.8065 % Model2 49.3590 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0557, Loss2: 0.0557 +Epoch [26/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0552, Loss2: 0.0546 +Epoch [26/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0467, Loss2: 0.0465 +Epoch [26/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0506, Loss2: 0.0504 +Epoch [26/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0633, Loss2: 0.0624 +Epoch [26/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.8750, Loss1: 0.0553, Loss2: 0.0525 +Epoch [26/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0602, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 49.8397 % Model2 49.7095 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0473, Loss2: 0.0482 +Epoch [27/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0551, Loss2: 0.0553 +Epoch [27/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.0625, Loss1: 0.0572, Loss2: 0.0620 +Epoch [27/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0534, Loss2: 0.0553 +Epoch [27/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0651, Loss2: 0.0643 +Epoch [27/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0588, Loss2: 0.0596 +Epoch [27/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0569, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 48.9383 % Model2 49.8297 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0521, Loss2: 0.0528 +Epoch [28/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0492, Loss2: 0.0488 +Epoch [28/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0681, Loss2: 0.0653 +Epoch [28/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0493, Loss2: 0.0489 +Epoch [28/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0540, Loss2: 0.0529 +Epoch [28/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.8750, Loss1: 0.0623, Loss2: 0.0607 +Epoch [28/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.9688, Loss1: 0.0552, Loss2: 0.0598 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 48.9984 % Model2 49.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0677, Loss2: 0.0679 +Epoch [29/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.0625, Loss1: 0.0488, Loss2: 0.0507 +Epoch [29/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0424, Loss2: 0.0437 +Epoch [29/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0538, Loss2: 0.0520 +Epoch [29/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0542, Loss2: 0.0540 +Epoch [29/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0532, Loss2: 0.0529 +Epoch [29/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0591, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 50.5509 % Model2 48.0168 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0602, Loss2: 0.0613 +Epoch [30/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0518, Loss2: 0.0517 +Epoch [30/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0524, Loss2: 0.0534 +Epoch [30/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0510, Loss2: 0.0526 +Epoch [30/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0681, Loss2: 0.0706 +Epoch [30/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0529, Loss2: 0.0532 +Epoch [30/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 42.1875, Loss1: 0.0510, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 51.4924 % Model2 52.4038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0558, Loss2: 0.0561 +Epoch [31/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0528, Loss2: 0.0525 +Epoch [31/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0632, Loss2: 0.0662 +Epoch [31/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0488, Loss2: 0.0493 +Epoch [31/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0435, Loss2: 0.0432 +Epoch [31/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 40.6250, Loss1: 0.0646, Loss2: 0.0681 +Epoch [31/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0557, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 50.4407 % Model2 51.5024 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0620, Loss2: 0.0639 +Epoch [32/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0598, Loss2: 0.0612 +Epoch [32/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.8750, Loss1: 0.0520, Loss2: 0.0542 +Epoch [32/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0524, Loss2: 0.0518 +Epoch [32/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0627, Loss2: 0.0639 +Epoch [32/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0585, Loss2: 0.0611 +Epoch [32/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0566, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 50.0801 % Model2 50.2304 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0590, Loss2: 0.0582 +Epoch [33/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0679, Loss2: 0.0678 +Epoch [33/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0689, Loss2: 0.0684 +Epoch [33/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0636, Loss2: 0.0638 +Epoch [33/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0486, Loss2: 0.0479 +Epoch [33/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0613, Loss2: 0.0606 +Epoch [33/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0683, Loss2: 0.0689 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 50.4407 % Model2 49.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0531, Loss2: 0.0546 +Epoch [34/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0493, Loss2: 0.0513 +Epoch [34/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 41.4062, Loss1: 0.0603, Loss2: 0.0624 +Epoch [34/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0566, Loss2: 0.0576 +Epoch [34/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0477, Loss2: 0.0517 +Epoch [34/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0450, Loss2: 0.0467 +Epoch [34/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 39.8438, Loss1: 0.0521, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 51.5825 % Model2 50.7512 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.8125, Loss1: 0.0624, Loss2: 0.0582 +Epoch [35/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0605, Loss2: 0.0613 +Epoch [35/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0513, Loss2: 0.0517 +Epoch [35/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0514, Loss2: 0.0510 +Epoch [35/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0488, Loss2: 0.0489 +Epoch [35/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0609, Loss2: 0.0598 +Epoch [35/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0618, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 50.1002 % Model2 50.5008 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 48.4375, Loss1: 0.0596, Loss2: 0.0547 +Epoch [36/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0511, Loss2: 0.0526 +Epoch [36/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0683, Loss2: 0.0694 +Epoch [36/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0715, Loss2: 0.0738 +Epoch [36/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0524, Loss2: 0.0545 +Epoch [36/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0603, Loss2: 0.0604 +Epoch [36/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0617, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 49.5192 % Model2 50.3906 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0583, Loss2: 0.0562 +Epoch [37/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0767, Loss2: 0.0791 +Epoch [37/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 38.2812, Loss1: 0.0569, Loss2: 0.0612 +Epoch [37/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0533, Loss2: 0.0543 +Epoch [37/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0579, Loss2: 0.0597 +Epoch [37/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0479, Loss2: 0.0481 +Epoch [37/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0455, Loss2: 0.0440 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 51.2620 % Model2 52.0633 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0515, Loss2: 0.0544 +Epoch [38/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0555, Loss2: 0.0538 +Epoch [38/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0490, Loss2: 0.0476 +Epoch [38/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0529, Loss2: 0.0553 +Epoch [38/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0685, Loss2: 0.0663 +Epoch [38/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 42.9688, Loss1: 0.0528, Loss2: 0.0581 +Epoch [38/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0511, Loss2: 0.0528 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 49.9099 % Model2 50.8413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0614, Loss2: 0.0621 +Epoch [39/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0657, Loss2: 0.0689 +Epoch [39/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0821, Loss2: 0.0790 +Epoch [39/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0574, Loss2: 0.0599 +Epoch [39/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0532, Loss2: 0.0534 +Epoch [39/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.8438, Loss1: 0.0517, Loss2: 0.0546 +Epoch [39/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0564, Loss2: 0.0528 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 51.2220 % Model2 51.1919 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.0938, Loss1: 0.0496, Loss2: 0.0533 +Epoch [40/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0588, Loss2: 0.0579 +Epoch [40/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0574, Loss2: 0.0555 +Epoch [40/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0582, Loss2: 0.0578 +Epoch [40/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0478, Loss2: 0.0490 +Epoch [40/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0662, Loss2: 0.0674 +Epoch [40/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0633, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 50.7212 % Model2 51.2119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0567, Loss2: 0.0585 +Epoch [41/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0565, Loss2: 0.0562 +Epoch [41/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0663, Loss2: 0.0703 +Epoch [41/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 44.5312, Loss1: 0.0524, Loss2: 0.0554 +Epoch [41/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0453, Loss2: 0.0463 +Epoch [41/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0619, Loss2: 0.0645 +Epoch [41/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0513, Loss2: 0.0513 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 48.7280 % Model2 49.0585 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0525, Loss2: 0.0492 +Epoch [42/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0539, Loss2: 0.0520 +Epoch [42/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0607, Loss2: 0.0601 +Epoch [42/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0550, Loss2: 0.0537 +Epoch [42/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0620, Loss2: 0.0610 +Epoch [42/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0558, Loss2: 0.0596 +Epoch [42/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0471, Loss2: 0.0483 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 50.6010 % Model2 50.5609 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0522, Loss2: 0.0538 +Epoch [43/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0515, Loss2: 0.0541 +Epoch [43/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0591, Loss2: 0.0669 +Epoch [43/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0526, Loss2: 0.0534 +Epoch [43/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0537, Loss2: 0.0547 +Epoch [43/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0485, Loss2: 0.0495 +Epoch [43/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0560, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 49.5994 % Model2 49.8798 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 43.7500, Loss1: 0.0512, Loss2: 0.0551 +Epoch [44/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0557, Loss2: 0.0544 +Epoch [44/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0496, Loss2: 0.0500 +Epoch [44/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0525, Loss2: 0.0545 +Epoch [44/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0572, Loss2: 0.0572 +Epoch [44/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0545, Loss2: 0.0565 +Epoch [44/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0665, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 51.0717 % Model2 50.8514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0560, Loss2: 0.0540 +Epoch [45/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0491, Loss2: 0.0501 +Epoch [45/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0573, Loss2: 0.0568 +Epoch [45/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0551, Loss2: 0.0567 +Epoch [45/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0484, Loss2: 0.0493 +Epoch [45/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0588, Loss2: 0.0591 +Epoch [45/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0489, Loss2: 0.0485 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 49.1787 % Model2 49.8197 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0532, Loss2: 0.0554 +Epoch [46/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0612, Loss2: 0.0624 +Epoch [46/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0567, Loss2: 0.0542 +Epoch [46/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0588, Loss2: 0.0575 +Epoch [46/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0602, Loss2: 0.0590 +Epoch [46/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0577, Loss2: 0.0567 +Epoch [46/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 42.1875, Loss1: 0.0462, Loss2: 0.0505 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 50.0401 % Model2 50.9716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0510, Loss2: 0.0526 +Epoch [47/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0777, Loss2: 0.0780 +Epoch [47/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0670, Loss2: 0.0665 +Epoch [47/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0570, Loss2: 0.0579 +Epoch [47/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0575, Loss2: 0.0592 +Epoch [47/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.0000, Loss1: 0.0663, Loss2: 0.0617 +Epoch [47/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0610, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 49.8197 % Model2 49.2688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0531, Loss2: 0.0563 +Epoch [48/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0556, Loss2: 0.0539 +Epoch [48/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0600, Loss2: 0.0609 +Epoch [48/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0680, Loss2: 0.0665 +Epoch [48/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0529, Loss2: 0.0510 +Epoch [48/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0510, Loss2: 0.0509 +Epoch [48/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0516, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 49.7196 % Model2 51.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0564, Loss2: 0.0552 +Epoch [49/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 39.0625, Loss1: 0.0521, Loss2: 0.0569 +Epoch [49/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0595, Loss2: 0.0607 +Epoch [49/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0520, Loss2: 0.0507 +Epoch [49/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0655, Loss2: 0.0691 +Epoch [49/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0617, Loss2: 0.0633 +Epoch [49/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 43.7500, Loss1: 0.0516, Loss2: 0.0578 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 49.0485 % Model2 50.8013 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0533, Loss2: 0.0521 +Epoch [50/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0471, Loss2: 0.0503 +Epoch [50/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0516 +Epoch [50/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.8750, Loss1: 0.0568, Loss2: 0.0629 +Epoch [50/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0609, Loss2: 0.0620 +Epoch [50/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0653, Loss2: 0.0659 +Epoch [50/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0663, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 49.3490 % Model2 49.9900 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0632, Loss2: 0.0635 +Epoch [51/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0615, Loss2: 0.0633 +Epoch [51/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0488, Loss2: 0.0498 +Epoch [51/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0532, Loss2: 0.0557 +Epoch [51/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0555, Loss2: 0.0583 +Epoch [51/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0527, Loss2: 0.0527 +Epoch [51/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 56.2500, Loss1: 0.0629, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 50.6811 % Model2 50.9315 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0539, Loss2: 0.0541 +Epoch [52/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0722, Loss2: 0.0733 +Epoch [52/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0580, Loss2: 0.0609 +Epoch [52/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0640, Loss2: 0.0623 +Epoch [52/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0746, Loss2: 0.0761 +Epoch [52/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0572, Loss2: 0.0565 +Epoch [52/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0562, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 49.3890 % Model2 51.1118 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0659, Loss2: 0.0666 +Epoch [53/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0642, Loss2: 0.0616 +Epoch [53/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0540, Loss2: 0.0529 +Epoch [53/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0540, Loss2: 0.0530 +Epoch [53/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0597, Loss2: 0.0586 +Epoch [53/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0457, Loss2: 0.0478 +Epoch [53/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0561, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 49.7596 % Model2 49.6494 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0656, Loss2: 0.0702 +Epoch [54/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0497, Loss2: 0.0499 +Epoch [54/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0555, Loss2: 0.0580 +Epoch [54/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0622, Loss2: 0.0594 +Epoch [54/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0565, Loss2: 0.0565 +Epoch [54/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0677, Loss2: 0.0720 +Epoch [54/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0642, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 49.0685 % Model2 49.2488 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0654, Loss2: 0.0651 +Epoch [55/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0612, Loss2: 0.0635 +Epoch [55/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0588, Loss2: 0.0580 +Epoch [55/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0633, Loss2: 0.0620 +Epoch [55/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0569, Loss2: 0.0573 +Epoch [55/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 45.3125, Loss1: 0.0514, Loss2: 0.0618 +Epoch [55/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0595, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 49.4591 % Model2 50.6510 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 55.4688, Loss1: 0.0518, Loss2: 0.0477 +Epoch [56/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0604, Loss2: 0.0570 +Epoch [56/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0651, Loss2: 0.0677 +Epoch [56/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0703, Loss2: 0.0728 +Epoch [56/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0460, Loss2: 0.0470 +Epoch [56/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0584, Loss2: 0.0590 +Epoch [56/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0475, Loss2: 0.0495 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 49.6194 % Model2 49.7796 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0586, Loss2: 0.0614 +Epoch [57/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0570, Loss2: 0.0586 +Epoch [57/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0630, Loss2: 0.0650 +Epoch [57/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0647, Loss2: 0.0642 +Epoch [57/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0557, Loss2: 0.0571 +Epoch [57/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0650, Loss2: 0.0691 +Epoch [57/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0514, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 48.3073 % Model2 49.4992 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.0000, Loss1: 0.0501, Loss2: 0.0531 +Epoch [58/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0624, Loss2: 0.0675 +Epoch [58/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0494, Loss2: 0.0526 +Epoch [58/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0460, Loss2: 0.0484 +Epoch [58/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0433, Loss2: 0.0425 +Epoch [58/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0540, Loss2: 0.0538 +Epoch [58/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 43.7500, Loss1: 0.0527, Loss2: 0.0561 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 49.1186 % Model2 50.1803 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0534, Loss2: 0.0556 +Epoch [59/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0509, Loss2: 0.0505 +Epoch [59/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0601, Loss2: 0.0568 +Epoch [59/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0522, Loss2: 0.0516 +Epoch [59/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 50.7812, Loss1: 0.0584, Loss2: 0.0650 +Epoch [59/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0523, Loss2: 0.0507 +Epoch [59/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0660, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 47.6663 % Model2 49.8898 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0641 +Epoch [60/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0619, Loss2: 0.0612 +Epoch [60/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0487, Loss2: 0.0475 +Epoch [60/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.8750, Loss1: 0.0479, Loss2: 0.0514 +Epoch [60/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0526, Loss2: 0.0512 +Epoch [60/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0467, Loss2: 0.0449 +Epoch [60/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0583, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 50.1302 % Model2 49.4091 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0685 +Epoch [61/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0591, Loss2: 0.0594 +Epoch [61/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0503, Loss2: 0.0504 +Epoch [61/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0566, Loss2: 0.0577 +Epoch [61/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0578, Loss2: 0.0585 +Epoch [61/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0540, Loss2: 0.0547 +Epoch [61/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0569, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 48.5076 % Model2 49.4892 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0554, Loss2: 0.0568 +Epoch [62/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0749, Loss2: 0.0791 +Epoch [62/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0673, Loss2: 0.0661 +Epoch [62/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0523, Loss2: 0.0530 +Epoch [62/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0565 +Epoch [62/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0547, Loss2: 0.0546 +Epoch [62/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0572, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 49.3690 % Model2 50.5008 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0571, Loss2: 0.0600 +Epoch [63/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0568, Loss2: 0.0583 +Epoch [63/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0543, Loss2: 0.0521 +Epoch [63/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0555, Loss2: 0.0558 +Epoch [63/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 48.4375, Loss1: 0.0470, Loss2: 0.0487 +Epoch [63/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0659, Loss2: 0.0618 +Epoch [63/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0614, Loss2: 0.0684 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 47.4659 % Model2 48.1971 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0594, Loss2: 0.0616 +Epoch [64/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0571, Loss2: 0.0559 +Epoch [64/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0561, Loss2: 0.0540 +Epoch [64/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0699, Loss2: 0.0698 +Epoch [64/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0457, Loss2: 0.0466 +Epoch [64/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0532, Loss2: 0.0538 +Epoch [64/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0678, Loss2: 0.0663 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 49.3189 % Model2 49.4692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0758, Loss2: 0.0760 +Epoch [65/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0532, Loss2: 0.0567 +Epoch [65/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0565, Loss2: 0.0589 +Epoch [65/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0570, Loss2: 0.0568 +Epoch [65/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 61.7188, Loss1: 0.0620, Loss2: 0.0567 +Epoch [65/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0654, Loss2: 0.0646 +Epoch [65/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.0000, Loss1: 0.0525, Loss2: 0.0556 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 49.5292 % Model2 51.2220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0492, Loss2: 0.0472 +Epoch [66/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0532, Loss2: 0.0547 +Epoch [66/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0550, Loss2: 0.0539 +Epoch [66/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0488, Loss2: 0.0489 +Epoch [66/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0620, Loss2: 0.0623 +Epoch [66/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0547, Loss2: 0.0539 +Epoch [66/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0542, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 48.1771 % Model2 49.2688 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0578, Loss2: 0.0587 +Epoch [67/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0595, Loss2: 0.0601 +Epoch [67/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0532, Loss2: 0.0563 +Epoch [67/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0654, Loss2: 0.0694 +Epoch [67/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0577, Loss2: 0.0590 +Epoch [67/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0536, Loss2: 0.0566 +Epoch [67/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0618, Loss2: 0.0631 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 48.3974 % Model2 49.1486 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0520, Loss2: 0.0501 +Epoch [68/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0590, Loss2: 0.0584 +Epoch [68/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0557, Loss2: 0.0538 +Epoch [68/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0578, Loss2: 0.0619 +Epoch [68/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0582, Loss2: 0.0570 +Epoch [68/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0751, Loss2: 0.0784 +Epoch [68/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 62.5000, Loss1: 0.0696, Loss2: 0.0777 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 47.9267 % Model2 48.4275 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0551, Loss2: 0.0555 +Epoch [69/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0585, Loss2: 0.0625 +Epoch [69/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0636, Loss2: 0.0607 +Epoch [69/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0603, Loss2: 0.0629 +Epoch [69/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0590, Loss2: 0.0575 +Epoch [69/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0658, Loss2: 0.0669 +Epoch [69/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0918, Loss2: 0.1006 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 48.5176 % Model2 48.8081 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0575, Loss2: 0.0562 +Epoch [70/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0490, Loss2: 0.0476 +Epoch [70/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0580, Loss2: 0.0577 +Epoch [70/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0621, Loss2: 0.0659 +Epoch [70/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0620, Loss2: 0.0659 +Epoch [70/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0589, Loss2: 0.0583 +Epoch [70/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0581, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 47.8265 % Model2 50.3506 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0693, Loss2: 0.0705 +Epoch [71/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0598, Loss2: 0.0603 +Epoch [71/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0560, Loss2: 0.0545 +Epoch [71/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0571, Loss2: 0.0554 +Epoch [71/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0557, Loss2: 0.0582 +Epoch [71/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0536, Loss2: 0.0519 +Epoch [71/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0744, Loss2: 0.0735 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 47.9567 % Model2 47.8165 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0633, Loss2: 0.0643 +Epoch [72/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0627, Loss2: 0.0646 +Epoch [72/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0501, Loss2: 0.0510 +Epoch [72/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0667, Loss2: 0.0684 +Epoch [72/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0611, Loss2: 0.0589 +Epoch [72/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0692, Loss2: 0.0675 +Epoch [72/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0596, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 47.7764 % Model2 49.5393 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0544, Loss2: 0.0547 +Epoch [73/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0503, Loss2: 0.0508 +Epoch [73/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0581, Loss2: 0.0600 +Epoch [73/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0592, Loss2: 0.0623 +Epoch [73/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.9062, Loss1: 0.0614, Loss2: 0.0680 +Epoch [73/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0592, Loss2: 0.0618 +Epoch [73/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0534, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 47.7264 % Model2 49.1486 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0598, Loss2: 0.0584 +Epoch [74/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0639, Loss2: 0.0703 +Epoch [74/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0775, Loss2: 0.0783 +Epoch [74/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0651, Loss2: 0.0652 +Epoch [74/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0669, Loss2: 0.0632 +Epoch [74/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0542, Loss2: 0.0557 +Epoch [74/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0488, Loss2: 0.0483 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 48.1170 % Model2 48.9683 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0489, Loss2: 0.0490 +Epoch [75/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0559, Loss2: 0.0532 +Epoch [75/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0685, Loss2: 0.0659 +Epoch [75/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0681 +Epoch [75/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0614, Loss2: 0.0618 +Epoch [75/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0511, Loss2: 0.0519 +Epoch [75/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0584, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 47.9968 % Model2 48.3574 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0749, Loss2: 0.0740 +Epoch [76/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0541, Loss2: 0.0557 +Epoch [76/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0600, Loss2: 0.0596 +Epoch [76/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0600, Loss2: 0.0610 +Epoch [76/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0634, Loss2: 0.0670 +Epoch [76/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0668, Loss2: 0.0661 +Epoch [76/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0514, Loss2: 0.0545 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 48.2873 % Model2 49.0084 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0495, Loss2: 0.0517 +Epoch [77/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0555, Loss2: 0.0526 +Epoch [77/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0541, Loss2: 0.0550 +Epoch [77/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0604, Loss2: 0.0605 +Epoch [77/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0512, Loss2: 0.0499 +Epoch [77/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0552, Loss2: 0.0513 +Epoch [77/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0735, Loss2: 0.0733 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 48.3273 % Model2 48.7881 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0745, Loss2: 0.0799 +Epoch [78/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0664, Loss2: 0.0666 +Epoch [78/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0522, Loss2: 0.0547 +Epoch [78/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0572, Loss2: 0.0538 +Epoch [78/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0623, Loss2: 0.0598 +Epoch [78/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0584, Loss2: 0.0600 +Epoch [78/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0618, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 47.6562 % Model2 49.3790 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0608, Loss2: 0.0666 +Epoch [79/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0522, Loss2: 0.0536 +Epoch [79/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0498, Loss2: 0.0530 +Epoch [79/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0709, Loss2: 0.0795 +Epoch [79/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0595, Loss2: 0.0565 +Epoch [79/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0478, Loss2: 0.0478 +Epoch [79/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 55.4688, Loss1: 0.0673, Loss2: 0.0626 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 47.9067 % Model2 48.5777 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0633, Loss2: 0.0598 +Epoch [80/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0604, Loss2: 0.0614 +Epoch [80/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0536, Loss2: 0.0560 +Epoch [80/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0535, Loss2: 0.0561 +Epoch [80/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0632, Loss2: 0.0628 +Epoch [80/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0609, Loss2: 0.0569 +Epoch [80/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0670, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 48.7179 % Model2 48.9483 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0642, Loss2: 0.0634 +Epoch [81/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 50.7812, Loss1: 0.0637, Loss2: 0.0693 +Epoch [81/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0593, Loss2: 0.0605 +Epoch [81/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0567 +Epoch [81/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0584, Loss2: 0.0601 +Epoch [81/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0604, Loss2: 0.0638 +Epoch [81/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0501, Loss2: 0.0482 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 48.4675 % Model2 48.8582 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0537, Loss2: 0.0540 +Epoch [82/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0622, Loss2: 0.0639 +Epoch [82/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0553, Loss2: 0.0514 +Epoch [82/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0756, Loss2: 0.0791 +Epoch [82/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0483, Loss2: 0.0478 +Epoch [82/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0497, Loss2: 0.0478 +Epoch [82/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0657, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 46.2740 % Model2 47.5962 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0569, Loss2: 0.0589 +Epoch [83/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 54.6875, Loss1: 0.0509, Loss2: 0.0538 +Epoch [83/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0543, Loss2: 0.0535 +Epoch [83/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0542, Loss2: 0.0561 +Epoch [83/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0646, Loss2: 0.0610 +Epoch [83/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 49.2188, Loss1: 0.0391, Loss2: 0.0450 +Epoch [83/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 48.4375, Loss1: 0.0443, Loss2: 0.0490 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 48.7881 % Model2 49.1186 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0595, Loss2: 0.0612 +Epoch [84/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0536, Loss2: 0.0532 +Epoch [84/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0576, Loss2: 0.0610 +Epoch [84/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0554, Loss2: 0.0584 +Epoch [84/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0543, Loss2: 0.0545 +Epoch [84/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0685, Loss2: 0.0664 +Epoch [84/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0623, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 48.5777 % Model2 48.7680 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 61.7188, Loss1: 0.0586, Loss2: 0.0525 +Epoch [85/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0686, Loss2: 0.0727 +Epoch [85/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0585, Loss2: 0.0567 +Epoch [85/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0489, Loss2: 0.0525 +Epoch [85/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0569, Loss2: 0.0576 +Epoch [85/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0550, Loss2: 0.0572 +Epoch [85/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0639, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 48.5677 % Model2 47.5761 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0586, Loss2: 0.0610 +Epoch [86/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0580, Loss2: 0.0576 +Epoch [86/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0720, Loss2: 0.0714 +Epoch [86/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0553, Loss2: 0.0581 +Epoch [86/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0560, Loss2: 0.0549 +Epoch [86/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0621, Loss2: 0.0576 +Epoch [86/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0688, Loss2: 0.0739 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 48.7580 % Model2 47.4659 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0631, Loss2: 0.0667 +Epoch [87/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0584, Loss2: 0.0590 +Epoch [87/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0509, Loss2: 0.0503 +Epoch [87/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0518, Loss2: 0.0472 +Epoch [87/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.0000, Loss1: 0.0601, Loss2: 0.0671 +Epoch [87/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0581, Loss2: 0.0560 +Epoch [87/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0619, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 48.7380 % Model2 48.0569 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0707, Loss2: 0.0687 +Epoch [88/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0575, Loss2: 0.0596 +Epoch [88/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0603, Loss2: 0.0571 +Epoch [88/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0618, Loss2: 0.0584 +Epoch [88/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0691, Loss2: 0.0747 +Epoch [88/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0772, Loss2: 0.0804 +Epoch [88/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0633, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 46.3542 % Model2 48.6979 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0616, Loss2: 0.0630 +Epoch [89/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0648 +Epoch [89/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0605, Loss2: 0.0578 +Epoch [89/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0476, Loss2: 0.0481 +Epoch [89/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0542, Loss2: 0.0545 +Epoch [89/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0496, Loss2: 0.0527 +Epoch [89/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0533, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 48.1370 % Model2 47.4559 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0642, Loss2: 0.0653 +Epoch [90/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0592, Loss2: 0.0549 +Epoch [90/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0540, Loss2: 0.0515 +Epoch [90/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0542, Loss2: 0.0552 +Epoch [90/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0700, Loss2: 0.0743 +Epoch [90/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0590, Loss2: 0.0560 +Epoch [90/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 48.5677 % Model2 48.6078 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0649, Loss2: 0.0700 +Epoch [91/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0648, Loss2: 0.0694 +Epoch [91/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0521, Loss2: 0.0541 +Epoch [91/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0643, Loss2: 0.0637 +Epoch [91/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0646, Loss2: 0.0671 +Epoch [91/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 54.6875, Loss1: 0.0624, Loss2: 0.0645 +Epoch [91/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0683, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 48.5677 % Model2 47.7364 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0644, Loss2: 0.0634 +Epoch [92/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0544, Loss2: 0.0592 +Epoch [92/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0622, Loss2: 0.0621 +Epoch [92/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0721, Loss2: 0.0730 +Epoch [92/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0696, Loss2: 0.0666 +Epoch [92/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0519, Loss2: 0.0548 +Epoch [92/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0479, Loss2: 0.0457 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 48.2171 % Model2 48.7680 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.0938, Loss1: 0.0582, Loss2: 0.0645 +Epoch [93/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0579, Loss2: 0.0576 +Epoch [93/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0521, Loss2: 0.0556 +Epoch [93/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 49.2188, Loss1: 0.0590, Loss2: 0.0645 +Epoch [93/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0629, Loss2: 0.0714 +Epoch [93/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0599, Loss2: 0.0587 +Epoch [93/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0539, Loss2: 0.0536 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 47.7764 % Model2 47.2957 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0530, Loss2: 0.0544 +Epoch [94/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0567, Loss2: 0.0592 +Epoch [94/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0520, Loss2: 0.0545 +Epoch [94/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.0000, Loss1: 0.0609, Loss2: 0.0687 +Epoch [94/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0593, Loss2: 0.0578 +Epoch [94/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0497, Loss2: 0.0476 +Epoch [94/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0553, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 48.1671 % Model2 48.4275 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0742, Loss2: 0.0757 +Epoch [95/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0561, Loss2: 0.0601 +Epoch [95/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0509, Loss2: 0.0549 +Epoch [95/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0709, Loss2: 0.0718 +Epoch [95/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0691, Loss2: 0.0640 +Epoch [95/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0667, Loss2: 0.0633 +Epoch [95/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0574, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 47.9868 % Model2 49.0184 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0597, Loss2: 0.0618 +Epoch [96/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0562, Loss2: 0.0544 +Epoch [96/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0588, Loss2: 0.0610 +Epoch [96/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0665, Loss2: 0.0639 +Epoch [96/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 50.7812, Loss1: 0.0582, Loss2: 0.0619 +Epoch [96/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 51.5625, Loss1: 0.0557, Loss2: 0.0607 +Epoch [96/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0796, Loss2: 0.0759 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 46.1839 % Model2 48.1470 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0676, Loss2: 0.0666 +Epoch [97/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0665, Loss2: 0.0654 +Epoch [97/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0645, Loss2: 0.0621 +Epoch [97/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0671, Loss2: 0.0656 +Epoch [97/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0531, Loss2: 0.0538 +Epoch [97/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0673, Loss2: 0.0632 +Epoch [97/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0521, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 48.1170 % Model2 48.2672 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0677, Loss2: 0.0681 +Epoch [98/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0809, Loss2: 0.0843 +Epoch [98/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0634, Loss2: 0.0619 +Epoch [98/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0702 +Epoch [98/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0467, Loss2: 0.0485 +Epoch [98/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0540, Loss2: 0.0551 +Epoch [98/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0654, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 48.7680 % Model2 47.6062 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0549, Loss2: 0.0587 +Epoch [99/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0867, Loss2: 0.0834 +Epoch [99/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0650, Loss2: 0.0663 +Epoch [99/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0583, Loss2: 0.0612 +Epoch [99/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0773, Loss2: 0.0727 +Epoch [99/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 57.8125, Loss1: 0.0601, Loss2: 0.0693 +Epoch [99/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0777, Loss2: 0.0785 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 47.7564 % Model2 48.0068 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0670, Loss2: 0.0656 +Epoch [100/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0696, Loss2: 0.0646 +Epoch [100/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0642, Loss2: 0.0648 +Epoch [100/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0693, Loss2: 0.0663 +Epoch [100/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0529, Loss2: 0.0537 +Epoch [100/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0533, Loss2: 0.0516 +Epoch [100/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0645, Loss2: 0.0690 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 47.8466 % Model2 47.7764 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0610, Loss2: 0.0613 +Epoch [101/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 47.6562, Loss1: 0.0562, Loss2: 0.0611 +Epoch [101/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.8125, Loss1: 0.0547, Loss2: 0.0605 +Epoch [101/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0530, Loss2: 0.0521 +Epoch [101/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0682, Loss2: 0.0665 +Epoch [101/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0822, Loss2: 0.0776 +Epoch [101/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0561, Loss2: 0.0574 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 47.4559 % Model2 47.8466 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0613, Loss2: 0.0571 +Epoch [102/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0626, Loss2: 0.0603 +Epoch [102/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0543, Loss2: 0.0578 +Epoch [102/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0536, Loss2: 0.0559 +Epoch [102/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.1562, Loss1: 0.0675, Loss2: 0.0795 +Epoch [102/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0535, Loss2: 0.0540 +Epoch [102/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0741, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 48.0869 % Model2 47.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0748, Loss2: 0.0753 +Epoch [103/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0557, Loss2: 0.0540 +Epoch [103/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0585, Loss2: 0.0613 +Epoch [103/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0663, Loss2: 0.0669 +Epoch [103/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0623, Loss2: 0.0598 +Epoch [103/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0597, Loss2: 0.0572 +Epoch [103/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0678, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 47.9067 % Model2 48.6178 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0540, Loss2: 0.0556 +Epoch [104/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0605, Loss2: 0.0653 +Epoch [104/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0698, Loss2: 0.0726 +Epoch [104/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0652, Loss2: 0.0643 +Epoch [104/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0721, Loss2: 0.0779 +Epoch [104/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 45.3125, Loss1: 0.0532, Loss2: 0.0585 +Epoch [104/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0611, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 48.5978 % Model2 49.1186 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0620, Loss2: 0.0670 +Epoch [105/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0847, Loss2: 0.0883 +Epoch [105/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0642, Loss2: 0.0690 +Epoch [105/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0550 +Epoch [105/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0528, Loss2: 0.0552 +Epoch [105/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0617, Loss2: 0.0619 +Epoch [105/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0617, Loss2: 0.0635 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 47.9167 % Model2 48.0569 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 52.3438, Loss1: 0.0510, Loss2: 0.0581 +Epoch [106/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0700, Loss2: 0.0674 +Epoch [106/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0614, Loss2: 0.0586 +Epoch [106/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0666, Loss2: 0.0668 +Epoch [106/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0581, Loss2: 0.0604 +Epoch [106/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0539, Loss2: 0.0523 +Epoch [106/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0497, Loss2: 0.0494 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 46.6146 % Model2 47.6062 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0682 +Epoch [107/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0533, Loss2: 0.0582 +Epoch [107/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0733, Loss2: 0.0712 +Epoch [107/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0675, Loss2: 0.0729 +Epoch [107/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0658, Loss2: 0.0695 +Epoch [107/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0657, Loss2: 0.0688 +Epoch [107/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0802, Loss2: 0.0809 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 47.7664 % Model2 47.8766 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0710, Loss2: 0.0735 +Epoch [108/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0762, Loss2: 0.0789 +Epoch [108/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0574, Loss2: 0.0550 +Epoch [108/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0587, Loss2: 0.0605 +Epoch [108/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0754, Loss2: 0.0798 +Epoch [108/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0694, Loss2: 0.0682 +Epoch [108/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0712, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 47.6262 % Model2 48.1070 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0560, Loss2: 0.0547 +Epoch [109/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0577, Loss2: 0.0603 +Epoch [109/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0635, Loss2: 0.0665 +Epoch [109/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0549, Loss2: 0.0545 +Epoch [109/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0520, Loss2: 0.0542 +Epoch [109/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0910, Loss2: 0.0915 +Epoch [109/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0924, Loss2: 0.0945 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 47.8666 % Model2 48.1671 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0627, Loss2: 0.0633 +Epoch [110/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0694, Loss2: 0.0703 +Epoch [110/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 56.2500, Loss1: 0.0653, Loss2: 0.0697 +Epoch [110/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0734, Loss2: 0.0737 +Epoch [110/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0693, Loss2: 0.0721 +Epoch [110/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0928, Loss2: 0.0886 +Epoch [110/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 60.1562, Loss1: 0.0479, Loss2: 0.0428 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 47.6863 % Model2 47.6663 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0651, Loss2: 0.0696 +Epoch [111/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0530, Loss2: 0.0539 +Epoch [111/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0613, Loss2: 0.0627 +Epoch [111/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0567, Loss2: 0.0613 +Epoch [111/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0542 +Epoch [111/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0762, Loss2: 0.0761 +Epoch [111/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0652, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 47.5962 % Model2 48.3073 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0601, Loss2: 0.0621 +Epoch [112/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0709, Loss2: 0.0708 +Epoch [112/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0647, Loss2: 0.0656 +Epoch [112/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.9062, Loss1: 0.0582, Loss2: 0.0645 +Epoch [112/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0669, Loss2: 0.0654 +Epoch [112/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0554, Loss2: 0.0561 +Epoch [112/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0758, Loss2: 0.0733 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 47.7163 % Model2 48.6579 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0885, Loss2: 0.0876 +Epoch [113/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0581, Loss2: 0.0638 +Epoch [113/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0613, Loss2: 0.0678 +Epoch [113/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0677, Loss2: 0.0754 +Epoch [113/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0745, Loss2: 0.0781 +Epoch [113/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0656, Loss2: 0.0651 +Epoch [113/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0824, Loss2: 0.0812 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 48.2272 % Model2 46.5845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0588, Loss2: 0.0612 +Epoch [114/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0756 +Epoch [114/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0875, Loss2: 0.0929 +Epoch [114/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0536, Loss2: 0.0552 +Epoch [114/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 55.4688, Loss1: 0.0540, Loss2: 0.0577 +Epoch [114/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0753, Loss2: 0.0702 +Epoch [114/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0746, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 47.7163 % Model2 47.2356 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0731, Loss2: 0.0712 +Epoch [115/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0625, Loss2: 0.0639 +Epoch [115/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0617, Loss2: 0.0617 +Epoch [115/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0587, Loss2: 0.0570 +Epoch [115/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 53.9062, Loss1: 0.0502, Loss2: 0.0564 +Epoch [115/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0532, Loss2: 0.0552 +Epoch [115/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0722, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 47.3758 % Model2 47.8666 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0818, Loss2: 0.0875 +Epoch [116/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0802, Loss2: 0.0837 +Epoch [116/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0638, Loss2: 0.0585 +Epoch [116/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0579, Loss2: 0.0603 +Epoch [116/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0577, Loss2: 0.0596 +Epoch [116/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0852, Loss2: 0.0823 +Epoch [116/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0551, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 48.4776 % Model2 47.7664 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0656, Loss2: 0.0724 +Epoch [117/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0804, Loss2: 0.0761 +Epoch [117/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.1875, Loss1: 0.0752, Loss2: 0.0663 +Epoch [117/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0601, Loss2: 0.0625 +Epoch [117/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0662, Loss2: 0.0612 +Epoch [117/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0704, Loss2: 0.0746 +Epoch [117/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0812, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 46.9952 % Model2 47.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0740, Loss2: 0.0776 +Epoch [118/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0748, Loss2: 0.0795 +Epoch [118/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0687, Loss2: 0.0643 +Epoch [118/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0838, Loss2: 0.0939 +Epoch [118/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0679, Loss2: 0.0709 +Epoch [118/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0604, Loss2: 0.0620 +Epoch [118/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.1074, Loss2: 0.1018 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 48.0369 % Model2 47.8165 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.7812, Loss1: 0.0652, Loss2: 0.0694 +Epoch [119/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0931, Loss2: 0.0876 +Epoch [119/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0849, Loss2: 0.0779 +Epoch [119/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0701, Loss2: 0.0766 +Epoch [119/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 53.9062, Loss1: 0.0504, Loss2: 0.0553 +Epoch [119/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0487, Loss2: 0.0521 +Epoch [119/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 55.4688, Loss1: 0.0584, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 47.0453 % Model2 47.0553 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0681, Loss2: 0.0666 +Epoch [120/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0788, Loss2: 0.0727 +Epoch [120/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0748, Loss2: 0.0692 +Epoch [120/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0691, Loss2: 0.0726 +Epoch [120/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0631, Loss2: 0.0597 +Epoch [120/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0779, Loss2: 0.0781 +Epoch [120/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0912, Loss2: 0.0852 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 47.7063 % Model2 47.5661 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0859, Loss2: 0.0751 +Epoch [121/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0734, Loss2: 0.0743 +Epoch [121/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0652, Loss2: 0.0615 +Epoch [121/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0708, Loss2: 0.0700 +Epoch [121/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0661, Loss2: 0.0691 +Epoch [121/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0527, Loss2: 0.0548 +Epoch [121/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0643, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 47.9667 % Model2 47.9267 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0604, Loss2: 0.0638 +Epoch [122/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0747, Loss2: 0.0801 +Epoch [122/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0838, Loss2: 0.0825 +Epoch [122/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0908, Loss2: 0.0871 +Epoch [122/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0513, Loss2: 0.0545 +Epoch [122/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 55.4688, Loss1: 0.0587, Loss2: 0.0618 +Epoch [122/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0583, Loss2: 0.0637 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 47.6062 % Model2 48.4675 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0729, Loss2: 0.0818 +Epoch [123/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0752, Loss2: 0.0753 +Epoch [123/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 56.2500, Loss1: 0.0706, Loss2: 0.0778 +Epoch [123/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 56.2500, Loss1: 0.0574, Loss2: 0.0641 +Epoch [123/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0766, Loss2: 0.0755 +Epoch [123/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0685, Loss2: 0.0724 +Epoch [123/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0605, Loss2: 0.0579 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 46.9551 % Model2 47.4960 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0691, Loss2: 0.0748 +Epoch [124/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0775, Loss2: 0.0748 +Epoch [124/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0675, Loss2: 0.0686 +Epoch [124/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0732, Loss2: 0.0759 +Epoch [124/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0575, Loss2: 0.0581 +Epoch [124/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0710, Loss2: 0.0660 +Epoch [124/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0560, Loss2: 0.0544 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 47.7464 % Model2 47.5160 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0789, Loss2: 0.0752 +Epoch [125/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0696, Loss2: 0.0726 +Epoch [125/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0662, Loss2: 0.0687 +Epoch [125/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0735 +Epoch [125/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0641, Loss2: 0.0653 +Epoch [125/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0821, Loss2: 0.0836 +Epoch [125/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0561, Loss2: 0.0582 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 47.5160 % Model2 47.9067 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0649, Loss2: 0.0634 +Epoch [126/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0856, Loss2: 0.0824 +Epoch [126/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0662, Loss2: 0.0635 +Epoch [126/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0685, Loss2: 0.0673 +Epoch [126/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0639, Loss2: 0.0644 +Epoch [126/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 55.4688, Loss1: 0.0563, Loss2: 0.0624 +Epoch [126/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0641, Loss2: 0.0648 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 47.2155 % Model2 47.8866 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0614, Loss2: 0.0621 +Epoch [127/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0691, Loss2: 0.0707 +Epoch [127/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0675, Loss2: 0.0673 +Epoch [127/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0741, Loss2: 0.0756 +Epoch [127/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0560, Loss2: 0.0591 +Epoch [127/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0678, Loss2: 0.0709 +Epoch [127/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0652, Loss2: 0.0643 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 47.0954 % Model2 47.5962 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0522, Loss2: 0.0556 +Epoch [128/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0652, Loss2: 0.0665 +Epoch [128/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0686, Loss2: 0.0634 +Epoch [128/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0711, Loss2: 0.0753 +Epoch [128/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0702, Loss2: 0.0690 +Epoch [128/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 65.6250, Loss1: 0.0590, Loss2: 0.0519 +Epoch [128/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0674, Loss2: 0.0689 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 46.7348 % Model2 47.4860 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0591, Loss2: 0.0624 +Epoch [129/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0612, Loss2: 0.0595 +Epoch [129/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0598, Loss2: 0.0579 +Epoch [129/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0801, Loss2: 0.0768 +Epoch [129/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0759, Loss2: 0.0763 +Epoch [129/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0630, Loss2: 0.0653 +Epoch [129/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0668, Loss2: 0.0718 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 47.4960 % Model2 47.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0648, Loss2: 0.0620 +Epoch [130/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0895, Loss2: 0.0808 +Epoch [130/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0837, Loss2: 0.0858 +Epoch [130/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.1032, Loss2: 0.1074 +Epoch [130/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0569, Loss2: 0.0624 +Epoch [130/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0706, Loss2: 0.0815 +Epoch [130/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0692, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 46.9551 % Model2 47.7865 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0972, Loss2: 0.0930 +Epoch [131/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0712, Loss2: 0.0687 +Epoch [131/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0806, Loss2: 0.0808 +Epoch [131/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1086, Loss2: 0.1114 +Epoch [131/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0747, Loss2: 0.0694 +Epoch [131/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0738, Loss2: 0.0734 +Epoch [131/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0772, Loss2: 0.0779 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 47.2957 % Model2 48.0970 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 58.5938, Loss1: 0.0558, Loss2: 0.0618 +Epoch [132/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0622, Loss2: 0.0680 +Epoch [132/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0660, Loss2: 0.0622 +Epoch [132/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0692, Loss2: 0.0648 +Epoch [132/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0917, Loss2: 0.0845 +Epoch [132/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0729, Loss2: 0.0743 +Epoch [132/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0818, Loss2: 0.0782 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 47.7464 % Model2 47.4259 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0745, Loss2: 0.0799 +Epoch [133/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0748, Loss2: 0.0711 +Epoch [133/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0635, Loss2: 0.0673 +Epoch [133/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0911, Loss2: 0.0879 +Epoch [133/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0701, Loss2: 0.0691 +Epoch [133/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0789, Loss2: 0.0795 +Epoch [133/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0749, Loss2: 0.0746 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 47.0853 % Model2 47.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0617, Loss2: 0.0614 +Epoch [134/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0790, Loss2: 0.0828 +Epoch [134/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0743, Loss2: 0.0762 +Epoch [134/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0728, Loss2: 0.0712 +Epoch [134/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0805, Loss2: 0.0798 +Epoch [134/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0635, Loss2: 0.0625 +Epoch [134/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0708, Loss2: 0.0683 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 47.2256 % Model2 47.2656 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.1229, Loss2: 0.1280 +Epoch [135/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0709, Loss2: 0.0709 +Epoch [135/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0733, Loss2: 0.0787 +Epoch [135/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0698, Loss2: 0.0671 +Epoch [135/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0732, Loss2: 0.0681 +Epoch [135/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0660, Loss2: 0.0726 +Epoch [135/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0784, Loss2: 0.0794 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 47.0954 % Model2 46.7648 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0801, Loss2: 0.0773 +Epoch [136/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0706, Loss2: 0.0718 +Epoch [136/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0883, Loss2: 0.0919 +Epoch [136/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0744, Loss2: 0.0814 +Epoch [136/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0733, Loss2: 0.0795 +Epoch [136/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0729, Loss2: 0.0677 +Epoch [136/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0766, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 47.2055 % Model2 47.6963 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0773, Loss2: 0.0738 +Epoch [137/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0809, Loss2: 0.0833 +Epoch [137/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0578, Loss2: 0.0608 +Epoch [137/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.0865, Loss2: 0.0984 +Epoch [137/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0843, Loss2: 0.0816 +Epoch [137/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0781, Loss2: 0.0770 +Epoch [137/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0847, Loss2: 0.0891 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 47.2656 % Model2 47.5160 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.8438, Loss1: 0.0748, Loss2: 0.0681 +Epoch [138/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0558, Loss2: 0.0584 +Epoch [138/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0766, Loss2: 0.0768 +Epoch [138/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0724, Loss2: 0.0753 +Epoch [138/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0804, Loss2: 0.0705 +Epoch [138/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0819, Loss2: 0.0922 +Epoch [138/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0831, Loss2: 0.0833 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 46.7648 % Model2 47.3658 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0895, Loss2: 0.0890 +Epoch [139/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0788, Loss2: 0.0761 +Epoch [139/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0716, Loss2: 0.0702 +Epoch [139/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0799, Loss2: 0.0791 +Epoch [139/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0648, Loss2: 0.0681 +Epoch [139/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0758, Loss2: 0.0699 +Epoch [139/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0771, Loss2: 0.0798 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 47.4659 % Model2 47.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0836, Loss2: 0.0822 +Epoch [140/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0698, Loss2: 0.0630 +Epoch [140/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0667, Loss2: 0.0706 +Epoch [140/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0795, Loss2: 0.0840 +Epoch [140/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0634, Loss2: 0.0643 +Epoch [140/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0737, Loss2: 0.0807 +Epoch [140/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0784, Loss2: 0.0775 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 47.4860 % Model2 48.0469 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0702, Loss2: 0.0651 +Epoch [141/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0906, Loss2: 0.0893 +Epoch [141/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0787, Loss2: 0.0783 +Epoch [141/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0663, Loss2: 0.0675 +Epoch [141/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.1007, Loss2: 0.0991 +Epoch [141/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0650, Loss2: 0.0654 +Epoch [141/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0745, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 47.1454 % Model2 47.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0775, Loss2: 0.0754 +Epoch [142/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0727, Loss2: 0.0718 +Epoch [142/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0967, Loss2: 0.1035 +Epoch [142/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0798, Loss2: 0.0826 +Epoch [142/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0664, Loss2: 0.0637 +Epoch [142/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0625, Loss2: 0.0604 +Epoch [142/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0716, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 47.0453 % Model2 47.4960 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0705, Loss2: 0.0686 +Epoch [143/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0778, Loss2: 0.0784 +Epoch [143/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0718, Loss2: 0.0717 +Epoch [143/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0829, Loss2: 0.0785 +Epoch [143/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 61.7188, Loss1: 0.0829, Loss2: 0.0907 +Epoch [143/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 53.1250, Loss1: 0.0580, Loss2: 0.0649 +Epoch [143/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0768, Loss2: 0.0791 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 47.1254 % Model2 46.9151 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0655, Loss2: 0.0691 +Epoch [144/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0672, Loss2: 0.0685 +Epoch [144/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0797, Loss2: 0.0835 +Epoch [144/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.8125, Loss1: 0.0642, Loss2: 0.0708 +Epoch [144/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0709, Loss2: 0.0705 +Epoch [144/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0667, Loss2: 0.0645 +Epoch [144/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0805, Loss2: 0.0793 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 46.7648 % Model2 46.6346 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0595, Loss2: 0.0622 +Epoch [145/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0651, Loss2: 0.0732 +Epoch [145/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0618, Loss2: 0.0618 +Epoch [145/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0738, Loss2: 0.0753 +Epoch [145/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0739, Loss2: 0.0773 +Epoch [145/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.1130, Loss2: 0.1086 +Epoch [145/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0669, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 47.3758 % Model2 47.4760 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0881, Loss2: 0.0855 +Epoch [146/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0833, Loss2: 0.0829 +Epoch [146/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0592, Loss2: 0.0591 +Epoch [146/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0776, Loss2: 0.0782 +Epoch [146/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0690, Loss2: 0.0687 +Epoch [146/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0854, Loss2: 0.0853 +Epoch [146/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 61.7188, Loss1: 0.0581, Loss2: 0.0624 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 46.9451 % Model2 46.9952 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0644, Loss2: 0.0638 +Epoch [147/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0883, Loss2: 0.0888 +Epoch [147/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0795, Loss2: 0.0751 +Epoch [147/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0749, Loss2: 0.0720 +Epoch [147/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0900, Loss2: 0.0900 +Epoch [147/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0870, Loss2: 0.0924 +Epoch [147/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0712, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 47.0553 % Model2 47.0853 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0757, Loss2: 0.0809 +Epoch [148/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0764, Loss2: 0.0808 +Epoch [148/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0796, Loss2: 0.0845 +Epoch [148/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0939, Loss2: 0.0940 +Epoch [148/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.1220, Loss2: 0.1049 +Epoch [148/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0653, Loss2: 0.0650 +Epoch [148/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1217, Loss2: 0.1184 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 47.1154 % Model2 47.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0915, Loss2: 0.0923 +Epoch [149/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0939, Loss2: 0.0976 +Epoch [149/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0697, Loss2: 0.0690 +Epoch [149/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0677, Loss2: 0.0734 +Epoch [149/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0693, Loss2: 0.0635 +Epoch [149/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0653, Loss2: 0.0719 +Epoch [149/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0721, Loss2: 0.0736 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 47.4159 % Model2 47.5160 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0784, Loss2: 0.0798 +Epoch [150/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0601, Loss2: 0.0567 +Epoch [150/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0689, Loss2: 0.0700 +Epoch [150/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0586, Loss2: 0.0588 +Epoch [150/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0780, Loss2: 0.0745 +Epoch [150/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0923, Loss2: 0.0955 +Epoch [150/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0669, Loss2: 0.0698 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 46.9952 % Model2 46.2941 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.0938, Loss1: 0.0876, Loss2: 0.0993 +Epoch [151/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 56.2500, Loss1: 0.0709, Loss2: 0.0793 +Epoch [151/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0977, Loss2: 0.0979 +Epoch [151/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1058, Loss2: 0.1015 +Epoch [151/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0758, Loss2: 0.0764 +Epoch [151/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0767, Loss2: 0.0735 +Epoch [151/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0739, Loss2: 0.0746 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 46.9651 % Model2 47.3257 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0958, Loss2: 0.0965 +Epoch [152/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0721, Loss2: 0.0672 +Epoch [152/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0911, Loss2: 0.0914 +Epoch [152/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0864, Loss2: 0.0890 +Epoch [152/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0803, Loss2: 0.0781 +Epoch [152/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0707, Loss2: 0.0682 +Epoch [152/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1086, Loss2: 0.1043 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 47.0353 % Model2 46.3542 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.1045, Loss2: 0.1138 +Epoch [153/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0762, Loss2: 0.0768 +Epoch [153/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0732, Loss2: 0.0717 +Epoch [153/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0623, Loss2: 0.0616 +Epoch [153/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0869, Loss2: 0.0852 +Epoch [153/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0596, Loss2: 0.0600 +Epoch [153/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.1018, Loss2: 0.0950 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 46.9050 % Model2 47.1955 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0712, Loss2: 0.0749 +Epoch [154/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0973, Loss2: 0.0935 +Epoch [154/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0655, Loss2: 0.0704 +Epoch [154/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0918, Loss2: 0.0930 +Epoch [154/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0764, Loss2: 0.0792 +Epoch [154/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0776, Loss2: 0.0817 +Epoch [154/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0767, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 46.7949 % Model2 47.0353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0791, Loss2: 0.0720 +Epoch [155/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0831, Loss2: 0.0811 +Epoch [155/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0764, Loss2: 0.0778 +Epoch [155/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0747, Loss2: 0.0824 +Epoch [155/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0753, Loss2: 0.0721 +Epoch [155/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.0625, Loss1: 0.0794, Loss2: 0.0917 +Epoch [155/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0701, Loss2: 0.0702 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 46.3942 % Model2 46.8750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0727, Loss2: 0.0736 +Epoch [156/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0697, Loss2: 0.0722 +Epoch [156/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0827, Loss2: 0.0859 +Epoch [156/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0641, Loss2: 0.0655 +Epoch [156/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0877, Loss2: 0.0887 +Epoch [156/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0817, Loss2: 0.0746 +Epoch [156/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0781, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 46.8950 % Model2 47.3057 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0687, Loss2: 0.0675 +Epoch [157/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 56.2500, Loss1: 0.0676, Loss2: 0.0744 +Epoch [157/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0830, Loss2: 0.0840 +Epoch [157/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1085, Loss2: 0.1003 +Epoch [157/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0751, Loss2: 0.0750 +Epoch [157/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0769, Loss2: 0.0740 +Epoch [157/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0784, Loss2: 0.0779 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 46.7047 % Model2 47.3357 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0715, Loss2: 0.0711 +Epoch [158/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1049, Loss2: 0.0996 +Epoch [158/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1041, Loss2: 0.1109 +Epoch [158/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0690, Loss2: 0.0639 +Epoch [158/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0884, Loss2: 0.0936 +Epoch [158/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0925, Loss2: 0.0972 +Epoch [158/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0742, Loss2: 0.0792 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 46.5745 % Model2 47.1054 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0695, Loss2: 0.0684 +Epoch [159/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0842, Loss2: 0.0854 +Epoch [159/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0767, Loss2: 0.0739 +Epoch [159/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0697, Loss2: 0.0731 +Epoch [159/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0672, Loss2: 0.0624 +Epoch [159/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0861, Loss2: 0.0888 +Epoch [159/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0710, Loss2: 0.0656 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 46.8650 % Model2 46.6546 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0764, Loss2: 0.0768 +Epoch [160/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0878, Loss2: 0.0898 +Epoch [160/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0699, Loss2: 0.0715 +Epoch [160/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0905, Loss2: 0.0933 +Epoch [160/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0909, Loss2: 0.0900 +Epoch [160/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.1061, Loss2: 0.0972 +Epoch [160/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0780, Loss2: 0.0809 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 46.9852 % Model2 46.9752 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0767, Loss2: 0.0854 +Epoch [161/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.1029, Loss2: 0.1159 +Epoch [161/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0828, Loss2: 0.0787 +Epoch [161/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.1071, Loss2: 0.1042 +Epoch [161/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0887, Loss2: 0.0912 +Epoch [161/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0829, Loss2: 0.0780 +Epoch [161/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0793, Loss2: 0.0714 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 47.1454 % Model2 47.1354 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0808, Loss2: 0.0793 +Epoch [162/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0846, Loss2: 0.0910 +Epoch [162/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0731, Loss2: 0.0735 +Epoch [162/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0655, Loss2: 0.0683 +Epoch [162/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0715, Loss2: 0.0744 +Epoch [162/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0903, Loss2: 0.0926 +Epoch [162/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0875, Loss2: 0.0944 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 47.0052 % Model2 47.0954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0935, Loss2: 0.0923 +Epoch [163/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0898, Loss2: 0.0906 +Epoch [163/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0877, Loss2: 0.0886 +Epoch [163/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0734, Loss2: 0.0779 +Epoch [163/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0671, Loss2: 0.0671 +Epoch [163/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.0844, Loss2: 0.0899 +Epoch [163/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0675, Loss2: 0.0641 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 46.8149 % Model2 47.2256 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0832, Loss2: 0.0869 +Epoch [164/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0883, Loss2: 0.0927 +Epoch [164/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0811, Loss2: 0.0814 +Epoch [164/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0870, Loss2: 0.0857 +Epoch [164/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0774, Loss2: 0.0745 +Epoch [164/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0808, Loss2: 0.0812 +Epoch [164/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0841, Loss2: 0.0833 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 46.2440 % Model2 47.2456 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0804 +Epoch [165/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0763, Loss2: 0.0799 +Epoch [165/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0877, Loss2: 0.0967 +Epoch [165/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0683, Loss2: 0.0730 +Epoch [165/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0843, Loss2: 0.0846 +Epoch [165/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0894, Loss2: 0.1008 +Epoch [165/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1087, Loss2: 0.1065 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 46.8349 % Model2 46.5445 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0762, Loss2: 0.0779 +Epoch [166/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0940, Loss2: 0.0861 +Epoch [166/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0786, Loss2: 0.0755 +Epoch [166/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0949, Loss2: 0.0966 +Epoch [166/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1096, Loss2: 0.1076 +Epoch [166/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 71.0938, Loss1: 0.0931, Loss2: 0.0818 +Epoch [166/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0877, Loss2: 0.0942 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 46.8450 % Model2 46.9050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0653, Loss2: 0.0637 +Epoch [167/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0792, Loss2: 0.0765 +Epoch [167/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1079, Loss2: 0.0983 +Epoch [167/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0665, Loss2: 0.0620 +Epoch [167/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1011, Loss2: 0.1021 +Epoch [167/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0602, Loss2: 0.0599 +Epoch [167/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0803, Loss2: 0.0754 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 46.6246 % Model2 47.3458 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0820, Loss2: 0.0712 +Epoch [168/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0774, Loss2: 0.0762 +Epoch [168/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0686, Loss2: 0.0711 +Epoch [168/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0972, Loss2: 0.0982 +Epoch [168/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0730, Loss2: 0.0744 +Epoch [168/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 59.3750, Loss1: 0.0694, Loss2: 0.0772 +Epoch [168/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1337, Loss2: 0.1333 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 46.7448 % Model2 46.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0880, Loss2: 0.0801 +Epoch [169/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0925, Loss2: 0.0837 +Epoch [169/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1034, Loss2: 0.1096 +Epoch [169/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1372, Loss2: 0.1488 +Epoch [169/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0810, Loss2: 0.0801 +Epoch [169/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0876, Loss2: 0.0923 +Epoch [169/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0863, Loss2: 0.0906 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 46.9050 % Model2 46.8650 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0994, Loss2: 0.1146 +Epoch [170/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0640, Loss2: 0.0638 +Epoch [170/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0896, Loss2: 0.0946 +Epoch [170/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0796, Loss2: 0.0750 +Epoch [170/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0816, Loss2: 0.0761 +Epoch [170/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0868, Loss2: 0.0821 +Epoch [170/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0928, Loss2: 0.0903 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 46.9151 % Model2 46.9852 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0678, Loss2: 0.0687 +Epoch [171/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0728, Loss2: 0.0729 +Epoch [171/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0861, Loss2: 0.0884 +Epoch [171/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0859, Loss2: 0.0924 +Epoch [171/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0872, Loss2: 0.0806 +Epoch [171/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0718, Loss2: 0.0762 +Epoch [171/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0964, Loss2: 0.0915 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 46.7047 % Model2 47.0353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0829, Loss2: 0.0852 +Epoch [172/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0817, Loss2: 0.0816 +Epoch [172/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0767, Loss2: 0.0738 +Epoch [172/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.1030, Loss2: 0.1116 +Epoch [172/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0927, Loss2: 0.0874 +Epoch [172/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 72.6562, Loss1: 0.0913, Loss2: 0.0768 +Epoch [172/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0805, Loss2: 0.0766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 46.6847 % Model2 46.7548 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0905, Loss2: 0.0851 +Epoch [173/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 66.4062, Loss1: 0.0762, Loss2: 0.0862 +Epoch [173/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0930, Loss2: 0.0934 +Epoch [173/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0783, Loss2: 0.0761 +Epoch [173/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 72.6562, Loss1: 0.0865, Loss2: 0.0928 +Epoch [173/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0760, Loss2: 0.0794 +Epoch [173/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0979, Loss2: 0.0868 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 46.7448 % Model2 46.4944 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0952, Loss2: 0.0843 +Epoch [174/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0824, Loss2: 0.0900 +Epoch [174/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0867, Loss2: 0.0942 +Epoch [174/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0709, Loss2: 0.0794 +Epoch [174/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1124, Loss2: 0.1124 +Epoch [174/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0991, Loss2: 0.0999 +Epoch [174/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0885, Loss2: 0.0944 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 46.4844 % Model2 46.6146 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0914, Loss2: 0.0925 +Epoch [175/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1008, Loss2: 0.1030 +Epoch [175/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.1463, Loss2: 0.1715 +Epoch [175/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0878, Loss2: 0.0901 +Epoch [175/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1049, Loss2: 0.1072 +Epoch [175/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0826, Loss2: 0.0763 +Epoch [175/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0863, Loss2: 0.0960 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 46.3542 % Model2 46.7047 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.0850, Loss2: 0.0958 +Epoch [176/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0786, Loss2: 0.0778 +Epoch [176/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0871, Loss2: 0.0827 +Epoch [176/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0918, Loss2: 0.0934 +Epoch [176/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0879, Loss2: 0.0892 +Epoch [176/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0793, Loss2: 0.0801 +Epoch [176/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0685, Loss2: 0.0686 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 46.4944 % Model2 46.8750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0917, Loss2: 0.0849 +Epoch [177/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1114, Loss2: 0.1102 +Epoch [177/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0836, Loss2: 0.0826 +Epoch [177/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1032, Loss2: 0.1031 +Epoch [177/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0902, Loss2: 0.0924 +Epoch [177/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0804, Loss2: 0.0811 +Epoch [177/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0801, Loss2: 0.0763 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 46.4744 % Model2 46.7147 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0657, Loss2: 0.0632 +Epoch [178/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1058, Loss2: 0.1069 +Epoch [178/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 64.8438, Loss1: 0.0815, Loss2: 0.0913 +Epoch [178/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0818, Loss2: 0.0823 +Epoch [178/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0917, Loss2: 0.0888 +Epoch [178/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0916, Loss2: 0.0966 +Epoch [178/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0654, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 46.2941 % Model2 46.6847 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0936, Loss2: 0.1030 +Epoch [179/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0724, Loss2: 0.0678 +Epoch [179/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0841, Loss2: 0.0796 +Epoch [179/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0771, Loss2: 0.0799 +Epoch [179/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1052, Loss2: 0.1154 +Epoch [179/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.0914, Loss2: 0.0798 +Epoch [179/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.1238, Loss2: 0.1295 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 46.2139 % Model2 46.7047 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0883, Loss2: 0.0810 +Epoch [180/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0868, Loss2: 0.0900 +Epoch [180/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0803, Loss2: 0.0792 +Epoch [180/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0648, Loss2: 0.0686 +Epoch [180/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 64.8438, Loss1: 0.0783, Loss2: 0.0856 +Epoch [180/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1052, Loss2: 0.1048 +Epoch [180/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0833, Loss2: 0.0807 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 46.0136 % Model2 46.6446 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0686, Loss2: 0.0677 +Epoch [181/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1396, Loss2: 0.1414 +Epoch [181/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0837, Loss2: 0.0808 +Epoch [181/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0819, Loss2: 0.0878 +Epoch [181/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1307, Loss2: 0.1258 +Epoch [181/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0651, Loss2: 0.0695 +Epoch [181/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0787, Loss2: 0.0781 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 46.4243 % Model2 46.9050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0958, Loss2: 0.0957 +Epoch [182/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0882, Loss2: 0.0967 +Epoch [182/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1144, Loss2: 0.1094 +Epoch [182/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0788, Loss2: 0.0781 +Epoch [182/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0699, Loss2: 0.0677 +Epoch [182/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0948, Loss2: 0.0947 +Epoch [182/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1006, Loss2: 0.1071 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 46.4543 % Model2 46.8850 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0908, Loss2: 0.0913 +Epoch [183/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0963, Loss2: 0.1097 +Epoch [183/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0810, Loss2: 0.0791 +Epoch [183/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0778, Loss2: 0.0852 +Epoch [183/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0988, Loss2: 0.0917 +Epoch [183/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1415, Loss2: 0.1630 +Epoch [183/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0660, Loss2: 0.0660 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 46.0236 % Model2 46.9651 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0792, Loss2: 0.0757 +Epoch [184/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0761, Loss2: 0.0803 +Epoch [184/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1053, Loss2: 0.1081 +Epoch [184/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0784, Loss2: 0.0830 +Epoch [184/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0603, Loss2: 0.0621 +Epoch [184/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0765, Loss2: 0.0766 +Epoch [184/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0786, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 46.0637 % Model2 47.2256 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1175, Loss2: 0.1138 +Epoch [185/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1053, Loss2: 0.1045 +Epoch [185/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0839, Loss2: 0.0837 +Epoch [185/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0852, Loss2: 0.0912 +Epoch [185/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1006, Loss2: 0.0914 +Epoch [185/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1023, Loss2: 0.0954 +Epoch [185/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0931, Loss2: 0.0933 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 46.0837 % Model2 47.1054 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1566, Loss2: 0.1595 +Epoch [186/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1147, Loss2: 0.1172 +Epoch [186/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.1875, Loss1: 0.1048, Loss2: 0.1130 +Epoch [186/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0862, Loss2: 0.0926 +Epoch [186/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0851, Loss2: 0.0916 +Epoch [186/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0761, Loss2: 0.0743 +Epoch [186/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.1047, Loss2: 0.1149 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 46.2740 % Model2 47.0152 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0989, Loss2: 0.0940 +Epoch [187/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.9688, Loss1: 0.1084, Loss2: 0.1237 +Epoch [187/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0955, Loss2: 0.0852 +Epoch [187/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.0727, Loss2: 0.0818 +Epoch [187/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.1076, Loss2: 0.1181 +Epoch [187/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0960, Loss2: 0.0934 +Epoch [187/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0981, Loss2: 0.0953 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 45.9034 % Model2 46.9151 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0927, Loss2: 0.0930 +Epoch [188/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0785, Loss2: 0.0846 +Epoch [188/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0811, Loss2: 0.0820 +Epoch [188/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0749, Loss2: 0.0759 +Epoch [188/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0795, Loss2: 0.0825 +Epoch [188/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0639, Loss2: 0.0697 +Epoch [188/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.0625, Loss1: 0.0901, Loss2: 0.1007 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 46.0136 % Model2 47.1054 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0770, Loss2: 0.0698 +Epoch [189/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0947, Loss2: 0.0927 +Epoch [189/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0678, Loss2: 0.0706 +Epoch [189/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.1008, Loss2: 0.1006 +Epoch [189/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0810, Loss2: 0.0851 +Epoch [189/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0825, Loss2: 0.0823 +Epoch [189/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0810, Loss2: 0.0809 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 46.2039 % Model2 46.9050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0896, Loss2: 0.0887 +Epoch [190/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1062, Loss2: 0.1072 +Epoch [190/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0837, Loss2: 0.0803 +Epoch [190/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1410, Loss2: 0.1320 +Epoch [190/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1127, Loss2: 0.1047 +Epoch [190/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0985, Loss2: 0.0901 +Epoch [190/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0932, Loss2: 0.0896 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 46.0437 % Model2 46.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0984, Loss2: 0.0962 +Epoch [191/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0769, Loss2: 0.0804 +Epoch [191/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0810, Loss2: 0.0834 +Epoch [191/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0853, Loss2: 0.0923 +Epoch [191/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0793, Loss2: 0.0787 +Epoch [191/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0900, Loss2: 0.0887 +Epoch [191/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0946, Loss2: 0.0853 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 46.2139 % Model2 46.7849 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1048, Loss2: 0.1006 +Epoch [192/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0915, Loss2: 0.0882 +Epoch [192/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1235, Loss2: 0.1120 +Epoch [192/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0746, Loss2: 0.0749 +Epoch [192/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.0925, Loss2: 0.1072 +Epoch [192/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0743, Loss2: 0.0705 +Epoch [192/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0745, Loss2: 0.0741 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 45.9936 % Model2 46.9852 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.1148, Loss2: 0.1121 +Epoch [193/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0865, Loss2: 0.0877 +Epoch [193/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0846, Loss2: 0.0787 +Epoch [193/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0738, Loss2: 0.0717 +Epoch [193/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0750, Loss2: 0.0726 +Epoch [193/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1027, Loss2: 0.0954 +Epoch [193/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0738, Loss2: 0.0687 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 46.0938 % Model2 46.9251 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1045, Loss2: 0.1121 +Epoch [194/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1121, Loss2: 0.1122 +Epoch [194/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.0852, Loss2: 0.0961 +Epoch [194/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0949, Loss2: 0.0917 +Epoch [194/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1025, Loss2: 0.1101 +Epoch [194/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0861, Loss2: 0.0927 +Epoch [194/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.1011, Loss2: 0.1115 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 46.1138 % Model2 46.6847 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0863, Loss2: 0.0830 +Epoch [195/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 66.4062, Loss1: 0.0808, Loss2: 0.0883 +Epoch [195/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1069, Loss2: 0.0942 +Epoch [195/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.0970, Loss2: 0.1016 +Epoch [195/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1323, Loss2: 0.1237 +Epoch [195/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0820, Loss2: 0.0834 +Epoch [195/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0990, Loss2: 0.0983 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 45.7933 % Model2 46.8650 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0900, Loss2: 0.0942 +Epoch [196/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1266, Loss2: 0.1206 +Epoch [196/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 68.7500, Loss1: 0.1002, Loss2: 0.1108 +Epoch [196/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1094, Loss2: 0.1123 +Epoch [196/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0690, Loss2: 0.0720 +Epoch [196/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0950, Loss2: 0.0917 +Epoch [196/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.1121, Loss2: 0.1027 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 45.9535 % Model2 46.9050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1023, Loss2: 0.0951 +Epoch [197/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0941, Loss2: 0.0934 +Epoch [197/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0828, Loss2: 0.0831 +Epoch [197/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.0961, Loss2: 0.0837 +Epoch [197/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 74.2188, Loss1: 0.1173, Loss2: 0.0963 +Epoch [197/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 67.1875, Loss1: 0.0887, Loss2: 0.1025 +Epoch [197/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.9688, Loss1: 0.1062, Loss2: 0.0917 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 45.9736 % Model2 46.6747 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0784, Loss2: 0.0800 +Epoch [198/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0851, Loss2: 0.0825 +Epoch [198/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0795, Loss2: 0.0749 +Epoch [198/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0852, Loss2: 0.0852 +Epoch [198/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1146, Loss2: 0.1238 +Epoch [198/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1202, Loss2: 0.1146 +Epoch [198/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0849, Loss2: 0.0800 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 46.0637 % Model2 46.7047 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1010, Loss2: 0.1049 +Epoch [199/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0946, Loss2: 0.0978 +Epoch [199/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0739, Loss2: 0.0778 +Epoch [199/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0866, Loss2: 0.0831 +Epoch [199/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0810, Loss2: 0.0795 +Epoch [199/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0771, Loss2: 0.0810 +Epoch [199/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1097, Loss2: 0.1078 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 46.1038 % Model2 46.6847 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0977, Loss2: 0.0899 +Epoch [200/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1176, Loss2: 0.1090 +Epoch [200/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0896, Loss2: 0.0931 +Epoch [200/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0949, Loss2: 0.0894 +Epoch [200/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0980, Loss2: 0.0899 +Epoch [200/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0948, Loss2: 0.0869 +Epoch [200/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0846, Loss2: 0.0829 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 46.0637 % Model2 46.7248 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_2_6.log b/other_methods/coteaching_plus/coteaching_plus_results/out_2_6.log new file mode 100644 index 0000000..78bc587 --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_2_6.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 16.4062, Training Accuracy2: 10.1562, Loss1: 0.0179, Loss2: 0.0180 +Epoch [2/200], Iter [100/390] Training Accuracy1: 12.5000, Training Accuracy2: 15.6250, Loss1: 0.0181, Loss2: 0.0180 +Epoch [2/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0171, Loss2: 0.0174 +Epoch [2/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 24.2188, Loss1: 0.0172, Loss2: 0.0173 +Epoch [2/200], Iter [250/390] Training Accuracy1: 21.8750, Training Accuracy2: 20.3125, Loss1: 0.0170, Loss2: 0.0171 +Epoch [2/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.7812, Loss1: 0.0163, Loss2: 0.0165 +Epoch [2/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 18.7500, Loss1: 0.0171, Loss2: 0.0171 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 16.1258 % Model2 17.0573 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.0000, Loss1: 0.0172, Loss2: 0.0174 +Epoch [3/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 19.5312, Loss1: 0.0172, Loss2: 0.0174 +Epoch [3/200], Iter [150/390] Training Accuracy1: 21.0938, Training Accuracy2: 17.9688, Loss1: 0.0171, Loss2: 0.0173 +Epoch [3/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 24.2188, Loss1: 0.0170, Loss2: 0.0171 +Epoch [3/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0161, Loss2: 0.0165 +Epoch [3/200], Iter [300/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.8750, Loss1: 0.0173, Loss2: 0.0174 +Epoch [3/200], Iter [350/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.1250, Loss1: 0.0162, Loss2: 0.0165 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 19.0304 % Model2 18.4495 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 21.0938, Loss1: 0.0170, Loss2: 0.0171 +Epoch [4/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0159, Loss2: 0.0157 +Epoch [4/200], Iter [150/390] Training Accuracy1: 25.0000, Training Accuracy2: 22.6562, Loss1: 0.0168, Loss2: 0.0169 +Epoch [4/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 21.0938, Loss1: 0.0168, Loss2: 0.0169 +Epoch [4/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0159, Loss2: 0.0161 +Epoch [4/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 22.6562, Loss1: 0.0168, Loss2: 0.0168 +Epoch [4/200], Iter [350/390] Training Accuracy1: 16.4062, Training Accuracy2: 17.1875, Loss1: 0.0178, Loss2: 0.0178 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 20.5729 % Model2 20.5629 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 26.5625, Loss1: 0.0159, Loss2: 0.0158 +Epoch [5/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.0938, Loss1: 0.0176, Loss2: 0.0176 +Epoch [5/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0160, Loss2: 0.0160 +Epoch [5/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 25.0000, Loss1: 0.0169, Loss2: 0.0171 +Epoch [5/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 25.0000, Loss1: 0.0162, Loss2: 0.0164 +Epoch [5/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.0000, Loss1: 0.0165, Loss2: 0.0171 +Epoch [5/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 28.9062, Loss1: 0.0162, Loss2: 0.0165 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 25.0100 % Model2 24.8998 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0168, Loss2: 0.0166 +Epoch [6/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.0000, Loss1: 0.0166, Loss2: 0.0168 +Epoch [6/200], Iter [150/390] Training Accuracy1: 26.5625, Training Accuracy2: 29.6875, Loss1: 0.0161, Loss2: 0.0159 +Epoch [6/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 24.2188, Loss1: 0.0176, Loss2: 0.0176 +Epoch [6/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0157, Loss2: 0.0157 +Epoch [6/200], Iter [300/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.0000, Loss1: 0.0170, Loss2: 0.0167 +Epoch [6/200], Iter [350/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.1250, Loss1: 0.0174, Loss2: 0.0175 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 24.9299 % Model2 24.8798 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 21.0938, Training Accuracy2: 25.0000, Loss1: 0.0178, Loss2: 0.0174 +Epoch [7/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 26.5625, Loss1: 0.0169, Loss2: 0.0167 +Epoch [7/200], Iter [150/390] Training Accuracy1: 19.5312, Training Accuracy2: 20.3125, Loss1: 0.0174, Loss2: 0.0173 +Epoch [7/200], Iter [200/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.8750, Loss1: 0.0182, Loss2: 0.0175 +Epoch [7/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 23.4375, Loss1: 0.0169, Loss2: 0.0169 +Epoch [7/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.1250, Loss1: 0.0160, Loss2: 0.0162 +Epoch [7/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0165, Loss2: 0.0169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 22.6663 % Model2 22.0453 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0160, Loss2: 0.0162 +Epoch [8/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 21.0938, Loss1: 0.0181, Loss2: 0.0180 +Epoch [8/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 29.6875, Loss1: 0.0171, Loss2: 0.0168 +Epoch [8/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0153 +Epoch [8/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 28.1250, Loss1: 0.0147, Loss2: 0.0155 +Epoch [8/200], Iter [300/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.7812, Loss1: 0.0174, Loss2: 0.0173 +Epoch [8/200], Iter [350/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.7812, Loss1: 0.0166, Loss2: 0.0167 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 22.0553 % Model2 22.1154 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0168, Loss2: 0.0167 +Epoch [9/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 26.5625, Loss1: 0.0176, Loss2: 0.0177 +Epoch [9/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.7812, Loss1: 0.0181, Loss2: 0.0184 +Epoch [9/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 21.8750, Loss1: 0.0165, Loss2: 0.0174 +Epoch [9/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0154, Loss2: 0.0151 +Epoch [9/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 21.8750, Loss1: 0.0164, Loss2: 0.0174 +Epoch [9/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0151, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 22.1955 % Model2 21.2240 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 26.5625, Loss1: 0.0163, Loss2: 0.0158 +Epoch [10/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.9062, Loss1: 0.0165, Loss2: 0.0168 +Epoch [10/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0165 +Epoch [10/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 19.5312, Loss1: 0.0168, Loss2: 0.0171 +Epoch [10/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0147 +Epoch [10/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0170, Loss2: 0.0170 +Epoch [10/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0157, Loss2: 0.0163 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 27.5541 % Model2 26.0617 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.0938, Loss1: 0.0170, Loss2: 0.0172 +Epoch [11/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0155, Loss2: 0.0149 +Epoch [11/200], Iter [150/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0177, Loss2: 0.0175 +Epoch [11/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 24.2188, Loss1: 0.0157, Loss2: 0.0155 +Epoch [11/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 28.9062, Loss1: 0.0163, Loss2: 0.0163 +Epoch [11/200], Iter [300/390] Training Accuracy1: 21.8750, Training Accuracy2: 26.5625, Loss1: 0.0157, Loss2: 0.0151 +Epoch [11/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 29.6875, Loss1: 0.0161, Loss2: 0.0163 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 23.7380 % Model2 22.3958 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0162, Loss2: 0.0168 +Epoch [12/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 30.4688, Loss1: 0.0158, Loss2: 0.0159 +Epoch [12/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.7812, Loss1: 0.0172, Loss2: 0.0167 +Epoch [12/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0157, Loss2: 0.0165 +Epoch [12/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.1250, Loss1: 0.0168, Loss2: 0.0161 +Epoch [12/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 26.5625, Loss1: 0.0160, Loss2: 0.0161 +Epoch [12/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 33.5938, Loss1: 0.0146, Loss2: 0.0145 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 23.4275 % Model2 22.7063 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0180, Loss2: 0.0186 +Epoch [13/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 28.9062, Loss1: 0.0175, Loss2: 0.0171 +Epoch [13/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0160, Loss2: 0.0170 +Epoch [13/200], Iter [200/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.0938, Loss1: 0.0173, Loss2: 0.0173 +Epoch [13/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0155, Loss2: 0.0157 +Epoch [13/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0173, Loss2: 0.0173 +Epoch [13/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 27.3438, Loss1: 0.0164, Loss2: 0.0166 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 22.4760 % Model2 23.1070 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0146, Loss2: 0.0147 +Epoch [14/200], Iter [100/390] Training Accuracy1: 25.7812, Training Accuracy2: 26.5625, Loss1: 0.0168, Loss2: 0.0171 +Epoch [14/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 27.3438, Loss1: 0.0169, Loss2: 0.0165 +Epoch [14/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0155, Loss2: 0.0157 +Epoch [14/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0154, Loss2: 0.0151 +Epoch [14/200], Iter [300/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0160, Loss2: 0.0157 +Epoch [14/200], Iter [350/390] Training Accuracy1: 25.0000, Training Accuracy2: 29.6875, Loss1: 0.0167, Loss2: 0.0166 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 27.1034 % Model2 25.6510 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 28.1250, Loss1: 0.0170, Loss2: 0.0169 +Epoch [15/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 21.8750, Loss1: 0.0150, Loss2: 0.0160 +Epoch [15/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 20.3125, Loss1: 0.0176, Loss2: 0.0174 +Epoch [15/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0152, Loss2: 0.0151 +Epoch [15/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0166, Loss2: 0.0160 +Epoch [15/200], Iter [300/390] Training Accuracy1: 22.6562, Training Accuracy2: 22.6562, Loss1: 0.0165, Loss2: 0.0165 +Epoch [15/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0173, Loss2: 0.0170 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 27.5841 % Model2 26.9331 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0159, Loss2: 0.0164 +Epoch [16/200], Iter [100/390] Training Accuracy1: 25.0000, Training Accuracy2: 25.7812, Loss1: 0.0155, Loss2: 0.0158 +Epoch [16/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 38.2812, Loss1: 0.0148, Loss2: 0.0144 +Epoch [16/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0154, Loss2: 0.0152 +Epoch [16/200], Iter [250/390] Training Accuracy1: 25.0000, Training Accuracy2: 30.4688, Loss1: 0.0156, Loss2: 0.0160 +Epoch [16/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 28.9062, Loss1: 0.0159, Loss2: 0.0156 +Epoch [16/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 26.5625, Loss1: 0.0164, Loss2: 0.0163 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 25.6410 % Model2 24.8598 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 27.3438, Loss1: 0.0161, Loss2: 0.0158 +Epoch [17/200], Iter [100/390] Training Accuracy1: 21.0938, Training Accuracy2: 23.4375, Loss1: 0.0173, Loss2: 0.0171 +Epoch [17/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0162, Loss2: 0.0161 +Epoch [17/200], Iter [200/390] Training Accuracy1: 24.2188, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0166 +Epoch [17/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 30.4688, Loss1: 0.0154, Loss2: 0.0156 +Epoch [17/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0154, Loss2: 0.0158 +Epoch [17/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0143, Loss2: 0.0145 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 27.7945 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0139, Loss2: 0.0141 +Epoch [18/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0165, Loss2: 0.0160 +Epoch [18/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.7812, Loss1: 0.0165, Loss2: 0.0170 +Epoch [18/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.1562, Loss1: 0.0145, Loss2: 0.0153 +Epoch [18/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 24.2188, Loss1: 0.0172, Loss2: 0.0167 +Epoch [18/200], Iter [300/390] Training Accuracy1: 28.1250, Training Accuracy2: 25.0000, Loss1: 0.0168, Loss2: 0.0163 +Epoch [18/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0163, Loss2: 0.0161 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 26.3221 % Model2 26.4423 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0163, Loss2: 0.0168 +Epoch [19/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.1250, Loss1: 0.0145, Loss2: 0.0155 +Epoch [19/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0148, Loss2: 0.0144 +Epoch [19/200], Iter [200/390] Training Accuracy1: 27.3438, Training Accuracy2: 25.7812, Loss1: 0.0157, Loss2: 0.0161 +Epoch [19/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.9062, Loss1: 0.0167, Loss2: 0.0161 +Epoch [19/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 27.3438, Loss1: 0.0167, Loss2: 0.0169 +Epoch [19/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0159, Loss2: 0.0154 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 22.1955 % Model2 23.6378 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.0625, Loss1: 0.0146, Loss2: 0.0146 +Epoch [20/200], Iter [100/390] Training Accuracy1: 24.2188, Training Accuracy2: 23.4375, Loss1: 0.0175, Loss2: 0.0177 +Epoch [20/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0154, Loss2: 0.0148 +Epoch [20/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 27.3438, Loss1: 0.0169, Loss2: 0.0165 +Epoch [20/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0140, Loss2: 0.0142 +Epoch [20/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0151, Loss2: 0.0147 +Epoch [20/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0160 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 26.3321 % Model2 25.5108 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.0312, Loss1: 0.0864, Loss2: 0.0893 +Epoch [21/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0619, Loss2: 0.0621 +Epoch [21/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 28.1250, Loss1: 0.0951, Loss2: 0.0903 +Epoch [21/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.7812, Loss1: 0.0941, Loss2: 0.0918 +Epoch [21/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.0000, Loss1: 0.0885, Loss2: 0.0844 +Epoch [21/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.0000, Loss1: 0.0745, Loss2: 0.0760 +Epoch [21/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0952, Loss2: 0.0924 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 22.3958 % Model2 20.8734 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.1026, Loss2: 0.1052 +Epoch [22/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 31.2500, Loss1: 0.0719, Loss2: 0.0750 +Epoch [22/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0892, Loss2: 0.0903 +Epoch [22/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 30.4688, Loss1: 0.0807, Loss2: 0.0772 +Epoch [22/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0625, Loss2: 0.0621 +Epoch [22/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0697, Loss2: 0.0698 +Epoch [22/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0697, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 24.1887 % Model2 24.1887 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0726, Loss2: 0.0724 +Epoch [23/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 28.9062, Loss1: 0.0854, Loss2: 0.0860 +Epoch [23/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.1250, Loss1: 0.0682, Loss2: 0.0674 +Epoch [23/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.9062, Loss1: 0.0585, Loss2: 0.0591 +Epoch [23/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0828, Loss2: 0.0829 +Epoch [23/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 31.2500, Loss1: 0.0657, Loss2: 0.0646 +Epoch [23/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 40.6250, Loss1: 0.0674, Loss2: 0.0653 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 24.2087 % Model2 22.8766 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 24.2188, Loss1: 0.0614, Loss2: 0.0622 +Epoch [24/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0540, Loss2: 0.0549 +Epoch [24/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0648, Loss2: 0.0634 +Epoch [24/200], Iter [200/390] Training Accuracy1: 26.5625, Training Accuracy2: 33.5938, Loss1: 0.0866, Loss2: 0.0806 +Epoch [24/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0713, Loss2: 0.0692 +Epoch [24/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0917, Loss2: 0.0896 +Epoch [24/200], Iter [350/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.9062, Loss1: 0.0683, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 25.9014 % Model2 25.8514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 27.3438, Loss1: 0.0576, Loss2: 0.0583 +Epoch [25/200], Iter [100/390] Training Accuracy1: 21.8750, Training Accuracy2: 27.3438, Loss1: 0.0820, Loss2: 0.0800 +Epoch [25/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 30.4688, Loss1: 0.0725, Loss2: 0.0682 +Epoch [25/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0730, Loss2: 0.0722 +Epoch [25/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.0312, Loss1: 0.0706, Loss2: 0.0681 +Epoch [25/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0609, Loss2: 0.0637 +Epoch [25/200], Iter [350/390] Training Accuracy1: 22.6562, Training Accuracy2: 24.2188, Loss1: 0.0805, Loss2: 0.0791 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 21.0837 % Model2 23.9784 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0727, Loss2: 0.0721 +Epoch [26/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0685, Loss2: 0.0675 +Epoch [26/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0524, Loss2: 0.0515 +Epoch [26/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 27.3438, Loss1: 0.0652, Loss2: 0.0668 +Epoch [26/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0500, Loss2: 0.0497 +Epoch [26/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0730, Loss2: 0.0725 +Epoch [26/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 25.7812, Loss1: 0.0594, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 23.8882 % Model2 23.6078 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 35.1562, Loss1: 0.0611, Loss2: 0.0585 +Epoch [27/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 28.1250, Loss1: 0.0667, Loss2: 0.0694 +Epoch [27/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 39.8438, Loss1: 0.0594, Loss2: 0.0543 +Epoch [27/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.1250, Loss1: 0.0656, Loss2: 0.0664 +Epoch [27/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0594, Loss2: 0.0583 +Epoch [27/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0866, Loss2: 0.0861 +Epoch [27/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0557, Loss2: 0.0552 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 27.0833 % Model2 26.2220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 30.4688, Training Accuracy2: 30.4688, Loss1: 0.0784, Loss2: 0.0780 +Epoch [28/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0641, Loss2: 0.0621 +Epoch [28/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0535, Loss2: 0.0537 +Epoch [28/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0585, Loss2: 0.0592 +Epoch [28/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0501, Loss2: 0.0493 +Epoch [28/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 29.6875, Loss1: 0.0554, Loss2: 0.0548 +Epoch [28/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0789, Loss2: 0.0790 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 25.4207 % Model2 26.2119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0615, Loss2: 0.0622 +Epoch [29/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0601, Loss2: 0.0583 +Epoch [29/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0596, Loss2: 0.0588 +Epoch [29/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 34.3750, Loss1: 0.0540, Loss2: 0.0550 +Epoch [29/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 31.2500, Loss1: 0.0645, Loss2: 0.0671 +Epoch [29/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0674, Loss2: 0.0658 +Epoch [29/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0592, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 28.8361 % Model2 27.4139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 39.0625, Loss1: 0.0799, Loss2: 0.0733 +Epoch [30/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 31.2500, Loss1: 0.0644, Loss2: 0.0634 +Epoch [30/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0554, Loss2: 0.0554 +Epoch [30/200], Iter [200/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.7812, Loss1: 0.0480, Loss2: 0.0476 +Epoch [30/200], Iter [250/390] Training Accuracy1: 25.0000, Training Accuracy2: 23.4375, Loss1: 0.0440, Loss2: 0.0437 +Epoch [30/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 34.3750, Loss1: 0.0551, Loss2: 0.0523 +Epoch [30/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 36.7188, Loss1: 0.0576, Loss2: 0.0547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 24.5693 % Model2 24.9499 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0562, Loss2: 0.0570 +Epoch [31/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0579, Loss2: 0.0569 +Epoch [31/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0535, Loss2: 0.0536 +Epoch [31/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0524, Loss2: 0.0541 +Epoch [31/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0561, Loss2: 0.0548 +Epoch [31/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0584, Loss2: 0.0584 +Epoch [31/200], Iter [350/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.7812, Loss1: 0.0507, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 24.9499 % Model2 24.9800 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0578, Loss2: 0.0601 +Epoch [32/200], Iter [100/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0707, Loss2: 0.0683 +Epoch [32/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.9375, Loss1: 0.0538, Loss2: 0.0530 +Epoch [32/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0645, Loss2: 0.0612 +Epoch [32/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.0312, Loss1: 0.0703, Loss2: 0.0655 +Epoch [32/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0558, Loss2: 0.0532 +Epoch [32/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 35.1562, Loss1: 0.0436, Loss2: 0.0411 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 26.9231 % Model2 27.6943 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 27.3438, Loss1: 0.0623, Loss2: 0.0595 +Epoch [33/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0612, Loss2: 0.0598 +Epoch [33/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0573, Loss2: 0.0578 +Epoch [33/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0701, Loss2: 0.0723 +Epoch [33/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0689, Loss2: 0.0676 +Epoch [33/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0642, Loss2: 0.0643 +Epoch [33/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0811, Loss2: 0.0777 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 25.2204 % Model2 25.8413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0538, Loss2: 0.0551 +Epoch [34/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 41.4062, Loss1: 0.0630, Loss2: 0.0594 +Epoch [34/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0731, Loss2: 0.0758 +Epoch [34/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0564, Loss2: 0.0566 +Epoch [34/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0694, Loss2: 0.0672 +Epoch [34/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0559, Loss2: 0.0551 +Epoch [34/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0615, Loss2: 0.0621 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 24.9499 % Model2 25.1803 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0649, Loss2: 0.0692 +Epoch [35/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0622, Loss2: 0.0605 +Epoch [35/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0646, Loss2: 0.0632 +Epoch [35/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0564, Loss2: 0.0561 +Epoch [35/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 23.4375, Loss1: 0.0470, Loss2: 0.0485 +Epoch [35/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0593, Loss2: 0.0594 +Epoch [35/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0502, Loss2: 0.0504 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 26.5024 % Model2 24.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0578, Loss2: 0.0567 +Epoch [36/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0511, Loss2: 0.0498 +Epoch [36/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0539, Loss2: 0.0542 +Epoch [36/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0552, Loss2: 0.0553 +Epoch [36/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 37.5000, Loss1: 0.0510, Loss2: 0.0488 +Epoch [36/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 30.4688, Loss1: 0.0490, Loss2: 0.0496 +Epoch [36/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0498, Loss2: 0.0509 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 27.1835 % Model2 26.5825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0567, Loss2: 0.0562 +Epoch [37/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0466, Loss2: 0.0468 +Epoch [37/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0495, Loss2: 0.0502 +Epoch [37/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0581, Loss2: 0.0586 +Epoch [37/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0548, Loss2: 0.0544 +Epoch [37/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0535, Loss2: 0.0541 +Epoch [37/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 30.4688, Loss1: 0.0493, Loss2: 0.0496 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 25.0801 % Model2 25.8413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 32.8125, Loss1: 0.0467, Loss2: 0.0488 +Epoch [38/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.1562, Loss1: 0.0472, Loss2: 0.0466 +Epoch [38/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 31.2500, Loss1: 0.0526, Loss2: 0.0516 +Epoch [38/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0682, Loss2: 0.0663 +Epoch [38/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 26.5625, Loss1: 0.0490, Loss2: 0.0507 +Epoch [38/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.9062, Loss1: 0.0609, Loss2: 0.0593 +Epoch [38/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 30.4688, Loss1: 0.0557, Loss2: 0.0561 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 27.1635 % Model2 23.6378 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0465, Loss2: 0.0473 +Epoch [39/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0480, Loss2: 0.0474 +Epoch [39/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 34.3750, Loss1: 0.0559, Loss2: 0.0574 +Epoch [39/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0526, Loss2: 0.0517 +Epoch [39/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0460, Loss2: 0.0457 +Epoch [39/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0522, Loss2: 0.0496 +Epoch [39/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0549, Loss2: 0.0528 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 26.0717 % Model2 26.9030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0531, Loss2: 0.0521 +Epoch [40/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0560, Loss2: 0.0553 +Epoch [40/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 38.2812, Loss1: 0.0582, Loss2: 0.0559 +Epoch [40/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0582, Loss2: 0.0602 +Epoch [40/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0454, Loss2: 0.0441 +Epoch [40/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0553, Loss2: 0.0532 +Epoch [40/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 31.2500, Loss1: 0.0480, Loss2: 0.0485 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 24.6995 % Model2 24.6394 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0592, Loss2: 0.0578 +Epoch [41/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0474, Loss2: 0.0470 +Epoch [41/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 41.4062, Loss1: 0.0597, Loss2: 0.0564 +Epoch [41/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0498, Loss2: 0.0490 +Epoch [41/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0498, Loss2: 0.0489 +Epoch [41/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0470, Loss2: 0.0471 +Epoch [41/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 42.1875, Loss1: 0.0678, Loss2: 0.0640 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 25.7512 % Model2 26.7528 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0498, Loss2: 0.0492 +Epoch [42/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.1562, Loss1: 0.0616, Loss2: 0.0613 +Epoch [42/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0614, Loss2: 0.0607 +Epoch [42/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0495, Loss2: 0.0504 +Epoch [42/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0509, Loss2: 0.0521 +Epoch [42/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0454, Loss2: 0.0451 +Epoch [42/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0606, Loss2: 0.0591 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 26.0617 % Model2 25.6711 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0608, Loss2: 0.0604 +Epoch [43/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0654, Loss2: 0.0666 +Epoch [43/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0677, Loss2: 0.0685 +Epoch [43/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0591, Loss2: 0.0619 +Epoch [43/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0584, Loss2: 0.0597 +Epoch [43/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0482, Loss2: 0.0468 +Epoch [43/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0620, Loss2: 0.0628 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 26.1118 % Model2 25.9816 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0571, Loss2: 0.0552 +Epoch [44/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0458, Loss2: 0.0456 +Epoch [44/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0567, Loss2: 0.0568 +Epoch [44/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0601, Loss2: 0.0595 +Epoch [44/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.1250, Loss1: 0.0628, Loss2: 0.0633 +Epoch [44/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0555, Loss2: 0.0550 +Epoch [44/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0552, Loss2: 0.0535 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 26.1018 % Model2 25.8113 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0531, Loss2: 0.0535 +Epoch [45/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0649, Loss2: 0.0617 +Epoch [45/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.9062, Loss1: 0.0620, Loss2: 0.0612 +Epoch [45/200], Iter [200/390] Training Accuracy1: 26.5625, Training Accuracy2: 32.8125, Loss1: 0.0568, Loss2: 0.0538 +Epoch [45/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0540, Loss2: 0.0515 +Epoch [45/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0694, Loss2: 0.0682 +Epoch [45/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0623, Loss2: 0.0600 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 25.7913 % Model2 25.0601 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0551, Loss2: 0.0534 +Epoch [46/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0488, Loss2: 0.0497 +Epoch [46/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 39.8438, Loss1: 0.0554, Loss2: 0.0507 +Epoch [46/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0500, Loss2: 0.0502 +Epoch [46/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0636, Loss2: 0.0653 +Epoch [46/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0482, Loss2: 0.0483 +Epoch [46/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 27.3438, Loss1: 0.0456, Loss2: 0.0461 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 25.6010 % Model2 25.6310 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0554, Loss2: 0.0547 +Epoch [47/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 37.5000, Loss1: 0.0522, Loss2: 0.0491 +Epoch [47/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0538, Loss2: 0.0515 +Epoch [47/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0572, Loss2: 0.0554 +Epoch [47/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0444, Loss2: 0.0432 +Epoch [47/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0438, Loss2: 0.0445 +Epoch [47/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0678, Loss2: 0.0693 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 24.5893 % Model2 25.3105 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0646, Loss2: 0.0655 +Epoch [48/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.0312, Loss1: 0.0534, Loss2: 0.0543 +Epoch [48/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0442, Loss2: 0.0445 +Epoch [48/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.0625, Loss1: 0.0506, Loss2: 0.0532 +Epoch [48/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 34.3750, Loss1: 0.0446, Loss2: 0.0458 +Epoch [48/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0443, Loss2: 0.0431 +Epoch [48/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0560, Loss2: 0.0544 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 25.1803 % Model2 25.1603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0512, Loss2: 0.0523 +Epoch [49/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0565, Loss2: 0.0554 +Epoch [49/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 41.4062, Loss1: 0.0443, Loss2: 0.0432 +Epoch [49/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 33.5938, Loss1: 0.0549, Loss2: 0.0533 +Epoch [49/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0484, Loss2: 0.0483 +Epoch [49/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 43.7500, Loss1: 0.0719, Loss2: 0.0689 +Epoch [49/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0456, Loss2: 0.0443 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 25.6110 % Model2 24.5393 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 40.6250, Loss1: 0.0509, Loss2: 0.0485 +Epoch [50/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0432, Loss2: 0.0430 +Epoch [50/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0416, Loss2: 0.0416 +Epoch [50/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0679, Loss2: 0.0669 +Epoch [50/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 26.5625, Loss1: 0.0500, Loss2: 0.0510 +Epoch [50/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0488, Loss2: 0.0477 +Epoch [50/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 50.7812, Loss1: 0.0572, Loss2: 0.0517 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 24.5693 % Model2 25.0401 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 30.4688, Loss1: 0.0583, Loss2: 0.0601 +Epoch [51/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0472, Loss2: 0.0464 +Epoch [51/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0472, Loss2: 0.0465 +Epoch [51/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0464, Loss2: 0.0473 +Epoch [51/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.9375, Loss1: 0.0571, Loss2: 0.0553 +Epoch [51/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.1562, Loss1: 0.0479, Loss2: 0.0475 +Epoch [51/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0513, Loss2: 0.0509 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 26.0517 % Model2 26.6526 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0611, Loss2: 0.0588 +Epoch [52/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0469, Loss2: 0.0466 +Epoch [52/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 47.6562, Loss1: 0.0533, Loss2: 0.0519 +Epoch [52/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.9688, Loss1: 0.0525, Loss2: 0.0501 +Epoch [52/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0553, Loss2: 0.0564 +Epoch [52/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0570, Loss2: 0.0563 +Epoch [52/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0456, Loss2: 0.0468 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 25.0200 % Model2 24.6294 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0574, Loss2: 0.0556 +Epoch [53/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0651, Loss2: 0.0663 +Epoch [53/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0531, Loss2: 0.0530 +Epoch [53/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0455, Loss2: 0.0437 +Epoch [53/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 32.0312, Loss1: 0.0477, Loss2: 0.0499 +Epoch [53/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0596, Loss2: 0.0619 +Epoch [53/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 42.1875, Loss1: 0.0513, Loss2: 0.0480 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 26.3622 % Model2 25.9515 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 42.1875, Loss1: 0.0542, Loss2: 0.0518 +Epoch [54/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0502, Loss2: 0.0501 +Epoch [54/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0497, Loss2: 0.0483 +Epoch [54/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0592, Loss2: 0.0603 +Epoch [54/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0533, Loss2: 0.0529 +Epoch [54/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0630, Loss2: 0.0617 +Epoch [54/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0581, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 23.8582 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0502, Loss2: 0.0532 +Epoch [55/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0487, Loss2: 0.0482 +Epoch [55/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 42.1875, Loss1: 0.0534, Loss2: 0.0504 +Epoch [55/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.9375, Loss1: 0.0461, Loss2: 0.0462 +Epoch [55/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 30.4688, Loss1: 0.0564, Loss2: 0.0599 +Epoch [55/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0531, Loss2: 0.0546 +Epoch [55/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0455, Loss2: 0.0470 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 24.1486 % Model2 25.3005 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0468, Loss2: 0.0463 +Epoch [56/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 29.6875, Loss1: 0.0437, Loss2: 0.0472 +Epoch [56/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0432, Loss2: 0.0435 +Epoch [56/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.8125, Loss1: 0.0478, Loss2: 0.0481 +Epoch [56/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0523, Loss2: 0.0530 +Epoch [56/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0506, Loss2: 0.0477 +Epoch [56/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0466, Loss2: 0.0465 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 24.8297 % Model2 25.0601 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0628, Loss2: 0.0626 +Epoch [57/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 54.6875, Loss1: 0.0560, Loss2: 0.0483 +Epoch [57/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0553, Loss2: 0.0550 +Epoch [57/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 41.4062, Loss1: 0.0560, Loss2: 0.0567 +Epoch [57/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0511, Loss2: 0.0520 +Epoch [57/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0507, Loss2: 0.0520 +Epoch [57/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0580, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 25.0401 % Model2 24.4792 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 45.3125, Loss1: 0.0489, Loss2: 0.0457 +Epoch [58/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0591, Loss2: 0.0594 +Epoch [58/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 31.2500, Loss1: 0.0520, Loss2: 0.0565 +Epoch [58/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0553, Loss2: 0.0557 +Epoch [58/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0492, Loss2: 0.0518 +Epoch [58/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 33.5938, Loss1: 0.0508, Loss2: 0.0532 +Epoch [58/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0506, Loss2: 0.0494 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 25.6310 % Model2 24.8998 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0401, Loss2: 0.0391 +Epoch [59/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0501, Loss2: 0.0491 +Epoch [59/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0436, Loss2: 0.0423 +Epoch [59/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0520, Loss2: 0.0530 +Epoch [59/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0542, Loss2: 0.0541 +Epoch [59/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0426, Loss2: 0.0418 +Epoch [59/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0506, Loss2: 0.0516 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 23.8381 % Model2 24.3189 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0498, Loss2: 0.0500 +Epoch [60/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0479, Loss2: 0.0490 +Epoch [60/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0467, Loss2: 0.0461 +Epoch [60/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0519, Loss2: 0.0494 +Epoch [60/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0466, Loss2: 0.0457 +Epoch [60/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 28.9062, Loss1: 0.0417, Loss2: 0.0419 +Epoch [60/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0599, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 25.6310 % Model2 25.3405 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0606, Loss2: 0.0618 +Epoch [61/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0492, Loss2: 0.0494 +Epoch [61/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0490, Loss2: 0.0480 +Epoch [61/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.1875, Loss1: 0.0451, Loss2: 0.0437 +Epoch [61/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0459, Loss2: 0.0459 +Epoch [61/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0548, Loss2: 0.0544 +Epoch [61/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0498, Loss2: 0.0498 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 24.9900 % Model2 24.5192 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0496, Loss2: 0.0476 +Epoch [62/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0489, Loss2: 0.0479 +Epoch [62/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 44.5312, Loss1: 0.0483, Loss2: 0.0452 +Epoch [62/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0477, Loss2: 0.0481 +Epoch [62/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0450, Loss2: 0.0460 +Epoch [62/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 28.9062, Loss1: 0.0488, Loss2: 0.0484 +Epoch [62/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0566, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 25.8413 % Model2 25.7412 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0518, Loss2: 0.0514 +Epoch [63/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0493, Loss2: 0.0477 +Epoch [63/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0487, Loss2: 0.0505 +Epoch [63/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0554, Loss2: 0.0539 +Epoch [63/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0620, Loss2: 0.0611 +Epoch [63/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0586, Loss2: 0.0574 +Epoch [63/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0611, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 23.8181 % Model2 25.3405 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0540, Loss2: 0.0541 +Epoch [64/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0475, Loss2: 0.0480 +Epoch [64/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0542, Loss2: 0.0539 +Epoch [64/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0554, Loss2: 0.0527 +Epoch [64/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0553, Loss2: 0.0566 +Epoch [64/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.9375, Loss1: 0.0537, Loss2: 0.0554 +Epoch [64/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0495, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 24.5393 % Model2 25.5409 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0516, Loss2: 0.0539 +Epoch [65/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0534, Loss2: 0.0547 +Epoch [65/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0549, Loss2: 0.0542 +Epoch [65/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0496, Loss2: 0.0515 +Epoch [65/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0501, Loss2: 0.0508 +Epoch [65/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0477, Loss2: 0.0476 +Epoch [65/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0580, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 25.1402 % Model2 24.6595 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0486, Loss2: 0.0496 +Epoch [66/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0494, Loss2: 0.0506 +Epoch [66/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0486, Loss2: 0.0479 +Epoch [66/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0485, Loss2: 0.0495 +Epoch [66/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0495, Loss2: 0.0502 +Epoch [66/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0546, Loss2: 0.0546 +Epoch [66/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0508, Loss2: 0.0505 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 24.9199 % Model2 25.0100 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0473, Loss2: 0.0482 +Epoch [67/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0476, Loss2: 0.0469 +Epoch [67/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0675, Loss2: 0.0669 +Epoch [67/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0549, Loss2: 0.0534 +Epoch [67/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0513, Loss2: 0.0506 +Epoch [67/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0546, Loss2: 0.0526 +Epoch [67/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0464, Loss2: 0.0438 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 24.4491 % Model2 23.8482 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0519, Loss2: 0.0514 +Epoch [68/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0552, Loss2: 0.0537 +Epoch [68/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0590, Loss2: 0.0540 +Epoch [68/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.1562, Loss1: 0.0609, Loss2: 0.0628 +Epoch [68/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0444, Loss2: 0.0442 +Epoch [68/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0493, Loss2: 0.0499 +Epoch [68/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0591, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 25.0000 % Model2 25.4607 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0488, Loss2: 0.0493 +Epoch [69/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 43.7500, Loss1: 0.0490, Loss2: 0.0478 +Epoch [69/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0788, Loss2: 0.0753 +Epoch [69/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0606, Loss2: 0.0609 +Epoch [69/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0616, Loss2: 0.0609 +Epoch [69/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0574, Loss2: 0.0560 +Epoch [69/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0540, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 24.6194 % Model2 25.5709 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0558, Loss2: 0.0533 +Epoch [70/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0494, Loss2: 0.0496 +Epoch [70/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0587, Loss2: 0.0595 +Epoch [70/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0520, Loss2: 0.0522 +Epoch [70/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 46.0938, Loss1: 0.0538, Loss2: 0.0485 +Epoch [70/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0693, Loss2: 0.0701 +Epoch [70/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0445, Loss2: 0.0450 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 25.1903 % Model2 24.6094 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0498, Loss2: 0.0484 +Epoch [71/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0526, Loss2: 0.0531 +Epoch [71/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0473, Loss2: 0.0478 +Epoch [71/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 40.6250, Loss1: 0.0422, Loss2: 0.0413 +Epoch [71/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0617, Loss2: 0.0616 +Epoch [71/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 45.3125, Loss1: 0.0605, Loss2: 0.0564 +Epoch [71/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0485, Loss2: 0.0474 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 25.6410 % Model2 24.9399 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0548, Loss2: 0.0550 +Epoch [72/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0508, Loss2: 0.0511 +Epoch [72/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0453, Loss2: 0.0465 +Epoch [72/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0484, Loss2: 0.0503 +Epoch [72/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0523, Loss2: 0.0518 +Epoch [72/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0459, Loss2: 0.0466 +Epoch [72/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0522, Loss2: 0.0531 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 24.2087 % Model2 25.2003 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 37.5000, Loss1: 0.0445, Loss2: 0.0478 +Epoch [73/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0522, Loss2: 0.0542 +Epoch [73/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0560, Loss2: 0.0559 +Epoch [73/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0495, Loss2: 0.0503 +Epoch [73/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0504, Loss2: 0.0518 +Epoch [73/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0614, Loss2: 0.0620 +Epoch [73/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.0938, Loss1: 0.0559, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 24.7997 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0596, Loss2: 0.0589 +Epoch [74/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 40.6250, Loss1: 0.0404, Loss2: 0.0422 +Epoch [74/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0516, Loss2: 0.0496 +Epoch [74/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0457, Loss2: 0.0440 +Epoch [74/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.8438, Loss1: 0.0539, Loss2: 0.0562 +Epoch [74/200], Iter [300/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.8125, Loss1: 0.0501, Loss2: 0.0488 +Epoch [74/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0547, Loss2: 0.0545 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 24.9900 % Model2 25.1603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0471, Loss2: 0.0476 +Epoch [75/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0549, Loss2: 0.0556 +Epoch [75/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0552, Loss2: 0.0527 +Epoch [75/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.0625, Loss1: 0.0640, Loss2: 0.0620 +Epoch [75/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0566, Loss2: 0.0579 +Epoch [75/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0425, Loss2: 0.0415 +Epoch [75/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0572, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 23.8482 % Model2 23.8682 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0640, Loss2: 0.0623 +Epoch [76/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 35.9375, Loss1: 0.0525, Loss2: 0.0553 +Epoch [76/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0514, Loss2: 0.0496 +Epoch [76/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 44.5312, Loss1: 0.0540, Loss2: 0.0496 +Epoch [76/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0531, Loss2: 0.0557 +Epoch [76/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0518, Loss2: 0.0511 +Epoch [76/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 37.5000, Loss1: 0.0473, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 24.1687 % Model2 24.9299 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0494, Loss2: 0.0518 +Epoch [77/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0490, Loss2: 0.0505 +Epoch [77/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0474, Loss2: 0.0466 +Epoch [77/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0505, Loss2: 0.0526 +Epoch [77/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0457, Loss2: 0.0453 +Epoch [77/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0530, Loss2: 0.0547 +Epoch [77/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0534, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 24.0785 % Model2 25.1002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0690, Loss2: 0.0674 +Epoch [78/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0541, Loss2: 0.0537 +Epoch [78/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0639, Loss2: 0.0645 +Epoch [78/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0583, Loss2: 0.0587 +Epoch [78/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0531 +Epoch [78/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0589, Loss2: 0.0582 +Epoch [78/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0568, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 24.6895 % Model2 25.1002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0522, Loss2: 0.0530 +Epoch [79/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 43.7500, Loss1: 0.0480, Loss2: 0.0496 +Epoch [79/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0534, Loss2: 0.0558 +Epoch [79/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0543, Loss2: 0.0544 +Epoch [79/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0436, Loss2: 0.0431 +Epoch [79/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 43.7500, Loss1: 0.0654, Loss2: 0.0619 +Epoch [79/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0503, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 25.7913 % Model2 24.3490 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0548, Loss2: 0.0562 +Epoch [80/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0520, Loss2: 0.0510 +Epoch [80/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0502, Loss2: 0.0519 +Epoch [80/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0629, Loss2: 0.0589 +Epoch [80/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0617, Loss2: 0.0623 +Epoch [80/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0503, Loss2: 0.0491 +Epoch [80/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0653, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 25.3305 % Model2 25.2504 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0452, Loss2: 0.0458 +Epoch [81/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.1875, Loss1: 0.0531, Loss2: 0.0550 +Epoch [81/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0503, Loss2: 0.0492 +Epoch [81/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 45.3125, Loss1: 0.0435, Loss2: 0.0407 +Epoch [81/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0494, Loss2: 0.0518 +Epoch [81/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0627, Loss2: 0.0612 +Epoch [81/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0537, Loss2: 0.0519 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 24.7997 % Model2 24.5292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0549, Loss2: 0.0552 +Epoch [82/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0551, Loss2: 0.0554 +Epoch [82/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0618, Loss2: 0.0578 +Epoch [82/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.0625, Loss1: 0.0428, Loss2: 0.0411 +Epoch [82/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0381, Loss2: 0.0388 +Epoch [82/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0539, Loss2: 0.0551 +Epoch [82/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0530, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 25.1903 % Model2 26.0817 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 34.3750, Loss1: 0.0419, Loss2: 0.0435 +Epoch [83/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0591, Loss2: 0.0597 +Epoch [83/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0483, Loss2: 0.0504 +Epoch [83/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0544, Loss2: 0.0558 +Epoch [83/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0551, Loss2: 0.0556 +Epoch [83/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0472, Loss2: 0.0465 +Epoch [83/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0476, Loss2: 0.0464 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 25.4006 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0429, Loss2: 0.0415 +Epoch [84/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0602, Loss2: 0.0595 +Epoch [84/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0392, Loss2: 0.0386 +Epoch [84/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 45.3125, Loss1: 0.0441, Loss2: 0.0412 +Epoch [84/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 35.9375, Loss1: 0.0529, Loss2: 0.0569 +Epoch [84/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 44.5312, Loss1: 0.0529, Loss2: 0.0502 +Epoch [84/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0496, Loss2: 0.0503 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 23.8281 % Model2 25.2103 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 36.7188, Loss1: 0.0501, Loss2: 0.0516 +Epoch [85/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0489, Loss2: 0.0473 +Epoch [85/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0498, Loss2: 0.0466 +Epoch [85/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0463, Loss2: 0.0458 +Epoch [85/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 50.7812, Loss1: 0.0470, Loss2: 0.0432 +Epoch [85/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0471, Loss2: 0.0481 +Epoch [85/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0603, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 24.8498 % Model2 24.9499 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0573, Loss2: 0.0577 +Epoch [86/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0513, Loss2: 0.0486 +Epoch [86/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0568, Loss2: 0.0547 +Epoch [86/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0492, Loss2: 0.0519 +Epoch [86/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0530, Loss2: 0.0537 +Epoch [86/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0539, Loss2: 0.0546 +Epoch [86/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0529, Loss2: 0.0499 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 24.9499 % Model2 25.6010 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0489, Loss2: 0.0481 +Epoch [87/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0491, Loss2: 0.0483 +Epoch [87/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0576, Loss2: 0.0597 +Epoch [87/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0438, Loss2: 0.0451 +Epoch [87/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.0625, Loss1: 0.0479, Loss2: 0.0497 +Epoch [87/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.7812, Loss1: 0.0471, Loss2: 0.0439 +Epoch [87/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0506, Loss2: 0.0503 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 24.0585 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0542, Loss2: 0.0556 +Epoch [88/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0504, Loss2: 0.0495 +Epoch [88/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 40.6250, Loss1: 0.0478, Loss2: 0.0504 +Epoch [88/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0545, Loss2: 0.0515 +Epoch [88/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 47.6562, Loss1: 0.0716, Loss2: 0.0661 +Epoch [88/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0562, Loss2: 0.0582 +Epoch [88/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0425, Loss2: 0.0429 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 23.8982 % Model2 25.0901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0453, Loss2: 0.0456 +Epoch [89/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0456, Loss2: 0.0463 +Epoch [89/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0503, Loss2: 0.0495 +Epoch [89/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0567, Loss2: 0.0590 +Epoch [89/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0477, Loss2: 0.0486 +Epoch [89/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0491, Loss2: 0.0498 +Epoch [89/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0502, Loss2: 0.0494 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 24.8598 % Model2 24.7396 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0569, Loss2: 0.0562 +Epoch [90/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 43.7500, Loss1: 0.0466, Loss2: 0.0442 +Epoch [90/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0514, Loss2: 0.0536 +Epoch [90/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0556, Loss2: 0.0530 +Epoch [90/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0498, Loss2: 0.0510 +Epoch [90/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0522, Loss2: 0.0506 +Epoch [90/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0507, Loss2: 0.0485 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 24.2688 % Model2 24.8998 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0515, Loss2: 0.0524 +Epoch [91/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0535, Loss2: 0.0524 +Epoch [91/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 47.6562, Loss1: 0.0537, Loss2: 0.0484 +Epoch [91/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0501, Loss2: 0.0505 +Epoch [91/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0547, Loss2: 0.0535 +Epoch [91/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0553, Loss2: 0.0580 +Epoch [91/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0552, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 24.6194 % Model2 25.1302 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0548, Loss2: 0.0546 +Epoch [92/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0575, Loss2: 0.0572 +Epoch [92/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.0625, Loss1: 0.0377, Loss2: 0.0380 +Epoch [92/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0456, Loss2: 0.0442 +Epoch [92/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0545, Loss2: 0.0547 +Epoch [92/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0495, Loss2: 0.0499 +Epoch [92/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0491, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 23.9183 % Model2 25.4808 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0458, Loss2: 0.0473 +Epoch [93/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0510, Loss2: 0.0524 +Epoch [93/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 53.1250, Loss1: 0.0512, Loss2: 0.0469 +Epoch [93/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0511, Loss2: 0.0486 +Epoch [93/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0487, Loss2: 0.0485 +Epoch [93/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0490, Loss2: 0.0489 +Epoch [93/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0552, Loss2: 0.0544 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 25.1703 % Model2 25.1102 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0443, Loss2: 0.0442 +Epoch [94/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0608, Loss2: 0.0611 +Epoch [94/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0498, Loss2: 0.0488 +Epoch [94/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0472, Loss2: 0.0454 +Epoch [94/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0446, Loss2: 0.0435 +Epoch [94/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0507, Loss2: 0.0540 +Epoch [94/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0495, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 24.6094 % Model2 24.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0491, Loss2: 0.0486 +Epoch [95/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0473, Loss2: 0.0493 +Epoch [95/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0461, Loss2: 0.0480 +Epoch [95/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0512, Loss2: 0.0492 +Epoch [95/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0571, Loss2: 0.0562 +Epoch [95/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.8750, Loss1: 0.0492, Loss2: 0.0472 +Epoch [95/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.7812, Loss1: 0.0458, Loss2: 0.0438 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 25.1803 % Model2 25.5609 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0586, Loss2: 0.0582 +Epoch [96/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 37.5000, Loss1: 0.0471, Loss2: 0.0491 +Epoch [96/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0482, Loss2: 0.0484 +Epoch [96/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0570, Loss2: 0.0573 +Epoch [96/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0456, Loss2: 0.0459 +Epoch [96/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0540, Loss2: 0.0556 +Epoch [96/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0680, Loss2: 0.0699 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 24.3089 % Model2 26.7528 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0443, Loss2: 0.0423 +Epoch [97/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0492, Loss2: 0.0472 +Epoch [97/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0463, Loss2: 0.0483 +Epoch [97/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0516, Loss2: 0.0518 +Epoch [97/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0500, Loss2: 0.0512 +Epoch [97/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0544, Loss2: 0.0575 +Epoch [97/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0652, Loss2: 0.0631 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 25.3305 % Model2 25.1002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0573, Loss2: 0.0552 +Epoch [98/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0465, Loss2: 0.0508 +Epoch [98/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0487, Loss2: 0.0489 +Epoch [98/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0459, Loss2: 0.0460 +Epoch [98/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0499, Loss2: 0.0524 +Epoch [98/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0518, Loss2: 0.0480 +Epoch [98/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 40.6250, Loss1: 0.0411, Loss2: 0.0389 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 25.5008 % Model2 25.2003 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0577, Loss2: 0.0565 +Epoch [99/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0564, Loss2: 0.0563 +Epoch [99/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.9375, Loss1: 0.0439, Loss2: 0.0427 +Epoch [99/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0460, Loss2: 0.0436 +Epoch [99/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 56.2500, Loss1: 0.0586, Loss2: 0.0535 +Epoch [99/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0554, Loss2: 0.0529 +Epoch [99/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0485, Loss2: 0.0464 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 24.6595 % Model2 25.2905 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0494, Loss2: 0.0504 +Epoch [100/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0503, Loss2: 0.0486 +Epoch [100/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0564, Loss2: 0.0572 +Epoch [100/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0464, Loss2: 0.0449 +Epoch [100/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 53.9062, Loss1: 0.0564, Loss2: 0.0508 +Epoch [100/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0588, Loss2: 0.0602 +Epoch [100/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 44.5312, Loss1: 0.0607, Loss2: 0.0640 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 24.9900 % Model2 25.4808 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.0625, Loss1: 0.0507, Loss2: 0.0522 +Epoch [101/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0488, Loss2: 0.0498 +Epoch [101/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0557, Loss2: 0.0557 +Epoch [101/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0500, Loss2: 0.0527 +Epoch [101/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 35.9375, Loss1: 0.0482, Loss2: 0.0506 +Epoch [101/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0542, Loss2: 0.0563 +Epoch [101/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0535, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 24.5292 % Model2 25.2704 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0494, Loss2: 0.0476 +Epoch [102/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0628, Loss2: 0.0633 +Epoch [102/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0550, Loss2: 0.0550 +Epoch [102/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0521, Loss2: 0.0522 +Epoch [102/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0521, Loss2: 0.0525 +Epoch [102/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0490, Loss2: 0.0468 +Epoch [102/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0520, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 24.3590 % Model2 24.4091 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 44.5312, Loss1: 0.0489, Loss2: 0.0516 +Epoch [103/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0482, Loss2: 0.0509 +Epoch [103/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0542, Loss2: 0.0533 +Epoch [103/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0491, Loss2: 0.0518 +Epoch [103/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0561, Loss2: 0.0556 +Epoch [103/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0499, Loss2: 0.0536 +Epoch [103/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 34.3750, Loss1: 0.0538, Loss2: 0.0588 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 24.7296 % Model2 24.5593 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0458, Loss2: 0.0461 +Epoch [104/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0544, Loss2: 0.0551 +Epoch [104/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0474, Loss2: 0.0445 +Epoch [104/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 40.6250, Loss1: 0.0584, Loss2: 0.0598 +Epoch [104/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0535, Loss2: 0.0515 +Epoch [104/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0548, Loss2: 0.0561 +Epoch [104/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0498, Loss2: 0.0485 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 24.9099 % Model2 24.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0706, Loss2: 0.0703 +Epoch [105/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0537, Loss2: 0.0518 +Epoch [105/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0491, Loss2: 0.0476 +Epoch [105/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0538, Loss2: 0.0514 +Epoch [105/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.8750, Loss1: 0.0616, Loss2: 0.0675 +Epoch [105/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0458, Loss2: 0.0466 +Epoch [105/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0678, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 24.5393 % Model2 24.3490 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0612, Loss2: 0.0623 +Epoch [106/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0525, Loss2: 0.0527 +Epoch [106/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0560, Loss2: 0.0531 +Epoch [106/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0638, Loss2: 0.0623 +Epoch [106/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 41.4062, Loss1: 0.0539, Loss2: 0.0548 +Epoch [106/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0558, Loss2: 0.0545 +Epoch [106/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0552, Loss2: 0.0552 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 24.5292 % Model2 25.4407 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0544, Loss2: 0.0552 +Epoch [107/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0601, Loss2: 0.0596 +Epoch [107/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0528, Loss2: 0.0510 +Epoch [107/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.1875, Loss1: 0.0579, Loss2: 0.0592 +Epoch [107/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0520, Loss2: 0.0516 +Epoch [107/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0596, Loss2: 0.0606 +Epoch [107/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0502, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 24.3490 % Model2 24.3990 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0601, Loss2: 0.0661 +Epoch [108/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0505, Loss2: 0.0503 +Epoch [108/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0493, Loss2: 0.0467 +Epoch [108/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0629, Loss2: 0.0675 +Epoch [108/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0516, Loss2: 0.0534 +Epoch [108/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0508, Loss2: 0.0492 +Epoch [108/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0676, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 24.3089 % Model2 24.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0535, Loss2: 0.0548 +Epoch [109/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0487, Loss2: 0.0477 +Epoch [109/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0547 +Epoch [109/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0509, Loss2: 0.0513 +Epoch [109/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0590, Loss2: 0.0562 +Epoch [109/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0705, Loss2: 0.0671 +Epoch [109/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 54.6875, Loss1: 0.0546, Loss2: 0.0505 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 24.1687 % Model2 25.0901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0570, Loss2: 0.0578 +Epoch [110/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0506, Loss2: 0.0510 +Epoch [110/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0542, Loss2: 0.0531 +Epoch [110/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0545, Loss2: 0.0517 +Epoch [110/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0630, Loss2: 0.0579 +Epoch [110/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0568, Loss2: 0.0580 +Epoch [110/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0504, Loss2: 0.0504 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 24.7796 % Model2 25.0000 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0553, Loss2: 0.0546 +Epoch [111/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0551, Loss2: 0.0557 +Epoch [111/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0569, Loss2: 0.0566 +Epoch [111/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0586, Loss2: 0.0572 +Epoch [111/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0560, Loss2: 0.0542 +Epoch [111/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 52.3438, Loss1: 0.0567, Loss2: 0.0511 +Epoch [111/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0452, Loss2: 0.0455 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 24.4391 % Model2 24.9399 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0557, Loss2: 0.0567 +Epoch [112/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0633, Loss2: 0.0615 +Epoch [112/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0567, Loss2: 0.0594 +Epoch [112/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0538, Loss2: 0.0538 +Epoch [112/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0521, Loss2: 0.0502 +Epoch [112/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0556, Loss2: 0.0534 +Epoch [112/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0524, Loss2: 0.0531 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 24.4391 % Model2 25.1302 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0686, Loss2: 0.0671 +Epoch [113/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0571, Loss2: 0.0580 +Epoch [113/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0637, Loss2: 0.0622 +Epoch [113/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.8750, Loss1: 0.0571, Loss2: 0.0529 +Epoch [113/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0490, Loss2: 0.0479 +Epoch [113/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0561, Loss2: 0.0567 +Epoch [113/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0618, Loss2: 0.0580 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 25.3806 % Model2 25.3205 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0435, Loss2: 0.0414 +Epoch [114/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0556, Loss2: 0.0525 +Epoch [114/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0634, Loss2: 0.0680 +Epoch [114/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0623, Loss2: 0.0587 +Epoch [114/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0499, Loss2: 0.0492 +Epoch [114/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0565, Loss2: 0.0550 +Epoch [114/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0535, Loss2: 0.0556 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 24.5192 % Model2 24.3790 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0517, Loss2: 0.0535 +Epoch [115/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0513, Loss2: 0.0493 +Epoch [115/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0712, Loss2: 0.0668 +Epoch [115/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0537, Loss2: 0.0529 +Epoch [115/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0509, Loss2: 0.0474 +Epoch [115/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0463, Loss2: 0.0468 +Epoch [115/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0578, Loss2: 0.0580 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 24.5693 % Model2 24.2488 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0614, Loss2: 0.0621 +Epoch [116/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0690, Loss2: 0.0718 +Epoch [116/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0536, Loss2: 0.0523 +Epoch [116/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0474, Loss2: 0.0490 +Epoch [116/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0551, Loss2: 0.0574 +Epoch [116/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0535, Loss2: 0.0503 +Epoch [116/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 54.6875, Loss1: 0.0694, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 24.7596 % Model2 24.4591 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0667, Loss2: 0.0637 +Epoch [117/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0487, Loss2: 0.0502 +Epoch [117/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0519, Loss2: 0.0528 +Epoch [117/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0555, Loss2: 0.0546 +Epoch [117/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 37.5000, Loss1: 0.0538, Loss2: 0.0600 +Epoch [117/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0601, Loss2: 0.0600 +Epoch [117/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0500, Loss2: 0.0494 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 25.4307 % Model2 25.3706 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0501, Loss2: 0.0487 +Epoch [118/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 50.0000, Loss1: 0.0608, Loss2: 0.0660 +Epoch [118/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0541, Loss2: 0.0547 +Epoch [118/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0504, Loss2: 0.0490 +Epoch [118/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0463, Loss2: 0.0454 +Epoch [118/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 46.0938, Loss1: 0.0482, Loss2: 0.0455 +Epoch [118/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0632, Loss2: 0.0631 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 25.4507 % Model2 25.0701 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 45.3125, Loss1: 0.0528, Loss2: 0.0550 +Epoch [119/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0709, Loss2: 0.0658 +Epoch [119/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0613, Loss2: 0.0612 +Epoch [119/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0544, Loss2: 0.0523 +Epoch [119/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0583, Loss2: 0.0574 +Epoch [119/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0466, Loss2: 0.0453 +Epoch [119/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0566, Loss2: 0.0572 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 24.5693 % Model2 24.7496 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0456, Loss2: 0.0474 +Epoch [120/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0580, Loss2: 0.0596 +Epoch [120/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0557, Loss2: 0.0551 +Epoch [120/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0529, Loss2: 0.0544 +Epoch [120/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0546, Loss2: 0.0527 +Epoch [120/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0597, Loss2: 0.0612 +Epoch [120/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0490, Loss2: 0.0483 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 25.1703 % Model2 24.4091 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0582, Loss2: 0.0613 +Epoch [121/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0542, Loss2: 0.0575 +Epoch [121/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0504, Loss2: 0.0505 +Epoch [121/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0725, Loss2: 0.0695 +Epoch [121/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0640, Loss2: 0.0649 +Epoch [121/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0606, Loss2: 0.0607 +Epoch [121/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0549, Loss2: 0.0582 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 25.3405 % Model2 24.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0602, Loss2: 0.0566 +Epoch [122/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0603, Loss2: 0.0624 +Epoch [122/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0518, Loss2: 0.0539 +Epoch [122/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0537, Loss2: 0.0570 +Epoch [122/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0509, Loss2: 0.0505 +Epoch [122/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 40.6250, Loss1: 0.0593, Loss2: 0.0630 +Epoch [122/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0548, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 25.2304 % Model2 24.8598 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0631, Loss2: 0.0591 +Epoch [123/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0598, Loss2: 0.0622 +Epoch [123/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0495, Loss2: 0.0493 +Epoch [123/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0524, Loss2: 0.0542 +Epoch [123/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0547, Loss2: 0.0552 +Epoch [123/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0615, Loss2: 0.0592 +Epoch [123/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0520, Loss2: 0.0532 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 25.8614 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0636, Loss2: 0.0634 +Epoch [124/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0498, Loss2: 0.0513 +Epoch [124/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0498, Loss2: 0.0483 +Epoch [124/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0638, Loss2: 0.0614 +Epoch [124/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0512, Loss2: 0.0530 +Epoch [124/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0558, Loss2: 0.0534 +Epoch [124/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0580, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 24.4091 % Model2 24.3389 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0484, Loss2: 0.0499 +Epoch [125/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0648, Loss2: 0.0650 +Epoch [125/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0516, Loss2: 0.0529 +Epoch [125/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 41.4062, Loss1: 0.0590, Loss2: 0.0621 +Epoch [125/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0684, Loss2: 0.0674 +Epoch [125/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0509, Loss2: 0.0500 +Epoch [125/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0626, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 25.5809 % Model2 24.7196 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0555, Loss2: 0.0570 +Epoch [126/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0489, Loss2: 0.0479 +Epoch [126/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0580, Loss2: 0.0569 +Epoch [126/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 44.5312, Loss1: 0.0593, Loss2: 0.0637 +Epoch [126/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0468, Loss2: 0.0446 +Epoch [126/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0435, Loss2: 0.0438 +Epoch [126/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0467, Loss2: 0.0476 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 24.8998 % Model2 24.9399 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0652, Loss2: 0.0638 +Epoch [127/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0440, Loss2: 0.0466 +Epoch [127/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0587, Loss2: 0.0593 +Epoch [127/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0614, Loss2: 0.0606 +Epoch [127/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0609, Loss2: 0.0595 +Epoch [127/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0657, Loss2: 0.0643 +Epoch [127/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0543, Loss2: 0.0559 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 25.3105 % Model2 25.2204 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0522, Loss2: 0.0519 +Epoch [128/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0666, Loss2: 0.0631 +Epoch [128/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0575, Loss2: 0.0624 +Epoch [128/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0640, Loss2: 0.0624 +Epoch [128/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0571, Loss2: 0.0552 +Epoch [128/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0800, Loss2: 0.0792 +Epoch [128/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0588, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 24.5092 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0626, Loss2: 0.0595 +Epoch [129/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0513, Loss2: 0.0495 +Epoch [129/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0498, Loss2: 0.0489 +Epoch [129/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0679, Loss2: 0.0736 +Epoch [129/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0590, Loss2: 0.0579 +Epoch [129/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0611, Loss2: 0.0611 +Epoch [129/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0644, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 24.2688 % Model2 24.3289 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0755, Loss2: 0.0749 +Epoch [130/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0539, Loss2: 0.0580 +Epoch [130/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0577, Loss2: 0.0557 +Epoch [130/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0539, Loss2: 0.0555 +Epoch [130/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0463, Loss2: 0.0467 +Epoch [130/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0748, Loss2: 0.0705 +Epoch [130/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0615, Loss2: 0.0574 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 25.2704 % Model2 24.5593 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0591, Loss2: 0.0604 +Epoch [131/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0740, Loss2: 0.0678 +Epoch [131/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0681, Loss2: 0.0685 +Epoch [131/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0658, Loss2: 0.0640 +Epoch [131/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0688, Loss2: 0.0717 +Epoch [131/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0526, Loss2: 0.0534 +Epoch [131/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0597, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 24.5192 % Model2 24.4692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0615, Loss2: 0.0576 +Epoch [132/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0528, Loss2: 0.0513 +Epoch [132/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0607, Loss2: 0.0596 +Epoch [132/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0562, Loss2: 0.0547 +Epoch [132/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0540, Loss2: 0.0554 +Epoch [132/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0622, Loss2: 0.0616 +Epoch [132/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0516, Loss2: 0.0536 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 24.3690 % Model2 24.1486 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0473, Loss2: 0.0469 +Epoch [133/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 47.6562, Loss1: 0.0603, Loss2: 0.0563 +Epoch [133/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0568, Loss2: 0.0559 +Epoch [133/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0556, Loss2: 0.0571 +Epoch [133/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0574, Loss2: 0.0583 +Epoch [133/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0578, Loss2: 0.0573 +Epoch [133/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0687, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 25.1402 % Model2 25.1102 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0712, Loss2: 0.0729 +Epoch [134/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0556, Loss2: 0.0558 +Epoch [134/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0640, Loss2: 0.0597 +Epoch [134/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0462, Loss2: 0.0435 +Epoch [134/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0628, Loss2: 0.0632 +Epoch [134/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0595, Loss2: 0.0621 +Epoch [134/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0606, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 24.8397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0573, Loss2: 0.0598 +Epoch [135/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0557, Loss2: 0.0553 +Epoch [135/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0494, Loss2: 0.0493 +Epoch [135/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0517, Loss2: 0.0511 +Epoch [135/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0560, Loss2: 0.0575 +Epoch [135/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0630, Loss2: 0.0633 +Epoch [135/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0666, Loss2: 0.0676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 24.5893 % Model2 24.3990 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0671, Loss2: 0.0637 +Epoch [136/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0530, Loss2: 0.0504 +Epoch [136/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 49.2188, Loss1: 0.0701, Loss2: 0.0789 +Epoch [136/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0487, Loss2: 0.0485 +Epoch [136/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 40.6250, Loss1: 0.0574, Loss2: 0.0617 +Epoch [136/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0702, Loss2: 0.0681 +Epoch [136/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0569, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 24.4692 % Model2 24.6494 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0600, Loss2: 0.0583 +Epoch [137/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0557, Loss2: 0.0556 +Epoch [137/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0645, Loss2: 0.0613 +Epoch [137/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0566, Loss2: 0.0553 +Epoch [137/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0602, Loss2: 0.0617 +Epoch [137/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0606, Loss2: 0.0605 +Epoch [137/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0581, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 24.9800 % Model2 24.4992 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0614, Loss2: 0.0607 +Epoch [138/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0670, Loss2: 0.0653 +Epoch [138/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0713, Loss2: 0.0702 +Epoch [138/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0533, Loss2: 0.0524 +Epoch [138/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0675, Loss2: 0.0664 +Epoch [138/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0898, Loss2: 0.0860 +Epoch [138/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0551, Loss2: 0.0556 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 24.9599 % Model2 24.6394 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0587, Loss2: 0.0597 +Epoch [139/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0613, Loss2: 0.0597 +Epoch [139/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0697, Loss2: 0.0660 +Epoch [139/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0750, Loss2: 0.0721 +Epoch [139/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0571, Loss2: 0.0534 +Epoch [139/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0531, Loss2: 0.0505 +Epoch [139/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0716, Loss2: 0.0698 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 24.7095 % Model2 24.0084 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0559, Loss2: 0.0550 +Epoch [140/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0595, Loss2: 0.0589 +Epoch [140/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0559, Loss2: 0.0542 +Epoch [140/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0621, Loss2: 0.0603 +Epoch [140/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0597, Loss2: 0.0595 +Epoch [140/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0658, Loss2: 0.0669 +Epoch [140/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0565, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 25.0100 % Model2 24.5092 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0685, Loss2: 0.0697 +Epoch [141/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0638, Loss2: 0.0656 +Epoch [141/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0569, Loss2: 0.0539 +Epoch [141/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0559, Loss2: 0.0550 +Epoch [141/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 51.5625, Loss1: 0.0620, Loss2: 0.0666 +Epoch [141/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0497, Loss2: 0.0504 +Epoch [141/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0619, Loss2: 0.0597 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 24.7997 % Model2 24.9700 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0756, Loss2: 0.0784 +Epoch [142/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0618, Loss2: 0.0596 +Epoch [142/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0576, Loss2: 0.0574 +Epoch [142/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0693, Loss2: 0.0701 +Epoch [142/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0489, Loss2: 0.0489 +Epoch [142/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0711, Loss2: 0.0689 +Epoch [142/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0584, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 24.4892 % Model2 25.3005 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0531, Loss2: 0.0512 +Epoch [143/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0692, Loss2: 0.0665 +Epoch [143/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0725, Loss2: 0.0697 +Epoch [143/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0697, Loss2: 0.0773 +Epoch [143/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0556, Loss2: 0.0552 +Epoch [143/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0622, Loss2: 0.0560 +Epoch [143/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0626, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 24.7796 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0536, Loss2: 0.0543 +Epoch [144/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0893, Loss2: 0.0944 +Epoch [144/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0550, Loss2: 0.0540 +Epoch [144/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0593, Loss2: 0.0553 +Epoch [144/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0625, Loss2: 0.0671 +Epoch [144/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0608, Loss2: 0.0595 +Epoch [144/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0513, Loss2: 0.0481 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 24.4692 % Model2 24.8798 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0659, Loss2: 0.0665 +Epoch [145/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0688, Loss2: 0.0641 +Epoch [145/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0600, Loss2: 0.0577 +Epoch [145/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0642, Loss2: 0.0610 +Epoch [145/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0535, Loss2: 0.0552 +Epoch [145/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0629, Loss2: 0.0602 +Epoch [145/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0617, Loss2: 0.0637 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 24.7396 % Model2 25.2304 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0582, Loss2: 0.0536 +Epoch [146/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0626, Loss2: 0.0618 +Epoch [146/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0532, Loss2: 0.0558 +Epoch [146/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0556, Loss2: 0.0552 +Epoch [146/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0478, Loss2: 0.0468 +Epoch [146/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0488, Loss2: 0.0484 +Epoch [146/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0617, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 24.3690 % Model2 25.0200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0678, Loss2: 0.0642 +Epoch [147/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0592, Loss2: 0.0628 +Epoch [147/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0625, Loss2: 0.0610 +Epoch [147/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0541, Loss2: 0.0571 +Epoch [147/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.1875, Loss1: 0.0594, Loss2: 0.0631 +Epoch [147/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0588, Loss2: 0.0585 +Epoch [147/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.7812, Loss1: 0.0582, Loss2: 0.0548 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 24.8798 % Model2 24.1787 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0733, Loss2: 0.0710 +Epoch [148/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0639, Loss2: 0.0633 +Epoch [148/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0697, Loss2: 0.0647 +Epoch [148/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0729, Loss2: 0.0680 +Epoch [148/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0655, Loss2: 0.0645 +Epoch [148/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0730, Loss2: 0.0682 +Epoch [148/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0549, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 24.3389 % Model2 24.1286 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0595, Loss2: 0.0603 +Epoch [149/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0796, Loss2: 0.0759 +Epoch [149/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0516, Loss2: 0.0539 +Epoch [149/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0659, Loss2: 0.0665 +Epoch [149/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0569, Loss2: 0.0565 +Epoch [149/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.0000, Loss1: 0.0540, Loss2: 0.0590 +Epoch [149/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0880, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 24.5493 % Model2 24.5393 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0628, Loss2: 0.0644 +Epoch [150/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0710, Loss2: 0.0747 +Epoch [150/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0625, Loss2: 0.0647 +Epoch [150/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0506, Loss2: 0.0516 +Epoch [150/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0679, Loss2: 0.0634 +Epoch [150/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0698, Loss2: 0.0738 +Epoch [150/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0574, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 24.8498 % Model2 24.4491 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0677, Loss2: 0.0660 +Epoch [151/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 54.6875, Loss1: 0.0598, Loss2: 0.0546 +Epoch [151/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.0938, Loss1: 0.0501, Loss2: 0.0522 +Epoch [151/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0631, Loss2: 0.0629 +Epoch [151/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0584, Loss2: 0.0562 +Epoch [151/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 42.9688, Loss1: 0.0538, Loss2: 0.0565 +Epoch [151/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0588, Loss2: 0.0630 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 24.8197 % Model2 24.4892 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0932, Loss2: 0.0893 +Epoch [152/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0632, Loss2: 0.0628 +Epoch [152/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0892, Loss2: 0.0922 +Epoch [152/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0634, Loss2: 0.0618 +Epoch [152/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0531, Loss2: 0.0552 +Epoch [152/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0541, Loss2: 0.0557 +Epoch [152/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0562, Loss2: 0.0547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 24.5793 % Model2 24.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0639, Loss2: 0.0606 +Epoch [153/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0750, Loss2: 0.0755 +Epoch [153/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0647, Loss2: 0.0651 +Epoch [153/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0600, Loss2: 0.0592 +Epoch [153/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 56.2500, Loss1: 0.0690, Loss2: 0.0599 +Epoch [153/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0559, Loss2: 0.0552 +Epoch [153/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0725, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 24.5893 % Model2 24.8498 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0723, Loss2: 0.0711 +Epoch [154/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0616, Loss2: 0.0623 +Epoch [154/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0691, Loss2: 0.0662 +Epoch [154/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0658, Loss2: 0.0644 +Epoch [154/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0642, Loss2: 0.0613 +Epoch [154/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0873, Loss2: 0.0862 +Epoch [154/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0648, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 24.5994 % Model2 24.6194 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0593, Loss2: 0.0616 +Epoch [155/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 53.1250, Loss1: 0.0599, Loss2: 0.0548 +Epoch [155/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0520, Loss2: 0.0499 +Epoch [155/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0684, Loss2: 0.0681 +Epoch [155/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0508, Loss2: 0.0492 +Epoch [155/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0689, Loss2: 0.0663 +Epoch [155/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0741, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 24.7095 % Model2 24.8798 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0657, Loss2: 0.0657 +Epoch [156/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0553, Loss2: 0.0577 +Epoch [156/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0636, Loss2: 0.0632 +Epoch [156/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0552, Loss2: 0.0568 +Epoch [156/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0526, Loss2: 0.0494 +Epoch [156/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0529, Loss2: 0.0540 +Epoch [156/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0707, Loss2: 0.0705 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 24.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0654, Loss2: 0.0663 +Epoch [157/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0579, Loss2: 0.0585 +Epoch [157/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0585, Loss2: 0.0566 +Epoch [157/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0570, Loss2: 0.0616 +Epoch [157/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0662, Loss2: 0.0638 +Epoch [157/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 43.7500, Loss1: 0.0722, Loss2: 0.0766 +Epoch [157/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0641, Loss2: 0.0625 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 24.5994 % Model2 24.5693 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0625, Loss2: 0.0597 +Epoch [158/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0499, Loss2: 0.0510 +Epoch [158/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0629, Loss2: 0.0585 +Epoch [158/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0863, Loss2: 0.0841 +Epoch [158/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0528, Loss2: 0.0564 +Epoch [158/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0679, Loss2: 0.0680 +Epoch [158/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0663, Loss2: 0.0658 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 25.0601 % Model2 25.1002 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0768, Loss2: 0.0797 +Epoch [159/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0596, Loss2: 0.0638 +Epoch [159/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0631, Loss2: 0.0593 +Epoch [159/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0819, Loss2: 0.0789 +Epoch [159/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0741, Loss2: 0.0765 +Epoch [159/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0793, Loss2: 0.0753 +Epoch [159/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0611, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 24.7496 % Model2 24.9900 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 51.5625, Loss1: 0.0539, Loss2: 0.0501 +Epoch [160/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0745, Loss2: 0.0736 +Epoch [160/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0513, Loss2: 0.0517 +Epoch [160/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0568, Loss2: 0.0556 +Epoch [160/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0698, Loss2: 0.0705 +Epoch [160/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0616, Loss2: 0.0597 +Epoch [160/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0621, Loss2: 0.0608 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 24.8197 % Model2 24.5593 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0640, Loss2: 0.0626 +Epoch [161/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0608, Loss2: 0.0632 +Epoch [161/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0668, Loss2: 0.0668 +Epoch [161/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0727, Loss2: 0.0730 +Epoch [161/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0710, Loss2: 0.0703 +Epoch [161/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0637, Loss2: 0.0622 +Epoch [161/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0574, Loss2: 0.0557 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 24.9299 % Model2 24.5994 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.0000, Loss1: 0.0590, Loss2: 0.0655 +Epoch [162/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0638, Loss2: 0.0609 +Epoch [162/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0696, Loss2: 0.0724 +Epoch [162/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0589, Loss2: 0.0558 +Epoch [162/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0571, Loss2: 0.0567 +Epoch [162/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0559, Loss2: 0.0542 +Epoch [162/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0759, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 24.8097 % Model2 24.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0603, Loss2: 0.0581 +Epoch [163/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0626, Loss2: 0.0646 +Epoch [163/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0697, Loss2: 0.0731 +Epoch [163/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0795, Loss2: 0.0747 +Epoch [163/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0608, Loss2: 0.0633 +Epoch [163/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0596, Loss2: 0.0616 +Epoch [163/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0630, Loss2: 0.0608 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 24.7296 % Model2 24.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0748, Loss2: 0.0745 +Epoch [164/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0821, Loss2: 0.0822 +Epoch [164/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0527, Loss2: 0.0548 +Epoch [164/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0644, Loss2: 0.0616 +Epoch [164/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0876, Loss2: 0.0817 +Epoch [164/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 41.4062, Loss1: 0.0610, Loss2: 0.0678 +Epoch [164/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0702, Loss2: 0.0692 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 24.8598 % Model2 24.7596 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0631, Loss2: 0.0604 +Epoch [165/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0750, Loss2: 0.0727 +Epoch [165/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0812, Loss2: 0.0753 +Epoch [165/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0498, Loss2: 0.0497 +Epoch [165/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0628, Loss2: 0.0638 +Epoch [165/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0615, Loss2: 0.0599 +Epoch [165/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0662, Loss2: 0.0672 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 24.7396 % Model2 24.7296 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0604, Loss2: 0.0586 +Epoch [166/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0504, Loss2: 0.0503 +Epoch [166/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0657, Loss2: 0.0670 +Epoch [166/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0451, Loss2: 0.0455 +Epoch [166/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0669, Loss2: 0.0678 +Epoch [166/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0573, Loss2: 0.0598 +Epoch [166/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0605, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 24.5493 % Model2 24.6294 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0631, Loss2: 0.0605 +Epoch [167/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.8125, Loss1: 0.0638, Loss2: 0.0586 +Epoch [167/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0583, Loss2: 0.0613 +Epoch [167/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0538, Loss2: 0.0561 +Epoch [167/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0594, Loss2: 0.0623 +Epoch [167/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0512, Loss2: 0.0505 +Epoch [167/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0611, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 24.6394 % Model2 24.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0586, Loss2: 0.0552 +Epoch [168/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0862, Loss2: 0.0855 +Epoch [168/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0748, Loss2: 0.0693 +Epoch [168/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0658, Loss2: 0.0663 +Epoch [168/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0719, Loss2: 0.0739 +Epoch [168/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0731, Loss2: 0.0679 +Epoch [168/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0615, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 24.7596 % Model2 24.4191 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0606, Loss2: 0.0615 +Epoch [169/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0727, Loss2: 0.0701 +Epoch [169/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0742, Loss2: 0.0757 +Epoch [169/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 68.7500, Loss1: 0.0791, Loss2: 0.0677 +Epoch [169/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 43.7500, Loss1: 0.0553, Loss2: 0.0564 +Epoch [169/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0620, Loss2: 0.0611 +Epoch [169/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0689, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 24.5092 % Model2 24.2588 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0628, Loss2: 0.0635 +Epoch [170/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0616, Loss2: 0.0619 +Epoch [170/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0638, Loss2: 0.0652 +Epoch [170/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0833, Loss2: 0.0843 +Epoch [170/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0612, Loss2: 0.0620 +Epoch [170/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0806, Loss2: 0.0783 +Epoch [170/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0538, Loss2: 0.0509 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 24.6695 % Model2 24.8498 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0732, Loss2: 0.0686 +Epoch [171/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0569, Loss2: 0.0560 +Epoch [171/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0771, Loss2: 0.0754 +Epoch [171/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0592, Loss2: 0.0580 +Epoch [171/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0633, Loss2: 0.0599 +Epoch [171/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0519, Loss2: 0.0525 +Epoch [171/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0933, Loss2: 0.0821 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 24.8297 % Model2 24.8798 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0612, Loss2: 0.0593 +Epoch [172/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0635, Loss2: 0.0610 +Epoch [172/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 62.5000, Loss1: 0.0734, Loss2: 0.0653 +Epoch [172/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0609, Loss2: 0.0624 +Epoch [172/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0717, Loss2: 0.0682 +Epoch [172/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0662, Loss2: 0.0658 +Epoch [172/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0732, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 24.7296 % Model2 24.7095 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0577, Loss2: 0.0552 +Epoch [173/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0713, Loss2: 0.0748 +Epoch [173/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0675, Loss2: 0.0650 +Epoch [173/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0640, Loss2: 0.0676 +Epoch [173/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0629, Loss2: 0.0630 +Epoch [173/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0634, Loss2: 0.0620 +Epoch [173/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0574, Loss2: 0.0577 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 24.7796 % Model2 24.3389 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0803, Loss2: 0.0756 +Epoch [174/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0755, Loss2: 0.0774 +Epoch [174/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0530, Loss2: 0.0535 +Epoch [174/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0695, Loss2: 0.0732 +Epoch [174/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0555, Loss2: 0.0549 +Epoch [174/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0810, Loss2: 0.0830 +Epoch [174/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0617, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 24.9399 % Model2 24.8998 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0626, Loss2: 0.0639 +Epoch [175/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0773, Loss2: 0.0826 +Epoch [175/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0746, Loss2: 0.0752 +Epoch [175/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0856, Loss2: 0.0918 +Epoch [175/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0525, Loss2: 0.0540 +Epoch [175/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0601, Loss2: 0.0564 +Epoch [175/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0664, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 25.1102 % Model2 24.8297 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0797, Loss2: 0.0796 +Epoch [176/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0599, Loss2: 0.0576 +Epoch [176/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0751, Loss2: 0.0770 +Epoch [176/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0649, Loss2: 0.0623 +Epoch [176/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0613, Loss2: 0.0608 +Epoch [176/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0628, Loss2: 0.0598 +Epoch [176/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0611, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 24.6294 % Model2 24.8397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0701, Loss2: 0.0675 +Epoch [177/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0654, Loss2: 0.0627 +Epoch [177/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0818, Loss2: 0.0809 +Epoch [177/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0725, Loss2: 0.0686 +Epoch [177/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0605, Loss2: 0.0609 +Epoch [177/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0745, Loss2: 0.0744 +Epoch [177/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0696, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 24.8397 % Model2 24.6494 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 45.3125, Loss1: 0.0646, Loss2: 0.0694 +Epoch [178/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0763, Loss2: 0.0779 +Epoch [178/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0699, Loss2: 0.0692 +Epoch [178/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0671, Loss2: 0.0643 +Epoch [178/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0768, Loss2: 0.0753 +Epoch [178/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0574, Loss2: 0.0559 +Epoch [178/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0688, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 25.0000 % Model2 24.5793 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0755, Loss2: 0.0783 +Epoch [179/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0552, Loss2: 0.0537 +Epoch [179/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 59.3750, Loss1: 0.0709, Loss2: 0.0767 +Epoch [179/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0586, Loss2: 0.0568 +Epoch [179/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0801, Loss2: 0.0836 +Epoch [179/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0819, Loss2: 0.0753 +Epoch [179/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0648, Loss2: 0.0660 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 24.8798 % Model2 25.0901 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0794, Loss2: 0.0794 +Epoch [180/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0762, Loss2: 0.0702 +Epoch [180/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0749, Loss2: 0.0689 +Epoch [180/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0548, Loss2: 0.0564 +Epoch [180/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0751, Loss2: 0.0693 +Epoch [180/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0654, Loss2: 0.0700 +Epoch [180/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0953, Loss2: 0.0939 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 25.0300 % Model2 24.5994 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0642, Loss2: 0.0646 +Epoch [181/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0596, Loss2: 0.0603 +Epoch [181/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0716, Loss2: 0.0705 +Epoch [181/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0598, Loss2: 0.0581 +Epoch [181/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0625, Loss2: 0.0639 +Epoch [181/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0639, Loss2: 0.0626 +Epoch [181/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0532, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 24.7596 % Model2 24.8898 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0526, Loss2: 0.0530 +Epoch [182/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0799, Loss2: 0.0768 +Epoch [182/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0633, Loss2: 0.0638 +Epoch [182/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0620, Loss2: 0.0592 +Epoch [182/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0645, Loss2: 0.0609 +Epoch [182/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0589, Loss2: 0.0599 +Epoch [182/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0785, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 24.7796 % Model2 24.5292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0659, Loss2: 0.0695 +Epoch [183/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0784, Loss2: 0.0730 +Epoch [183/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0759, Loss2: 0.0725 +Epoch [183/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0674, Loss2: 0.0645 +Epoch [183/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0572, Loss2: 0.0548 +Epoch [183/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0725, Loss2: 0.0749 +Epoch [183/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0810, Loss2: 0.0809 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 24.6695 % Model2 24.7095 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0611, Loss2: 0.0639 +Epoch [184/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0695, Loss2: 0.0696 +Epoch [184/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0765, Loss2: 0.0793 +Epoch [184/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0712, Loss2: 0.0710 +Epoch [184/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0720, Loss2: 0.0643 +Epoch [184/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0693, Loss2: 0.0675 +Epoch [184/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0629, Loss2: 0.0611 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 25.0000 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0597, Loss2: 0.0588 +Epoch [185/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0640, Loss2: 0.0605 +Epoch [185/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0749, Loss2: 0.0745 +Epoch [185/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0760, Loss2: 0.0750 +Epoch [185/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0685, Loss2: 0.0715 +Epoch [185/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0641, Loss2: 0.0620 +Epoch [185/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0649, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 24.7696 % Model2 24.8998 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0657, Loss2: 0.0671 +Epoch [186/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0619, Loss2: 0.0626 +Epoch [186/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0626, Loss2: 0.0633 +Epoch [186/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0616, Loss2: 0.0606 +Epoch [186/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0700, Loss2: 0.0712 +Epoch [186/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0620, Loss2: 0.0620 +Epoch [186/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0707, Loss2: 0.0728 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 24.4992 % Model2 24.7496 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0687, Loss2: 0.0712 +Epoch [187/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0855, Loss2: 0.0881 +Epoch [187/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0761, Loss2: 0.0724 +Epoch [187/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0928, Loss2: 0.0926 +Epoch [187/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0658, Loss2: 0.0608 +Epoch [187/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0750, Loss2: 0.0782 +Epoch [187/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0747, Loss2: 0.0739 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 24.4892 % Model2 24.8698 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0809, Loss2: 0.0721 +Epoch [188/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0616, Loss2: 0.0641 +Epoch [188/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0636, Loss2: 0.0607 +Epoch [188/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0721, Loss2: 0.0684 +Epoch [188/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0568, Loss2: 0.0592 +Epoch [188/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0672, Loss2: 0.0667 +Epoch [188/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0622, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 24.7095 % Model2 24.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0565, Loss2: 0.0580 +Epoch [189/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0812, Loss2: 0.0811 +Epoch [189/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0501, Loss2: 0.0501 +Epoch [189/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0800, Loss2: 0.0719 +Epoch [189/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0642, Loss2: 0.0662 +Epoch [189/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0545, Loss2: 0.0539 +Epoch [189/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0650, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 24.6595 % Model2 24.9700 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0763, Loss2: 0.0777 +Epoch [190/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0626, Loss2: 0.0597 +Epoch [190/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 59.3750, Loss1: 0.0802, Loss2: 0.0691 +Epoch [190/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0762, Loss2: 0.0762 +Epoch [190/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0815, Loss2: 0.0765 +Epoch [190/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0832, Loss2: 0.0797 +Epoch [190/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0677, Loss2: 0.0699 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 24.5994 % Model2 24.5493 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0538, Loss2: 0.0564 +Epoch [191/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0695, Loss2: 0.0693 +Epoch [191/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0715, Loss2: 0.0706 +Epoch [191/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0715, Loss2: 0.0705 +Epoch [191/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0781, Loss2: 0.0759 +Epoch [191/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0648, Loss2: 0.0631 +Epoch [191/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0710, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 24.7396 % Model2 24.6394 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0696, Loss2: 0.0671 +Epoch [192/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0593, Loss2: 0.0586 +Epoch [192/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0849, Loss2: 0.0822 +Epoch [192/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0566, Loss2: 0.0526 +Epoch [192/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0635, Loss2: 0.0608 +Epoch [192/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0637, Loss2: 0.0633 +Epoch [192/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0663, Loss2: 0.0669 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 24.6494 % Model2 24.6294 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0630, Loss2: 0.0665 +Epoch [193/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0615, Loss2: 0.0565 +Epoch [193/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0634, Loss2: 0.0667 +Epoch [193/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0548, Loss2: 0.0534 +Epoch [193/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0601, Loss2: 0.0615 +Epoch [193/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0539, Loss2: 0.0588 +Epoch [193/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0709, Loss2: 0.0678 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 24.5493 % Model2 24.7396 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0781, Loss2: 0.0758 +Epoch [194/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0748, Loss2: 0.0733 +Epoch [194/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0492, Loss2: 0.0490 +Epoch [194/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0916, Loss2: 0.0898 +Epoch [194/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 60.1562, Loss1: 0.0755, Loss2: 0.0646 +Epoch [194/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0604, Loss2: 0.0618 +Epoch [194/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0697, Loss2: 0.0674 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 24.5994 % Model2 24.7496 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0785, Loss2: 0.0726 +Epoch [195/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0659, Loss2: 0.0642 +Epoch [195/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0786, Loss2: 0.0818 +Epoch [195/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0738, Loss2: 0.0755 +Epoch [195/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0772, Loss2: 0.0788 +Epoch [195/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0707, Loss2: 0.0721 +Epoch [195/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0693, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 24.5793 % Model2 24.6995 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0555, Loss2: 0.0566 +Epoch [196/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0648, Loss2: 0.0682 +Epoch [196/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0602, Loss2: 0.0576 +Epoch [196/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0691, Loss2: 0.0671 +Epoch [196/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0639, Loss2: 0.0608 +Epoch [196/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0661, Loss2: 0.0642 +Epoch [196/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0750, Loss2: 0.0775 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 24.6194 % Model2 24.8197 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0574, Loss2: 0.0549 +Epoch [197/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0633, Loss2: 0.0648 +Epoch [197/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0620, Loss2: 0.0630 +Epoch [197/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0791, Loss2: 0.0752 +Epoch [197/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0638, Loss2: 0.0639 +Epoch [197/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0628, Loss2: 0.0609 +Epoch [197/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0582, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 24.5793 % Model2 24.7196 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0771, Loss2: 0.0773 +Epoch [198/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0791, Loss2: 0.0773 +Epoch [198/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0694, Loss2: 0.0667 +Epoch [198/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0789, Loss2: 0.0834 +Epoch [198/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0800, Loss2: 0.0819 +Epoch [198/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0772, Loss2: 0.0781 +Epoch [198/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0603, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 24.5693 % Model2 24.6394 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0578, Loss2: 0.0601 +Epoch [199/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0641, Loss2: 0.0638 +Epoch [199/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0495, Loss2: 0.0470 +Epoch [199/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0682, Loss2: 0.0664 +Epoch [199/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0598, Loss2: 0.0591 +Epoch [199/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0601, Loss2: 0.0600 +Epoch [199/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0735, Loss2: 0.0687 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 24.6194 % Model2 24.6795 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0687, Loss2: 0.0740 +Epoch [200/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0726, Loss2: 0.0728 +Epoch [200/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0694, Loss2: 0.0707 +Epoch [200/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0633, Loss2: 0.0635 +Epoch [200/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0607, Loss2: 0.0649 +Epoch [200/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0697, Loss2: 0.0681 +Epoch [200/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0626, Loss2: 0.0640 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 24.5893 % Model2 24.7196 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_4_2.log b/other_methods/coteaching_plus/coteaching_plus_results/out_4_2.log new file mode 100644 index 0000000..2dce1fc --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_4_2.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 13.2812, Training Accuracy2: 18.7500, Loss1: 0.0182, Loss2: 0.0176 +Epoch [2/200], Iter [100/390] Training Accuracy1: 16.4062, Training Accuracy2: 16.4062, Loss1: 0.0175, Loss2: 0.0168 +Epoch [2/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0156, Loss2: 0.0159 +Epoch [2/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0159, Loss2: 0.0160 +Epoch [2/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.1250, Loss1: 0.0149, Loss2: 0.0151 +Epoch [2/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0141, Loss2: 0.0146 +Epoch [2/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 38.2512 % Model2 37.0092 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0150, Loss2: 0.0155 +Epoch [3/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0152, Loss2: 0.0151 +Epoch [3/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0156, Loss2: 0.0155 +Epoch [3/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0149, Loss2: 0.0148 +Epoch [3/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0136, Loss2: 0.0135 +Epoch [3/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 34.3750, Loss1: 0.0144, Loss2: 0.0147 +Epoch [3/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0134, Loss2: 0.0134 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 43.5697 % Model2 42.1775 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0142, Loss2: 0.0139 +Epoch [4/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0125 +Epoch [4/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0140, Loss2: 0.0139 +Epoch [4/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0126, Loss2: 0.0131 +Epoch [4/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 35.9375, Loss1: 0.0129, Loss2: 0.0133 +Epoch [4/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0138, Loss2: 0.0140 +Epoch [4/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0160, Loss2: 0.0162 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 46.2841 % Model2 45.1322 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0135, Loss2: 0.0129 +Epoch [5/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0140, Loss2: 0.0134 +Epoch [5/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0123, Loss2: 0.0117 +Epoch [5/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0147, Loss2: 0.0143 +Epoch [5/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0132, Loss2: 0.0134 +Epoch [5/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0138, Loss2: 0.0131 +Epoch [5/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0142, Loss2: 0.0139 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 46.1939 % Model2 49.5793 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0112, Loss2: 0.0110 +Epoch [6/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0115, Loss2: 0.0114 +Epoch [6/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0116, Loss2: 0.0113 +Epoch [6/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0151, Loss2: 0.0140 +Epoch [6/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.9688, Loss1: 0.0135, Loss2: 0.0126 +Epoch [6/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0118, Loss2: 0.0115 +Epoch [6/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0124, Loss2: 0.0127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 50.7011 % Model2 52.6042 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0109, Loss2: 0.0100 +Epoch [7/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0114, Loss2: 0.0113 +Epoch [7/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 41.4062, Loss1: 0.0126, Loss2: 0.0132 +Epoch [7/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0131, Loss2: 0.0124 +Epoch [7/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0132, Loss2: 0.0127 +Epoch [7/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0109, Loss2: 0.0107 +Epoch [7/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0122, Loss2: 0.0114 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 51.6326 % Model2 52.7444 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0110, Loss2: 0.0109 +Epoch [8/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0135, Loss2: 0.0137 +Epoch [8/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0101, Loss2: 0.0100 +Epoch [8/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0113, Loss2: 0.0113 +Epoch [8/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0094, Loss2: 0.0091 +Epoch [8/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0117, Loss2: 0.0113 +Epoch [8/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0119, Loss2: 0.0110 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 52.6142 % Model2 55.4587 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0114, Loss2: 0.0123 +Epoch [9/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0119, Loss2: 0.0125 +Epoch [9/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0124, Loss2: 0.0122 +Epoch [9/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 42.9688, Loss1: 0.0114, Loss2: 0.0112 +Epoch [9/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0106, Loss2: 0.0099 +Epoch [9/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0113, Loss2: 0.0112 +Epoch [9/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.8438, Loss1: 0.0114, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 53.8161 % Model2 54.6875 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 39.8438, Loss1: 0.0119, Loss2: 0.0117 +Epoch [10/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0090, Loss2: 0.0092 +Epoch [10/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0115, Loss2: 0.0108 +Epoch [10/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0121, Loss2: 0.0117 +Epoch [10/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.8750, Loss1: 0.0092, Loss2: 0.0101 +Epoch [10/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0099, Loss2: 0.0101 +Epoch [10/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 46.8750, Loss1: 0.0097, Loss2: 0.0100 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 53.9764 % Model2 55.5589 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0110, Loss2: 0.0098 +Epoch [11/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0094, Loss2: 0.0093 +Epoch [11/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0132, Loss2: 0.0120 +Epoch [11/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0087, Loss2: 0.0085 +Epoch [11/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0099, Loss2: 0.0098 +Epoch [11/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0090, Loss2: 0.0088 +Epoch [11/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 51.5625, Loss1: 0.0116, Loss2: 0.0096 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 55.4287 % Model2 55.7292 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0097, Loss2: 0.0090 +Epoch [12/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0111, Loss2: 0.0110 +Epoch [12/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0089, Loss2: 0.0096 +Epoch [12/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0112, Loss2: 0.0109 +Epoch [12/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0122, Loss2: 0.0114 +Epoch [12/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0096, Loss2: 0.0082 +Epoch [12/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0092, Loss2: 0.0087 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 54.7376 % Model2 56.1198 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0103, Loss2: 0.0102 +Epoch [13/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0113, Loss2: 0.0107 +Epoch [13/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0097, Loss2: 0.0093 +Epoch [13/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0101, Loss2: 0.0097 +Epoch [13/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0084, Loss2: 0.0080 +Epoch [13/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0087, Loss2: 0.0075 +Epoch [13/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0099, Loss2: 0.0098 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 56.1699 % Model2 57.1815 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0089, Loss2: 0.0095 +Epoch [14/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0106, Loss2: 0.0114 +Epoch [14/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0099, Loss2: 0.0086 +Epoch [14/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0108, Loss2: 0.0109 +Epoch [14/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0104, Loss2: 0.0099 +Epoch [14/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0088, Loss2: 0.0090 +Epoch [14/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0097, Loss2: 0.0095 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 54.7476 % Model2 57.0613 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0090, Loss2: 0.0098 +Epoch [15/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0107, Loss2: 0.0102 +Epoch [15/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0105, Loss2: 0.0106 +Epoch [15/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0100, Loss2: 0.0100 +Epoch [15/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0093, Loss2: 0.0090 +Epoch [15/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0101, Loss2: 0.0100 +Epoch [15/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0099, Loss2: 0.0101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 57.5321 % Model2 57.9828 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0119, Loss2: 0.0108 +Epoch [16/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0090, Loss2: 0.0090 +Epoch [16/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0085, Loss2: 0.0087 +Epoch [16/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0082, Loss2: 0.0082 +Epoch [16/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0097, Loss2: 0.0101 +Epoch [16/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0112, Loss2: 0.0106 +Epoch [16/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.7812, Loss1: 0.0085, Loss2: 0.0093 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 57.2616 % Model2 58.8041 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0094, Loss2: 0.0088 +Epoch [17/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0085, Loss2: 0.0082 +Epoch [17/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0101, Loss2: 0.0089 +Epoch [17/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0092, Loss2: 0.0078 +Epoch [17/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0086, Loss2: 0.0085 +Epoch [17/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0089, Loss2: 0.0091 +Epoch [17/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0090, Loss2: 0.0087 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 56.8910 % Model2 58.7941 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0084, Loss2: 0.0070 +Epoch [18/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0098, Loss2: 0.0091 +Epoch [18/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0096, Loss2: 0.0085 +Epoch [18/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0103, Loss2: 0.0099 +Epoch [18/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0088, Loss2: 0.0078 +Epoch [18/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0093, Loss2: 0.0086 +Epoch [18/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 61.7188, Loss1: 0.0089, Loss2: 0.0074 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 55.9996 % Model2 58.0829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0102, Loss2: 0.0108 +Epoch [19/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0096, Loss2: 0.0092 +Epoch [19/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0075, Loss2: 0.0078 +Epoch [19/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0098, Loss2: 0.0093 +Epoch [19/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0076, Loss2: 0.0077 +Epoch [19/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0101, Loss2: 0.0095 +Epoch [19/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0084, Loss2: 0.0089 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 56.9912 % Model2 58.3433 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0084, Loss2: 0.0080 +Epoch [20/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0098, Loss2: 0.0099 +Epoch [20/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0059, Loss2: 0.0064 +Epoch [20/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0088, Loss2: 0.0100 +Epoch [20/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0073, Loss2: 0.0064 +Epoch [20/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0074, Loss2: 0.0071 +Epoch [20/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0087, Loss2: 0.0088 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 57.1815 % Model2 57.8926 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0528, Loss2: 0.0522 +Epoch [21/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0635 +Epoch [21/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0602, Loss2: 0.0619 +Epoch [21/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0621, Loss2: 0.0606 +Epoch [21/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 40.6250, Loss1: 0.0396, Loss2: 0.0420 +Epoch [21/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0604, Loss2: 0.0592 +Epoch [21/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0486, Loss2: 0.0461 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 53.3353 % Model2 54.4671 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0519, Loss2: 0.0515 +Epoch [22/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0588, Loss2: 0.0570 +Epoch [22/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0708, Loss2: 0.0715 +Epoch [22/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 53.1250, Loss1: 0.0697, Loss2: 0.0620 +Epoch [22/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0587, Loss2: 0.0596 +Epoch [22/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.9062, Loss1: 0.0614, Loss2: 0.0573 +Epoch [22/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0601, Loss2: 0.0598 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 55.9996 % Model2 56.9411 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0576, Loss2: 0.0599 +Epoch [23/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0544, Loss2: 0.0538 +Epoch [23/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0567, Loss2: 0.0558 +Epoch [23/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 59.3750, Loss1: 0.0595, Loss2: 0.0551 +Epoch [23/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0629, Loss2: 0.0637 +Epoch [23/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0801, Loss2: 0.0764 +Epoch [23/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0607, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 56.3001 % Model2 58.0228 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0663, Loss2: 0.0668 +Epoch [24/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0528, Loss2: 0.0502 +Epoch [24/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0713, Loss2: 0.0708 +Epoch [24/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0751, Loss2: 0.0794 +Epoch [24/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 57.0312, Loss1: 0.0727, Loss2: 0.0665 +Epoch [24/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 36.7188, Loss1: 0.0662, Loss2: 0.0691 +Epoch [24/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0658, Loss2: 0.0661 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 57.0813 % Model2 58.2232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0547, Loss2: 0.0538 +Epoch [25/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0597, Loss2: 0.0571 +Epoch [25/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0795, Loss2: 0.0805 +Epoch [25/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0602, Loss2: 0.0576 +Epoch [25/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 40.6250, Loss1: 0.0454, Loss2: 0.0495 +Epoch [25/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0700, Loss2: 0.0698 +Epoch [25/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0567, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 57.1715 % Model2 56.4403 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 43.7500, Loss1: 0.0460, Loss2: 0.0475 +Epoch [26/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0633, Loss2: 0.0636 +Epoch [26/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0675, Loss2: 0.0702 +Epoch [26/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0656, Loss2: 0.0647 +Epoch [26/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0747, Loss2: 0.0728 +Epoch [26/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0590, Loss2: 0.0568 +Epoch [26/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0450, Loss2: 0.0445 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 56.8810 % Model2 57.2316 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 55.4688, Loss1: 0.0679, Loss2: 0.0744 +Epoch [27/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0528, Loss2: 0.0532 +Epoch [27/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0671, Loss2: 0.0621 +Epoch [27/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0607, Loss2: 0.0602 +Epoch [27/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0790, Loss2: 0.0764 +Epoch [27/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0459, Loss2: 0.0429 +Epoch [27/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0510, Loss2: 0.0537 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 56.4103 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0642, Loss2: 0.0610 +Epoch [28/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0544, Loss2: 0.0534 +Epoch [28/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0521, Loss2: 0.0490 +Epoch [28/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0549, Loss2: 0.0555 +Epoch [28/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0744, Loss2: 0.0759 +Epoch [28/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0672, Loss2: 0.0685 +Epoch [28/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0647, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 57.3718 % Model2 59.0946 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0705, Loss2: 0.0728 +Epoch [29/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0722, Loss2: 0.0728 +Epoch [29/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0658, Loss2: 0.0632 +Epoch [29/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0668, Loss2: 0.0671 +Epoch [29/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0590, Loss2: 0.0568 +Epoch [29/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0599, Loss2: 0.0624 +Epoch [29/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0539, Loss2: 0.0554 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 57.4319 % Model2 59.1947 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0662, Loss2: 0.0651 +Epoch [30/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0744, Loss2: 0.0708 +Epoch [30/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0654, Loss2: 0.0633 +Epoch [30/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0660, Loss2: 0.0635 +Epoch [30/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0484 +Epoch [30/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0544, Loss2: 0.0525 +Epoch [30/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0558, Loss2: 0.0530 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 57.1314 % Model2 59.5553 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0634, Loss2: 0.0625 +Epoch [31/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0666, Loss2: 0.0642 +Epoch [31/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0649, Loss2: 0.0612 +Epoch [31/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 52.3438, Loss1: 0.0616, Loss2: 0.0658 +Epoch [31/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 50.7812, Loss1: 0.0491, Loss2: 0.0454 +Epoch [31/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0636, Loss2: 0.0631 +Epoch [31/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0778, Loss2: 0.0792 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 57.7825 % Model2 59.6354 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0488, Loss2: 0.0508 +Epoch [32/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0819, Loss2: 0.0809 +Epoch [32/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0520, Loss2: 0.0514 +Epoch [32/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0628, Loss2: 0.0637 +Epoch [32/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0642, Loss2: 0.0671 +Epoch [32/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0559, Loss2: 0.0556 +Epoch [32/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0620, Loss2: 0.0643 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 57.6723 % Model2 56.8510 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0499, Loss2: 0.0528 +Epoch [33/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0782, Loss2: 0.0749 +Epoch [33/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0574, Loss2: 0.0595 +Epoch [33/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0743, Loss2: 0.0697 +Epoch [33/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0550, Loss2: 0.0593 +Epoch [33/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0764, Loss2: 0.0797 +Epoch [33/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0661, Loss2: 0.0680 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 57.2416 % Model2 57.7925 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0779, Loss2: 0.0779 +Epoch [34/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0703, Loss2: 0.0691 +Epoch [34/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0598, Loss2: 0.0594 +Epoch [34/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0648, Loss2: 0.0621 +Epoch [34/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0509, Loss2: 0.0508 +Epoch [34/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0790, Loss2: 0.0747 +Epoch [34/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0727, Loss2: 0.0730 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 58.0329 % Model2 58.1530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0708, Loss2: 0.0701 +Epoch [35/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0716, Loss2: 0.0663 +Epoch [35/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0678, Loss2: 0.0653 +Epoch [35/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0553, Loss2: 0.0575 +Epoch [35/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0599, Loss2: 0.0617 +Epoch [35/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 64.0625, Loss1: 0.0662, Loss2: 0.0599 +Epoch [35/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0974, Loss2: 0.0941 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 58.1831 % Model2 59.6454 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0667, Loss2: 0.0637 +Epoch [36/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 49.2188, Loss1: 0.0543, Loss2: 0.0575 +Epoch [36/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0685, Loss2: 0.0607 +Epoch [36/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0677, Loss2: 0.0657 +Epoch [36/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0583, Loss2: 0.0579 +Epoch [36/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0527, Loss2: 0.0523 +Epoch [36/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0711, Loss2: 0.0767 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 57.3818 % Model2 58.5938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0937, Loss2: 0.0961 +Epoch [37/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0744, Loss2: 0.0737 +Epoch [37/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0579, Loss2: 0.0573 +Epoch [37/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0789, Loss2: 0.0793 +Epoch [37/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0683, Loss2: 0.0659 +Epoch [37/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0655, Loss2: 0.0668 +Epoch [37/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0491, Loss2: 0.0505 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 57.3918 % Model2 59.2248 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0615, Loss2: 0.0573 +Epoch [38/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0601, Loss2: 0.0626 +Epoch [38/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0568, Loss2: 0.0576 +Epoch [38/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0711, Loss2: 0.0689 +Epoch [38/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0996, Loss2: 0.0937 +Epoch [38/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0663, Loss2: 0.0702 +Epoch [38/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0700, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 57.1514 % Model2 59.5553 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0720, Loss2: 0.0670 +Epoch [39/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.1000, Loss2: 0.0961 +Epoch [39/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0908, Loss2: 0.0815 +Epoch [39/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0897, Loss2: 0.0894 +Epoch [39/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 60.1562, Loss1: 0.0590, Loss2: 0.0538 +Epoch [39/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0544, Loss2: 0.0587 +Epoch [39/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0589, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 58.1130 % Model2 57.8025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 59.3750, Loss1: 0.0673, Loss2: 0.0614 +Epoch [40/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0660, Loss2: 0.0667 +Epoch [40/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0728, Loss2: 0.0738 +Epoch [40/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0660, Loss2: 0.0669 +Epoch [40/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0726, Loss2: 0.0701 +Epoch [40/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0805, Loss2: 0.0728 +Epoch [40/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0568, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 57.8125 % Model2 59.4151 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0696 +Epoch [41/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 58.5938, Loss1: 0.0690, Loss2: 0.0629 +Epoch [41/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0688, Loss2: 0.0619 +Epoch [41/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0660, Loss2: 0.0707 +Epoch [41/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0530, Loss2: 0.0526 +Epoch [41/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0597, Loss2: 0.0572 +Epoch [41/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 53.9062, Loss1: 0.0518, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 56.8710 % Model2 59.1046 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0627, Loss2: 0.0615 +Epoch [42/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0888, Loss2: 0.0889 +Epoch [42/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 49.2188, Loss1: 0.0553, Loss2: 0.0567 +Epoch [42/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0569, Loss2: 0.0558 +Epoch [42/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 67.1875, Loss1: 0.0649, Loss2: 0.0579 +Epoch [42/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0611, Loss2: 0.0647 +Epoch [42/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0639, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 57.7524 % Model2 59.3950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.9375, Loss1: 0.0727, Loss2: 0.0782 +Epoch [43/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0691, Loss2: 0.0706 +Epoch [43/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0557, Loss2: 0.0523 +Epoch [43/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0648, Loss2: 0.0652 +Epoch [43/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0613, Loss2: 0.0582 +Epoch [43/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0594, Loss2: 0.0597 +Epoch [43/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0881, Loss2: 0.0887 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 57.2015 % Model2 59.8257 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0599, Loss2: 0.0641 +Epoch [44/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0739, Loss2: 0.0721 +Epoch [44/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0613, Loss2: 0.0600 +Epoch [44/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0825, Loss2: 0.0832 +Epoch [44/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0762, Loss2: 0.0776 +Epoch [44/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 62.5000, Loss1: 0.0616, Loss2: 0.0557 +Epoch [44/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 50.7812, Loss1: 0.0555, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 58.0228 % Model2 59.8758 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0858, Loss2: 0.0852 +Epoch [45/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0566, Loss2: 0.0618 +Epoch [45/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0851, Loss2: 0.0923 +Epoch [45/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0697, Loss2: 0.0718 +Epoch [45/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0602, Loss2: 0.0565 +Epoch [45/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0528, Loss2: 0.0501 +Epoch [45/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0648 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 56.7408 % Model2 58.6438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0678, Loss2: 0.0677 +Epoch [46/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0937, Loss2: 0.0948 +Epoch [46/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0885, Loss2: 0.0878 +Epoch [46/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0609, Loss2: 0.0625 +Epoch [46/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0549, Loss2: 0.0587 +Epoch [46/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0744, Loss2: 0.0733 +Epoch [46/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0723, Loss2: 0.0719 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 58.1030 % Model2 60.7372 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0630, Loss2: 0.0609 +Epoch [47/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0633, Loss2: 0.0621 +Epoch [47/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0766, Loss2: 0.0794 +Epoch [47/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0527, Loss2: 0.0543 +Epoch [47/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0610, Loss2: 0.0572 +Epoch [47/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0750, Loss2: 0.0773 +Epoch [47/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0571, Loss2: 0.0568 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 57.0513 % Model2 59.3049 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0766, Loss2: 0.0692 +Epoch [48/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0620, Loss2: 0.0672 +Epoch [48/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0751, Loss2: 0.0801 +Epoch [48/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0607, Loss2: 0.0587 +Epoch [48/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 51.5625, Loss1: 0.0512, Loss2: 0.0560 +Epoch [48/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0789, Loss2: 0.0755 +Epoch [48/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0631, Loss2: 0.0675 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 57.9427 % Model2 58.7841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0691, Loss2: 0.0664 +Epoch [49/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0855, Loss2: 0.0764 +Epoch [49/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0701, Loss2: 0.0787 +Epoch [49/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 53.9062, Loss1: 0.0609, Loss2: 0.0655 +Epoch [49/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0894, Loss2: 0.0849 +Epoch [49/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0762, Loss2: 0.0724 +Epoch [49/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0510, Loss2: 0.0536 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 58.1931 % Model2 58.7139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0625, Loss2: 0.0660 +Epoch [50/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0627, Loss2: 0.0635 +Epoch [50/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0577, Loss2: 0.0586 +Epoch [50/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0716, Loss2: 0.0743 +Epoch [50/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0772, Loss2: 0.0746 +Epoch [50/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0601, Loss2: 0.0570 +Epoch [50/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0726, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 57.7424 % Model2 57.9327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0622, Loss2: 0.0622 +Epoch [51/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0719, Loss2: 0.0662 +Epoch [51/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0676, Loss2: 0.0669 +Epoch [51/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0537, Loss2: 0.0541 +Epoch [51/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0599, Loss2: 0.0639 +Epoch [51/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0567, Loss2: 0.0589 +Epoch [51/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.9375, Loss1: 0.0609, Loss2: 0.0625 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 57.4920 % Model2 59.0946 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0611, Loss2: 0.0596 +Epoch [52/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0674, Loss2: 0.0724 +Epoch [52/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0606, Loss2: 0.0572 +Epoch [52/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 66.4062, Loss1: 0.0552, Loss2: 0.0508 +Epoch [52/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0773, Loss2: 0.0779 +Epoch [52/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0576, Loss2: 0.0547 +Epoch [52/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0760, Loss2: 0.0717 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 57.0312 % Model2 59.5152 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0628, Loss2: 0.0625 +Epoch [53/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0655, Loss2: 0.0663 +Epoch [53/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0902, Loss2: 0.0807 +Epoch [53/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0855, Loss2: 0.0815 +Epoch [53/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0927, Loss2: 0.1002 +Epoch [53/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0682, Loss2: 0.0676 +Epoch [53/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0551, Loss2: 0.0574 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 56.6506 % Model2 58.4635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0700, Loss2: 0.0665 +Epoch [54/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0627, Loss2: 0.0639 +Epoch [54/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0637, Loss2: 0.0690 +Epoch [54/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0687, Loss2: 0.0743 +Epoch [54/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0593, Loss2: 0.0566 +Epoch [54/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0644, Loss2: 0.0630 +Epoch [54/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0803, Loss2: 0.0782 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 57.9026 % Model2 59.5753 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0610, Loss2: 0.0612 +Epoch [55/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0745, Loss2: 0.0744 +Epoch [55/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0724, Loss2: 0.0667 +Epoch [55/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0574, Loss2: 0.0605 +Epoch [55/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0745, Loss2: 0.0760 +Epoch [55/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0722, Loss2: 0.0740 +Epoch [55/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0997, Loss2: 0.0936 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 57.1114 % Model2 59.3750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0865, Loss2: 0.0826 +Epoch [56/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0776 +Epoch [56/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0747, Loss2: 0.0785 +Epoch [56/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0637, Loss2: 0.0632 +Epoch [56/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0680, Loss2: 0.0675 +Epoch [56/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0551, Loss2: 0.0556 +Epoch [56/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 70.3125, Loss1: 0.0619, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 58.0829 % Model2 59.5353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0635, Loss2: 0.0619 +Epoch [57/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0885, Loss2: 0.0852 +Epoch [57/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0615, Loss2: 0.0637 +Epoch [57/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0605, Loss2: 0.0630 +Epoch [57/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0626, Loss2: 0.0622 +Epoch [57/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0652, Loss2: 0.0681 +Epoch [57/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0747, Loss2: 0.0696 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 57.6522 % Model2 59.6354 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0763, Loss2: 0.0697 +Epoch [58/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 55.4688, Loss1: 0.0620, Loss2: 0.0706 +Epoch [58/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 59.3750, Loss1: 0.0628, Loss2: 0.0557 +Epoch [58/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0639, Loss2: 0.0627 +Epoch [58/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0661, Loss2: 0.0666 +Epoch [58/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0520, Loss2: 0.0509 +Epoch [58/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0633, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 56.9311 % Model2 58.7540 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0675, Loss2: 0.0680 +Epoch [59/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0792, Loss2: 0.0733 +Epoch [59/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0433, Loss2: 0.0425 +Epoch [59/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0869, Loss2: 0.0840 +Epoch [59/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0676, Loss2: 0.0688 +Epoch [59/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0716, Loss2: 0.0707 +Epoch [59/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0651, Loss2: 0.0663 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 57.2015 % Model2 59.3349 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0575, Loss2: 0.0597 +Epoch [60/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0673, Loss2: 0.0727 +Epoch [60/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0609, Loss2: 0.0649 +Epoch [60/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0762, Loss2: 0.0790 +Epoch [60/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0743, Loss2: 0.0743 +Epoch [60/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0604, Loss2: 0.0642 +Epoch [60/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1009, Loss2: 0.0988 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 56.9712 % Model2 58.5938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0588, Loss2: 0.0549 +Epoch [61/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0581, Loss2: 0.0574 +Epoch [61/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0771, Loss2: 0.0735 +Epoch [61/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0550, Loss2: 0.0531 +Epoch [61/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0595, Loss2: 0.0554 +Epoch [61/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0570, Loss2: 0.0581 +Epoch [61/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 56.2500, Loss1: 0.0624, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 56.8409 % Model2 59.0745 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0648, Loss2: 0.0620 +Epoch [62/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0595, Loss2: 0.0567 +Epoch [62/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0593, Loss2: 0.0581 +Epoch [62/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0743, Loss2: 0.0729 +Epoch [62/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0714, Loss2: 0.0682 +Epoch [62/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0653, Loss2: 0.0613 +Epoch [62/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0564, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 57.6222 % Model2 58.4635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0665, Loss2: 0.0691 +Epoch [63/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 65.6250, Loss1: 0.0715, Loss2: 0.0641 +Epoch [63/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0613, Loss2: 0.0596 +Epoch [63/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1077, Loss2: 0.1084 +Epoch [63/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0690, Loss2: 0.0640 +Epoch [63/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0721, Loss2: 0.0718 +Epoch [63/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0653, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 57.2115 % Model2 59.5353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0716, Loss2: 0.0747 +Epoch [64/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0800, Loss2: 0.0726 +Epoch [64/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0785, Loss2: 0.0737 +Epoch [64/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0859, Loss2: 0.0822 +Epoch [64/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0465, Loss2: 0.0449 +Epoch [64/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0557, Loss2: 0.0555 +Epoch [64/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0825, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 57.2716 % Model2 58.0629 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0624, Loss2: 0.0643 +Epoch [65/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0691, Loss2: 0.0706 +Epoch [65/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0635, Loss2: 0.0672 +Epoch [65/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0757, Loss2: 0.0782 +Epoch [65/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0746, Loss2: 0.0760 +Epoch [65/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0699, Loss2: 0.0656 +Epoch [65/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0786, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 57.1214 % Model2 58.4736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0725, Loss2: 0.0698 +Epoch [66/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0742, Loss2: 0.0693 +Epoch [66/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0566, Loss2: 0.0558 +Epoch [66/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0655, Loss2: 0.0638 +Epoch [66/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0564, Loss2: 0.0526 +Epoch [66/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0632, Loss2: 0.0658 +Epoch [66/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 72.6562, Loss1: 0.0862, Loss2: 0.0741 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 57.3818 % Model2 58.9944 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0563, Loss2: 0.0560 +Epoch [67/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.0625, Loss2: 0.0672 +Epoch [67/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0714, Loss2: 0.0715 +Epoch [67/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0812, Loss2: 0.0809 +Epoch [67/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0696, Loss2: 0.0685 +Epoch [67/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0798, Loss2: 0.0878 +Epoch [67/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0605, Loss2: 0.0605 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 58.3433 % Model2 58.8942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0650, Loss2: 0.0638 +Epoch [68/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0662, Loss2: 0.0696 +Epoch [68/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.0654, Loss2: 0.0700 +Epoch [68/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1036, Loss2: 0.1049 +Epoch [68/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0663, Loss2: 0.0680 +Epoch [68/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0880, Loss2: 0.0861 +Epoch [68/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0744, Loss2: 0.0709 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 57.3017 % Model2 58.8241 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 60.9375, Loss1: 0.0570, Loss2: 0.0672 +Epoch [69/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0791, Loss2: 0.0706 +Epoch [69/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0628, Loss2: 0.0669 +Epoch [69/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0814, Loss2: 0.0816 +Epoch [69/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0615, Loss2: 0.0605 +Epoch [69/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0835, Loss2: 0.0857 +Epoch [69/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0714, Loss2: 0.0706 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 57.0012 % Model2 58.8942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 67.9688, Loss1: 0.0736, Loss2: 0.0637 +Epoch [70/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0674, Loss2: 0.0633 +Epoch [70/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0783, Loss2: 0.0784 +Epoch [70/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0789, Loss2: 0.0759 +Epoch [70/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0777, Loss2: 0.0745 +Epoch [70/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.0827, Loss2: 0.0892 +Epoch [70/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 62.5000, Loss1: 0.0541, Loss2: 0.0494 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 56.7608 % Model2 58.6939 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0579, Loss2: 0.0597 +Epoch [71/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0950, Loss2: 0.0963 +Epoch [71/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0811, Loss2: 0.0847 +Epoch [71/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0697, Loss2: 0.0679 +Epoch [71/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0728, Loss2: 0.0683 +Epoch [71/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0566, Loss2: 0.0591 +Epoch [71/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0792, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 57.1014 % Model2 58.4435 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1004, Loss2: 0.0938 +Epoch [72/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0552, Loss2: 0.0509 +Epoch [72/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0653, Loss2: 0.0678 +Epoch [72/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0895, Loss2: 0.0823 +Epoch [72/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0727, Loss2: 0.0705 +Epoch [72/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0885, Loss2: 0.0903 +Epoch [72/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0823, Loss2: 0.0831 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 57.2115 % Model2 58.5437 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0880, Loss2: 0.0873 +Epoch [73/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0863, Loss2: 0.0813 +Epoch [73/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0755, Loss2: 0.0780 +Epoch [73/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0639, Loss2: 0.0669 +Epoch [73/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0573, Loss2: 0.0577 +Epoch [73/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0784, Loss2: 0.0771 +Epoch [73/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.0625, Loss1: 0.0543, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 57.4419 % Model2 59.1046 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0855, Loss2: 0.0941 +Epoch [74/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0848, Loss2: 0.0889 +Epoch [74/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0743, Loss2: 0.0788 +Epoch [74/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0721, Loss2: 0.0746 +Epoch [74/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0659, Loss2: 0.0660 +Epoch [74/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 58.5938, Loss1: 0.0674, Loss2: 0.0766 +Epoch [74/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0733, Loss2: 0.0750 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 57.3618 % Model2 59.0144 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0897, Loss2: 0.0863 +Epoch [75/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0650, Loss2: 0.0647 +Epoch [75/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0684, Loss2: 0.0687 +Epoch [75/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0704, Loss2: 0.0705 +Epoch [75/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0986, Loss2: 0.0930 +Epoch [75/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0684, Loss2: 0.0666 +Epoch [75/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0745, Loss2: 0.0699 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 56.8710 % Model2 58.9643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0810, Loss2: 0.0759 +Epoch [76/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0585, Loss2: 0.0615 +Epoch [76/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1029, Loss2: 0.1073 +Epoch [76/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0554, Loss2: 0.0571 +Epoch [76/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0625, Loss2: 0.0685 +Epoch [76/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0688, Loss2: 0.0701 +Epoch [76/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0832, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 57.1214 % Model2 59.3049 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0850, Loss2: 0.0842 +Epoch [77/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 72.6562, Loss1: 0.0793, Loss2: 0.0694 +Epoch [77/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0657, Loss2: 0.0640 +Epoch [77/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0602, Loss2: 0.0615 +Epoch [77/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0617, Loss2: 0.0669 +Epoch [77/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0623, Loss2: 0.0667 +Epoch [77/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0870, Loss2: 0.0893 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 56.9912 % Model2 58.1931 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0869, Loss2: 0.0848 +Epoch [78/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0639, Loss2: 0.0620 +Epoch [78/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0621, Loss2: 0.0592 +Epoch [78/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0660, Loss2: 0.0633 +Epoch [78/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0750, Loss2: 0.0698 +Epoch [78/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0762, Loss2: 0.0712 +Epoch [78/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0783, Loss2: 0.0816 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 57.2516 % Model2 58.9844 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0614, Loss2: 0.0597 +Epoch [79/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 69.5312, Loss1: 0.0978, Loss2: 0.1065 +Epoch [79/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0741, Loss2: 0.0758 +Epoch [79/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1109, Loss2: 0.1130 +Epoch [79/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0761, Loss2: 0.0716 +Epoch [79/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0713, Loss2: 0.0766 +Epoch [79/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.1875, Loss1: 0.0632, Loss2: 0.0596 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 56.8409 % Model2 58.4535 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0993, Loss2: 0.0910 +Epoch [80/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0876, Loss2: 0.0876 +Epoch [80/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 65.6250, Loss1: 0.0622, Loss2: 0.0581 +Epoch [80/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0983, Loss2: 0.1024 +Epoch [80/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0647, Loss2: 0.0630 +Epoch [80/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 61.7188, Loss1: 0.0674, Loss2: 0.0719 +Epoch [80/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0827, Loss2: 0.0805 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 57.3417 % Model2 58.9443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0923, Loss2: 0.0831 +Epoch [81/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.1034, Loss2: 0.0963 +Epoch [81/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0665, Loss2: 0.0629 +Epoch [81/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0713, Loss2: 0.0709 +Epoch [81/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0671, Loss2: 0.0738 +Epoch [81/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0801, Loss2: 0.0778 +Epoch [81/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0668, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 56.5004 % Model2 58.2532 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0697, Loss2: 0.0690 +Epoch [82/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0863, Loss2: 0.0908 +Epoch [82/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0767, Loss2: 0.0755 +Epoch [82/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0788, Loss2: 0.0854 +Epoch [82/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0630, Loss2: 0.0684 +Epoch [82/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0747, Loss2: 0.0681 +Epoch [82/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0738, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 57.7324 % Model2 58.4836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0520, Loss2: 0.0531 +Epoch [83/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0903, Loss2: 0.0957 +Epoch [83/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0937, Loss2: 0.0913 +Epoch [83/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0729, Loss2: 0.0758 +Epoch [83/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0591, Loss2: 0.0615 +Epoch [83/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0981, Loss2: 0.0949 +Epoch [83/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 68.7500, Loss1: 0.0745, Loss2: 0.0666 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 56.9411 % Model2 57.9327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.1036, Loss2: 0.0924 +Epoch [84/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0663, Loss2: 0.0633 +Epoch [84/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 65.6250, Loss1: 0.0620, Loss2: 0.0549 +Epoch [84/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0575, Loss2: 0.0585 +Epoch [84/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0807, Loss2: 0.0744 +Epoch [84/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0712, Loss2: 0.0798 +Epoch [84/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0665, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 57.5120 % Model2 58.3634 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0687, Loss2: 0.0689 +Epoch [85/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0843, Loss2: 0.0876 +Epoch [85/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.0772, Loss2: 0.0701 +Epoch [85/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0739, Loss2: 0.0722 +Epoch [85/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0658 +Epoch [85/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0735, Loss2: 0.0669 +Epoch [85/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0615, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 57.2115 % Model2 58.1731 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0726, Loss2: 0.0749 +Epoch [86/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0781, Loss2: 0.0755 +Epoch [86/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0583, Loss2: 0.0561 +Epoch [86/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.0647, Loss2: 0.0597 +Epoch [86/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0625, Loss2: 0.0614 +Epoch [86/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0830, Loss2: 0.0825 +Epoch [86/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0786, Loss2: 0.0829 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 56.8009 % Model2 58.6639 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0766, Loss2: 0.0691 +Epoch [87/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0707, Loss2: 0.0692 +Epoch [87/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0799, Loss2: 0.0748 +Epoch [87/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0715, Loss2: 0.0724 +Epoch [87/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.1006, Loss2: 0.1003 +Epoch [87/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 67.1875, Loss1: 0.0682, Loss2: 0.0763 +Epoch [87/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1053, Loss2: 0.1117 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 56.9511 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0652, Loss2: 0.0629 +Epoch [88/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0896, Loss2: 0.0906 +Epoch [88/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0721, Loss2: 0.0672 +Epoch [88/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0901, Loss2: 0.0861 +Epoch [88/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0937, Loss2: 0.0910 +Epoch [88/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 61.7188, Loss1: 0.0794, Loss2: 0.0864 +Epoch [88/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0720, Loss2: 0.0766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 56.7007 % Model2 58.0829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0723, Loss2: 0.0825 +Epoch [89/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0657, Loss2: 0.0634 +Epoch [89/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0743, Loss2: 0.0725 +Epoch [89/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0816, Loss2: 0.0767 +Epoch [89/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0801, Loss2: 0.0767 +Epoch [89/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0861, Loss2: 0.0857 +Epoch [89/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0731, Loss2: 0.0740 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 56.8610 % Model2 57.9127 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0719, Loss2: 0.0737 +Epoch [90/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1002, Loss2: 0.0956 +Epoch [90/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 55.4688, Loss1: 0.0689, Loss2: 0.0816 +Epoch [90/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0989, Loss2: 0.1125 +Epoch [90/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0661, Loss2: 0.0622 +Epoch [90/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0637, Loss2: 0.0658 +Epoch [90/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0835, Loss2: 0.0873 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 57.0913 % Model2 58.1230 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0900, Loss2: 0.0894 +Epoch [91/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 68.7500, Loss1: 0.0661, Loss2: 0.0577 +Epoch [91/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0866, Loss2: 0.0795 +Epoch [91/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0796, Loss2: 0.0713 +Epoch [91/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0696, Loss2: 0.0722 +Epoch [91/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0723, Loss2: 0.0726 +Epoch [91/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0755, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 56.5805 % Model2 57.0012 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0795, Loss2: 0.0786 +Epoch [92/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0793, Loss2: 0.0776 +Epoch [92/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0894, Loss2: 0.0871 +Epoch [92/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0866, Loss2: 0.0832 +Epoch [92/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0732, Loss2: 0.0741 +Epoch [92/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0732, Loss2: 0.0735 +Epoch [92/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0595, Loss2: 0.0613 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 57.2616 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0758, Loss2: 0.0666 +Epoch [93/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0987, Loss2: 0.0991 +Epoch [93/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0684, Loss2: 0.0748 +Epoch [93/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0619, Loss2: 0.0668 +Epoch [93/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1211, Loss2: 0.1132 +Epoch [93/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0633, Loss2: 0.0604 +Epoch [93/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0839, Loss2: 0.0808 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 54.8778 % Model2 57.9828 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.8438, Loss1: 0.0619, Loss2: 0.0589 +Epoch [94/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.1125, Loss2: 0.0996 +Epoch [94/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1054, Loss2: 0.1034 +Epoch [94/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0766, Loss2: 0.0774 +Epoch [94/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0913, Loss2: 0.0839 +Epoch [94/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.9688, Loss1: 0.0733, Loss2: 0.0631 +Epoch [94/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0661, Loss2: 0.0660 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 57.0913 % Model2 58.9443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0841, Loss2: 0.0893 +Epoch [95/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0552, Loss2: 0.0548 +Epoch [95/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 60.1562, Loss1: 0.0809, Loss2: 0.0951 +Epoch [95/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0749, Loss2: 0.0690 +Epoch [95/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0760, Loss2: 0.0747 +Epoch [95/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0859, Loss2: 0.0857 +Epoch [95/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0761, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 56.8510 % Model2 56.9611 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0670, Loss2: 0.0701 +Epoch [96/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0893, Loss2: 0.0985 +Epoch [96/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0776, Loss2: 0.0685 +Epoch [96/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0831, Loss2: 0.0822 +Epoch [96/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0676, Loss2: 0.0729 +Epoch [96/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0631, Loss2: 0.0629 +Epoch [96/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0964, Loss2: 0.0980 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 56.5805 % Model2 58.1030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.1562, Loss1: 0.0779, Loss2: 0.0882 +Epoch [97/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0779, Loss2: 0.0739 +Epoch [97/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.0856, Loss2: 0.0779 +Epoch [97/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 63.2812, Loss1: 0.0709, Loss2: 0.0633 +Epoch [97/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0843, Loss2: 0.0798 +Epoch [97/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0781, Loss2: 0.0793 +Epoch [97/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0957, Loss2: 0.1023 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 56.7408 % Model2 57.9928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0844, Loss2: 0.0818 +Epoch [98/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0763, Loss2: 0.0792 +Epoch [98/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.0756, Loss2: 0.0718 +Epoch [98/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0654, Loss2: 0.0698 +Epoch [98/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0916, Loss2: 0.0881 +Epoch [98/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0878, Loss2: 0.0866 +Epoch [98/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0548, Loss2: 0.0503 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 57.2416 % Model2 58.0128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0696, Loss2: 0.0681 +Epoch [99/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0822, Loss2: 0.0738 +Epoch [99/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0797, Loss2: 0.0776 +Epoch [99/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0722, Loss2: 0.0738 +Epoch [99/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0855, Loss2: 0.0841 +Epoch [99/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0954, Loss2: 0.0979 +Epoch [99/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1056, Loss2: 0.1086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 57.5120 % Model2 58.6839 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0716, Loss2: 0.0708 +Epoch [100/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0878, Loss2: 0.0860 +Epoch [100/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1008, Loss2: 0.1008 +Epoch [100/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0844, Loss2: 0.0953 +Epoch [100/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0696, Loss2: 0.0656 +Epoch [100/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0691, Loss2: 0.0657 +Epoch [100/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0994, Loss2: 0.0921 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 56.4002 % Model2 58.7240 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.1034, Loss2: 0.0949 +Epoch [101/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0708, Loss2: 0.0695 +Epoch [101/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0564, Loss2: 0.0535 +Epoch [101/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0816, Loss2: 0.0861 +Epoch [101/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0861, Loss2: 0.0843 +Epoch [101/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0761, Loss2: 0.0731 +Epoch [101/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0747, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 56.7808 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0845, Loss2: 0.0807 +Epoch [102/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0763, Loss2: 0.0759 +Epoch [102/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0852, Loss2: 0.0850 +Epoch [102/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1190, Loss2: 0.1150 +Epoch [102/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0773, Loss2: 0.0719 +Epoch [102/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0648, Loss2: 0.0687 +Epoch [102/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0959, Loss2: 0.0983 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 56.3301 % Model2 58.3133 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.1562, Loss1: 0.0672, Loss2: 0.0767 +Epoch [103/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0860, Loss2: 0.0885 +Epoch [103/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.0627, Loss2: 0.0598 +Epoch [103/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.0904, Loss2: 0.0854 +Epoch [103/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0860, Loss2: 0.0800 +Epoch [103/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0711, Loss2: 0.0701 +Epoch [103/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0772, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 56.7808 % Model2 57.4820 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0838, Loss2: 0.0862 +Epoch [104/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.0695, Loss2: 0.0753 +Epoch [104/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0815, Loss2: 0.0767 +Epoch [104/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 65.6250, Loss1: 0.0654, Loss2: 0.0706 +Epoch [104/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0813, Loss2: 0.0822 +Epoch [104/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0773, Loss2: 0.0745 +Epoch [104/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0848, Loss2: 0.0876 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 57.1715 % Model2 58.4936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0867, Loss2: 0.0906 +Epoch [105/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0808, Loss2: 0.0871 +Epoch [105/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0937, Loss2: 0.0958 +Epoch [105/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0885, Loss2: 0.0876 +Epoch [105/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0786, Loss2: 0.0730 +Epoch [105/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0556, Loss2: 0.0597 +Epoch [105/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0833, Loss2: 0.0867 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 56.4303 % Model2 58.1230 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 62.5000, Loss1: 0.0563, Loss2: 0.0648 +Epoch [106/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0786, Loss2: 0.0747 +Epoch [106/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0773, Loss2: 0.0806 +Epoch [106/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0941, Loss2: 0.0988 +Epoch [106/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0730, Loss2: 0.0709 +Epoch [106/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0799, Loss2: 0.0813 +Epoch [106/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0826, Loss2: 0.0859 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 56.7308 % Model2 58.6438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0916, Loss2: 0.0904 +Epoch [107/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0881, Loss2: 0.0916 +Epoch [107/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1032, Loss2: 0.1075 +Epoch [107/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0654, Loss2: 0.0664 +Epoch [107/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0944, Loss2: 0.0986 +Epoch [107/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0665, Loss2: 0.0627 +Epoch [107/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0732, Loss2: 0.0692 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 57.0913 % Model2 58.1831 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0963, Loss2: 0.0887 +Epoch [108/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0881, Loss2: 0.0912 +Epoch [108/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0950, Loss2: 0.0935 +Epoch [108/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0771, Loss2: 0.0829 +Epoch [108/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0887, Loss2: 0.0854 +Epoch [108/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0907, Loss2: 0.0863 +Epoch [108/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0735, Loss2: 0.0745 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 57.3217 % Model2 58.2131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0744, Loss2: 0.0769 +Epoch [109/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.0000, Loss1: 0.1147, Loss2: 0.1035 +Epoch [109/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0783, Loss2: 0.0833 +Epoch [109/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0881, Loss2: 0.0801 +Epoch [109/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0628, Loss2: 0.0643 +Epoch [109/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0773, Loss2: 0.0723 +Epoch [109/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0933, Loss2: 0.0903 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 56.8109 % Model2 58.5236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0906, Loss2: 0.0922 +Epoch [110/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0907, Loss2: 0.0918 +Epoch [110/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0871, Loss2: 0.0869 +Epoch [110/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0868, Loss2: 0.0886 +Epoch [110/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.0776, Loss2: 0.0759 +Epoch [110/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1027, Loss2: 0.1024 +Epoch [110/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0916, Loss2: 0.0995 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 56.3702 % Model2 57.7524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0916, Loss2: 0.0980 +Epoch [111/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1246, Loss2: 0.1229 +Epoch [111/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0858, Loss2: 0.0879 +Epoch [111/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0656, Loss2: 0.0622 +Epoch [111/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0915, Loss2: 0.0832 +Epoch [111/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.0779, Loss2: 0.0834 +Epoch [111/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1023, Loss2: 0.1108 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 56.7708 % Model2 58.1530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0902, Loss2: 0.0899 +Epoch [112/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1125, Loss2: 0.1025 +Epoch [112/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 71.0938, Loss1: 0.0763, Loss2: 0.0666 +Epoch [112/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0957, Loss2: 0.0909 +Epoch [112/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0835, Loss2: 0.0765 +Epoch [112/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0736, Loss2: 0.0649 +Epoch [112/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1411, Loss2: 0.1551 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 57.1014 % Model2 57.3918 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0858, Loss2: 0.0860 +Epoch [113/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.0843, Loss2: 0.0782 +Epoch [113/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0758, Loss2: 0.0743 +Epoch [113/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0897, Loss2: 0.0836 +Epoch [113/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0874, Loss2: 0.0885 +Epoch [113/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0909, Loss2: 0.0945 +Epoch [113/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0888, Loss2: 0.0863 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 57.2015 % Model2 58.2532 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.0830, Loss2: 0.0951 +Epoch [114/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0775, Loss2: 0.0853 +Epoch [114/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 71.0938, Loss1: 0.0849, Loss2: 0.0773 +Epoch [114/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0728, Loss2: 0.0768 +Epoch [114/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0916, Loss2: 0.0914 +Epoch [114/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0824, Loss2: 0.0931 +Epoch [114/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0817, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 56.5705 % Model2 58.6839 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1074, Loss2: 0.1024 +Epoch [115/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0660, Loss2: 0.0654 +Epoch [115/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.0961, Loss2: 0.0867 +Epoch [115/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0874, Loss2: 0.0942 +Epoch [115/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 76.5625, Loss1: 0.1171, Loss2: 0.1031 +Epoch [115/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0783, Loss2: 0.0799 +Epoch [115/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0807, Loss2: 0.0779 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 56.3001 % Model2 58.1530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 69.5312, Loss1: 0.0987, Loss2: 0.1101 +Epoch [116/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0956, Loss2: 0.1011 +Epoch [116/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0905, Loss2: 0.0828 +Epoch [116/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0681, Loss2: 0.0711 +Epoch [116/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0908, Loss2: 0.0898 +Epoch [116/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0834, Loss2: 0.0875 +Epoch [116/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0895, Loss2: 0.0881 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 57.2616 % Model2 58.2732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0839, Loss2: 0.0855 +Epoch [117/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.9688, Loss1: 0.0724, Loss2: 0.0783 +Epoch [117/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1273, Loss2: 0.1322 +Epoch [117/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0939, Loss2: 0.0872 +Epoch [117/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.0893, Loss2: 0.0929 +Epoch [117/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0799, Loss2: 0.0807 +Epoch [117/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.1053, Loss2: 0.1092 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 56.8910 % Model2 57.9928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0893, Loss2: 0.0878 +Epoch [118/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1182, Loss2: 0.1134 +Epoch [118/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 72.6562, Loss1: 0.0884, Loss2: 0.0947 +Epoch [118/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1566, Loss2: 0.1345 +Epoch [118/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 62.5000, Loss1: 0.0896, Loss2: 0.0977 +Epoch [118/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0958, Loss2: 0.0958 +Epoch [118/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0888, Loss2: 0.0853 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 56.7208 % Model2 58.4836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0929, Loss2: 0.0819 +Epoch [119/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1037, Loss2: 0.1008 +Epoch [119/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1090, Loss2: 0.1007 +Epoch [119/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1113, Loss2: 0.1000 +Epoch [119/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 60.9375, Loss1: 0.0516, Loss2: 0.0635 +Epoch [119/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0666, Loss2: 0.0638 +Epoch [119/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0730, Loss2: 0.0707 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 56.3301 % Model2 58.0128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1044, Loss2: 0.1035 +Epoch [120/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.2555, Loss2: 0.2356 +Epoch [120/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0729, Loss2: 0.0732 +Epoch [120/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1027, Loss2: 0.0996 +Epoch [120/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0943, Loss2: 0.0973 +Epoch [120/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0683, Loss2: 0.0641 +Epoch [120/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0741, Loss2: 0.0720 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 56.1298 % Model2 58.6739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1326, Loss2: 0.1193 +Epoch [121/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0642, Loss2: 0.0645 +Epoch [121/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 58.5938, Loss1: 0.0732, Loss2: 0.0916 +Epoch [121/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0913, Loss2: 0.0918 +Epoch [121/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1089, Loss2: 0.0981 +Epoch [121/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0988, Loss2: 0.0938 +Epoch [121/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0964, Loss2: 0.0894 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 56.6206 % Model2 58.2732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 66.4062, Loss1: 0.0802, Loss2: 0.0915 +Epoch [122/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0905, Loss2: 0.0858 +Epoch [122/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0943, Loss2: 0.0980 +Epoch [122/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.1042, Loss2: 0.0980 +Epoch [122/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0793, Loss2: 0.0783 +Epoch [122/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1245, Loss2: 0.1067 +Epoch [122/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0880, Loss2: 0.0897 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 56.5204 % Model2 58.0128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0861, Loss2: 0.0914 +Epoch [123/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0858, Loss2: 0.0839 +Epoch [123/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1117, Loss2: 0.1084 +Epoch [123/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.1041, Loss2: 0.1096 +Epoch [123/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 74.2188, Loss1: 0.0715, Loss2: 0.0614 +Epoch [123/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0903, Loss2: 0.0903 +Epoch [123/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0882, Loss2: 0.0901 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 55.9395 % Model2 57.9627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1039, Loss2: 0.1117 +Epoch [124/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1185, Loss2: 0.1227 +Epoch [124/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1328, Loss2: 0.1347 +Epoch [124/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1052, Loss2: 0.0994 +Epoch [124/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1010, Loss2: 0.1044 +Epoch [124/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.1007, Loss2: 0.1149 +Epoch [124/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0822, Loss2: 0.0802 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 56.2400 % Model2 57.6723 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0801, Loss2: 0.0815 +Epoch [125/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0879, Loss2: 0.0933 +Epoch [125/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0927, Loss2: 0.0915 +Epoch [125/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0998, Loss2: 0.1079 +Epoch [125/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1218, Loss2: 0.1226 +Epoch [125/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0849, Loss2: 0.0788 +Epoch [125/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0875, Loss2: 0.0878 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 55.6891 % Model2 57.7925 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0787, Loss2: 0.0785 +Epoch [126/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 67.9688, Loss1: 0.0646, Loss2: 0.0763 +Epoch [126/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0936, Loss2: 0.0918 +Epoch [126/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1014, Loss2: 0.1078 +Epoch [126/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0845, Loss2: 0.0844 +Epoch [126/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.0815, Loss2: 0.0874 +Epoch [126/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1323, Loss2: 0.1387 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 56.3001 % Model2 58.4335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0705, Loss2: 0.0753 +Epoch [127/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1192, Loss2: 0.1102 +Epoch [127/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 66.4062, Loss1: 0.0991, Loss2: 0.1128 +Epoch [127/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0905, Loss2: 0.0963 +Epoch [127/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1032, Loss2: 0.1018 +Epoch [127/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1312, Loss2: 0.1306 +Epoch [127/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0747, Loss2: 0.0788 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 57.9828 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0945, Loss2: 0.1031 +Epoch [128/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0732, Loss2: 0.0783 +Epoch [128/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0867, Loss2: 0.0897 +Epoch [128/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0947, Loss2: 0.0899 +Epoch [128/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0785, Loss2: 0.0836 +Epoch [128/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0943, Loss2: 0.0953 +Epoch [128/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0794, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 56.0897 % Model2 58.1430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.1048, Loss2: 0.1016 +Epoch [129/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0812, Loss2: 0.0852 +Epoch [129/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0864, Loss2: 0.0846 +Epoch [129/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1355, Loss2: 0.1251 +Epoch [129/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1474, Loss2: 0.1591 +Epoch [129/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1527, Loss2: 0.1606 +Epoch [129/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0842, Loss2: 0.0852 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 56.6306 % Model2 58.3834 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1069, Loss2: 0.1004 +Epoch [130/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0986, Loss2: 0.0945 +Epoch [130/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.0895, Loss2: 0.0792 +Epoch [130/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1074, Loss2: 0.1081 +Epoch [130/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0753, Loss2: 0.0710 +Epoch [130/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.1035, Loss2: 0.1165 +Epoch [130/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1318, Loss2: 0.1228 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 56.6506 % Model2 57.7825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0892, Loss2: 0.0848 +Epoch [131/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 71.8750, Loss1: 0.0959, Loss2: 0.0791 +Epoch [131/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0793, Loss2: 0.0751 +Epoch [131/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0993, Loss2: 0.0914 +Epoch [131/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.1533, Loss2: 0.1875 +Epoch [131/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1283, Loss2: 0.1127 +Epoch [131/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.8750, Loss1: 0.1190, Loss2: 0.1312 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 55.5489 % Model2 58.4936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1110, Loss2: 0.1086 +Epoch [132/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1057, Loss2: 0.1038 +Epoch [132/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.1034, Loss2: 0.0961 +Epoch [132/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0874, Loss2: 0.0835 +Epoch [132/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1371, Loss2: 0.1416 +Epoch [132/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1388, Loss2: 0.1291 +Epoch [132/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1019, Loss2: 0.1074 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 57.0813 % Model2 57.9627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0786, Loss2: 0.0880 +Epoch [133/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1286, Loss2: 0.1316 +Epoch [133/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1139, Loss2: 0.1159 +Epoch [133/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1298, Loss2: 0.1371 +Epoch [133/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1010, Loss2: 0.1032 +Epoch [133/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0977, Loss2: 0.0898 +Epoch [133/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1026, Loss2: 0.0991 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 56.6406 % Model2 58.3133 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1132, Loss2: 0.1053 +Epoch [134/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0969, Loss2: 0.0992 +Epoch [134/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0872, Loss2: 0.0905 +Epoch [134/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1210, Loss2: 0.1231 +Epoch [134/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0840, Loss2: 0.0863 +Epoch [134/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 72.6562, Loss1: 0.0912, Loss2: 0.0756 +Epoch [134/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0938, Loss2: 0.0881 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 56.8910 % Model2 58.4135 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.9688, Loss1: 0.0923, Loss2: 0.0997 +Epoch [135/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1135, Loss2: 0.1095 +Epoch [135/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1037, Loss2: 0.0986 +Epoch [135/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1316, Loss2: 0.1353 +Epoch [135/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0790, Loss2: 0.0783 +Epoch [135/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0947, Loss2: 0.0999 +Epoch [135/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1011, Loss2: 0.1020 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 56.6206 % Model2 57.8025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1008, Loss2: 0.0931 +Epoch [136/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0956, Loss2: 0.1042 +Epoch [136/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0993, Loss2: 0.0895 +Epoch [136/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 65.6250, Loss1: 0.1208, Loss2: 0.1343 +Epoch [136/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.1679, Loss2: 0.1392 +Epoch [136/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0776, Loss2: 0.0756 +Epoch [136/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0913, Loss2: 0.0913 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 56.6206 % Model2 58.2732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1078, Loss2: 0.1034 +Epoch [137/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1337, Loss2: 0.1304 +Epoch [137/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1001, Loss2: 0.0982 +Epoch [137/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1113, Loss2: 0.1176 +Epoch [137/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1297, Loss2: 0.1400 +Epoch [137/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1212, Loss2: 0.1085 +Epoch [137/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1300, Loss2: 0.1391 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 56.7708 % Model2 58.1931 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0761, Loss2: 0.0777 +Epoch [138/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0893, Loss2: 0.0855 +Epoch [138/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1638, Loss2: 0.1740 +Epoch [138/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0967, Loss2: 0.0992 +Epoch [138/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 66.4062, Loss1: 0.0820, Loss2: 0.0894 +Epoch [138/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0984, Loss2: 0.1018 +Epoch [138/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1612, Loss2: 0.1411 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 56.4704 % Model2 57.9327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 61.7188, Loss1: 0.0736, Loss2: 0.0778 +Epoch [139/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0957, Loss2: 0.0989 +Epoch [139/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0911, Loss2: 0.0927 +Epoch [139/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0875, Loss2: 0.0866 +Epoch [139/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1298, Loss2: 0.1274 +Epoch [139/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0782, Loss2: 0.0808 +Epoch [139/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1023, Loss2: 0.1010 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 56.4002 % Model2 58.1130 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1056, Loss2: 0.1043 +Epoch [140/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1111, Loss2: 0.1114 +Epoch [140/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1041, Loss2: 0.1096 +Epoch [140/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.0938, Loss1: 0.0902, Loss2: 0.1032 +Epoch [140/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0895, Loss2: 0.0953 +Epoch [140/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 72.6562, Loss1: 0.1001, Loss2: 0.1072 +Epoch [140/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0595, Loss2: 0.0608 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 56.6406 % Model2 57.9026 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1313, Loss2: 0.1285 +Epoch [141/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.1268, Loss2: 0.1283 +Epoch [141/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.0975, Loss2: 0.1057 +Epoch [141/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1080, Loss2: 0.1072 +Epoch [141/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1046, Loss2: 0.1016 +Epoch [141/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1036, Loss2: 0.1053 +Epoch [141/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.1622, Loss2: 0.1358 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 56.1599 % Model2 58.4235 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1061, Loss2: 0.1048 +Epoch [142/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1110, Loss2: 0.1017 +Epoch [142/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1243, Loss2: 0.1207 +Epoch [142/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1114, Loss2: 0.1109 +Epoch [142/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1093, Loss2: 0.1053 +Epoch [142/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0961, Loss2: 0.0903 +Epoch [142/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0905, Loss2: 0.0894 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 56.5204 % Model2 57.8325 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0758, Loss2: 0.0708 +Epoch [143/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.1112, Loss2: 0.1000 +Epoch [143/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0914, Loss2: 0.0905 +Epoch [143/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1006, Loss2: 0.1087 +Epoch [143/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1561, Loss2: 0.1645 +Epoch [143/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0837, Loss2: 0.0795 +Epoch [143/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1011, Loss2: 0.1016 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 56.7007 % Model2 57.9627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1368, Loss2: 0.1320 +Epoch [144/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.1153, Loss2: 0.1253 +Epoch [144/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1203, Loss2: 0.1212 +Epoch [144/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1081, Loss2: 0.1045 +Epoch [144/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1183, Loss2: 0.1249 +Epoch [144/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1327, Loss2: 0.1376 +Epoch [144/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1088, Loss2: 0.1042 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 55.9295 % Model2 58.0429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1422, Loss2: 0.1460 +Epoch [145/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0789, Loss2: 0.0815 +Epoch [145/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1480, Loss2: 0.1430 +Epoch [145/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0731, Loss2: 0.0732 +Epoch [145/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.1078, Loss2: 0.1163 +Epoch [145/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.0961, Loss2: 0.0897 +Epoch [145/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1026, Loss2: 0.0935 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 56.2300 % Model2 57.8325 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1049, Loss2: 0.1069 +Epoch [146/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1760, Loss2: 0.1822 +Epoch [146/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1140, Loss2: 0.1145 +Epoch [146/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0743, Loss2: 0.0797 +Epoch [146/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1334, Loss2: 0.1238 +Epoch [146/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0752, Loss2: 0.0749 +Epoch [146/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0875, Loss2: 0.0854 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 56.4704 % Model2 58.1430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1605, Loss2: 0.1651 +Epoch [147/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1149, Loss2: 0.1083 +Epoch [147/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1062, Loss2: 0.1022 +Epoch [147/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 80.4688, Loss1: 0.1239, Loss2: 0.1019 +Epoch [147/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0858, Loss2: 0.0896 +Epoch [147/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1059, Loss2: 0.1179 +Epoch [147/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1263, Loss2: 0.1253 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 56.8810 % Model2 57.7224 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0924, Loss2: 0.0840 +Epoch [148/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1085, Loss2: 0.1094 +Epoch [148/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1047, Loss2: 0.1136 +Epoch [148/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.1181, Loss2: 0.1033 +Epoch [148/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1130, Loss2: 0.1152 +Epoch [148/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0976, Loss2: 0.0952 +Epoch [148/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.1144, Loss2: 0.0997 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 56.5505 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0927, Loss2: 0.0925 +Epoch [149/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1072, Loss2: 0.0976 +Epoch [149/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1304, Loss2: 0.1295 +Epoch [149/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1187, Loss2: 0.1054 +Epoch [149/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0820, Loss2: 0.0822 +Epoch [149/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0929, Loss2: 0.0931 +Epoch [149/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0912, Loss2: 0.0886 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 56.3201 % Model2 58.2131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0939, Loss2: 0.0941 +Epoch [150/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1210, Loss2: 0.1093 +Epoch [150/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1223, Loss2: 0.1209 +Epoch [150/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1108, Loss2: 0.1061 +Epoch [150/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1054, Loss2: 0.0941 +Epoch [150/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0858, Loss2: 0.0916 +Epoch [150/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0928, Loss2: 0.0957 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 56.0597 % Model2 57.8726 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.0000, Loss1: 0.0940, Loss2: 0.1039 +Epoch [151/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 62.5000, Loss1: 0.0962, Loss2: 0.1100 +Epoch [151/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1056, Loss2: 0.1071 +Epoch [151/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.1348, Loss2: 0.1413 +Epoch [151/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1085, Loss2: 0.1147 +Epoch [151/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 72.6562, Loss1: 0.0809, Loss2: 0.0724 +Epoch [151/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1489, Loss2: 0.1458 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 56.2700 % Model2 57.8526 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1135, Loss2: 0.1111 +Epoch [152/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1113, Loss2: 0.0996 +Epoch [152/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1359, Loss2: 0.1247 +Epoch [152/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.1200, Loss2: 0.1256 +Epoch [152/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0998, Loss2: 0.1044 +Epoch [152/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0972, Loss2: 0.0889 +Epoch [152/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1487, Loss2: 0.1549 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 56.0897 % Model2 57.4720 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1937, Loss2: 0.1798 +Epoch [153/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1289, Loss2: 0.1204 +Epoch [153/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1122, Loss2: 0.1163 +Epoch [153/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1417, Loss2: 0.1316 +Epoch [153/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1335, Loss2: 0.1283 +Epoch [153/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0768, Loss2: 0.0777 +Epoch [153/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.1266, Loss2: 0.1122 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 54.9679 % Model2 57.6823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.1240, Loss2: 0.1114 +Epoch [154/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0963, Loss2: 0.0947 +Epoch [154/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1847, Loss2: 0.1771 +Epoch [154/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.1012, Loss2: 0.1139 +Epoch [154/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0901, Loss2: 0.0854 +Epoch [154/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1059, Loss2: 0.1139 +Epoch [154/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0975, Loss2: 0.0979 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 56.1599 % Model2 57.5921 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 77.3438, Loss1: 0.1831, Loss2: 0.1971 +Epoch [155/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1722, Loss2: 0.1566 +Epoch [155/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1163, Loss2: 0.1076 +Epoch [155/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.1196, Loss2: 0.1251 +Epoch [155/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1203, Loss2: 0.1170 +Epoch [155/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1200, Loss2: 0.1222 +Epoch [155/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1236, Loss2: 0.1256 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 56.5204 % Model2 58.0729 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0814, Loss2: 0.0790 +Epoch [156/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1221, Loss2: 0.1348 +Epoch [156/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.2025, Loss2: 0.1908 +Epoch [156/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0684, Loss2: 0.0675 +Epoch [156/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0948, Loss2: 0.0975 +Epoch [156/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0830, Loss2: 0.0792 +Epoch [156/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1094, Loss2: 0.1153 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 55.7192 % Model2 58.3333 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0968, Loss2: 0.0933 +Epoch [157/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1060, Loss2: 0.1113 +Epoch [157/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1183, Loss2: 0.1244 +Epoch [157/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1034, Loss2: 0.1022 +Epoch [157/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1399, Loss2: 0.1289 +Epoch [157/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1393, Loss2: 0.1255 +Epoch [157/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1205, Loss2: 0.1293 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 56.2300 % Model2 58.3233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1750, Loss2: 0.1624 +Epoch [158/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0918, Loss2: 0.0894 +Epoch [158/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1699, Loss2: 0.1750 +Epoch [158/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0923, Loss2: 0.0929 +Epoch [158/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1330, Loss2: 0.1321 +Epoch [158/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1547, Loss2: 0.1559 +Epoch [158/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1358, Loss2: 0.1417 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 56.1098 % Model2 57.5721 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1279, Loss2: 0.1238 +Epoch [159/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1104, Loss2: 0.1079 +Epoch [159/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1050, Loss2: 0.1002 +Epoch [159/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1408, Loss2: 0.1298 +Epoch [159/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1766, Loss2: 0.1804 +Epoch [159/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1222, Loss2: 0.1208 +Epoch [159/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1262, Loss2: 0.1288 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 56.0797 % Model2 57.7524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1316, Loss2: 0.1435 +Epoch [160/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.0956, Loss2: 0.1044 +Epoch [160/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1490, Loss2: 0.1417 +Epoch [160/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1286, Loss2: 0.1269 +Epoch [160/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1086, Loss2: 0.1118 +Epoch [160/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.0942, Loss2: 0.0879 +Epoch [160/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1420, Loss2: 0.1369 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 56.1699 % Model2 58.0629 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1360, Loss2: 0.1341 +Epoch [161/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1422, Loss2: 0.1413 +Epoch [161/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1302, Loss2: 0.1190 +Epoch [161/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1367, Loss2: 0.1349 +Epoch [161/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1410, Loss2: 0.1538 +Epoch [161/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1145, Loss2: 0.1111 +Epoch [161/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1342, Loss2: 0.1400 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 56.5304 % Model2 57.8225 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 72.6562, Loss1: 0.1472, Loss2: 0.1262 +Epoch [162/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1126, Loss2: 0.1032 +Epoch [162/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1415, Loss2: 0.1365 +Epoch [162/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1139, Loss2: 0.1235 +Epoch [162/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1052, Loss2: 0.1173 +Epoch [162/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.1116, Loss2: 0.0982 +Epoch [162/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0941, Loss2: 0.0979 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 56.0897 % Model2 57.8726 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0846, Loss2: 0.0882 +Epoch [163/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.1161, Loss2: 0.1186 +Epoch [163/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1376, Loss2: 0.1526 +Epoch [163/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1476, Loss2: 0.1486 +Epoch [163/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1268, Loss2: 0.1282 +Epoch [163/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1454, Loss2: 0.1260 +Epoch [163/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0882, Loss2: 0.0883 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 55.9395 % Model2 57.5721 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1111, Loss2: 0.1138 +Epoch [164/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1507, Loss2: 0.1344 +Epoch [164/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.9062, Loss1: 0.1594, Loss2: 0.1549 +Epoch [164/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1300, Loss2: 0.1350 +Epoch [164/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1376, Loss2: 0.1491 +Epoch [164/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1366, Loss2: 0.1420 +Epoch [164/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1545, Loss2: 0.1332 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 55.5789 % Model2 57.8225 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1172, Loss2: 0.1078 +Epoch [165/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1322, Loss2: 0.1370 +Epoch [165/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1523, Loss2: 0.1493 +Epoch [165/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1835, Loss2: 0.1662 +Epoch [165/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.2247, Loss2: 0.2354 +Epoch [165/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1081, Loss2: 0.1063 +Epoch [165/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1541, Loss2: 0.1411 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 55.8994 % Model2 57.6222 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.1161, Loss2: 0.1041 +Epoch [166/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1388, Loss2: 0.1480 +Epoch [166/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.1754, Loss2: 0.1553 +Epoch [166/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0943, Loss2: 0.1051 +Epoch [166/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1503, Loss2: 0.1503 +Epoch [166/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1113, Loss2: 0.1063 +Epoch [166/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1367, Loss2: 0.1400 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 56.1498 % Model2 57.4619 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0868, Loss2: 0.0868 +Epoch [167/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1330, Loss2: 0.1214 +Epoch [167/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1253, Loss2: 0.1230 +Epoch [167/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1816, Loss2: 0.1901 +Epoch [167/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1079, Loss2: 0.1048 +Epoch [167/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0935, Loss2: 0.0918 +Epoch [167/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1059, Loss2: 0.0960 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 56.2200 % Model2 57.8025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 82.0312, Training Accuracy2: 78.1250, Loss1: 0.1437, Loss2: 0.1696 +Epoch [168/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1282, Loss2: 0.1434 +Epoch [168/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1577, Loss2: 0.1650 +Epoch [168/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1140, Loss2: 0.1159 +Epoch [168/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1430, Loss2: 0.1476 +Epoch [168/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0891, Loss2: 0.0916 +Epoch [168/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 69.5312, Loss1: 0.1029, Loss2: 0.1169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 56.0096 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1446, Loss2: 0.1500 +Epoch [169/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1071, Loss2: 0.1126 +Epoch [169/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1147, Loss2: 0.1212 +Epoch [169/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1179, Loss2: 0.1126 +Epoch [169/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1470, Loss2: 0.1295 +Epoch [169/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1317, Loss2: 0.1492 +Epoch [169/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1329, Loss2: 0.1196 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 57.5521 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1065, Loss2: 0.1075 +Epoch [170/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.9062, Loss1: 0.1039, Loss2: 0.1125 +Epoch [170/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1177, Loss2: 0.1181 +Epoch [170/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1495, Loss2: 0.1384 +Epoch [170/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1327, Loss2: 0.1396 +Epoch [170/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 81.2500, Loss1: 0.1243, Loss2: 0.1130 +Epoch [170/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1021, Loss2: 0.1045 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 56.2200 % Model2 57.4619 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1194, Loss2: 0.1160 +Epoch [171/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1355, Loss2: 0.1340 +Epoch [171/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1453, Loss2: 0.1453 +Epoch [171/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1838, Loss2: 0.1789 +Epoch [171/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1087, Loss2: 0.1057 +Epoch [171/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1521, Loss2: 0.1652 +Epoch [171/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1192, Loss2: 0.1219 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 56.0597 % Model2 57.1314 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.1326, Loss2: 0.1402 +Epoch [172/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.1860, Loss2: 0.1768 +Epoch [172/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1407, Loss2: 0.1294 +Epoch [172/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1311, Loss2: 0.1272 +Epoch [172/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1178, Loss2: 0.1085 +Epoch [172/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1297, Loss2: 0.1228 +Epoch [172/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1743, Loss2: 0.1882 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1181, Loss2: 0.1091 +Epoch [173/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0978, Loss2: 0.0904 +Epoch [173/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1387, Loss2: 0.1262 +Epoch [173/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.1370, Loss2: 0.1237 +Epoch [173/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.1831, Loss2: 0.1739 +Epoch [173/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1242, Loss2: 0.1159 +Epoch [173/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1810, Loss2: 0.1813 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 55.9696 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 80.4688, Training Accuracy2: 79.6875, Loss1: 0.1737, Loss2: 0.1832 +Epoch [174/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1342, Loss2: 0.1145 +Epoch [174/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1243, Loss2: 0.1210 +Epoch [174/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0994, Loss2: 0.0835 +Epoch [174/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1691, Loss2: 0.1588 +Epoch [174/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1576, Loss2: 0.1600 +Epoch [174/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1477, Loss2: 0.1378 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 55.8794 % Model2 57.6623 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1325, Loss2: 0.1283 +Epoch [175/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1410, Loss2: 0.1400 +Epoch [175/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 78.9062, Loss1: 0.1609, Loss2: 0.1366 +Epoch [175/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1842, Loss2: 0.1841 +Epoch [175/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1373, Loss2: 0.1348 +Epoch [175/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1110, Loss2: 0.1099 +Epoch [175/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1969, Loss2: 0.1766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 55.8293 % Model2 57.6923 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1839, Loss2: 0.1738 +Epoch [176/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1396, Loss2: 0.1229 +Epoch [176/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1311, Loss2: 0.1283 +Epoch [176/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1374, Loss2: 0.1235 +Epoch [176/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1858, Loss2: 0.1759 +Epoch [176/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1633, Loss2: 0.1868 +Epoch [176/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1500, Loss2: 0.1416 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 55.9495 % Model2 57.7123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1124, Loss2: 0.1053 +Epoch [177/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.2007, Loss2: 0.2162 +Epoch [177/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1393, Loss2: 0.1539 +Epoch [177/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 83.5938, Loss1: 0.1525, Loss2: 0.1284 +Epoch [177/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1326, Loss2: 0.1303 +Epoch [177/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 81.2500, Loss1: 0.1990, Loss2: 0.1899 +Epoch [177/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1339, Loss2: 0.1434 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 55.8594 % Model2 57.6022 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1074, Loss2: 0.1010 +Epoch [178/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.1850, Loss2: 0.1736 +Epoch [178/200], Iter [150/390] Training Accuracy1: 82.0312, Training Accuracy2: 83.5938, Loss1: 0.1903, Loss2: 0.1976 +Epoch [178/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.2028, Loss2: 0.1848 +Epoch [178/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1050, Loss2: 0.1125 +Epoch [178/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1149, Loss2: 0.1142 +Epoch [178/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1357, Loss2: 0.1426 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 55.9095 % Model2 57.9127 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1625, Loss2: 0.1501 +Epoch [179/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1154, Loss2: 0.1158 +Epoch [179/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 65.6250, Loss1: 0.1177, Loss2: 0.1261 +Epoch [179/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 78.1250, Loss1: 0.2075, Loss2: 0.2382 +Epoch [179/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1206, Loss2: 0.1128 +Epoch [179/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.1396, Loss2: 0.1571 +Epoch [179/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1273, Loss2: 0.1346 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 57.4219 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1217, Loss2: 0.1123 +Epoch [180/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1452, Loss2: 0.1374 +Epoch [180/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1797, Loss2: 0.1634 +Epoch [180/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1290, Loss2: 0.1218 +Epoch [180/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.1566, Loss2: 0.1420 +Epoch [180/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1485, Loss2: 0.1521 +Epoch [180/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1896, Loss2: 0.2060 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 56.0296 % Model2 57.3518 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1056, Loss2: 0.1195 +Epoch [181/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1698, Loss2: 0.1546 +Epoch [181/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1183, Loss2: 0.1140 +Epoch [181/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1737, Loss2: 0.1566 +Epoch [181/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.2370, Loss2: 0.2538 +Epoch [181/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1367, Loss2: 0.1379 +Epoch [181/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1413, Loss2: 0.1337 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 55.2484 % Model2 57.5421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.8750, Loss1: 0.1371, Loss2: 0.1563 +Epoch [182/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.2012, Loss2: 0.1699 +Epoch [182/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1221, Loss2: 0.1207 +Epoch [182/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1401, Loss2: 0.1376 +Epoch [182/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1060, Loss2: 0.1010 +Epoch [182/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1642, Loss2: 0.1543 +Epoch [182/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1311, Loss2: 0.1273 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 55.7091 % Model2 56.9912 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1829, Loss2: 0.1643 +Epoch [183/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1490, Loss2: 0.1453 +Epoch [183/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1001, Loss2: 0.1005 +Epoch [183/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1372, Loss2: 0.1485 +Epoch [183/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1759, Loss2: 0.1719 +Epoch [183/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1653, Loss2: 0.1502 +Epoch [183/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.1151, Loss2: 0.1230 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 55.4888 % Model2 57.4018 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1445, Loss2: 0.1338 +Epoch [184/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1424, Loss2: 0.1381 +Epoch [184/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 73.4375, Loss1: 0.1225, Loss2: 0.1431 +Epoch [184/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1694, Loss2: 0.1708 +Epoch [184/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1133, Loss2: 0.1127 +Epoch [184/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1485, Loss2: 0.1338 +Epoch [184/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1320, Loss2: 0.1317 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 55.5288 % Model2 57.4820 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.1587, Loss2: 0.1330 +Epoch [185/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1643, Loss2: 0.1707 +Epoch [185/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1334, Loss2: 0.1272 +Epoch [185/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1088, Loss2: 0.1125 +Epoch [185/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1349, Loss2: 0.1374 +Epoch [185/200], Iter [300/390] Training Accuracy1: 79.6875, Training Accuracy2: 83.5938, Loss1: 0.1497, Loss2: 0.1330 +Epoch [185/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1289, Loss2: 0.1375 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 55.5889 % Model2 57.4920 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.1813, Loss2: 0.1628 +Epoch [186/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1495, Loss2: 0.1416 +Epoch [186/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 70.3125, Loss1: 0.1050, Loss2: 0.1248 +Epoch [186/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1177, Loss2: 0.1152 +Epoch [186/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1006, Loss2: 0.0983 +Epoch [186/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1736, Loss2: 0.1775 +Epoch [186/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1267, Loss2: 0.1341 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 55.4387 % Model2 57.4319 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1286, Loss2: 0.1305 +Epoch [187/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1762, Loss2: 0.2035 +Epoch [187/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1374, Loss2: 0.1334 +Epoch [187/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1530, Loss2: 0.1716 +Epoch [187/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.1545, Loss2: 0.1495 +Epoch [187/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1639, Loss2: 0.1817 +Epoch [187/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 81.2500, Loss1: 0.1705, Loss2: 0.1593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 57.4119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1050, Loss2: 0.1098 +Epoch [188/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.1293, Loss2: 0.1451 +Epoch [188/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1171, Loss2: 0.1183 +Epoch [188/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1194, Loss2: 0.1027 +Epoch [188/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1149, Loss2: 0.1119 +Epoch [188/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1172, Loss2: 0.1153 +Epoch [188/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1744, Loss2: 0.1545 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 57.5220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1599, Loss2: 0.1595 +Epoch [189/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1958, Loss2: 0.1833 +Epoch [189/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1317, Loss2: 0.1322 +Epoch [189/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.1452, Loss2: 0.1345 +Epoch [189/200], Iter [250/390] Training Accuracy1: 81.2500, Training Accuracy2: 75.0000, Loss1: 0.1263, Loss2: 0.1488 +Epoch [189/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1298, Loss2: 0.1240 +Epoch [189/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1187, Loss2: 0.1202 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 57.5621 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.1596, Loss2: 0.1548 +Epoch [190/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1271, Loss2: 0.1303 +Epoch [190/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0970, Loss2: 0.0923 +Epoch [190/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1501, Loss2: 0.1441 +Epoch [190/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1414, Loss2: 0.1412 +Epoch [190/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1097, Loss2: 0.1233 +Epoch [190/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1373, Loss2: 0.1384 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1710, Loss2: 0.1562 +Epoch [191/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 81.2500, Loss1: 0.1697, Loss2: 0.1348 +Epoch [191/200], Iter [150/390] Training Accuracy1: 79.6875, Training Accuracy2: 79.6875, Loss1: 0.1495, Loss2: 0.1544 +Epoch [191/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1493, Loss2: 0.1532 +Epoch [191/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1118, Loss2: 0.1053 +Epoch [191/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1139, Loss2: 0.1202 +Epoch [191/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1258, Loss2: 0.1230 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 55.3686 % Model2 57.2316 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.1375, Loss2: 0.1262 +Epoch [192/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1147, Loss2: 0.1154 +Epoch [192/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.2164, Loss2: 0.2296 +Epoch [192/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1336, Loss2: 0.1417 +Epoch [192/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1594, Loss2: 0.1503 +Epoch [192/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1257, Loss2: 0.1224 +Epoch [192/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1095, Loss2: 0.1151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 57.3618 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1073, Loss2: 0.1126 +Epoch [193/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1177, Loss2: 0.1132 +Epoch [193/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1220, Loss2: 0.1371 +Epoch [193/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1096, Loss2: 0.1069 +Epoch [193/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1016, Loss2: 0.1078 +Epoch [193/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.1447, Loss2: 0.1570 +Epoch [193/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1640, Loss2: 0.1521 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 55.5188 % Model2 57.4419 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1337, Loss2: 0.1411 +Epoch [194/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.1250, Loss1: 0.1332, Loss2: 0.1408 +Epoch [194/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1360, Loss2: 0.1453 +Epoch [194/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1650, Loss2: 0.1707 +Epoch [194/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.7812, Loss1: 0.1404, Loss2: 0.1544 +Epoch [194/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1412, Loss2: 0.1343 +Epoch [194/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1980, Loss2: 0.2096 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 55.5689 % Model2 57.3317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 77.3438, Loss1: 0.2754, Loss2: 0.2899 +Epoch [195/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.1271, Loss2: 0.1133 +Epoch [195/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.2075, Loss2: 0.2060 +Epoch [195/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.1490, Loss2: 0.1328 +Epoch [195/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1434, Loss2: 0.1466 +Epoch [195/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1373, Loss2: 0.1306 +Epoch [195/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.9062, Loss1: 0.1906, Loss2: 0.1751 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 55.3786 % Model2 57.1514 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0903, Loss2: 0.0942 +Epoch [196/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1393, Loss2: 0.1416 +Epoch [196/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1416, Loss2: 0.1372 +Epoch [196/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0990, Loss2: 0.1070 +Epoch [196/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0987, Loss2: 0.0910 +Epoch [196/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1899, Loss2: 0.1840 +Epoch [196/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.1704, Loss2: 0.1508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 55.2484 % Model2 57.2716 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1236, Loss2: 0.1263 +Epoch [197/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1270, Loss2: 0.1336 +Epoch [197/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1098, Loss2: 0.1041 +Epoch [197/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1301, Loss2: 0.1193 +Epoch [197/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1415, Loss2: 0.1367 +Epoch [197/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.9062, Loss1: 0.1302, Loss2: 0.1267 +Epoch [197/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1176, Loss2: 0.1113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 55.4587 % Model2 57.3117 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1444, Loss2: 0.1450 +Epoch [198/200], Iter [100/390] Training Accuracy1: 79.6875, Training Accuracy2: 74.2188, Loss1: 0.1480, Loss2: 0.1810 +Epoch [198/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.1302, Loss2: 0.1402 +Epoch [198/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1173, Loss2: 0.1133 +Epoch [198/200], Iter [250/390] Training Accuracy1: 82.8125, Training Accuracy2: 78.9062, Loss1: 0.1339, Loss2: 0.1489 +Epoch [198/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 79.6875, Loss1: 0.2850, Loss2: 0.2709 +Epoch [198/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1975, Loss2: 0.1901 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 57.2516 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1241, Loss2: 0.1263 +Epoch [199/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1416, Loss2: 0.1344 +Epoch [199/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1169, Loss2: 0.1124 +Epoch [199/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1121, Loss2: 0.1079 +Epoch [199/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1323, Loss2: 0.1116 +Epoch [199/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1372, Loss2: 0.1381 +Epoch [199/200], Iter [350/390] Training Accuracy1: 79.6875, Training Accuracy2: 76.5625, Loss1: 0.1675, Loss2: 0.1959 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 55.0982 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1593, Loss2: 0.1738 +Epoch [200/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1641, Loss2: 0.1594 +Epoch [200/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1449, Loss2: 0.1344 +Epoch [200/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1262, Loss2: 0.1329 +Epoch [200/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1370, Loss2: 0.1417 +Epoch [200/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1075, Loss2: 0.1097 +Epoch [200/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.1499, Loss2: 0.1411 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 55.1683 % Model2 57.1014 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_4_4.log b/other_methods/coteaching_plus/coteaching_plus_results/out_4_4.log new file mode 100644 index 0000000..56a336d --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_4_4.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 21.8750, Training Accuracy2: 16.4062, Loss1: 0.0177, Loss2: 0.0176 +Epoch [2/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 24.2188, Loss1: 0.0164, Loss2: 0.0167 +Epoch [2/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 22.6562, Loss1: 0.0157, Loss2: 0.0159 +Epoch [2/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0158, Loss2: 0.0160 +Epoch [2/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 23.4375, Loss1: 0.0147, Loss2: 0.0153 +Epoch [2/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0144, Loss2: 0.0147 +Epoch [2/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 23.4375, Loss1: 0.0156, Loss2: 0.0163 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 29.1967 % Model2 28.0649 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0144, Loss2: 0.0149 +Epoch [3/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0157 +Epoch [3/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 24.2188, Loss1: 0.0155, Loss2: 0.0153 +Epoch [3/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.9062, Loss1: 0.0161, Loss2: 0.0166 +Epoch [3/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 31.2500, Loss1: 0.0142, Loss2: 0.0139 +Epoch [3/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0147, Loss2: 0.0143 +Epoch [3/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 30.4688, Loss1: 0.0136, Loss2: 0.0142 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 36.6887 % Model2 35.3766 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0153, Loss2: 0.0152 +Epoch [4/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0131, Loss2: 0.0137 +Epoch [4/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0158, Loss2: 0.0160 +Epoch [4/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0147, Loss2: 0.0148 +Epoch [4/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.9062, Loss1: 0.0144, Loss2: 0.0141 +Epoch [4/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0129, Loss2: 0.0134 +Epoch [4/200], Iter [350/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0150, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 39.0224 % Model2 36.7388 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 28.9062, Loss1: 0.0138, Loss2: 0.0142 +Epoch [5/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 28.1250, Loss1: 0.0142, Loss2: 0.0148 +Epoch [5/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0125, Loss2: 0.0126 +Epoch [5/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0148, Loss2: 0.0147 +Epoch [5/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0139 +Epoch [5/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0141, Loss2: 0.0138 +Epoch [5/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0148, Loss2: 0.0148 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 40.0040 % Model2 42.2376 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0131, Loss2: 0.0128 +Epoch [6/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0123, Loss2: 0.0119 +Epoch [6/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0134, Loss2: 0.0130 +Epoch [6/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.8125, Loss1: 0.0142, Loss2: 0.0141 +Epoch [6/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0137, Loss2: 0.0138 +Epoch [6/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0139, Loss2: 0.0141 +Epoch [6/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0149, Loss2: 0.0146 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 41.2660 % Model2 40.9155 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0126, Loss2: 0.0127 +Epoch [7/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0132 +Epoch [7/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.1562, Loss1: 0.0133, Loss2: 0.0133 +Epoch [7/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0143, Loss2: 0.0140 +Epoch [7/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0133 +Epoch [7/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0125, Loss2: 0.0125 +Epoch [7/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0126, Loss2: 0.0127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 44.1206 % Model2 43.6599 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0126, Loss2: 0.0131 +Epoch [8/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0144, Loss2: 0.0141 +Epoch [8/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0121, Loss2: 0.0125 +Epoch [8/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0128, Loss2: 0.0126 +Epoch [8/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0135, Loss2: 0.0142 +Epoch [8/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0131, Loss2: 0.0131 +Epoch [8/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0125, Loss2: 0.0136 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 45.6631 % Model2 45.2825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0135, Loss2: 0.0137 +Epoch [9/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 34.3750, Loss1: 0.0140, Loss2: 0.0135 +Epoch [9/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0133, Loss2: 0.0134 +Epoch [9/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0122, Loss2: 0.0129 +Epoch [9/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0126, Loss2: 0.0115 +Epoch [9/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0122, Loss2: 0.0122 +Epoch [9/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0111, Loss2: 0.0117 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 42.9387 % Model2 42.5982 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0121, Loss2: 0.0129 +Epoch [10/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0126, Loss2: 0.0126 +Epoch [10/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0146, Loss2: 0.0145 +Epoch [10/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0131, Loss2: 0.0131 +Epoch [10/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0131, Loss2: 0.0138 +Epoch [10/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0110 +Epoch [10/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0117, Loss2: 0.0121 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 45.8433 % Model2 46.2039 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0111, Loss2: 0.0107 +Epoch [11/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0107, Loss2: 0.0109 +Epoch [11/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0140, Loss2: 0.0136 +Epoch [11/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0112, Loss2: 0.0121 +Epoch [11/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0129 +Epoch [11/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0156, Loss2: 0.0153 +Epoch [11/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0126, Loss2: 0.0125 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 45.3225 % Model2 45.5128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0121, Loss2: 0.0125 +Epoch [12/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0136, Loss2: 0.0135 +Epoch [12/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0130, Loss2: 0.0122 +Epoch [12/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0141, Loss2: 0.0144 +Epoch [12/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0120, Loss2: 0.0116 +Epoch [12/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0116, Loss2: 0.0115 +Epoch [12/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0114, Loss2: 0.0116 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 46.1839 % Model2 48.4675 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0112, Loss2: 0.0111 +Epoch [13/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0134, Loss2: 0.0127 +Epoch [13/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0129, Loss2: 0.0119 +Epoch [13/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0113, Loss2: 0.0117 +Epoch [13/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0109, Loss2: 0.0103 +Epoch [13/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0106, Loss2: 0.0111 +Epoch [13/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0124, Loss2: 0.0128 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 46.1238 % Model2 46.1739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0119, Loss2: 0.0116 +Epoch [14/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0132, Loss2: 0.0134 +Epoch [14/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0128, Loss2: 0.0134 +Epoch [14/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0120, Loss2: 0.0117 +Epoch [14/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0130, Loss2: 0.0123 +Epoch [14/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0124, Loss2: 0.0125 +Epoch [14/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0123, Loss2: 0.0127 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 45.5529 % Model2 45.5729 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0124, Loss2: 0.0133 +Epoch [15/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0103, Loss2: 0.0108 +Epoch [15/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0126, Loss2: 0.0117 +Epoch [15/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0121, Loss2: 0.0116 +Epoch [15/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0110, Loss2: 0.0112 +Epoch [15/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 37.5000, Loss1: 0.0136, Loss2: 0.0127 +Epoch [15/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0110, Loss2: 0.0108 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 48.7580 % Model2 48.5276 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 37.5000, Loss1: 0.0125, Loss2: 0.0128 +Epoch [16/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0119, Loss2: 0.0110 +Epoch [16/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0122, Loss2: 0.0130 +Epoch [16/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0120, Loss2: 0.0127 +Epoch [16/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0129, Loss2: 0.0130 +Epoch [16/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0110 +Epoch [16/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0102, Loss2: 0.0101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 48.6178 % Model2 48.0669 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0110, Loss2: 0.0107 +Epoch [17/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0125, Loss2: 0.0133 +Epoch [17/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0116, Loss2: 0.0117 +Epoch [17/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0123, Loss2: 0.0117 +Epoch [17/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0104, Loss2: 0.0104 +Epoch [17/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0108, Loss2: 0.0107 +Epoch [17/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0127, Loss2: 0.0120 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 46.4443 % Model2 48.3273 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0113, Loss2: 0.0111 +Epoch [18/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0109, Loss2: 0.0112 +Epoch [18/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0114, Loss2: 0.0114 +Epoch [18/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0124, Loss2: 0.0121 +Epoch [18/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0102, Loss2: 0.0105 +Epoch [18/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0100, Loss2: 0.0105 +Epoch [18/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0114, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 47.6963 % Model2 47.8566 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0121, Loss2: 0.0115 +Epoch [19/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0120, Loss2: 0.0120 +Epoch [19/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0096, Loss2: 0.0091 +Epoch [19/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0114, Loss2: 0.0108 +Epoch [19/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0105, Loss2: 0.0108 +Epoch [19/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0112, Loss2: 0.0119 +Epoch [19/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0122, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 50.5108 % Model2 51.0016 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0092, Loss2: 0.0094 +Epoch [20/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 37.5000, Loss1: 0.0126, Loss2: 0.0131 +Epoch [20/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0109, Loss2: 0.0114 +Epoch [20/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 36.7188, Loss1: 0.0127, Loss2: 0.0133 +Epoch [20/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0108, Loss2: 0.0108 +Epoch [20/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0109, Loss2: 0.0102 +Epoch [20/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.1875, Loss1: 0.0123, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 49.5192 % Model2 48.9583 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0546, Loss2: 0.0544 +Epoch [21/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0686, Loss2: 0.0663 +Epoch [21/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0596, Loss2: 0.0590 +Epoch [21/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0862, Loss2: 0.0860 +Epoch [21/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0571, Loss2: 0.0590 +Epoch [21/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0802, Loss2: 0.0795 +Epoch [21/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0734, Loss2: 0.0788 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 45.0321 % Model2 46.6146 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0674, Loss2: 0.0653 +Epoch [22/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0846, Loss2: 0.0847 +Epoch [22/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0710, Loss2: 0.0712 +Epoch [22/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0641, Loss2: 0.0644 +Epoch [22/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0490, Loss2: 0.0501 +Epoch [22/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0658, Loss2: 0.0629 +Epoch [22/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0488, Loss2: 0.0499 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 45.5429 % Model2 46.9251 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0520, Loss2: 0.0506 +Epoch [23/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 46.0938, Loss1: 0.0789, Loss2: 0.0730 +Epoch [23/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0665, Loss2: 0.0644 +Epoch [23/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0608, Loss2: 0.0605 +Epoch [23/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.1562, Loss1: 0.0795, Loss2: 0.0829 +Epoch [23/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0803, Loss2: 0.0829 +Epoch [23/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0567, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 44.8718 % Model2 45.7532 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.1012, Loss2: 0.1038 +Epoch [24/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0481, Loss2: 0.0478 +Epoch [24/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0720, Loss2: 0.0703 +Epoch [24/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0711, Loss2: 0.0683 +Epoch [24/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0628, Loss2: 0.0635 +Epoch [24/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0516, Loss2: 0.0507 +Epoch [24/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0623, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 41.6967 % Model2 44.6615 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0916, Loss2: 0.0960 +Epoch [25/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0816, Loss2: 0.0825 +Epoch [25/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0815, Loss2: 0.0837 +Epoch [25/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0777, Loss2: 0.0828 +Epoch [25/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 38.2812, Loss1: 0.0648, Loss2: 0.0672 +Epoch [25/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0684, Loss2: 0.0707 +Epoch [25/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0966, Loss2: 0.0918 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 47.8265 % Model2 49.3389 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0587, Loss2: 0.0600 +Epoch [26/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0567, Loss2: 0.0563 +Epoch [26/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0772, Loss2: 0.0771 +Epoch [26/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0751, Loss2: 0.0764 +Epoch [26/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0872, Loss2: 0.0853 +Epoch [26/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0730, Loss2: 0.0694 +Epoch [26/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0808, Loss2: 0.0830 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 47.0052 % Model2 46.3942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0577, Loss2: 0.0562 +Epoch [27/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0676, Loss2: 0.0660 +Epoch [27/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.8438, Loss1: 0.0725, Loss2: 0.0770 +Epoch [27/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0529, Loss2: 0.0525 +Epoch [27/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0848, Loss2: 0.0800 +Epoch [27/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0678, Loss2: 0.0683 +Epoch [27/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0668, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 46.9151 % Model2 47.0052 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 31.2500, Loss1: 0.0484, Loss2: 0.0504 +Epoch [28/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0589, Loss2: 0.0551 +Epoch [28/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0603, Loss2: 0.0607 +Epoch [28/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0608, Loss2: 0.0632 +Epoch [28/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.8750, Loss1: 0.0765, Loss2: 0.0822 +Epoch [28/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0593, Loss2: 0.0582 +Epoch [28/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0576, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 46.8049 % Model2 48.1370 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0837, Loss2: 0.0846 +Epoch [29/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0659, Loss2: 0.0678 +Epoch [29/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 37.5000, Loss1: 0.0620, Loss2: 0.0674 +Epoch [29/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0637, Loss2: 0.0650 +Epoch [29/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.9375, Loss1: 0.0619, Loss2: 0.0638 +Epoch [29/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0592, Loss2: 0.0574 +Epoch [29/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0629, Loss2: 0.0609 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 46.0337 % Model2 47.9567 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0587, Loss2: 0.0570 +Epoch [30/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0698, Loss2: 0.0715 +Epoch [30/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0557, Loss2: 0.0568 +Epoch [30/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0559, Loss2: 0.0542 +Epoch [30/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0617, Loss2: 0.0625 +Epoch [30/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0521, Loss2: 0.0530 +Epoch [30/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0738, Loss2: 0.0767 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 47.3958 % Model2 46.5545 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.0938, Loss1: 0.0515, Loss2: 0.0534 +Epoch [31/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0541, Loss2: 0.0547 +Epoch [31/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0600, Loss2: 0.0613 +Epoch [31/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0726, Loss2: 0.0762 +Epoch [31/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0805, Loss2: 0.0794 +Epoch [31/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0804, Loss2: 0.0829 +Epoch [31/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 48.4375, Loss1: 0.0555, Loss2: 0.0527 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 46.3742 % Model2 47.2756 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0676, Loss2: 0.0641 +Epoch [32/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0737, Loss2: 0.0720 +Epoch [32/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 44.5312, Loss1: 0.0565, Loss2: 0.0594 +Epoch [32/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0557, Loss2: 0.0575 +Epoch [32/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 43.7500, Loss1: 0.0559, Loss2: 0.0586 +Epoch [32/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0585, Loss2: 0.0588 +Epoch [32/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0596, Loss2: 0.0613 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 45.6230 % Model2 47.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0594, Loss2: 0.0593 +Epoch [33/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0734, Loss2: 0.0689 +Epoch [33/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 36.7188, Loss1: 0.0534, Loss2: 0.0561 +Epoch [33/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 41.4062, Loss1: 0.0612, Loss2: 0.0625 +Epoch [33/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0628, Loss2: 0.0641 +Epoch [33/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.1875, Loss1: 0.0740, Loss2: 0.0701 +Epoch [33/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0549, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 44.7516 % Model2 47.7464 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0581, Loss2: 0.0583 +Epoch [34/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0742, Loss2: 0.0765 +Epoch [34/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0855, Loss2: 0.0812 +Epoch [34/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0639, Loss2: 0.0694 +Epoch [34/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0640, Loss2: 0.0639 +Epoch [34/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0617, Loss2: 0.0587 +Epoch [34/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0585, Loss2: 0.0556 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 44.9619 % Model2 45.8834 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0627, Loss2: 0.0634 +Epoch [35/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0702, Loss2: 0.0705 +Epoch [35/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0742, Loss2: 0.0719 +Epoch [35/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0502, Loss2: 0.0519 +Epoch [35/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0720, Loss2: 0.0744 +Epoch [35/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0704, Loss2: 0.0701 +Epoch [35/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0823, Loss2: 0.0821 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 45.9635 % Model2 47.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0459, Loss2: 0.0479 +Epoch [36/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0665, Loss2: 0.0688 +Epoch [36/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0633, Loss2: 0.0633 +Epoch [36/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0604, Loss2: 0.0585 +Epoch [36/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0567, Loss2: 0.0569 +Epoch [36/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0656, Loss2: 0.0607 +Epoch [36/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0600, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 45.9635 % Model2 48.4976 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0822, Loss2: 0.0819 +Epoch [37/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0685, Loss2: 0.0718 +Epoch [37/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0662, Loss2: 0.0692 +Epoch [37/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0899, Loss2: 0.0920 +Epoch [37/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0656, Loss2: 0.0645 +Epoch [37/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0563, Loss2: 0.0579 +Epoch [37/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0568, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 46.4643 % Model2 48.1871 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 48.4375, Loss1: 0.0638, Loss2: 0.0679 +Epoch [38/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0557, Loss2: 0.0573 +Epoch [38/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0626, Loss2: 0.0648 +Epoch [38/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0701, Loss2: 0.0681 +Epoch [38/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0626, Loss2: 0.0644 +Epoch [38/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0822, Loss2: 0.0821 +Epoch [38/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0687, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 46.8049 % Model2 48.1771 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0610, Loss2: 0.0598 +Epoch [39/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0728, Loss2: 0.0748 +Epoch [39/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0602, Loss2: 0.0617 +Epoch [39/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0720, Loss2: 0.0704 +Epoch [39/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 38.2812, Loss1: 0.0770, Loss2: 0.0810 +Epoch [39/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0661, Loss2: 0.0649 +Epoch [39/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0561, Loss2: 0.0519 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 47.2356 % Model2 48.7780 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0760, Loss2: 0.0778 +Epoch [40/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0896, Loss2: 0.0919 +Epoch [40/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0631, Loss2: 0.0607 +Epoch [40/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0600, Loss2: 0.0611 +Epoch [40/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0614, Loss2: 0.0631 +Epoch [40/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0629, Loss2: 0.0607 +Epoch [40/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0702, Loss2: 0.0721 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 47.8065 % Model2 48.7981 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0730, Loss2: 0.0731 +Epoch [41/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0616, Loss2: 0.0607 +Epoch [41/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0668, Loss2: 0.0659 +Epoch [41/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0586, Loss2: 0.0591 +Epoch [41/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0590, Loss2: 0.0599 +Epoch [41/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0823, Loss2: 0.0793 +Epoch [41/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0514, Loss2: 0.0547 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 48.5276 % Model2 48.7580 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0653, Loss2: 0.0611 +Epoch [42/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0784, Loss2: 0.0762 +Epoch [42/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0707, Loss2: 0.0735 +Epoch [42/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0849, Loss2: 0.0822 +Epoch [42/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0771, Loss2: 0.0776 +Epoch [42/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0654, Loss2: 0.0649 +Epoch [42/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0609, Loss2: 0.0630 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 48.1671 % Model2 45.6330 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0618, Loss2: 0.0610 +Epoch [43/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0544, Loss2: 0.0551 +Epoch [43/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0706, Loss2: 0.0682 +Epoch [43/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0656, Loss2: 0.0666 +Epoch [43/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0606, Loss2: 0.0602 +Epoch [43/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 37.5000, Loss1: 0.0669, Loss2: 0.0755 +Epoch [43/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0482, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 46.6146 % Model2 45.9635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0592, Loss2: 0.0611 +Epoch [44/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0582, Loss2: 0.0570 +Epoch [44/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0654, Loss2: 0.0662 +Epoch [44/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 45.3125, Loss1: 0.0546, Loss2: 0.0584 +Epoch [44/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0597, Loss2: 0.0597 +Epoch [44/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0549, Loss2: 0.0553 +Epoch [44/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0514, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 46.3642 % Model2 46.6446 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0629, Loss2: 0.0623 +Epoch [45/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0756, Loss2: 0.0777 +Epoch [45/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0517, Loss2: 0.0539 +Epoch [45/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0610, Loss2: 0.0607 +Epoch [45/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.1875, Loss1: 0.0567, Loss2: 0.0600 +Epoch [45/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0580, Loss2: 0.0595 +Epoch [45/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0644, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 45.7432 % Model2 47.0052 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0653, Loss2: 0.0666 +Epoch [46/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0764, Loss2: 0.0780 +Epoch [46/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0606, Loss2: 0.0584 +Epoch [46/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0650, Loss2: 0.0714 +Epoch [46/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0647, Loss2: 0.0641 +Epoch [46/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0668, Loss2: 0.0692 +Epoch [46/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.0000, Loss1: 0.0629, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 48.1070 % Model2 47.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0614, Loss2: 0.0623 +Epoch [47/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0676, Loss2: 0.0715 +Epoch [47/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.0000, Loss1: 0.0560, Loss2: 0.0607 +Epoch [47/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0637, Loss2: 0.0625 +Epoch [47/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0585, Loss2: 0.0575 +Epoch [47/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0623, Loss2: 0.0635 +Epoch [47/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 42.9688, Loss1: 0.0435, Loss2: 0.0467 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 48.2973 % Model2 46.6246 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0638, Loss2: 0.0654 +Epoch [48/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0634, Loss2: 0.0667 +Epoch [48/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0727, Loss2: 0.0713 +Epoch [48/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0564, Loss2: 0.0563 +Epoch [48/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0571, Loss2: 0.0573 +Epoch [48/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0571, Loss2: 0.0549 +Epoch [48/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0674, Loss2: 0.0684 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 46.5144 % Model2 47.4760 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0698, Loss2: 0.0777 +Epoch [49/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0561, Loss2: 0.0545 +Epoch [49/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0546, Loss2: 0.0566 +Epoch [49/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0575, Loss2: 0.0577 +Epoch [49/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0595, Loss2: 0.0594 +Epoch [49/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0593, Loss2: 0.0571 +Epoch [49/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0591, Loss2: 0.0598 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 48.3173 % Model2 46.8349 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0584, Loss2: 0.0614 +Epoch [50/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0645, Loss2: 0.0618 +Epoch [50/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0571, Loss2: 0.0563 +Epoch [50/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0593, Loss2: 0.0582 +Epoch [50/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0642, Loss2: 0.0609 +Epoch [50/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0670, Loss2: 0.0656 +Epoch [50/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0503, Loss2: 0.0521 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 46.8249 % Model2 48.0168 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0579, Loss2: 0.0575 +Epoch [51/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0881, Loss2: 0.0874 +Epoch [51/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0753, Loss2: 0.0754 +Epoch [51/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0575, Loss2: 0.0616 +Epoch [51/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0735, Loss2: 0.0784 +Epoch [51/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0563, Loss2: 0.0562 +Epoch [51/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0579, Loss2: 0.0609 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 48.5777 % Model2 48.5577 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 51.5625, Loss1: 0.0533, Loss2: 0.0577 +Epoch [52/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0645, Loss2: 0.0680 +Epoch [52/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0496, Loss2: 0.0486 +Epoch [52/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0571, Loss2: 0.0601 +Epoch [52/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 58.5938, Loss1: 0.0754, Loss2: 0.0664 +Epoch [52/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0557, Loss2: 0.0584 +Epoch [52/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0562, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 48.4075 % Model2 47.9067 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0681, Loss2: 0.0696 +Epoch [53/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0567, Loss2: 0.0561 +Epoch [53/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0601, Loss2: 0.0575 +Epoch [53/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0560 +Epoch [53/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0669, Loss2: 0.0684 +Epoch [53/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0750, Loss2: 0.0733 +Epoch [53/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0610, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 45.8333 % Model2 46.9852 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0689, Loss2: 0.0663 +Epoch [54/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0578, Loss2: 0.0542 +Epoch [54/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0565, Loss2: 0.0531 +Epoch [54/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0662, Loss2: 0.0714 +Epoch [54/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0632, Loss2: 0.0593 +Epoch [54/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0573, Loss2: 0.0568 +Epoch [54/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0657, Loss2: 0.0669 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 46.7548 % Model2 47.5260 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0651, Loss2: 0.0667 +Epoch [55/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0746, Loss2: 0.0751 +Epoch [55/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0683, Loss2: 0.0654 +Epoch [55/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0827, Loss2: 0.0842 +Epoch [55/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0541, Loss2: 0.0548 +Epoch [55/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0668, Loss2: 0.0668 +Epoch [55/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0653, Loss2: 0.0648 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 47.9768 % Model2 49.2788 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0694, Loss2: 0.0669 +Epoch [56/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0684, Loss2: 0.0658 +Epoch [56/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0634, Loss2: 0.0676 +Epoch [56/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0592, Loss2: 0.0589 +Epoch [56/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0530, Loss2: 0.0538 +Epoch [56/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0632, Loss2: 0.0618 +Epoch [56/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0471, Loss2: 0.0450 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 47.6863 % Model2 48.8582 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0615, Loss2: 0.0629 +Epoch [57/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0691, Loss2: 0.0682 +Epoch [57/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.0938, Loss1: 0.0499, Loss2: 0.0527 +Epoch [57/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0636, Loss2: 0.0624 +Epoch [57/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0573, Loss2: 0.0560 +Epoch [57/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0733, Loss2: 0.0731 +Epoch [57/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 39.8438, Loss1: 0.0532, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 48.8081 % Model2 46.8249 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0623, Loss2: 0.0629 +Epoch [58/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0639, Loss2: 0.0629 +Epoch [58/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0680, Loss2: 0.0707 +Epoch [58/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0608, Loss2: 0.0622 +Epoch [58/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0762, Loss2: 0.0787 +Epoch [58/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0612, Loss2: 0.0602 +Epoch [58/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0536, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 48.2272 % Model2 47.3958 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0659, Loss2: 0.0651 +Epoch [59/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 46.8750, Loss1: 0.0516, Loss2: 0.0547 +Epoch [59/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0692, Loss2: 0.0744 +Epoch [59/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0555, Loss2: 0.0557 +Epoch [59/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0620, Loss2: 0.0565 +Epoch [59/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0582, Loss2: 0.0585 +Epoch [59/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0578, Loss2: 0.0578 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 47.8466 % Model2 48.0869 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0667, Loss2: 0.0669 +Epoch [60/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0583, Loss2: 0.0571 +Epoch [60/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0616, Loss2: 0.0588 +Epoch [60/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0472, Loss2: 0.0482 +Epoch [60/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0568, Loss2: 0.0596 +Epoch [60/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0622, Loss2: 0.0614 +Epoch [60/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0703, Loss2: 0.0694 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 47.0553 % Model2 48.1370 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0916, Loss2: 0.0937 +Epoch [61/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0596, Loss2: 0.0583 +Epoch [61/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0488, Loss2: 0.0490 +Epoch [61/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0642, Loss2: 0.0612 +Epoch [61/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0602, Loss2: 0.0591 +Epoch [61/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0660, Loss2: 0.0668 +Epoch [61/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0724, Loss2: 0.0740 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 47.6462 % Model2 47.0954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0614, Loss2: 0.0613 +Epoch [62/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0516, Loss2: 0.0520 +Epoch [62/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0598, Loss2: 0.0606 +Epoch [62/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0668, Loss2: 0.0646 +Epoch [62/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0644, Loss2: 0.0668 +Epoch [62/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0630, Loss2: 0.0604 +Epoch [62/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0692, Loss2: 0.0662 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 46.1839 % Model2 47.8566 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 49.2188, Loss1: 0.0561, Loss2: 0.0633 +Epoch [63/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0547, Loss2: 0.0568 +Epoch [63/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0578, Loss2: 0.0569 +Epoch [63/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0692, Loss2: 0.0747 +Epoch [63/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0671, Loss2: 0.0645 +Epoch [63/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0729, Loss2: 0.0741 +Epoch [63/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 56.2500, Loss1: 0.0692, Loss2: 0.0742 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 47.6262 % Model2 47.7965 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0903, Loss2: 0.0855 +Epoch [64/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0638, Loss2: 0.0598 +Epoch [64/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0681, Loss2: 0.0706 +Epoch [64/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0615, Loss2: 0.0647 +Epoch [64/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0545, Loss2: 0.0551 +Epoch [64/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0709, Loss2: 0.0683 +Epoch [64/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0651, Loss2: 0.0661 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 47.4159 % Model2 48.2272 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0610, Loss2: 0.0639 +Epoch [65/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0569, Loss2: 0.0536 +Epoch [65/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0685, Loss2: 0.0675 +Epoch [65/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0576, Loss2: 0.0576 +Epoch [65/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0560, Loss2: 0.0570 +Epoch [65/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0558, Loss2: 0.0568 +Epoch [65/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 48.4375, Loss1: 0.0588, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 47.8466 % Model2 47.8966 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0855, Loss2: 0.0836 +Epoch [66/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0565, Loss2: 0.0587 +Epoch [66/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0608, Loss2: 0.0595 +Epoch [66/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 59.3750, Loss1: 0.0550, Loss2: 0.0508 +Epoch [66/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0706, Loss2: 0.0728 +Epoch [66/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.0625, Loss1: 0.0650, Loss2: 0.0683 +Epoch [66/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0669, Loss2: 0.0663 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 47.0954 % Model2 47.7764 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0688, Loss2: 0.0659 +Epoch [67/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0712, Loss2: 0.0769 +Epoch [67/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0518, Loss2: 0.0524 +Epoch [67/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0550, Loss2: 0.0582 +Epoch [67/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0731, Loss2: 0.0722 +Epoch [67/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0574, Loss2: 0.0550 +Epoch [67/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0543, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 46.1538 % Model2 47.4559 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0640, Loss2: 0.0619 +Epoch [68/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0656, Loss2: 0.0665 +Epoch [68/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0739, Loss2: 0.0717 +Epoch [68/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0800, Loss2: 0.0776 +Epoch [68/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0569, Loss2: 0.0616 +Epoch [68/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0574, Loss2: 0.0523 +Epoch [68/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 59.3750, Loss1: 0.0617, Loss2: 0.0668 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 46.3442 % Model2 46.8850 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0571, Loss2: 0.0538 +Epoch [69/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0713, Loss2: 0.0709 +Epoch [69/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0571, Loss2: 0.0594 +Epoch [69/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0594, Loss2: 0.0559 +Epoch [69/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0779, Loss2: 0.0730 +Epoch [69/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0638, Loss2: 0.0630 +Epoch [69/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0658, Loss2: 0.0667 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 46.9351 % Model2 47.8866 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0632, Loss2: 0.0611 +Epoch [70/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0624, Loss2: 0.0664 +Epoch [70/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0778, Loss2: 0.0770 +Epoch [70/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0510, Loss2: 0.0497 +Epoch [70/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0670, Loss2: 0.0670 +Epoch [70/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0650, Loss2: 0.0661 +Epoch [70/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0641, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 46.4343 % Model2 47.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0608, Loss2: 0.0662 +Epoch [71/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0759, Loss2: 0.0715 +Epoch [71/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0658, Loss2: 0.0647 +Epoch [71/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0558, Loss2: 0.0546 +Epoch [71/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0564, Loss2: 0.0572 +Epoch [71/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0621, Loss2: 0.0626 +Epoch [71/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0562, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 47.2356 % Model2 47.9467 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0650, Loss2: 0.0707 +Epoch [72/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0637, Loss2: 0.0612 +Epoch [72/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.8750, Loss1: 0.0533, Loss2: 0.0575 +Epoch [72/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0690, Loss2: 0.0650 +Epoch [72/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0571, Loss2: 0.0570 +Epoch [72/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0834, Loss2: 0.0811 +Epoch [72/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0747, Loss2: 0.0761 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 47.8566 % Model2 47.2356 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0708, Loss2: 0.0697 +Epoch [73/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0658, Loss2: 0.0606 +Epoch [73/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0682, Loss2: 0.0667 +Epoch [73/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0475, Loss2: 0.0488 +Epoch [73/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0675, Loss2: 0.0663 +Epoch [73/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0836, Loss2: 0.0778 +Epoch [73/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0607, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 48.4275 % Model2 47.3157 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0685, Loss2: 0.0676 +Epoch [74/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0775, Loss2: 0.0750 +Epoch [74/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0678, Loss2: 0.0665 +Epoch [74/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0745, Loss2: 0.0686 +Epoch [74/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0616, Loss2: 0.0648 +Epoch [74/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0569, Loss2: 0.0581 +Epoch [74/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0536, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 46.8750 % Model2 47.3057 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0585, Loss2: 0.0589 +Epoch [75/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.0000, Loss1: 0.0571, Loss2: 0.0599 +Epoch [75/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0603, Loss2: 0.0626 +Epoch [75/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0647, Loss2: 0.0692 +Epoch [75/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.1044, Loss2: 0.1111 +Epoch [75/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0651, Loss2: 0.0688 +Epoch [75/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0768, Loss2: 0.0723 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 47.7664 % Model2 48.8782 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0560, Loss2: 0.0583 +Epoch [76/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0512, Loss2: 0.0517 +Epoch [76/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0578, Loss2: 0.0599 +Epoch [76/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0621, Loss2: 0.0605 +Epoch [76/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0660, Loss2: 0.0650 +Epoch [76/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0458, Loss2: 0.0430 +Epoch [76/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0506, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 46.4443 % Model2 46.6246 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0579 +Epoch [77/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0663, Loss2: 0.0682 +Epoch [77/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0663, Loss2: 0.0642 +Epoch [77/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0575, Loss2: 0.0553 +Epoch [77/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0585, Loss2: 0.0562 +Epoch [77/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0760, Loss2: 0.0768 +Epoch [77/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0701, Loss2: 0.0706 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 45.6831 % Model2 45.9936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0535, Loss2: 0.0510 +Epoch [78/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0641, Loss2: 0.0619 +Epoch [78/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.0000, Loss1: 0.0622, Loss2: 0.0667 +Epoch [78/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0536, Loss2: 0.0535 +Epoch [78/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0628, Loss2: 0.0683 +Epoch [78/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0595, Loss2: 0.0587 +Epoch [78/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0722, Loss2: 0.0732 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 48.2973 % Model2 47.4659 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0644, Loss2: 0.0641 +Epoch [79/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0613, Loss2: 0.0690 +Epoch [79/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0574, Loss2: 0.0587 +Epoch [79/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0743 +Epoch [79/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0675, Loss2: 0.0654 +Epoch [79/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0597, Loss2: 0.0610 +Epoch [79/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0527, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 47.5361 % Model2 47.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0573, Loss2: 0.0605 +Epoch [80/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0604, Loss2: 0.0634 +Epoch [80/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0543, Loss2: 0.0581 +Epoch [80/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0567, Loss2: 0.0582 +Epoch [80/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0742, Loss2: 0.0723 +Epoch [80/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 64.0625, Loss1: 0.0798, Loss2: 0.0694 +Epoch [80/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0592, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 46.0337 % Model2 47.4760 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0722, Loss2: 0.0662 +Epoch [81/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0610, Loss2: 0.0612 +Epoch [81/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0577, Loss2: 0.0619 +Epoch [81/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.8125, Loss1: 0.0615, Loss2: 0.0556 +Epoch [81/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0571, Loss2: 0.0551 +Epoch [81/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0617, Loss2: 0.0610 +Epoch [81/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0649, Loss2: 0.0643 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 46.1639 % Model2 46.7648 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0602, Loss2: 0.0622 +Epoch [82/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0753, Loss2: 0.0745 +Epoch [82/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0669, Loss2: 0.0708 +Epoch [82/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0677, Loss2: 0.0672 +Epoch [82/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0584, Loss2: 0.0581 +Epoch [82/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0551, Loss2: 0.0530 +Epoch [82/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.9062, Loss1: 0.0708, Loss2: 0.0782 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 46.8550 % Model2 46.6346 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 59.3750, Loss1: 0.0675, Loss2: 0.0611 +Epoch [83/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0642, Loss2: 0.0659 +Epoch [83/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0538, Loss2: 0.0539 +Epoch [83/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.8750, Loss1: 0.0559, Loss2: 0.0532 +Epoch [83/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 47.6562, Loss1: 0.0573, Loss2: 0.0622 +Epoch [83/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0698, Loss2: 0.0709 +Epoch [83/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 51.5625, Loss1: 0.0548, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 46.5345 % Model2 47.3458 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0760, Loss2: 0.0784 +Epoch [84/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0575, Loss2: 0.0575 +Epoch [84/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0814, Loss2: 0.0884 +Epoch [84/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0656, Loss2: 0.0622 +Epoch [84/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0622, Loss2: 0.0652 +Epoch [84/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0555, Loss2: 0.0512 +Epoch [84/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0507, Loss2: 0.0496 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 46.7448 % Model2 46.7748 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0510, Loss2: 0.0506 +Epoch [85/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.9062, Loss1: 0.0520, Loss2: 0.0563 +Epoch [85/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0589, Loss2: 0.0572 +Epoch [85/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0670, Loss2: 0.0679 +Epoch [85/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0916, Loss2: 0.0803 +Epoch [85/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0722, Loss2: 0.0774 +Epoch [85/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0591, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 45.8834 % Model2 46.8149 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0741, Loss2: 0.0713 +Epoch [86/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0539, Loss2: 0.0529 +Epoch [86/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0485, Loss2: 0.0487 +Epoch [86/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0576, Loss2: 0.0590 +Epoch [86/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0742, Loss2: 0.0733 +Epoch [86/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0708, Loss2: 0.0701 +Epoch [86/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0606, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 48.1270 % Model2 47.7063 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0544, Loss2: 0.0550 +Epoch [87/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0493, Loss2: 0.0469 +Epoch [87/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0542, Loss2: 0.0506 +Epoch [87/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0624, Loss2: 0.0635 +Epoch [87/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0568, Loss2: 0.0568 +Epoch [87/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0673, Loss2: 0.0710 +Epoch [87/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0582, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 46.0337 % Model2 45.7432 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0640, Loss2: 0.0697 +Epoch [88/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0776, Loss2: 0.0717 +Epoch [88/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0750, Loss2: 0.0731 +Epoch [88/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0579, Loss2: 0.0601 +Epoch [88/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0493, Loss2: 0.0479 +Epoch [88/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0737, Loss2: 0.0721 +Epoch [88/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0804, Loss2: 0.0814 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 47.3558 % Model2 47.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0737, Loss2: 0.0694 +Epoch [89/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0889, Loss2: 0.0847 +Epoch [89/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 54.6875, Loss1: 0.0507, Loss2: 0.0451 +Epoch [89/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0598, Loss2: 0.0589 +Epoch [89/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0579, Loss2: 0.0601 +Epoch [89/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0648, Loss2: 0.0703 +Epoch [89/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0527, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 46.7849 % Model2 47.5861 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0563, Loss2: 0.0532 +Epoch [90/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0572, Loss2: 0.0545 +Epoch [90/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0610, Loss2: 0.0583 +Epoch [90/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0770, Loss2: 0.0695 +Epoch [90/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0613, Loss2: 0.0580 +Epoch [90/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0486, Loss2: 0.0463 +Epoch [90/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0680, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 47.7264 % Model2 46.2139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 63.2812, Loss1: 0.0666, Loss2: 0.0590 +Epoch [91/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0579, Loss2: 0.0574 +Epoch [91/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0835, Loss2: 0.0850 +Epoch [91/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0845, Loss2: 0.0832 +Epoch [91/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0777, Loss2: 0.0730 +Epoch [91/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0557, Loss2: 0.0586 +Epoch [91/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0656, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 47.1354 % Model2 47.2556 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0657, Loss2: 0.0649 +Epoch [92/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0670, Loss2: 0.0640 +Epoch [92/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0784, Loss2: 0.0741 +Epoch [92/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0746, Loss2: 0.0804 +Epoch [92/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0849, Loss2: 0.0838 +Epoch [92/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0651, Loss2: 0.0632 +Epoch [92/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0610, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 47.8866 % Model2 46.4543 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0582, Loss2: 0.0585 +Epoch [93/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0561, Loss2: 0.0539 +Epoch [93/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0681, Loss2: 0.0657 +Epoch [93/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0586, Loss2: 0.0579 +Epoch [93/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0640, Loss2: 0.0653 +Epoch [93/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0572, Loss2: 0.0562 +Epoch [93/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0522, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 47.9968 % Model2 47.6963 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0640, Loss2: 0.0619 +Epoch [94/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0816, Loss2: 0.0855 +Epoch [94/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0690, Loss2: 0.0676 +Epoch [94/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0767, Loss2: 0.0773 +Epoch [94/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0658, Loss2: 0.0637 +Epoch [94/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0619, Loss2: 0.0620 +Epoch [94/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0634, Loss2: 0.0653 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 45.4227 % Model2 45.1522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0801, Loss2: 0.0777 +Epoch [95/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0605, Loss2: 0.0590 +Epoch [95/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0529, Loss2: 0.0498 +Epoch [95/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0871, Loss2: 0.0866 +Epoch [95/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0630, Loss2: 0.0668 +Epoch [95/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0643, Loss2: 0.0639 +Epoch [95/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0666, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 47.2857 % Model2 47.0653 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0599, Loss2: 0.0581 +Epoch [96/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0556, Loss2: 0.0520 +Epoch [96/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0639, Loss2: 0.0649 +Epoch [96/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0824, Loss2: 0.0733 +Epoch [96/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0768, Loss2: 0.0774 +Epoch [96/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0688, Loss2: 0.0669 +Epoch [96/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0704, Loss2: 0.0672 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 47.5461 % Model2 46.4243 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0727, Loss2: 0.0750 +Epoch [97/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0472, Loss2: 0.0454 +Epoch [97/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0628, Loss2: 0.0591 +Epoch [97/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0655 +Epoch [97/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0663, Loss2: 0.0642 +Epoch [97/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0631, Loss2: 0.0636 +Epoch [97/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0683, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 46.6046 % Model2 45.5829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0530, Loss2: 0.0515 +Epoch [98/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 64.8438, Loss1: 0.0731, Loss2: 0.0636 +Epoch [98/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.0895, Loss2: 0.0827 +Epoch [98/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0643 +Epoch [98/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0659, Loss2: 0.0681 +Epoch [98/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0657, Loss2: 0.0643 +Epoch [98/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0554, Loss2: 0.0512 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 47.2556 % Model2 47.3858 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0618, Loss2: 0.0612 +Epoch [99/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.1107, Loss2: 0.1107 +Epoch [99/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0703, Loss2: 0.0658 +Epoch [99/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0539, Loss2: 0.0536 +Epoch [99/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0627, Loss2: 0.0654 +Epoch [99/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0640, Loss2: 0.0701 +Epoch [99/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0785, Loss2: 0.0763 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 46.5445 % Model2 45.7732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0658, Loss2: 0.0713 +Epoch [100/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0755, Loss2: 0.0740 +Epoch [100/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0790, Loss2: 0.0732 +Epoch [100/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0765, Loss2: 0.0724 +Epoch [100/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0742, Loss2: 0.0698 +Epoch [100/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0553 +Epoch [100/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0732, Loss2: 0.0763 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 46.9451 % Model2 47.3458 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0688, Loss2: 0.0659 +Epoch [101/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0717, Loss2: 0.0728 +Epoch [101/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0703, Loss2: 0.0756 +Epoch [101/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0678, Loss2: 0.0710 +Epoch [101/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0685, Loss2: 0.0671 +Epoch [101/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0870, Loss2: 0.0835 +Epoch [101/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0640, Loss2: 0.0674 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 46.6146 % Model2 46.0236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0629, Loss2: 0.0652 +Epoch [102/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0531, Loss2: 0.0528 +Epoch [102/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 53.9062, Loss1: 0.0573, Loss2: 0.0627 +Epoch [102/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 51.5625, Loss1: 0.0591, Loss2: 0.0624 +Epoch [102/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0653, Loss2: 0.0717 +Epoch [102/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0615, Loss2: 0.0657 +Epoch [102/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 61.7188, Loss1: 0.0648, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 47.0753 % Model2 47.6462 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0607, Loss2: 0.0644 +Epoch [103/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0770, Loss2: 0.0828 +Epoch [103/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0639, Loss2: 0.0644 +Epoch [103/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0802, Loss2: 0.0814 +Epoch [103/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0671, Loss2: 0.0647 +Epoch [103/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0590, Loss2: 0.0577 +Epoch [103/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0658, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 46.6546 % Model2 46.8349 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0614, Loss2: 0.0601 +Epoch [104/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0632, Loss2: 0.0681 +Epoch [104/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0769, Loss2: 0.0710 +Epoch [104/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0707, Loss2: 0.0667 +Epoch [104/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0742, Loss2: 0.0692 +Epoch [104/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0584, Loss2: 0.0594 +Epoch [104/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0771, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 46.7648 % Model2 46.6046 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0633, Loss2: 0.0660 +Epoch [105/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0798, Loss2: 0.0785 +Epoch [105/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0743, Loss2: 0.0809 +Epoch [105/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0648, Loss2: 0.0697 +Epoch [105/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 54.6875, Loss1: 0.0563, Loss2: 0.0635 +Epoch [105/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 56.2500, Loss1: 0.0574, Loss2: 0.0639 +Epoch [105/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0772, Loss2: 0.0763 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 45.2724 % Model2 46.1338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0649, Loss2: 0.0672 +Epoch [106/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0685, Loss2: 0.0645 +Epoch [106/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0770, Loss2: 0.0751 +Epoch [106/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0603, Loss2: 0.0548 +Epoch [106/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0636, Loss2: 0.0665 +Epoch [106/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0649, Loss2: 0.0633 +Epoch [106/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0598, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 47.2656 % Model2 46.5044 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0589, Loss2: 0.0547 +Epoch [107/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0680, Loss2: 0.0636 +Epoch [107/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0609, Loss2: 0.0589 +Epoch [107/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0594, Loss2: 0.0576 +Epoch [107/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.0580, Loss2: 0.0631 +Epoch [107/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0880, Loss2: 0.0934 +Epoch [107/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0691, Loss2: 0.0663 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 47.3357 % Model2 45.9235 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1102, Loss2: 0.1079 +Epoch [108/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0952, Loss2: 0.0908 +Epoch [108/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0555, Loss2: 0.0592 +Epoch [108/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0828, Loss2: 0.0782 +Epoch [108/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0818, Loss2: 0.0795 +Epoch [108/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0785, Loss2: 0.0817 +Epoch [108/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0733, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 46.1038 % Model2 46.8750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0749, Loss2: 0.0737 +Epoch [109/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0762, Loss2: 0.0816 +Epoch [109/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0723, Loss2: 0.0715 +Epoch [109/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0598, Loss2: 0.0610 +Epoch [109/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0641 +Epoch [109/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 63.2812, Loss1: 0.0622, Loss2: 0.0556 +Epoch [109/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0725, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 46.0537 % Model2 45.7332 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 63.2812, Loss1: 0.0662, Loss2: 0.0592 +Epoch [110/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0642, Loss2: 0.0590 +Epoch [110/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0740, Loss2: 0.0711 +Epoch [110/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0815, Loss2: 0.0737 +Epoch [110/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0757, Loss2: 0.0734 +Epoch [110/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0781, Loss2: 0.0777 +Epoch [110/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0694, Loss2: 0.0703 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 46.1538 % Model2 46.8450 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0658, Loss2: 0.0702 +Epoch [111/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0634, Loss2: 0.0614 +Epoch [111/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0528, Loss2: 0.0502 +Epoch [111/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0684, Loss2: 0.0705 +Epoch [111/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0660, Loss2: 0.0683 +Epoch [111/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0733, Loss2: 0.0721 +Epoch [111/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0812, Loss2: 0.0794 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 46.2139 % Model2 46.6947 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0863, Loss2: 0.0819 +Epoch [112/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0613, Loss2: 0.0604 +Epoch [112/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0768, Loss2: 0.0716 +Epoch [112/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0698, Loss2: 0.0653 +Epoch [112/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0655, Loss2: 0.0614 +Epoch [112/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0607, Loss2: 0.0590 +Epoch [112/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 54.6875, Loss1: 0.0531, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 46.5946 % Model2 47.0954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0612, Loss2: 0.0650 +Epoch [113/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0751, Loss2: 0.0732 +Epoch [113/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0890, Loss2: 0.0961 +Epoch [113/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0862, Loss2: 0.0912 +Epoch [113/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0680, Loss2: 0.0666 +Epoch [113/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0855, Loss2: 0.0788 +Epoch [113/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0809, Loss2: 0.0826 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 46.3642 % Model2 47.4159 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0705, Loss2: 0.0691 +Epoch [114/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0855, Loss2: 0.0812 +Epoch [114/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0571, Loss2: 0.0544 +Epoch [114/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0749, Loss2: 0.0688 +Epoch [114/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0667, Loss2: 0.0712 +Epoch [114/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.9688, Loss1: 0.0886, Loss2: 0.0813 +Epoch [114/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 67.9688, Loss1: 0.0714, Loss2: 0.0618 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 45.8634 % Model2 46.7047 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0821, Loss2: 0.0745 +Epoch [115/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0614, Loss2: 0.0622 +Epoch [115/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0735, Loss2: 0.0765 +Epoch [115/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0735, Loss2: 0.0801 +Epoch [115/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0673, Loss2: 0.0677 +Epoch [115/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0763, Loss2: 0.0728 +Epoch [115/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0653, Loss2: 0.0655 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 47.0553 % Model2 44.8017 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0690, Loss2: 0.0705 +Epoch [116/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0629, Loss2: 0.0668 +Epoch [116/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0854, Loss2: 0.0949 +Epoch [116/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0773, Loss2: 0.0695 +Epoch [116/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 52.3438, Loss1: 0.0513, Loss2: 0.0547 +Epoch [116/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0776, Loss2: 0.0761 +Epoch [116/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0742, Loss2: 0.0808 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 46.9651 % Model2 47.0453 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0695, Loss2: 0.0660 +Epoch [117/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0745, Loss2: 0.0746 +Epoch [117/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0660, Loss2: 0.0667 +Epoch [117/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0859, Loss2: 0.0887 +Epoch [117/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0777, Loss2: 0.0764 +Epoch [117/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0774, Loss2: 0.0782 +Epoch [117/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0740, Loss2: 0.0804 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 46.9351 % Model2 47.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0609, Loss2: 0.0648 +Epoch [118/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0780, Loss2: 0.0775 +Epoch [118/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0819, Loss2: 0.0874 +Epoch [118/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0694, Loss2: 0.0683 +Epoch [118/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0597, Loss2: 0.0589 +Epoch [118/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0734, Loss2: 0.0731 +Epoch [118/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0659, Loss2: 0.0671 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 46.9752 % Model2 46.0537 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0468, Loss2: 0.0482 +Epoch [119/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 68.7500, Loss1: 0.0758, Loss2: 0.0658 +Epoch [119/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0919, Loss2: 0.0895 +Epoch [119/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0608, Loss2: 0.0618 +Epoch [119/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0516, Loss2: 0.0503 +Epoch [119/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0691, Loss2: 0.0677 +Epoch [119/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0685, Loss2: 0.0686 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 46.8850 % Model2 46.6146 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0683, Loss2: 0.0652 +Epoch [120/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0858, Loss2: 0.0845 +Epoch [120/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0594, Loss2: 0.0580 +Epoch [120/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0754, Loss2: 0.0764 +Epoch [120/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0607, Loss2: 0.0597 +Epoch [120/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0957, Loss2: 0.0945 +Epoch [120/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0598, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 47.0052 % Model2 46.6747 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0885, Loss2: 0.0904 +Epoch [121/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0702, Loss2: 0.0704 +Epoch [121/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0619, Loss2: 0.0615 +Epoch [121/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0659, Loss2: 0.0698 +Epoch [121/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0762, Loss2: 0.0753 +Epoch [121/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1012, Loss2: 0.0983 +Epoch [121/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0670, Loss2: 0.0702 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 45.9535 % Model2 46.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0679, Loss2: 0.0685 +Epoch [122/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0657, Loss2: 0.0691 +Epoch [122/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0705, Loss2: 0.0686 +Epoch [122/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0822, Loss2: 0.0893 +Epoch [122/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0686, Loss2: 0.0707 +Epoch [122/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0635, Loss2: 0.0634 +Epoch [122/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0664, Loss2: 0.0676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 46.2540 % Model2 47.1955 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0784, Loss2: 0.0784 +Epoch [123/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0695, Loss2: 0.0713 +Epoch [123/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0779, Loss2: 0.0767 +Epoch [123/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0617, Loss2: 0.0657 +Epoch [123/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0617, Loss2: 0.0636 +Epoch [123/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0859, Loss2: 0.0779 +Epoch [123/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0700, Loss2: 0.0678 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 46.7448 % Model2 47.1955 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0696, Loss2: 0.0647 +Epoch [124/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0630, Loss2: 0.0657 +Epoch [124/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0869, Loss2: 0.0823 +Epoch [124/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0778, Loss2: 0.0830 +Epoch [124/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0643, Loss2: 0.0670 +Epoch [124/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0679, Loss2: 0.0691 +Epoch [124/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.1027, Loss2: 0.0959 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 46.3642 % Model2 45.8033 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0676, Loss2: 0.0718 +Epoch [125/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.1250, Loss1: 0.0579, Loss2: 0.0635 +Epoch [125/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0699, Loss2: 0.0685 +Epoch [125/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0760, Loss2: 0.0765 +Epoch [125/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0781, Loss2: 0.0773 +Epoch [125/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0764, Loss2: 0.0785 +Epoch [125/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 69.5312, Loss1: 0.0651, Loss2: 0.0596 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 45.9435 % Model2 46.0637 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0706, Loss2: 0.0718 +Epoch [126/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0966, Loss2: 0.0939 +Epoch [126/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0804, Loss2: 0.0757 +Epoch [126/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.1090, Loss2: 0.1004 +Epoch [126/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0720, Loss2: 0.0800 +Epoch [126/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1142, Loss2: 0.1036 +Epoch [126/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0766, Loss2: 0.0815 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 46.7648 % Model2 46.7748 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.9375, Loss1: 0.1003, Loss2: 0.1150 +Epoch [127/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0735, Loss2: 0.0747 +Epoch [127/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0676, Loss2: 0.0643 +Epoch [127/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0656, Loss2: 0.0682 +Epoch [127/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.1038, Loss2: 0.1051 +Epoch [127/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0758, Loss2: 0.0803 +Epoch [127/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0665, Loss2: 0.0628 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 45.8834 % Model2 46.8149 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0688, Loss2: 0.0677 +Epoch [128/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0667, Loss2: 0.0649 +Epoch [128/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0852, Loss2: 0.0824 +Epoch [128/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1105, Loss2: 0.1052 +Epoch [128/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0897, Loss2: 0.0871 +Epoch [128/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0675, Loss2: 0.0716 +Epoch [128/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0605, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 46.2640 % Model2 46.1238 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0638, Loss2: 0.0644 +Epoch [129/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0745, Loss2: 0.0761 +Epoch [129/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.1875, Loss1: 0.0813, Loss2: 0.0903 +Epoch [129/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0973, Loss2: 0.1048 +Epoch [129/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0876, Loss2: 0.0856 +Epoch [129/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0653, Loss2: 0.0625 +Epoch [129/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0996, Loss2: 0.0956 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 46.4443 % Model2 46.1538 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0748, Loss2: 0.0706 +Epoch [130/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0816, Loss2: 0.0799 +Epoch [130/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0646, Loss2: 0.0641 +Epoch [130/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0812, Loss2: 0.0745 +Epoch [130/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0671, Loss2: 0.0673 +Epoch [130/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0956, Loss2: 0.0966 +Epoch [130/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0892, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 46.7748 % Model2 47.0152 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0961, Loss2: 0.0836 +Epoch [131/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0899, Loss2: 0.0930 +Epoch [131/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0653, Loss2: 0.0638 +Epoch [131/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1016, Loss2: 0.1080 +Epoch [131/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0822, Loss2: 0.0795 +Epoch [131/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0844, Loss2: 0.0880 +Epoch [131/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1008, Loss2: 0.0981 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 45.8033 % Model2 45.3726 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0844, Loss2: 0.0835 +Epoch [132/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0925, Loss2: 0.0799 +Epoch [132/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0752, Loss2: 0.0789 +Epoch [132/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0789, Loss2: 0.0775 +Epoch [132/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0690 +Epoch [132/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0948, Loss2: 0.0916 +Epoch [132/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0920, Loss2: 0.0875 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 45.9836 % Model2 46.4443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0726, Loss2: 0.0782 +Epoch [133/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.1051, Loss2: 0.1041 +Epoch [133/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0530, Loss2: 0.0544 +Epoch [133/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0737, Loss2: 0.0753 +Epoch [133/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0748, Loss2: 0.0711 +Epoch [133/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0725, Loss2: 0.0760 +Epoch [133/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0770, Loss2: 0.0789 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 46.1238 % Model2 46.2540 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0889, Loss2: 0.0893 +Epoch [134/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0898, Loss2: 0.0824 +Epoch [134/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0721, Loss2: 0.0789 +Epoch [134/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 65.6250, Loss1: 0.0718, Loss2: 0.0781 +Epoch [134/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0772, Loss2: 0.0728 +Epoch [134/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0813, Loss2: 0.0754 +Epoch [134/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0744, Loss2: 0.0750 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 46.6046 % Model2 46.5946 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0681, Loss2: 0.0708 +Epoch [135/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0789, Loss2: 0.0747 +Epoch [135/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0755, Loss2: 0.0739 +Epoch [135/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0798, Loss2: 0.0782 +Epoch [135/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0820, Loss2: 0.0823 +Epoch [135/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.1004, Loss2: 0.0973 +Epoch [135/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0876, Loss2: 0.0916 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 46.6246 % Model2 45.9936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0797, Loss2: 0.0811 +Epoch [136/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0736, Loss2: 0.0702 +Epoch [136/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0760, Loss2: 0.0819 +Epoch [136/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0641, Loss2: 0.0642 +Epoch [136/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0626, Loss2: 0.0654 +Epoch [136/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0960, Loss2: 0.0983 +Epoch [136/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.1043, Loss2: 0.0955 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 46.3942 % Model2 45.5128 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 71.0938, Loss1: 0.0870, Loss2: 0.0787 +Epoch [137/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0678, Loss2: 0.0662 +Epoch [137/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.0801, Loss2: 0.0703 +Epoch [137/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0753, Loss2: 0.0764 +Epoch [137/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1008, Loss2: 0.0973 +Epoch [137/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0769, Loss2: 0.0736 +Epoch [137/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0701, Loss2: 0.0689 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 45.8534 % Model2 46.4643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0930, Loss2: 0.0910 +Epoch [138/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0932, Loss2: 0.0967 +Epoch [138/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.1087, Loss2: 0.1199 +Epoch [138/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.1067, Loss2: 0.0935 +Epoch [138/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0808, Loss2: 0.0834 +Epoch [138/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0694, Loss2: 0.0675 +Epoch [138/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0912, Loss2: 0.0915 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 46.1739 % Model2 45.7632 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0668, Loss2: 0.0659 +Epoch [139/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0775, Loss2: 0.0812 +Epoch [139/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0939, Loss2: 0.0875 +Epoch [139/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0810, Loss2: 0.0870 +Epoch [139/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0861, Loss2: 0.0794 +Epoch [139/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 62.5000, Loss1: 0.0609, Loss2: 0.0540 +Epoch [139/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0771, Loss2: 0.0807 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 46.0637 % Model2 45.8133 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0666, Loss2: 0.0664 +Epoch [140/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0912, Loss2: 0.0835 +Epoch [140/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0608, Loss2: 0.0629 +Epoch [140/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0887, Loss2: 0.0943 +Epoch [140/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0721, Loss2: 0.0742 +Epoch [140/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0940, Loss2: 0.0916 +Epoch [140/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0710, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 46.1238 % Model2 46.3742 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0836, Loss2: 0.0843 +Epoch [141/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0852, Loss2: 0.0820 +Epoch [141/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0699, Loss2: 0.0657 +Epoch [141/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0954, Loss2: 0.0937 +Epoch [141/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 55.4688, Loss1: 0.0626, Loss2: 0.0678 +Epoch [141/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0620, Loss2: 0.0621 +Epoch [141/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0857, Loss2: 0.0852 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 45.6330 % Model2 46.5845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0837, Loss2: 0.0790 +Epoch [142/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0907, Loss2: 0.0870 +Epoch [142/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1069, Loss2: 0.1094 +Epoch [142/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1012, Loss2: 0.1034 +Epoch [142/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0898, Loss2: 0.0925 +Epoch [142/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0789, Loss2: 0.0760 +Epoch [142/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0766, Loss2: 0.0740 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 46.0938 % Model2 46.0938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0995, Loss2: 0.0980 +Epoch [143/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0753, Loss2: 0.0793 +Epoch [143/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0770, Loss2: 0.0834 +Epoch [143/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0787, Loss2: 0.0825 +Epoch [143/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0773, Loss2: 0.0762 +Epoch [143/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0659, Loss2: 0.0712 +Epoch [143/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0772, Loss2: 0.0790 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 46.1639 % Model2 45.4627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0796, Loss2: 0.0747 +Epoch [144/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.1093, Loss2: 0.1080 +Epoch [144/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0745, Loss2: 0.0823 +Epoch [144/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0906, Loss2: 0.0835 +Epoch [144/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0669, Loss2: 0.0644 +Epoch [144/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.0939, Loss2: 0.0844 +Epoch [144/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0871, Loss2: 0.0790 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 46.0437 % Model2 45.7432 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0772, Loss2: 0.0731 +Epoch [145/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0877, Loss2: 0.0875 +Epoch [145/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1005, Loss2: 0.0916 +Epoch [145/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0799, Loss2: 0.0777 +Epoch [145/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 59.3750, Loss1: 0.0720, Loss2: 0.0790 +Epoch [145/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1008, Loss2: 0.0976 +Epoch [145/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0891, Loss2: 0.0927 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 46.2039 % Model2 46.1138 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0938, Loss2: 0.0964 +Epoch [146/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.1122, Loss2: 0.1056 +Epoch [146/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 57.0312, Loss1: 0.0898, Loss2: 0.0976 +Epoch [146/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0758, Loss2: 0.0842 +Epoch [146/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0884, Loss2: 0.0922 +Epoch [146/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0721, Loss2: 0.0762 +Epoch [146/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0741, Loss2: 0.0732 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 46.0938 % Model2 45.7732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0781, Loss2: 0.0773 +Epoch [147/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1183, Loss2: 0.1103 +Epoch [147/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0654, Loss2: 0.0680 +Epoch [147/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0707, Loss2: 0.0692 +Epoch [147/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1321, Loss2: 0.1442 +Epoch [147/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0942, Loss2: 0.0912 +Epoch [147/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0935, Loss2: 0.0909 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 46.1939 % Model2 45.9335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0753, Loss2: 0.0714 +Epoch [148/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0939, Loss2: 0.0864 +Epoch [148/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0947, Loss2: 0.0982 +Epoch [148/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.1036, Loss2: 0.0889 +Epoch [148/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1317, Loss2: 0.1367 +Epoch [148/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.1091, Loss2: 0.1034 +Epoch [148/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0842, Loss2: 0.0783 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 45.8233 % Model2 45.2925 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1543, Loss2: 0.1690 +Epoch [149/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0858, Loss2: 0.0812 +Epoch [149/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0811, Loss2: 0.0768 +Epoch [149/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0705, Loss2: 0.0761 +Epoch [149/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0773, Loss2: 0.0746 +Epoch [149/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0689, Loss2: 0.0680 +Epoch [149/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.1034, Loss2: 0.0982 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 46.2039 % Model2 45.7031 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0870, Loss2: 0.0828 +Epoch [150/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0659, Loss2: 0.0668 +Epoch [150/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0912, Loss2: 0.0929 +Epoch [150/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0867, Loss2: 0.0869 +Epoch [150/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0735, Loss2: 0.0717 +Epoch [150/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.1080, Loss2: 0.1201 +Epoch [150/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0839, Loss2: 0.0838 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 44.8818 % Model2 45.9435 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1038, Loss2: 0.0995 +Epoch [151/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0960, Loss2: 0.0990 +Epoch [151/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 72.6562, Loss1: 0.0868, Loss2: 0.0772 +Epoch [151/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0767, Loss2: 0.0782 +Epoch [151/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0774, Loss2: 0.0807 +Epoch [151/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0825, Loss2: 0.0836 +Epoch [151/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0949, Loss2: 0.0947 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 45.7732 % Model2 46.0337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0775, Loss2: 0.0823 +Epoch [152/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0778, Loss2: 0.0758 +Epoch [152/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0779, Loss2: 0.0806 +Epoch [152/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0849, Loss2: 0.0814 +Epoch [152/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1130, Loss2: 0.1109 +Epoch [152/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0821, Loss2: 0.0777 +Epoch [152/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0875, Loss2: 0.0839 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 45.9135 % Model2 45.9936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0835, Loss2: 0.0825 +Epoch [153/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0872, Loss2: 0.0811 +Epoch [153/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0952, Loss2: 0.0889 +Epoch [153/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0654, Loss2: 0.0645 +Epoch [153/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0900, Loss2: 0.0893 +Epoch [153/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0799, Loss2: 0.0757 +Epoch [153/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0803, Loss2: 0.0708 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 46.2640 % Model2 45.6230 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0909, Loss2: 0.0949 +Epoch [154/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1010, Loss2: 0.0976 +Epoch [154/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0754, Loss2: 0.0775 +Epoch [154/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0797, Loss2: 0.0834 +Epoch [154/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0983, Loss2: 0.0922 +Epoch [154/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0930, Loss2: 0.0873 +Epoch [154/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0884, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 46.0938 % Model2 46.4744 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0879, Loss2: 0.0868 +Epoch [155/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0911, Loss2: 0.0898 +Epoch [155/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0973, Loss2: 0.1032 +Epoch [155/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0695, Loss2: 0.0706 +Epoch [155/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0884, Loss2: 0.0929 +Epoch [155/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1145, Loss2: 0.1098 +Epoch [155/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0983, Loss2: 0.0903 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 46.1739 % Model2 46.2740 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0909, Loss2: 0.0802 +Epoch [156/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0951, Loss2: 0.0896 +Epoch [156/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0921, Loss2: 0.0921 +Epoch [156/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0725, Loss2: 0.0797 +Epoch [156/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1196, Loss2: 0.1127 +Epoch [156/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0832, Loss2: 0.0799 +Epoch [156/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0899, Loss2: 0.0912 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 46.3842 % Model2 46.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0726, Loss2: 0.0769 +Epoch [157/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0616, Loss2: 0.0621 +Epoch [157/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1117, Loss2: 0.1225 +Epoch [157/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0735, Loss2: 0.0693 +Epoch [157/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0775, Loss2: 0.0689 +Epoch [157/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0893, Loss2: 0.0900 +Epoch [157/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0771, Loss2: 0.0814 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 46.3742 % Model2 46.6947 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1005, Loss2: 0.1125 +Epoch [158/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0771, Loss2: 0.0801 +Epoch [158/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 67.9688, Loss1: 0.0620, Loss2: 0.0702 +Epoch [158/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0671, Loss2: 0.0681 +Epoch [158/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0736, Loss2: 0.0695 +Epoch [158/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0857, Loss2: 0.0844 +Epoch [158/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0767, Loss2: 0.0731 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 45.7933 % Model2 46.2039 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0883, Loss2: 0.0864 +Epoch [159/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.0850, Loss2: 0.0762 +Epoch [159/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0880, Loss2: 0.0899 +Epoch [159/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1056, Loss2: 0.1062 +Epoch [159/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0684, Loss2: 0.0721 +Epoch [159/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1018, Loss2: 0.1004 +Epoch [159/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0968, Loss2: 0.1032 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 45.3125 % Model2 45.7933 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1044, Loss2: 0.1101 +Epoch [160/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1095, Loss2: 0.1014 +Epoch [160/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0980, Loss2: 0.0848 +Epoch [160/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.1007, Loss2: 0.0885 +Epoch [160/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0751, Loss2: 0.0718 +Epoch [160/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0747, Loss2: 0.0742 +Epoch [160/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0889, Loss2: 0.0928 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 45.5228 % Model2 46.2841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0657, Loss2: 0.0668 +Epoch [161/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.1030, Loss2: 0.1147 +Epoch [161/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0928, Loss2: 0.0894 +Epoch [161/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1065, Loss2: 0.1085 +Epoch [161/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0842, Loss2: 0.0798 +Epoch [161/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0914, Loss2: 0.0882 +Epoch [161/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0817, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 45.7632 % Model2 46.2640 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0849, Loss2: 0.0845 +Epoch [162/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0758, Loss2: 0.0749 +Epoch [162/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0794, Loss2: 0.0782 +Epoch [162/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0952, Loss2: 0.0981 +Epoch [162/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0703, Loss2: 0.0657 +Epoch [162/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0741, Loss2: 0.0801 +Epoch [162/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0820, Loss2: 0.0798 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 45.6631 % Model2 46.1338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0997, Loss2: 0.0940 +Epoch [163/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1094, Loss2: 0.1098 +Epoch [163/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0788, Loss2: 0.0757 +Epoch [163/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0843, Loss2: 0.0787 +Epoch [163/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0754, Loss2: 0.0801 +Epoch [163/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0930, Loss2: 0.0974 +Epoch [163/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0752, Loss2: 0.0842 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 46.0136 % Model2 46.0737 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0833, Loss2: 0.0834 +Epoch [164/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0946, Loss2: 0.0911 +Epoch [164/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0929, Loss2: 0.0889 +Epoch [164/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0846, Loss2: 0.0847 +Epoch [164/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0727, Loss2: 0.0729 +Epoch [164/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1130, Loss2: 0.1095 +Epoch [164/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0974, Loss2: 0.1004 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 45.9135 % Model2 46.4944 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0753, Loss2: 0.0708 +Epoch [165/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0886, Loss2: 0.0927 +Epoch [165/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0936, Loss2: 0.0919 +Epoch [165/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1167, Loss2: 0.1115 +Epoch [165/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1052, Loss2: 0.1119 +Epoch [165/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1504, Loss2: 0.1336 +Epoch [165/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0657, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 45.4828 % Model2 45.9335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 70.3125, Loss1: 0.1032, Loss2: 0.0917 +Epoch [166/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0818, Loss2: 0.0853 +Epoch [166/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.8750, Loss1: 0.0808, Loss2: 0.0715 +Epoch [166/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0767, Loss2: 0.0796 +Epoch [166/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1145, Loss2: 0.0998 +Epoch [166/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0745, Loss2: 0.0686 +Epoch [166/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0759, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 45.5429 % Model2 46.1639 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0813, Loss2: 0.0806 +Epoch [167/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0937, Loss2: 0.0881 +Epoch [167/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0861, Loss2: 0.0799 +Epoch [167/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0767, Loss2: 0.0775 +Epoch [167/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1107, Loss2: 0.1159 +Epoch [167/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0698, Loss2: 0.0716 +Epoch [167/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0920, Loss2: 0.0909 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 45.1022 % Model2 45.8233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0760, Loss2: 0.0715 +Epoch [168/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0860, Loss2: 0.0807 +Epoch [168/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1167, Loss2: 0.1080 +Epoch [168/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0983, Loss2: 0.0993 +Epoch [168/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0972, Loss2: 0.1008 +Epoch [168/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0997, Loss2: 0.0983 +Epoch [168/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0913, Loss2: 0.0983 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 45.1322 % Model2 46.1739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0903, Loss2: 0.0909 +Epoch [169/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0797, Loss2: 0.0737 +Epoch [169/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1009, Loss2: 0.0966 +Epoch [169/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.1261, Loss2: 0.1330 +Epoch [169/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.0769, Loss2: 0.0814 +Epoch [169/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0739, Loss2: 0.0784 +Epoch [169/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0752, Loss2: 0.0753 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 45.5829 % Model2 45.2524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.1139, Loss2: 0.1095 +Epoch [170/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0893, Loss2: 0.0911 +Epoch [170/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0962, Loss2: 0.1001 +Epoch [170/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0819, Loss2: 0.0801 +Epoch [170/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.1129, Loss2: 0.1095 +Epoch [170/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0890, Loss2: 0.0860 +Epoch [170/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0809, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 45.6330 % Model2 46.0136 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0850, Loss2: 0.0832 +Epoch [171/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0865, Loss2: 0.0874 +Epoch [171/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0967, Loss2: 0.0843 +Epoch [171/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0741, Loss2: 0.0770 +Epoch [171/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0961, Loss2: 0.1048 +Epoch [171/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0834, Loss2: 0.0755 +Epoch [171/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1323, Loss2: 0.1405 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 45.5128 % Model2 45.6530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0967, Loss2: 0.0954 +Epoch [172/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.1042, Loss2: 0.0936 +Epoch [172/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0849, Loss2: 0.0881 +Epoch [172/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 65.6250, Loss1: 0.0883, Loss2: 0.0950 +Epoch [172/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0953, Loss2: 0.0952 +Epoch [172/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0610, Loss2: 0.0621 +Epoch [172/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0970, Loss2: 0.0921 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 45.8433 % Model2 46.1038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0814, Loss2: 0.0847 +Epoch [173/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1286, Loss2: 0.1229 +Epoch [173/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0980, Loss2: 0.0980 +Epoch [173/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.0825, Loss2: 0.0944 +Epoch [173/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1107, Loss2: 0.0962 +Epoch [173/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0965, Loss2: 0.0893 +Epoch [173/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0825, Loss2: 0.0788 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 45.5929 % Model2 46.2941 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1197, Loss2: 0.1205 +Epoch [174/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0931, Loss2: 0.0916 +Epoch [174/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0902, Loss2: 0.0985 +Epoch [174/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0677, Loss2: 0.0726 +Epoch [174/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0994, Loss2: 0.1021 +Epoch [174/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1121, Loss2: 0.1030 +Epoch [174/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0896, Loss2: 0.0969 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 45.3225 % Model2 45.8233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0893, Loss2: 0.0857 +Epoch [175/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0829, Loss2: 0.0824 +Epoch [175/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.1013, Loss2: 0.0903 +Epoch [175/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0851, Loss2: 0.0900 +Epoch [175/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1113, Loss2: 0.1097 +Epoch [175/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1089, Loss2: 0.1024 +Epoch [175/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.0625, Loss1: 0.0862, Loss2: 0.1016 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 45.2724 % Model2 45.5429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0990, Loss2: 0.0997 +Epoch [176/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0863, Loss2: 0.0811 +Epoch [176/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0916, Loss2: 0.0910 +Epoch [176/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0865, Loss2: 0.0792 +Epoch [176/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1381, Loss2: 0.1374 +Epoch [176/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.1076, Loss2: 0.1063 +Epoch [176/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0785, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 45.3025 % Model2 45.3726 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0772, Loss2: 0.0786 +Epoch [177/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1147, Loss2: 0.1019 +Epoch [177/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0877, Loss2: 0.0929 +Epoch [177/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0703, Loss2: 0.0714 +Epoch [177/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0841, Loss2: 0.0883 +Epoch [177/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1099, Loss2: 0.1123 +Epoch [177/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0941, Loss2: 0.0886 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 45.4327 % Model2 45.9435 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.1079, Loss2: 0.1095 +Epoch [178/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1261, Loss2: 0.1209 +Epoch [178/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0770, Loss2: 0.0716 +Epoch [178/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.1168, Loss2: 0.1034 +Epoch [178/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0840, Loss2: 0.0775 +Epoch [178/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0927, Loss2: 0.0947 +Epoch [178/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0784, Loss2: 0.0737 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 45.3726 % Model2 45.8433 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1147, Loss2: 0.1107 +Epoch [179/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0872, Loss2: 0.0862 +Epoch [179/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0808, Loss2: 0.0745 +Epoch [179/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0874, Loss2: 0.0867 +Epoch [179/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0752, Loss2: 0.0779 +Epoch [179/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0811, Loss2: 0.0819 +Epoch [179/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0909, Loss2: 0.0836 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 45.3025 % Model2 45.5529 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0977, Loss2: 0.0866 +Epoch [180/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1124, Loss2: 0.1074 +Epoch [180/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.1102, Loss2: 0.1136 +Epoch [180/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0837, Loss2: 0.0823 +Epoch [180/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0934, Loss2: 0.0932 +Epoch [180/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.1268, Loss2: 0.1398 +Epoch [180/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1498, Loss2: 0.1454 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 45.5128 % Model2 45.3225 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0654, Loss2: 0.0685 +Epoch [181/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1267, Loss2: 0.1206 +Epoch [181/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1076, Loss2: 0.1064 +Epoch [181/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.1216, Loss2: 0.1222 +Epoch [181/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.0875, Loss2: 0.0917 +Epoch [181/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0991, Loss2: 0.0977 +Epoch [181/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.1032, Loss2: 0.1086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 45.5028 % Model2 45.6030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0932, Loss2: 0.0902 +Epoch [182/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0942, Loss2: 0.0901 +Epoch [182/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0901, Loss2: 0.0867 +Epoch [182/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.1031, Loss2: 0.1006 +Epoch [182/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0809, Loss2: 0.0761 +Epoch [182/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0931, Loss2: 0.0940 +Epoch [182/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0743, Loss2: 0.0835 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 45.3025 % Model2 45.6931 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.0938, Loss1: 0.1029, Loss2: 0.0928 +Epoch [183/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0845 +Epoch [183/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0813, Loss2: 0.0884 +Epoch [183/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.1255, Loss2: 0.1413 +Epoch [183/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0864, Loss2: 0.0818 +Epoch [183/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1161, Loss2: 0.1235 +Epoch [183/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0801, Loss2: 0.0825 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 45.2724 % Model2 45.9135 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0950, Loss2: 0.1038 +Epoch [184/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0899, Loss2: 0.0867 +Epoch [184/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1376, Loss2: 0.1379 +Epoch [184/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0912, Loss2: 0.0927 +Epoch [184/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0760, Loss2: 0.0795 +Epoch [184/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0959, Loss2: 0.0888 +Epoch [184/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.1435, Loss2: 0.1301 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 45.2825 % Model2 45.8233 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1170, Loss2: 0.1120 +Epoch [185/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0952, Loss2: 0.0889 +Epoch [185/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0935, Loss2: 0.0914 +Epoch [185/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0672, Loss2: 0.0693 +Epoch [185/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0978, Loss2: 0.0989 +Epoch [185/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0917, Loss2: 0.0918 +Epoch [185/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0822, Loss2: 0.0801 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 45.4227 % Model2 45.4728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 78.1250, Loss1: 0.0976, Loss2: 0.0815 +Epoch [186/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1333, Loss2: 0.1288 +Epoch [186/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1011, Loss2: 0.1009 +Epoch [186/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0986, Loss2: 0.1060 +Epoch [186/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0823, Loss2: 0.0854 +Epoch [186/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.0625, Loss1: 0.1030, Loss2: 0.1187 +Epoch [186/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1031, Loss2: 0.1042 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 45.3526 % Model2 45.7332 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0946, Loss2: 0.1021 +Epoch [187/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1072, Loss2: 0.1118 +Epoch [187/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1178, Loss2: 0.1129 +Epoch [187/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0836, Loss2: 0.0818 +Epoch [187/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.1050, Loss2: 0.0957 +Epoch [187/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.1145, Loss2: 0.1170 +Epoch [187/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.1369, Loss2: 0.1267 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 45.0521 % Model2 45.4828 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0951, Loss2: 0.0864 +Epoch [188/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.1372, Loss2: 0.1306 +Epoch [188/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0862, Loss2: 0.0837 +Epoch [188/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1066, Loss2: 0.1053 +Epoch [188/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1014, Loss2: 0.0959 +Epoch [188/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0944, Loss2: 0.0988 +Epoch [188/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1560, Loss2: 0.1606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 44.8918 % Model2 45.7031 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0840, Loss2: 0.0837 +Epoch [189/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0828, Loss2: 0.0855 +Epoch [189/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0970, Loss2: 0.1013 +Epoch [189/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.1875, Loss1: 0.0853, Loss2: 0.0775 +Epoch [189/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1106, Loss2: 0.1078 +Epoch [189/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.1075, Loss2: 0.1086 +Epoch [189/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 67.1875, Loss1: 0.0760, Loss2: 0.0862 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 45.2624 % Model2 45.4327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0977, Loss2: 0.0943 +Epoch [190/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 61.7188, Loss1: 0.0800, Loss2: 0.0868 +Epoch [190/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0792, Loss2: 0.0803 +Epoch [190/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1074, Loss2: 0.1029 +Epoch [190/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1039, Loss2: 0.1094 +Epoch [190/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1189, Loss2: 0.1213 +Epoch [190/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1159, Loss2: 0.1169 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 45.3425 % Model2 45.6530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0991, Loss2: 0.1048 +Epoch [191/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1208, Loss2: 0.1210 +Epoch [191/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0762, Loss2: 0.0797 +Epoch [191/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0933, Loss2: 0.0925 +Epoch [191/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0915, Loss2: 0.0849 +Epoch [191/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.1030, Loss2: 0.0993 +Epoch [191/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0895, Loss2: 0.0849 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 44.9419 % Model2 45.4627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1209, Loss2: 0.1159 +Epoch [192/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0835, Loss2: 0.0846 +Epoch [192/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0999, Loss2: 0.0967 +Epoch [192/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1056, Loss2: 0.0920 +Epoch [192/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0866, Loss2: 0.0922 +Epoch [192/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0772, Loss2: 0.0820 +Epoch [192/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0966, Loss2: 0.0933 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 45.0921 % Model2 45.6831 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1229, Loss2: 0.1086 +Epoch [193/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.1026, Loss2: 0.0989 +Epoch [193/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0785, Loss2: 0.0830 +Epoch [193/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0846, Loss2: 0.0861 +Epoch [193/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.1032, Loss2: 0.1016 +Epoch [193/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0842, Loss2: 0.0848 +Epoch [193/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0951, Loss2: 0.0953 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 45.0020 % Model2 45.3526 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1296, Loss2: 0.1299 +Epoch [194/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1125, Loss2: 0.1161 +Epoch [194/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1114, Loss2: 0.1146 +Epoch [194/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0967, Loss2: 0.0960 +Epoch [194/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1097, Loss2: 0.0979 +Epoch [194/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0900, Loss2: 0.0927 +Epoch [194/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 64.8438, Loss1: 0.0915, Loss2: 0.1046 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 45.0321 % Model2 45.4026 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0949, Loss2: 0.0930 +Epoch [195/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1451, Loss2: 0.1397 +Epoch [195/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0831, Loss2: 0.0805 +Epoch [195/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1081, Loss2: 0.1108 +Epoch [195/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0950, Loss2: 0.0910 +Epoch [195/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0634, Loss2: 0.0699 +Epoch [195/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1100, Loss2: 0.1073 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 44.8918 % Model2 45.5329 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0928, Loss2: 0.0914 +Epoch [196/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0952, Loss2: 0.0927 +Epoch [196/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1145, Loss2: 0.1112 +Epoch [196/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0914, Loss2: 0.0883 +Epoch [196/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0891, Loss2: 0.0983 +Epoch [196/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 74.2188, Loss1: 0.1139, Loss2: 0.0942 +Epoch [196/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0929, Loss2: 0.0879 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 44.8317 % Model2 45.4728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1238, Loss2: 0.1235 +Epoch [197/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0932, Loss2: 0.0949 +Epoch [197/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0970, Loss2: 0.1000 +Epoch [197/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1124, Loss2: 0.1195 +Epoch [197/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.1010, Loss2: 0.1120 +Epoch [197/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1158, Loss2: 0.1102 +Epoch [197/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0819, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 44.9119 % Model2 45.3025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.1064, Loss2: 0.1135 +Epoch [198/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.1463, Loss2: 0.1411 +Epoch [198/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0934, Loss2: 0.0926 +Epoch [198/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.1123, Loss2: 0.1172 +Epoch [198/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1125, Loss2: 0.1062 +Epoch [198/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.1124, Loss2: 0.1195 +Epoch [198/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.1136, Loss2: 0.1055 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 44.9720 % Model2 45.2123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0847, Loss2: 0.0748 +Epoch [199/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1020, Loss2: 0.0991 +Epoch [199/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1175, Loss2: 0.1123 +Epoch [199/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0830, Loss2: 0.0928 +Epoch [199/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0704, Loss2: 0.0678 +Epoch [199/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0808, Loss2: 0.0873 +Epoch [199/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1200, Loss2: 0.1150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 44.9920 % Model2 45.2724 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1547, Loss2: 0.1585 +Epoch [200/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.1202, Loss2: 0.1197 +Epoch [200/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1174, Loss2: 0.1221 +Epoch [200/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1028, Loss2: 0.1088 +Epoch [200/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0964, Loss2: 0.0921 +Epoch [200/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1413, Loss2: 0.1410 +Epoch [200/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1282, Loss2: 0.1148 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 44.9018 % Model2 45.3325 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_4_6.log b/other_methods/coteaching_plus/coteaching_plus_results/out_4_6.log new file mode 100644 index 0000000..41135c3 --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_4_6.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 8.5938, Training Accuracy2: 20.3125, Loss1: 0.0176, Loss2: 0.0176 +Epoch [2/200], Iter [100/390] Training Accuracy1: 14.8438, Training Accuracy2: 16.4062, Loss1: 0.0173, Loss2: 0.0174 +Epoch [2/200], Iter [150/390] Training Accuracy1: 21.8750, Training Accuracy2: 24.2188, Loss1: 0.0166, Loss2: 0.0167 +Epoch [2/200], Iter [200/390] Training Accuracy1: 21.0938, Training Accuracy2: 25.7812, Loss1: 0.0167, Loss2: 0.0167 +Epoch [2/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0157, Loss2: 0.0157 +Epoch [2/200], Iter [300/390] Training Accuracy1: 23.4375, Training Accuracy2: 28.1250, Loss1: 0.0163, Loss2: 0.0164 +Epoch [2/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.0000, Loss1: 0.0163, Loss2: 0.0162 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 13.0809 % Model2 13.6418 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.0000, Loss1: 0.0158, Loss2: 0.0160 +Epoch [3/200], Iter [100/390] Training Accuracy1: 23.4375, Training Accuracy2: 26.5625, Loss1: 0.0160, Loss2: 0.0158 +Epoch [3/200], Iter [150/390] Training Accuracy1: 21.0938, Training Accuracy2: 19.5312, Loss1: 0.0166, Loss2: 0.0163 +Epoch [3/200], Iter [200/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.0938, Loss1: 0.0166, Loss2: 0.0165 +Epoch [3/200], Iter [250/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.1562, Loss1: 0.0152, Loss2: 0.0148 +Epoch [3/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 21.8750, Loss1: 0.0157, Loss2: 0.0159 +Epoch [3/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 24.2188, Loss1: 0.0161, Loss2: 0.0158 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 13.5216 % Model2 14.4631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.0000, Loss1: 0.0167, Loss2: 0.0168 +Epoch [4/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0153, Loss2: 0.0154 +Epoch [4/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0154, Loss2: 0.0150 +Epoch [4/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 27.3438, Loss1: 0.0152, Loss2: 0.0152 +Epoch [4/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 36.7188, Loss1: 0.0155, Loss2: 0.0151 +Epoch [4/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.7812, Loss1: 0.0156, Loss2: 0.0152 +Epoch [4/200], Iter [350/390] Training Accuracy1: 21.8750, Training Accuracy2: 22.6562, Loss1: 0.0167, Loss2: 0.0162 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 17.6983 % Model2 17.4179 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 28.1250, Loss1: 0.0155, Loss2: 0.0153 +Epoch [5/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 30.4688, Loss1: 0.0168, Loss2: 0.0163 +Epoch [5/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 33.5938, Loss1: 0.0147, Loss2: 0.0143 +Epoch [5/200], Iter [200/390] Training Accuracy1: 25.0000, Training Accuracy2: 28.1250, Loss1: 0.0156, Loss2: 0.0151 +Epoch [5/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0149, Loss2: 0.0146 +Epoch [5/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 25.7812, Loss1: 0.0168, Loss2: 0.0168 +Epoch [5/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 30.4688, Loss1: 0.0155, Loss2: 0.0153 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 17.0172 % Model2 17.4679 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 31.2500, Loss1: 0.0146, Loss2: 0.0140 +Epoch [6/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0158, Loss2: 0.0147 +Epoch [6/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0145, Loss2: 0.0144 +Epoch [6/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 25.0000, Loss1: 0.0167, Loss2: 0.0164 +Epoch [6/200], Iter [250/390] Training Accuracy1: 25.7812, Training Accuracy2: 32.0312, Loss1: 0.0157, Loss2: 0.0149 +Epoch [6/200], Iter [300/390] Training Accuracy1: 23.4375, Training Accuracy2: 24.2188, Loss1: 0.0169, Loss2: 0.0167 +Epoch [6/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0161, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 21.5144 % Model2 22.2556 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 26.5625, Training Accuracy2: 25.7812, Loss1: 0.0171, Loss2: 0.0176 +Epoch [7/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0151, Loss2: 0.0148 +Epoch [7/200], Iter [150/390] Training Accuracy1: 24.2188, Training Accuracy2: 22.6562, Loss1: 0.0167, Loss2: 0.0163 +Epoch [7/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 22.6562, Loss1: 0.0161, Loss2: 0.0158 +Epoch [7/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0154, Loss2: 0.0149 +Epoch [7/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0148, Loss2: 0.0145 +Epoch [7/200], Iter [350/390] Training Accuracy1: 23.4375, Training Accuracy2: 31.2500, Loss1: 0.0163, Loss2: 0.0155 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 20.3025 % Model2 20.1422 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0158, Loss2: 0.0152 +Epoch [8/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 22.6562, Loss1: 0.0174, Loss2: 0.0175 +Epoch [8/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 25.0000, Loss1: 0.0168, Loss2: 0.0155 +Epoch [8/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0144, Loss2: 0.0138 +Epoch [8/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0157, Loss2: 0.0153 +Epoch [8/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0155, Loss2: 0.0151 +Epoch [8/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0147, Loss2: 0.0145 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 19.1707 % Model2 19.8818 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0149 +Epoch [9/200], Iter [100/390] Training Accuracy1: 21.8750, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0152 +Epoch [9/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 32.0312, Loss1: 0.0164, Loss2: 0.0154 +Epoch [9/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0145, Loss2: 0.0147 +Epoch [9/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0137, Loss2: 0.0141 +Epoch [9/200], Iter [300/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0147 +Epoch [9/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0150, Loss2: 0.0147 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 18.2893 % Model2 18.0288 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0143, Loss2: 0.0137 +Epoch [10/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0145, Loss2: 0.0140 +Epoch [10/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.9375, Loss1: 0.0149, Loss2: 0.0144 +Epoch [10/200], Iter [200/390] Training Accuracy1: 27.3438, Training Accuracy2: 26.5625, Loss1: 0.0166, Loss2: 0.0160 +Epoch [10/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.9062, Loss1: 0.0147, Loss2: 0.0149 +Epoch [10/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0139, Loss2: 0.0140 +Epoch [10/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 32.8125, Loss1: 0.0135, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 22.4760 % Model2 21.7648 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0159, Loss2: 0.0150 +Epoch [11/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 28.9062, Loss1: 0.0144, Loss2: 0.0141 +Epoch [11/200], Iter [150/390] Training Accuracy1: 23.4375, Training Accuracy2: 28.1250, Loss1: 0.0160, Loss2: 0.0159 +Epoch [11/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 31.2500, Loss1: 0.0158, Loss2: 0.0147 +Epoch [11/200], Iter [250/390] Training Accuracy1: 27.3438, Training Accuracy2: 34.3750, Loss1: 0.0149, Loss2: 0.0142 +Epoch [11/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 31.2500, Loss1: 0.0151, Loss2: 0.0145 +Epoch [11/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0142, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 21.3542 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0136, Loss2: 0.0134 +Epoch [12/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0136, Loss2: 0.0130 +Epoch [12/200], Iter [150/390] Training Accuracy1: 27.3438, Training Accuracy2: 28.9062, Loss1: 0.0147, Loss2: 0.0147 +Epoch [12/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 33.5938, Loss1: 0.0147, Loss2: 0.0141 +Epoch [12/200], Iter [250/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0144, Loss2: 0.0139 +Epoch [12/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0140, Loss2: 0.0131 +Epoch [12/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0141, Loss2: 0.0140 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 20.6530 % Model2 19.7115 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 29.6875, Loss1: 0.0165, Loss2: 0.0159 +Epoch [13/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0144, Loss2: 0.0153 +Epoch [13/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0148, Loss2: 0.0142 +Epoch [13/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 34.3750, Loss1: 0.0148, Loss2: 0.0154 +Epoch [13/200], Iter [250/390] Training Accuracy1: 23.4375, Training Accuracy2: 25.0000, Loss1: 0.0166, Loss2: 0.0163 +Epoch [13/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 37.5000, Loss1: 0.0140, Loss2: 0.0134 +Epoch [13/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0135, Loss2: 0.0135 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 20.9936 % Model2 22.2556 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 36.7188, Loss1: 0.0137, Loss2: 0.0128 +Epoch [14/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.9062, Loss1: 0.0140, Loss2: 0.0145 +Epoch [14/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0139, Loss2: 0.0132 +Epoch [14/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 28.1250, Loss1: 0.0150, Loss2: 0.0148 +Epoch [14/200], Iter [250/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0146, Loss2: 0.0144 +Epoch [14/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0135, Loss2: 0.0132 +Epoch [14/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0147, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 21.4443 % Model2 22.2957 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0142, Loss2: 0.0136 +Epoch [15/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 26.5625, Loss1: 0.0151, Loss2: 0.0149 +Epoch [15/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 36.7188, Loss1: 0.0139, Loss2: 0.0137 +Epoch [15/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 30.4688, Loss1: 0.0143, Loss2: 0.0145 +Epoch [15/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0135, Loss2: 0.0126 +Epoch [15/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0153, Loss2: 0.0146 +Epoch [15/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 41.4062, Loss1: 0.0131, Loss2: 0.0122 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 24.1286 % Model2 23.9583 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0137 +Epoch [16/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0142, Loss2: 0.0138 +Epoch [16/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0132, Loss2: 0.0130 +Epoch [16/200], Iter [200/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0147, Loss2: 0.0145 +Epoch [16/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0145 +Epoch [16/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0136, Loss2: 0.0125 +Epoch [16/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0140, Loss2: 0.0136 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 23.6979 % Model2 22.8766 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0147, Loss2: 0.0138 +Epoch [17/200], Iter [100/390] Training Accuracy1: 29.6875, Training Accuracy2: 29.6875, Loss1: 0.0147, Loss2: 0.0146 +Epoch [17/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0137, Loss2: 0.0135 +Epoch [17/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0136 +Epoch [17/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0129, Loss2: 0.0135 +Epoch [17/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 34.3750, Loss1: 0.0140, Loss2: 0.0134 +Epoch [17/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0141, Loss2: 0.0139 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 21.5044 % Model2 22.6663 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0128, Loss2: 0.0123 +Epoch [18/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 30.4688, Loss1: 0.0142, Loss2: 0.0147 +Epoch [18/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.8125, Loss1: 0.0148, Loss2: 0.0144 +Epoch [18/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0138, Loss2: 0.0137 +Epoch [18/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0129, Loss2: 0.0127 +Epoch [18/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 39.8438, Loss1: 0.0126, Loss2: 0.0120 +Epoch [18/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0142, Loss2: 0.0134 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 23.0569 % Model2 24.2388 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 41.4062, Loss1: 0.0143, Loss2: 0.0127 +Epoch [19/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 34.3750, Loss1: 0.0148, Loss2: 0.0141 +Epoch [19/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.0625, Loss1: 0.0132, Loss2: 0.0124 +Epoch [19/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0147, Loss2: 0.0148 +Epoch [19/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0146, Loss2: 0.0145 +Epoch [19/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 28.9062, Loss1: 0.0156, Loss2: 0.0156 +Epoch [19/200], Iter [350/390] Training Accuracy1: 24.2188, Training Accuracy2: 32.8125, Loss1: 0.0154, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 20.1723 % Model2 20.9736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0128, Loss2: 0.0129 +Epoch [20/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0137, Loss2: 0.0138 +Epoch [20/200], Iter [150/390] Training Accuracy1: 32.0312, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0122 +Epoch [20/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0154, Loss2: 0.0157 +Epoch [20/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0159, Loss2: 0.0152 +Epoch [20/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 29.6875, Loss1: 0.0128, Loss2: 0.0129 +Epoch [20/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 37.5000, Loss1: 0.0144, Loss2: 0.0132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 22.5761 % Model2 22.2456 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0521, Loss2: 0.0501 +Epoch [21/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 34.3750, Loss1: 0.0540, Loss2: 0.0526 +Epoch [21/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 26.5625, Loss1: 0.0493, Loss2: 0.0491 +Epoch [21/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0612, Loss2: 0.0590 +Epoch [21/200], Iter [250/390] Training Accuracy1: 22.6562, Training Accuracy2: 26.5625, Loss1: 0.0477, Loss2: 0.0467 +Epoch [21/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0495, Loss2: 0.0493 +Epoch [21/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0689, Loss2: 0.0680 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 22.5260 % Model2 22.0653 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0528, Loss2: 0.0529 +Epoch [22/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 46.0938, Loss1: 0.0713, Loss2: 0.0668 +Epoch [22/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0625, Loss2: 0.0629 +Epoch [22/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 35.9375, Loss1: 0.0529, Loss2: 0.0505 +Epoch [22/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0663, Loss2: 0.0672 +Epoch [22/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0914, Loss2: 0.0891 +Epoch [22/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0483, Loss2: 0.0478 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 23.1070 % Model2 21.3742 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.9375, Loss1: 0.0615, Loss2: 0.0596 +Epoch [23/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0692, Loss2: 0.0668 +Epoch [23/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0514, Loss2: 0.0513 +Epoch [23/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 29.6875, Loss1: 0.0590, Loss2: 0.0590 +Epoch [23/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0699, Loss2: 0.0685 +Epoch [23/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.1562, Loss1: 0.0758, Loss2: 0.0737 +Epoch [23/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0563, Loss2: 0.0548 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 22.7264 % Model2 22.2756 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0567, Loss2: 0.0568 +Epoch [24/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 35.9375, Loss1: 0.0514, Loss2: 0.0505 +Epoch [24/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 32.0312, Loss1: 0.0510, Loss2: 0.0502 +Epoch [24/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0466, Loss2: 0.0466 +Epoch [24/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0598, Loss2: 0.0565 +Epoch [24/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0519, Loss2: 0.0516 +Epoch [24/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0673, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 21.2841 % Model2 20.4627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 30.4688, Loss1: 0.0533, Loss2: 0.0551 +Epoch [25/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 35.9375, Loss1: 0.0605, Loss2: 0.0612 +Epoch [25/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 30.4688, Loss1: 0.0473, Loss2: 0.0480 +Epoch [25/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0571, Loss2: 0.0566 +Epoch [25/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 29.6875, Loss1: 0.0442, Loss2: 0.0456 +Epoch [25/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 34.3750, Loss1: 0.0555, Loss2: 0.0557 +Epoch [25/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 43.7500, Loss1: 0.0606, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 19.5613 % Model2 22.6763 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.8125, Loss1: 0.0513, Loss2: 0.0521 +Epoch [26/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 36.7188, Loss1: 0.0509, Loss2: 0.0510 +Epoch [26/200], Iter [150/390] Training Accuracy1: 25.7812, Training Accuracy2: 31.2500, Loss1: 0.0483, Loss2: 0.0468 +Epoch [26/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 31.2500, Loss1: 0.0560, Loss2: 0.0565 +Epoch [26/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0522, Loss2: 0.0531 +Epoch [26/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0605, Loss2: 0.0608 +Epoch [26/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 42.1875, Loss1: 0.0571, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 22.9167 % Model2 23.4876 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 34.3750, Loss1: 0.0636, Loss2: 0.0635 +Epoch [27/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0494, Loss2: 0.0492 +Epoch [27/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0598, Loss2: 0.0595 +Epoch [27/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 32.0312, Loss1: 0.0481, Loss2: 0.0473 +Epoch [27/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0731, Loss2: 0.0719 +Epoch [27/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0551, Loss2: 0.0546 +Epoch [27/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0612, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 22.2556 % Model2 23.2372 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0512, Loss2: 0.0485 +Epoch [28/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 34.3750, Loss1: 0.0519, Loss2: 0.0497 +Epoch [28/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0703, Loss2: 0.0710 +Epoch [28/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 40.6250, Loss1: 0.0566, Loss2: 0.0541 +Epoch [28/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0685, Loss2: 0.0680 +Epoch [28/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0584, Loss2: 0.0585 +Epoch [28/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0485, Loss2: 0.0486 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 21.6546 % Model2 22.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0480, Loss2: 0.0468 +Epoch [29/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0553, Loss2: 0.0534 +Epoch [29/200], Iter [150/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0507, Loss2: 0.0507 +Epoch [29/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0520, Loss2: 0.0507 +Epoch [29/200], Iter [250/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.9375, Loss1: 0.0466, Loss2: 0.0451 +Epoch [29/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0512, Loss2: 0.0491 +Epoch [29/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 34.3750, Loss1: 0.0578, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 22.2356 % Model2 22.7464 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0515, Loss2: 0.0519 +Epoch [30/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 33.5938, Loss1: 0.0620, Loss2: 0.0602 +Epoch [30/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 36.7188, Loss1: 0.0532, Loss2: 0.0534 +Epoch [30/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0434, Loss2: 0.0430 +Epoch [30/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0523, Loss2: 0.0508 +Epoch [30/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0554, Loss2: 0.0570 +Epoch [30/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0486, Loss2: 0.0471 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 23.0970 % Model2 23.0869 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.9375, Loss1: 0.0518, Loss2: 0.0492 +Epoch [31/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.1562, Loss1: 0.0526, Loss2: 0.0504 +Epoch [31/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0468, Loss2: 0.0465 +Epoch [31/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 41.4062, Loss1: 0.0501, Loss2: 0.0490 +Epoch [31/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 31.2500, Loss1: 0.0424, Loss2: 0.0414 +Epoch [31/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0486, Loss2: 0.0491 +Epoch [31/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.8438, Loss1: 0.0485, Loss2: 0.0484 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 23.6779 % Model2 23.8582 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0512, Loss2: 0.0518 +Epoch [32/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.1562, Loss1: 0.0577, Loss2: 0.0578 +Epoch [32/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0524, Loss2: 0.0504 +Epoch [32/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0455, Loss2: 0.0446 +Epoch [32/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 35.1562, Loss1: 0.0460, Loss2: 0.0462 +Epoch [32/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0489, Loss2: 0.0492 +Epoch [32/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0490, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 22.5962 % Model2 22.5461 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0462, Loss2: 0.0472 +Epoch [33/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0558, Loss2: 0.0551 +Epoch [33/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0545, Loss2: 0.0520 +Epoch [33/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0491, Loss2: 0.0478 +Epoch [33/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0500, Loss2: 0.0479 +Epoch [33/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.0312, Loss1: 0.0435, Loss2: 0.0448 +Epoch [33/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0489, Loss2: 0.0467 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 23.6178 % Model2 22.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0605, Loss2: 0.0588 +Epoch [34/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 38.2812, Loss1: 0.0455, Loss2: 0.0429 +Epoch [34/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0543, Loss2: 0.0524 +Epoch [34/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0598, Loss2: 0.0609 +Epoch [34/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0639, Loss2: 0.0616 +Epoch [34/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0515, Loss2: 0.0499 +Epoch [34/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 42.1875, Loss1: 0.0419, Loss2: 0.0400 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 23.1270 % Model2 24.0084 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0661, Loss2: 0.0654 +Epoch [35/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0521, Loss2: 0.0488 +Epoch [35/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0494, Loss2: 0.0470 +Epoch [35/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 34.3750, Loss1: 0.0428, Loss2: 0.0420 +Epoch [35/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.8125, Loss1: 0.0467, Loss2: 0.0469 +Epoch [35/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0437, Loss2: 0.0435 +Epoch [35/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0604, Loss2: 0.0591 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 22.9567 % Model2 22.1755 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0461, Loss2: 0.0457 +Epoch [36/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0473, Loss2: 0.0488 +Epoch [36/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 45.3125, Loss1: 0.0579, Loss2: 0.0533 +Epoch [36/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 47.6562, Loss1: 0.0502, Loss2: 0.0461 +Epoch [36/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0543, Loss2: 0.0534 +Epoch [36/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 43.7500, Loss1: 0.0556, Loss2: 0.0533 +Epoch [36/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0496, Loss2: 0.0483 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 22.8365 % Model2 22.8966 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0557, Loss2: 0.0529 +Epoch [37/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.8438, Loss1: 0.0647, Loss2: 0.0682 +Epoch [37/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 46.0938, Loss1: 0.0475, Loss2: 0.0443 +Epoch [37/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0506, Loss2: 0.0516 +Epoch [37/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.1875, Loss1: 0.0512, Loss2: 0.0512 +Epoch [37/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.8125, Loss1: 0.0525, Loss2: 0.0546 +Epoch [37/200], Iter [350/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0555, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 22.5962 % Model2 24.0184 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 42.1875, Loss1: 0.0498, Loss2: 0.0468 +Epoch [38/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0464, Loss2: 0.0454 +Epoch [38/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 32.8125, Loss1: 0.0409, Loss2: 0.0421 +Epoch [38/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0485, Loss2: 0.0463 +Epoch [38/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0477, Loss2: 0.0465 +Epoch [38/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0498, Loss2: 0.0475 +Epoch [38/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.9688, Loss1: 0.0591, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 20.2524 % Model2 22.1254 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0489, Loss2: 0.0477 +Epoch [39/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.9688, Loss1: 0.0505, Loss2: 0.0484 +Epoch [39/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 37.5000, Loss1: 0.0497, Loss2: 0.0480 +Epoch [39/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0492, Loss2: 0.0499 +Epoch [39/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0424, Loss2: 0.0417 +Epoch [39/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0437, Loss2: 0.0429 +Epoch [39/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.1875, Loss1: 0.0446, Loss2: 0.0436 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 24.0585 % Model2 24.4692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0607, Loss2: 0.0602 +Epoch [40/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0461, Loss2: 0.0434 +Epoch [40/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0563, Loss2: 0.0598 +Epoch [40/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 42.1875, Loss1: 0.0454, Loss2: 0.0430 +Epoch [40/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0524, Loss2: 0.0501 +Epoch [40/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 30.4688, Loss1: 0.0410, Loss2: 0.0429 +Epoch [40/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0541, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 21.9952 % Model2 22.9167 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0541, Loss2: 0.0543 +Epoch [41/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0494, Loss2: 0.0484 +Epoch [41/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0575, Loss2: 0.0577 +Epoch [41/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.8438, Loss1: 0.0519, Loss2: 0.0539 +Epoch [41/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0462, Loss2: 0.0447 +Epoch [41/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0547, Loss2: 0.0519 +Epoch [41/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0433, Loss2: 0.0434 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 23.3974 % Model2 22.5361 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0448, Loss2: 0.0455 +Epoch [42/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0445, Loss2: 0.0437 +Epoch [42/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0427, Loss2: 0.0409 +Epoch [42/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0474, Loss2: 0.0466 +Epoch [42/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0489, Loss2: 0.0495 +Epoch [42/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0455, Loss2: 0.0451 +Epoch [42/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 38.2812, Loss1: 0.0422, Loss2: 0.0403 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 22.3257 % Model2 22.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0424, Loss2: 0.0426 +Epoch [43/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0462, Loss2: 0.0475 +Epoch [43/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0604, Loss2: 0.0609 +Epoch [43/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0411, Loss2: 0.0424 +Epoch [43/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 50.7812, Loss1: 0.0462, Loss2: 0.0438 +Epoch [43/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.1562, Loss1: 0.0437, Loss2: 0.0417 +Epoch [43/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0604, Loss2: 0.0577 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 21.9551 % Model2 22.4058 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0489, Loss2: 0.0461 +Epoch [44/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 36.7188, Loss1: 0.0468, Loss2: 0.0494 +Epoch [44/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0517, Loss2: 0.0499 +Epoch [44/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0483, Loss2: 0.0478 +Epoch [44/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0501, Loss2: 0.0499 +Epoch [44/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 45.3125, Loss1: 0.0484, Loss2: 0.0468 +Epoch [44/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0498, Loss2: 0.0490 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 23.2472 % Model2 23.5677 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0564, Loss2: 0.0565 +Epoch [45/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0566, Loss2: 0.0557 +Epoch [45/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0510, Loss2: 0.0487 +Epoch [45/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0544, Loss2: 0.0549 +Epoch [45/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0649, Loss2: 0.0637 +Epoch [45/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.8438, Loss1: 0.0516, Loss2: 0.0496 +Epoch [45/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0418, Loss2: 0.0407 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 22.8966 % Model2 23.9683 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 48.4375, Loss1: 0.0466, Loss2: 0.0438 +Epoch [46/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0590, Loss2: 0.0563 +Epoch [46/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0448, Loss2: 0.0439 +Epoch [46/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0538, Loss2: 0.0517 +Epoch [46/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 47.6562, Loss1: 0.0424, Loss2: 0.0394 +Epoch [46/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0563, Loss2: 0.0537 +Epoch [46/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 47.6562, Loss1: 0.0493, Loss2: 0.0455 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 23.9283 % Model2 23.0669 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0502, Loss2: 0.0501 +Epoch [47/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.9688, Loss1: 0.0452, Loss2: 0.0442 +Epoch [47/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0546, Loss2: 0.0526 +Epoch [47/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0445, Loss2: 0.0445 +Epoch [47/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0448, Loss2: 0.0438 +Epoch [47/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0491, Loss2: 0.0500 +Epoch [47/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0519, Loss2: 0.0544 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 21.8550 % Model2 21.4643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0509, Loss2: 0.0536 +Epoch [48/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0471, Loss2: 0.0471 +Epoch [48/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 37.5000, Loss1: 0.0523, Loss2: 0.0544 +Epoch [48/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 44.5312, Loss1: 0.0560, Loss2: 0.0525 +Epoch [48/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0534, Loss2: 0.0537 +Epoch [48/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 44.5312, Loss1: 0.0479, Loss2: 0.0447 +Epoch [48/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0447, Loss2: 0.0447 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 21.6146 % Model2 22.6663 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0666, Loss2: 0.0634 +Epoch [49/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0597, Loss2: 0.0599 +Epoch [49/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0424, Loss2: 0.0424 +Epoch [49/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0543, Loss2: 0.0561 +Epoch [49/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0514, Loss2: 0.0500 +Epoch [49/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 32.0312, Loss1: 0.0402, Loss2: 0.0423 +Epoch [49/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0437, Loss2: 0.0428 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 22.5561 % Model2 23.4375 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0490, Loss2: 0.0485 +Epoch [50/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0501, Loss2: 0.0498 +Epoch [50/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.8750, Loss1: 0.0518, Loss2: 0.0491 +Epoch [50/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 49.2188, Loss1: 0.0534, Loss2: 0.0501 +Epoch [50/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0489, Loss2: 0.0491 +Epoch [50/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0406, Loss2: 0.0385 +Epoch [50/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.8750, Loss1: 0.0579, Loss2: 0.0600 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 22.9567 % Model2 23.7480 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0522, Loss2: 0.0533 +Epoch [51/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0484, Loss2: 0.0484 +Epoch [51/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0519, Loss2: 0.0499 +Epoch [51/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0467, Loss2: 0.0462 +Epoch [51/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0532, Loss2: 0.0519 +Epoch [51/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 32.0312, Loss1: 0.0411, Loss2: 0.0422 +Epoch [51/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0441, Loss2: 0.0441 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 20.9135 % Model2 22.6663 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.8438, Loss1: 0.0417, Loss2: 0.0429 +Epoch [52/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0491, Loss2: 0.0504 +Epoch [52/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0465, Loss2: 0.0469 +Epoch [52/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0545, Loss2: 0.0525 +Epoch [52/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0525, Loss2: 0.0506 +Epoch [52/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0461, Loss2: 0.0440 +Epoch [52/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 44.5312, Loss1: 0.0576, Loss2: 0.0624 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 23.1771 % Model2 22.7163 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0568, Loss2: 0.0540 +Epoch [53/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0540, Loss2: 0.0532 +Epoch [53/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0440, Loss2: 0.0425 +Epoch [53/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0554, Loss2: 0.0556 +Epoch [53/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0468, Loss2: 0.0469 +Epoch [53/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0452, Loss2: 0.0446 +Epoch [53/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0473, Loss2: 0.0459 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 23.2873 % Model2 22.7063 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0566, Loss2: 0.0534 +Epoch [54/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0605, Loss2: 0.0566 +Epoch [54/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0508, Loss2: 0.0482 +Epoch [54/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0453, Loss2: 0.0448 +Epoch [54/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0548, Loss2: 0.0542 +Epoch [54/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 53.9062, Loss1: 0.0461, Loss2: 0.0424 +Epoch [54/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0501, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 22.5661 % Model2 22.7664 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0594, Loss2: 0.0567 +Epoch [55/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0492, Loss2: 0.0470 +Epoch [55/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0501, Loss2: 0.0481 +Epoch [55/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0489, Loss2: 0.0481 +Epoch [55/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 50.0000, Loss1: 0.0446, Loss2: 0.0416 +Epoch [55/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0448, Loss2: 0.0430 +Epoch [55/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 43.7500, Loss1: 0.0536, Loss2: 0.0549 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.7364 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.8750, Loss1: 0.0533, Loss2: 0.0489 +Epoch [56/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.0938, Loss1: 0.0441, Loss2: 0.0450 +Epoch [56/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0520, Loss2: 0.0490 +Epoch [56/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0472, Loss2: 0.0470 +Epoch [56/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0483, Loss2: 0.0475 +Epoch [56/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0552, Loss2: 0.0525 +Epoch [56/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0417, Loss2: 0.0423 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 22.2155 % Model2 22.8165 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 59.3750, Loss1: 0.0557, Loss2: 0.0509 +Epoch [57/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0652, Loss2: 0.0623 +Epoch [57/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0444, Loss2: 0.0430 +Epoch [57/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0572, Loss2: 0.0586 +Epoch [57/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0493, Loss2: 0.0508 +Epoch [57/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0450, Loss2: 0.0444 +Epoch [57/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0458, Loss2: 0.0477 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 22.5361 % Model2 23.8782 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0451, Loss2: 0.0433 +Epoch [58/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0509, Loss2: 0.0523 +Epoch [58/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 49.2188, Loss1: 0.0508, Loss2: 0.0479 +Epoch [58/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0488, Loss2: 0.0476 +Epoch [58/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 42.9688, Loss1: 0.0434, Loss2: 0.0461 +Epoch [58/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 35.1562, Loss1: 0.0414, Loss2: 0.0463 +Epoch [58/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0537, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 22.3357 % Model2 23.0168 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0513, Loss2: 0.0495 +Epoch [59/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0473, Loss2: 0.0465 +Epoch [59/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0559, Loss2: 0.0547 +Epoch [59/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 41.4062, Loss1: 0.0392, Loss2: 0.0387 +Epoch [59/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0624, Loss2: 0.0605 +Epoch [59/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0448, Loss2: 0.0468 +Epoch [59/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 40.6250, Loss1: 0.0474, Loss2: 0.0506 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 21.8349 % Model2 21.6246 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0415, Loss2: 0.0402 +Epoch [60/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0489, Loss2: 0.0509 +Epoch [60/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0447, Loss2: 0.0432 +Epoch [60/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.7812, Loss1: 0.0462, Loss2: 0.0437 +Epoch [60/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0444, Loss2: 0.0462 +Epoch [60/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0425, Loss2: 0.0427 +Epoch [60/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0518, Loss2: 0.0516 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 22.6963 % Model2 23.1070 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0540, Loss2: 0.0529 +Epoch [61/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0619, Loss2: 0.0601 +Epoch [61/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.0625, Loss1: 0.0503, Loss2: 0.0524 +Epoch [61/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0467, Loss2: 0.0469 +Epoch [61/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0495, Loss2: 0.0479 +Epoch [61/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0528, Loss2: 0.0546 +Epoch [61/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.0000, Loss1: 0.0590, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 24.2688 % Model2 23.8582 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.9688, Loss1: 0.0505, Loss2: 0.0535 +Epoch [62/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0448, Loss2: 0.0438 +Epoch [62/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0476, Loss2: 0.0497 +Epoch [62/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 41.4062, Loss1: 0.0468, Loss2: 0.0479 +Epoch [62/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0568, Loss2: 0.0543 +Epoch [62/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0504, Loss2: 0.0514 +Epoch [62/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0438, Loss2: 0.0440 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 23.1270 % Model2 23.4475 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.8750, Loss1: 0.0500, Loss2: 0.0520 +Epoch [63/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0606, Loss2: 0.0599 +Epoch [63/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 39.8438, Loss1: 0.0468, Loss2: 0.0476 +Epoch [63/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0532, Loss2: 0.0540 +Epoch [63/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0478, Loss2: 0.0462 +Epoch [63/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0478, Loss2: 0.0472 +Epoch [63/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0578, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 22.5361 % Model2 22.9768 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0508, Loss2: 0.0479 +Epoch [64/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.9062, Loss1: 0.0619, Loss2: 0.0574 +Epoch [64/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0533, Loss2: 0.0568 +Epoch [64/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0500, Loss2: 0.0501 +Epoch [64/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0519, Loss2: 0.0510 +Epoch [64/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0488, Loss2: 0.0490 +Epoch [64/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0606, Loss2: 0.0638 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 22.4259 % Model2 22.2556 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0504, Loss2: 0.0481 +Epoch [65/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0470, Loss2: 0.0464 +Epoch [65/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 41.4062, Loss1: 0.0530, Loss2: 0.0528 +Epoch [65/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0493, Loss2: 0.0467 +Epoch [65/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0441, Loss2: 0.0433 +Epoch [65/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0461, Loss2: 0.0484 +Epoch [65/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0531, Loss2: 0.0506 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 22.0853 % Model2 22.5861 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0602, Loss2: 0.0645 +Epoch [66/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0588, Loss2: 0.0581 +Epoch [66/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0496, Loss2: 0.0491 +Epoch [66/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0545, Loss2: 0.0524 +Epoch [66/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0489, Loss2: 0.0497 +Epoch [66/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0427, Loss2: 0.0424 +Epoch [66/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.1250, Loss1: 0.0555, Loss2: 0.0527 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 23.3373 % Model2 22.3157 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0520, Loss2: 0.0517 +Epoch [67/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0427, Loss2: 0.0421 +Epoch [67/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0524, Loss2: 0.0527 +Epoch [67/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0498, Loss2: 0.0486 +Epoch [67/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0532, Loss2: 0.0536 +Epoch [67/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 40.6250, Loss1: 0.0477, Loss2: 0.0501 +Epoch [67/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0476, Loss2: 0.0463 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 22.8766 % Model2 23.2272 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0577, Loss2: 0.0551 +Epoch [68/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0474, Loss2: 0.0478 +Epoch [68/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0490, Loss2: 0.0471 +Epoch [68/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0422, Loss2: 0.0414 +Epoch [68/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0487, Loss2: 0.0463 +Epoch [68/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.1250, Loss1: 0.0509, Loss2: 0.0482 +Epoch [68/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0521, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 22.9267 % Model2 23.0669 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0463, Loss2: 0.0468 +Epoch [69/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0533, Loss2: 0.0533 +Epoch [69/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0499, Loss2: 0.0496 +Epoch [69/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0453, Loss2: 0.0449 +Epoch [69/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0466, Loss2: 0.0479 +Epoch [69/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0686, Loss2: 0.0721 +Epoch [69/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0443, Loss2: 0.0429 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.2857 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0477, Loss2: 0.0457 +Epoch [70/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0534, Loss2: 0.0541 +Epoch [70/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0561, Loss2: 0.0561 +Epoch [70/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0534, Loss2: 0.0498 +Epoch [70/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0433, Loss2: 0.0447 +Epoch [70/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 45.3125, Loss1: 0.0460, Loss2: 0.0498 +Epoch [70/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0601, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 22.7865 % Model2 22.4659 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0622, Loss2: 0.0603 +Epoch [71/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0429, Loss2: 0.0401 +Epoch [71/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0516, Loss2: 0.0537 +Epoch [71/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0485, Loss2: 0.0499 +Epoch [71/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0529, Loss2: 0.0551 +Epoch [71/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0454, Loss2: 0.0451 +Epoch [71/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0465, Loss2: 0.0455 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 22.7063 % Model2 21.9952 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0533, Loss2: 0.0524 +Epoch [72/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0514, Loss2: 0.0499 +Epoch [72/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0474, Loss2: 0.0476 +Epoch [72/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 52.3438, Loss1: 0.0537, Loss2: 0.0521 +Epoch [72/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0418, Loss2: 0.0427 +Epoch [72/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0572, Loss2: 0.0540 +Epoch [72/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0528, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 21.6647 % Model2 22.7965 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0555, Loss2: 0.0586 +Epoch [73/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.9688, Loss1: 0.0506, Loss2: 0.0534 +Epoch [73/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0531, Loss2: 0.0516 +Epoch [73/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0531, Loss2: 0.0527 +Epoch [73/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0489, Loss2: 0.0495 +Epoch [73/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0483, Loss2: 0.0472 +Epoch [73/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0527, Loss2: 0.0519 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 22.3157 % Model2 22.5461 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0527, Loss2: 0.0551 +Epoch [74/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0539, Loss2: 0.0522 +Epoch [74/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0601, Loss2: 0.0609 +Epoch [74/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0515, Loss2: 0.0519 +Epoch [74/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0475, Loss2: 0.0462 +Epoch [74/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 43.7500, Loss1: 0.0509, Loss2: 0.0545 +Epoch [74/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0445, Loss2: 0.0433 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 22.5661 % Model2 22.9267 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0534, Loss2: 0.0533 +Epoch [75/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0504, Loss2: 0.0491 +Epoch [75/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0483, Loss2: 0.0472 +Epoch [75/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0525, Loss2: 0.0529 +Epoch [75/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0487, Loss2: 0.0493 +Epoch [75/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0513, Loss2: 0.0521 +Epoch [75/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0584, Loss2: 0.0553 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 21.9651 % Model2 22.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0531, Loss2: 0.0496 +Epoch [76/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0507, Loss2: 0.0523 +Epoch [76/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0548, Loss2: 0.0520 +Epoch [76/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0518, Loss2: 0.0522 +Epoch [76/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0547, Loss2: 0.0553 +Epoch [76/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 46.0938, Loss1: 0.0539, Loss2: 0.0511 +Epoch [76/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0500, Loss2: 0.0476 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 22.1955 % Model2 22.3057 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0492, Loss2: 0.0505 +Epoch [77/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0545, Loss2: 0.0561 +Epoch [77/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0484, Loss2: 0.0491 +Epoch [77/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0472, Loss2: 0.0493 +Epoch [77/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0473, Loss2: 0.0491 +Epoch [77/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0528 +Epoch [77/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0764, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 22.0052 % Model2 22.6963 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0510, Loss2: 0.0498 +Epoch [78/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0497, Loss2: 0.0501 +Epoch [78/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0437, Loss2: 0.0469 +Epoch [78/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0474, Loss2: 0.0474 +Epoch [78/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0558, Loss2: 0.0513 +Epoch [78/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0548, Loss2: 0.0534 +Epoch [78/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0533, Loss2: 0.0514 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 21.8149 % Model2 21.9050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0486, Loss2: 0.0508 +Epoch [79/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0526, Loss2: 0.0493 +Epoch [79/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0430, Loss2: 0.0445 +Epoch [79/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0472, Loss2: 0.0475 +Epoch [79/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0620, Loss2: 0.0594 +Epoch [79/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0528, Loss2: 0.0554 +Epoch [79/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0499, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 22.4960 % Model2 21.5545 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0523, Loss2: 0.0548 +Epoch [80/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0529, Loss2: 0.0551 +Epoch [80/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0480, Loss2: 0.0445 +Epoch [80/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.9062, Loss1: 0.0530, Loss2: 0.0496 +Epoch [80/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0458, Loss2: 0.0451 +Epoch [80/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0520, Loss2: 0.0485 +Epoch [80/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0510, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 21.8550 % Model2 21.8650 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0488, Loss2: 0.0476 +Epoch [81/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0514, Loss2: 0.0489 +Epoch [81/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0501, Loss2: 0.0489 +Epoch [81/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.7812, Loss1: 0.0536, Loss2: 0.0502 +Epoch [81/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0482, Loss2: 0.0474 +Epoch [81/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0625, Loss2: 0.0589 +Epoch [81/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0482, Loss2: 0.0500 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 22.7063 % Model2 21.8750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0553, Loss2: 0.0531 +Epoch [82/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0508, Loss2: 0.0550 +Epoch [82/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0489, Loss2: 0.0498 +Epoch [82/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0438, Loss2: 0.0451 +Epoch [82/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0524, Loss2: 0.0530 +Epoch [82/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0512, Loss2: 0.0516 +Epoch [82/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.0938, Loss1: 0.0432, Loss2: 0.0462 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 21.9451 % Model2 22.7965 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0499, Loss2: 0.0480 +Epoch [83/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0477, Loss2: 0.0465 +Epoch [83/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0545, Loss2: 0.0532 +Epoch [83/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 43.7500, Loss1: 0.0415, Loss2: 0.0458 +Epoch [83/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0463, Loss2: 0.0446 +Epoch [83/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0500, Loss2: 0.0512 +Epoch [83/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0494, Loss2: 0.0487 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 21.8650 % Model2 22.0553 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0622, Loss2: 0.0619 +Epoch [84/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0426, Loss2: 0.0436 +Epoch [84/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0523, Loss2: 0.0550 +Epoch [84/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0428, Loss2: 0.0441 +Epoch [84/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0457, Loss2: 0.0436 +Epoch [84/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0421, Loss2: 0.0415 +Epoch [84/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 43.7500, Loss1: 0.0406, Loss2: 0.0436 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 22.9067 % Model2 22.5561 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0532, Loss2: 0.0534 +Epoch [85/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.7812, Loss1: 0.0411, Loss2: 0.0383 +Epoch [85/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0481, Loss2: 0.0489 +Epoch [85/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0468, Loss2: 0.0461 +Epoch [85/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0500, Loss2: 0.0516 +Epoch [85/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0484, Loss2: 0.0507 +Epoch [85/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0609, Loss2: 0.0591 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 21.3442 % Model2 21.4443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0582, Loss2: 0.0554 +Epoch [86/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0530, Loss2: 0.0537 +Epoch [86/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 48.4375, Loss1: 0.0449, Loss2: 0.0484 +Epoch [86/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0540, Loss2: 0.0550 +Epoch [86/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0557, Loss2: 0.0526 +Epoch [86/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0493, Loss2: 0.0487 +Epoch [86/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0497, Loss2: 0.0496 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 22.2256 % Model2 22.2456 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0535, Loss2: 0.0526 +Epoch [87/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0417, Loss2: 0.0391 +Epoch [87/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0481, Loss2: 0.0466 +Epoch [87/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0579, Loss2: 0.0599 +Epoch [87/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0475, Loss2: 0.0470 +Epoch [87/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0433, Loss2: 0.0437 +Epoch [87/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0459, Loss2: 0.0434 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 22.8466 % Model2 22.6663 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0524, Loss2: 0.0534 +Epoch [88/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0624, Loss2: 0.0600 +Epoch [88/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0423, Loss2: 0.0434 +Epoch [88/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0481, Loss2: 0.0487 +Epoch [88/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0499, Loss2: 0.0529 +Epoch [88/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0597, Loss2: 0.0615 +Epoch [88/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0529, Loss2: 0.0549 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 22.1755 % Model2 22.5761 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0566, Loss2: 0.0551 +Epoch [89/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 42.9688, Loss1: 0.0436, Loss2: 0.0463 +Epoch [89/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 49.2188, Loss1: 0.0468, Loss2: 0.0434 +Epoch [89/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 44.5312, Loss1: 0.0500, Loss2: 0.0519 +Epoch [89/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0455, Loss2: 0.0454 +Epoch [89/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.9688, Loss1: 0.0503, Loss2: 0.0521 +Epoch [89/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0489, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 22.8065 % Model2 21.6346 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 45.3125, Loss1: 0.0529, Loss2: 0.0571 +Epoch [90/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0587, Loss2: 0.0570 +Epoch [90/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0521, Loss2: 0.0537 +Epoch [90/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0579, Loss2: 0.0563 +Epoch [90/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 47.6562, Loss1: 0.0453, Loss2: 0.0490 +Epoch [90/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 45.3125, Loss1: 0.0548, Loss2: 0.0551 +Epoch [90/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0495, Loss2: 0.0498 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 22.2155 % Model2 21.6146 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0607, Loss2: 0.0651 +Epoch [91/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0497, Loss2: 0.0509 +Epoch [91/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0532, Loss2: 0.0509 +Epoch [91/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0545, Loss2: 0.0542 +Epoch [91/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0535, Loss2: 0.0532 +Epoch [91/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0451, Loss2: 0.0473 +Epoch [91/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0503, Loss2: 0.0481 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 21.0437 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0573, Loss2: 0.0560 +Epoch [92/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0462, Loss2: 0.0459 +Epoch [92/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0558, Loss2: 0.0543 +Epoch [92/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0434, Loss2: 0.0448 +Epoch [92/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 55.4688, Loss1: 0.0498, Loss2: 0.0457 +Epoch [92/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0556, Loss2: 0.0532 +Epoch [92/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0580, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 21.6246 % Model2 22.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0554, Loss2: 0.0554 +Epoch [93/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0540, Loss2: 0.0523 +Epoch [93/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 51.5625, Loss1: 0.0691, Loss2: 0.0737 +Epoch [93/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 46.0938, Loss1: 0.0510, Loss2: 0.0560 +Epoch [93/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0655, Loss2: 0.0619 +Epoch [93/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0601, Loss2: 0.0566 +Epoch [93/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0439, Loss2: 0.0436 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.3858 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0485, Loss2: 0.0494 +Epoch [94/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.0938, Loss1: 0.0434, Loss2: 0.0418 +Epoch [94/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0495, Loss2: 0.0490 +Epoch [94/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0460, Loss2: 0.0451 +Epoch [94/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0611, Loss2: 0.0631 +Epoch [94/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0542, Loss2: 0.0558 +Epoch [94/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0510, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 22.0954 % Model2 22.6262 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 46.0938, Loss1: 0.0479, Loss2: 0.0583 +Epoch [95/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0448, Loss2: 0.0457 +Epoch [95/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0530, Loss2: 0.0528 +Epoch [95/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0565, Loss2: 0.0528 +Epoch [95/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0551, Loss2: 0.0504 +Epoch [95/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0557, Loss2: 0.0505 +Epoch [95/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0416, Loss2: 0.0419 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 22.0753 % Model2 22.2556 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0464, Loss2: 0.0478 +Epoch [96/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.0938, Loss1: 0.0442, Loss2: 0.0449 +Epoch [96/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0480, Loss2: 0.0490 +Epoch [96/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0564, Loss2: 0.0602 +Epoch [96/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0512, Loss2: 0.0511 +Epoch [96/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0646, Loss2: 0.0646 +Epoch [96/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0594, Loss2: 0.0548 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 21.8950 % Model2 22.3858 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0576, Loss2: 0.0566 +Epoch [97/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0490, Loss2: 0.0503 +Epoch [97/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0521, Loss2: 0.0512 +Epoch [97/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0584, Loss2: 0.0624 +Epoch [97/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0503, Loss2: 0.0478 +Epoch [97/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0513, Loss2: 0.0536 +Epoch [97/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0622, Loss2: 0.0592 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 22.3958 % Model2 22.1655 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0525, Loss2: 0.0553 +Epoch [98/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0559, Loss2: 0.0551 +Epoch [98/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0465, Loss2: 0.0458 +Epoch [98/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0475, Loss2: 0.0467 +Epoch [98/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0599, Loss2: 0.0585 +Epoch [98/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0547, Loss2: 0.0510 +Epoch [98/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0527, Loss2: 0.0515 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 22.7764 % Model2 22.5661 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0500, Loss2: 0.0497 +Epoch [99/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0663, Loss2: 0.0698 +Epoch [99/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0523, Loss2: 0.0506 +Epoch [99/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0509, Loss2: 0.0479 +Epoch [99/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0605, Loss2: 0.0625 +Epoch [99/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0477, Loss2: 0.0492 +Epoch [99/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0616, Loss2: 0.0582 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 22.2756 % Model2 22.3257 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0719, Loss2: 0.0694 +Epoch [100/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.1250, Loss1: 0.0541, Loss2: 0.0499 +Epoch [100/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0524, Loss2: 0.0525 +Epoch [100/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0529, Loss2: 0.0496 +Epoch [100/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0533, Loss2: 0.0535 +Epoch [100/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0523, Loss2: 0.0499 +Epoch [100/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0569, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 22.6362 % Model2 20.9736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.0312, Loss1: 0.0544, Loss2: 0.0504 +Epoch [101/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0473, Loss2: 0.0496 +Epoch [101/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0527, Loss2: 0.0512 +Epoch [101/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0650, Loss2: 0.0630 +Epoch [101/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0694, Loss2: 0.0664 +Epoch [101/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0575, Loss2: 0.0609 +Epoch [101/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0479, Loss2: 0.0477 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 23.1971 % Model2 21.7348 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0555, Loss2: 0.0550 +Epoch [102/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0543, Loss2: 0.0555 +Epoch [102/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0500, Loss2: 0.0522 +Epoch [102/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0498, Loss2: 0.0489 +Epoch [102/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0531, Loss2: 0.0502 +Epoch [102/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0417, Loss2: 0.0406 +Epoch [102/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0501, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 21.9651 % Model2 22.2456 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0711, Loss2: 0.0736 +Epoch [103/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0483, Loss2: 0.0465 +Epoch [103/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0485, Loss2: 0.0478 +Epoch [103/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0584, Loss2: 0.0585 +Epoch [103/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0528, Loss2: 0.0507 +Epoch [103/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0475, Loss2: 0.0481 +Epoch [103/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0505, Loss2: 0.0486 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 22.4659 % Model2 21.8249 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 61.7188, Loss1: 0.0566, Loss2: 0.0492 +Epoch [104/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0577, Loss2: 0.0580 +Epoch [104/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0612, Loss2: 0.0614 +Epoch [104/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0527, Loss2: 0.0491 +Epoch [104/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0667, Loss2: 0.0637 +Epoch [104/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0508, Loss2: 0.0489 +Epoch [104/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0535, Loss2: 0.0517 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 21.9451 % Model2 21.2941 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.0659, Loss2: 0.0662 +Epoch [105/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0587, Loss2: 0.0602 +Epoch [105/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0583, Loss2: 0.0621 +Epoch [105/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0537, Loss2: 0.0525 +Epoch [105/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0478, Loss2: 0.0475 +Epoch [105/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 52.3438, Loss1: 0.0475, Loss2: 0.0449 +Epoch [105/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0554, Loss2: 0.0533 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 22.5361 % Model2 22.5561 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0513, Loss2: 0.0507 +Epoch [106/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 51.5625, Loss1: 0.0640, Loss2: 0.0554 +Epoch [106/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0598, Loss2: 0.0599 +Epoch [106/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0509, Loss2: 0.0523 +Epoch [106/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0549, Loss2: 0.0547 +Epoch [106/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0606, Loss2: 0.0570 +Epoch [106/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0522, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 22.2656 % Model2 22.5861 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0506, Loss2: 0.0512 +Epoch [107/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0598, Loss2: 0.0608 +Epoch [107/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0543, Loss2: 0.0560 +Epoch [107/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0487, Loss2: 0.0464 +Epoch [107/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0548, Loss2: 0.0588 +Epoch [107/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0664, Loss2: 0.0662 +Epoch [107/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.0312, Loss1: 0.0595, Loss2: 0.0563 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 21.8249 % Model2 21.5845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0617, Loss2: 0.0619 +Epoch [108/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0614, Loss2: 0.0657 +Epoch [108/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0533, Loss2: 0.0541 +Epoch [108/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0634, Loss2: 0.0607 +Epoch [108/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0523, Loss2: 0.0551 +Epoch [108/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0663, Loss2: 0.0634 +Epoch [108/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0573, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 22.3057 % Model2 21.4744 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0590, Loss2: 0.0574 +Epoch [109/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0516, Loss2: 0.0503 +Epoch [109/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0446, Loss2: 0.0448 +Epoch [109/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0511, Loss2: 0.0519 +Epoch [109/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0500, Loss2: 0.0496 +Epoch [109/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0571, Loss2: 0.0551 +Epoch [109/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0534, Loss2: 0.0492 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 22.5160 % Model2 22.5861 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0503, Loss2: 0.0490 +Epoch [110/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0700, Loss2: 0.0710 +Epoch [110/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0632, Loss2: 0.0667 +Epoch [110/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0630, Loss2: 0.0626 +Epoch [110/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0572, Loss2: 0.0597 +Epoch [110/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0486, Loss2: 0.0476 +Epoch [110/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0538, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 22.4559 % Model2 22.1855 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0618, Loss2: 0.0618 +Epoch [111/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 52.3438, Loss1: 0.0481, Loss2: 0.0441 +Epoch [111/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0507, Loss2: 0.0481 +Epoch [111/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 49.2188, Loss1: 0.0501, Loss2: 0.0542 +Epoch [111/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0545, Loss2: 0.0514 +Epoch [111/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0580, Loss2: 0.0594 +Epoch [111/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0552, Loss2: 0.0518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 23.0769 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0619, Loss2: 0.0613 +Epoch [112/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0726, Loss2: 0.0636 +Epoch [112/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0513, Loss2: 0.0511 +Epoch [112/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0541, Loss2: 0.0524 +Epoch [112/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0509, Loss2: 0.0505 +Epoch [112/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0464, Loss2: 0.0462 +Epoch [112/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0539, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 23.0669 % Model2 22.3257 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0521, Loss2: 0.0480 +Epoch [113/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0692, Loss2: 0.0682 +Epoch [113/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0622, Loss2: 0.0605 +Epoch [113/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0548, Loss2: 0.0501 +Epoch [113/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0489, Loss2: 0.0503 +Epoch [113/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0586, Loss2: 0.0586 +Epoch [113/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0477, Loss2: 0.0472 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 22.3758 % Model2 22.9567 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0500, Loss2: 0.0516 +Epoch [114/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0585, Loss2: 0.0557 +Epoch [114/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0430, Loss2: 0.0450 +Epoch [114/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0519, Loss2: 0.0516 +Epoch [114/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0665, Loss2: 0.0690 +Epoch [114/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0512, Loss2: 0.0526 +Epoch [114/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0542, Loss2: 0.0543 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 21.8049 % Model2 21.8149 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0625, Loss2: 0.0582 +Epoch [115/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0553, Loss2: 0.0563 +Epoch [115/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0595, Loss2: 0.0587 +Epoch [115/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0555, Loss2: 0.0544 +Epoch [115/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0557, Loss2: 0.0552 +Epoch [115/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0579, Loss2: 0.0563 +Epoch [115/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0776, Loss2: 0.0726 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 22.2957 % Model2 21.9852 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0532, Loss2: 0.0561 +Epoch [116/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0649, Loss2: 0.0659 +Epoch [116/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0619, Loss2: 0.0631 +Epoch [116/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.9062, Loss1: 0.0628, Loss2: 0.0577 +Epoch [116/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0675, Loss2: 0.0685 +Epoch [116/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0495, Loss2: 0.0492 +Epoch [116/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0517, Loss2: 0.0522 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 22.6262 % Model2 22.2957 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0552, Loss2: 0.0541 +Epoch [117/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0534, Loss2: 0.0537 +Epoch [117/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0507, Loss2: 0.0533 +Epoch [117/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.1875, Loss1: 0.0420, Loss2: 0.0437 +Epoch [117/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0599, Loss2: 0.0627 +Epoch [117/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0563, Loss2: 0.0572 +Epoch [117/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 51.5625, Loss1: 0.0571, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 22.3257 % Model2 21.7648 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0605, Loss2: 0.0618 +Epoch [118/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 62.5000, Loss1: 0.0633, Loss2: 0.0578 +Epoch [118/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0656, Loss2: 0.0653 +Epoch [118/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0591, Loss2: 0.0603 +Epoch [118/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0466, Loss2: 0.0465 +Epoch [118/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0546, Loss2: 0.0493 +Epoch [118/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0734, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 22.7464 % Model2 21.9651 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0613, Loss2: 0.0605 +Epoch [119/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0570, Loss2: 0.0578 +Epoch [119/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0571, Loss2: 0.0580 +Epoch [119/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0604, Loss2: 0.0580 +Epoch [119/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0510, Loss2: 0.0523 +Epoch [119/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0434, Loss2: 0.0446 +Epoch [119/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0546, Loss2: 0.0546 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 22.0453 % Model2 22.0853 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0609, Loss2: 0.0590 +Epoch [120/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0490, Loss2: 0.0482 +Epoch [120/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 46.8750, Loss1: 0.0483, Loss2: 0.0517 +Epoch [120/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0578, Loss2: 0.0581 +Epoch [120/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0559, Loss2: 0.0512 +Epoch [120/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0602, Loss2: 0.0591 +Epoch [120/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0523, Loss2: 0.0559 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 22.4259 % Model2 22.2155 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0526, Loss2: 0.0554 +Epoch [121/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0518, Loss2: 0.0501 +Epoch [121/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0538, Loss2: 0.0499 +Epoch [121/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0646, Loss2: 0.0685 +Epoch [121/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0640, Loss2: 0.0666 +Epoch [121/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0624, Loss2: 0.0607 +Epoch [121/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0520, Loss2: 0.0485 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 22.2456 % Model2 22.0853 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0588, Loss2: 0.0610 +Epoch [122/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0605, Loss2: 0.0592 +Epoch [122/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0507, Loss2: 0.0538 +Epoch [122/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0581, Loss2: 0.0601 +Epoch [122/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 54.6875, Loss1: 0.0520, Loss2: 0.0474 +Epoch [122/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0719, Loss2: 0.0720 +Epoch [122/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0594, Loss2: 0.0599 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 22.1154 % Model2 22.1855 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0650, Loss2: 0.0689 +Epoch [123/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0602, Loss2: 0.0562 +Epoch [123/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0541, Loss2: 0.0560 +Epoch [123/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 48.4375, Loss1: 0.0548, Loss2: 0.0579 +Epoch [123/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0603, Loss2: 0.0556 +Epoch [123/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0568 +Epoch [123/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0538, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 22.1855 % Model2 22.2756 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0475, Loss2: 0.0460 +Epoch [124/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0443, Loss2: 0.0437 +Epoch [124/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0624, Loss2: 0.0582 +Epoch [124/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0683, Loss2: 0.0683 +Epoch [124/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0576, Loss2: 0.0582 +Epoch [124/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0563, Loss2: 0.0558 +Epoch [124/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0533, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 22.3257 % Model2 22.3558 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0554, Loss2: 0.0585 +Epoch [125/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0578, Loss2: 0.0604 +Epoch [125/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0617, Loss2: 0.0639 +Epoch [125/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0584, Loss2: 0.0559 +Epoch [125/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0625, Loss2: 0.0570 +Epoch [125/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0595, Loss2: 0.0594 +Epoch [125/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0591, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 21.9251 % Model2 22.4359 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0636, Loss2: 0.0666 +Epoch [126/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0592, Loss2: 0.0583 +Epoch [126/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0549, Loss2: 0.0556 +Epoch [126/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 56.2500, Loss1: 0.0530, Loss2: 0.0576 +Epoch [126/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0667, Loss2: 0.0631 +Epoch [126/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0516, Loss2: 0.0503 +Epoch [126/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0621, Loss2: 0.0585 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 22.2356 % Model2 21.1538 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0687, Loss2: 0.0693 +Epoch [127/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0644, Loss2: 0.0675 +Epoch [127/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0625, Loss2: 0.0677 +Epoch [127/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0513, Loss2: 0.0518 +Epoch [127/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0614, Loss2: 0.0555 +Epoch [127/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0662, Loss2: 0.0675 +Epoch [127/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0479, Loss2: 0.0482 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 21.8249 % Model2 22.4659 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0504, Loss2: 0.0489 +Epoch [128/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0609, Loss2: 0.0584 +Epoch [128/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0669, Loss2: 0.0642 +Epoch [128/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0573, Loss2: 0.0550 +Epoch [128/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0573, Loss2: 0.0567 +Epoch [128/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0621, Loss2: 0.0627 +Epoch [128/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0650, Loss2: 0.0607 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 22.0653 % Model2 22.0954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0651, Loss2: 0.0631 +Epoch [129/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0804, Loss2: 0.0725 +Epoch [129/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0551, Loss2: 0.0538 +Epoch [129/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0639, Loss2: 0.0648 +Epoch [129/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0530, Loss2: 0.0537 +Epoch [129/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0558, Loss2: 0.0573 +Epoch [129/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0619, Loss2: 0.0557 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 22.1755 % Model2 21.8750 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0653, Loss2: 0.0631 +Epoch [130/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0753, Loss2: 0.0751 +Epoch [130/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0488, Loss2: 0.0470 +Epoch [130/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.8750, Loss1: 0.0568, Loss2: 0.0612 +Epoch [130/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0494, Loss2: 0.0515 +Epoch [130/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0581, Loss2: 0.0570 +Epoch [130/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0694, Loss2: 0.0708 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 22.4960 % Model2 21.9651 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0647, Loss2: 0.0678 +Epoch [131/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0676, Loss2: 0.0685 +Epoch [131/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0602, Loss2: 0.0583 +Epoch [131/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0737, Loss2: 0.0710 +Epoch [131/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0600, Loss2: 0.0606 +Epoch [131/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0658, Loss2: 0.0647 +Epoch [131/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0553, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 22.4659 % Model2 21.6546 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0515, Loss2: 0.0531 +Epoch [132/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0571, Loss2: 0.0541 +Epoch [132/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0522, Loss2: 0.0521 +Epoch [132/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 51.5625, Loss1: 0.0503, Loss2: 0.0551 +Epoch [132/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0560, Loss2: 0.0590 +Epoch [132/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0618, Loss2: 0.0586 +Epoch [132/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0502, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 22.3758 % Model2 21.8850 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0623, Loss2: 0.0615 +Epoch [133/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0618, Loss2: 0.0616 +Epoch [133/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 51.5625, Loss1: 0.0473, Loss2: 0.0524 +Epoch [133/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0625, Loss2: 0.0614 +Epoch [133/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0475, Loss2: 0.0458 +Epoch [133/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0525, Loss2: 0.0516 +Epoch [133/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0628, Loss2: 0.0572 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 23.0369 % Model2 22.2356 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0599, Loss2: 0.0612 +Epoch [134/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0660, Loss2: 0.0661 +Epoch [134/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0597, Loss2: 0.0583 +Epoch [134/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0639, Loss2: 0.0629 +Epoch [134/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0646, Loss2: 0.0653 +Epoch [134/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0676, Loss2: 0.0653 +Epoch [134/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0562, Loss2: 0.0539 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 22.7865 % Model2 22.0353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0687, Loss2: 0.0664 +Epoch [135/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0609, Loss2: 0.0614 +Epoch [135/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0553, Loss2: 0.0526 +Epoch [135/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0559, Loss2: 0.0523 +Epoch [135/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0589, Loss2: 0.0620 +Epoch [135/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0596, Loss2: 0.0571 +Epoch [135/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0709, Loss2: 0.0762 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 22.2155 % Model2 22.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0922, Loss2: 0.0871 +Epoch [136/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0661, Loss2: 0.0693 +Epoch [136/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0705, Loss2: 0.0703 +Epoch [136/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0597, Loss2: 0.0623 +Epoch [136/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0635, Loss2: 0.0607 +Epoch [136/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0753, Loss2: 0.0713 +Epoch [136/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0651, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 22.5561 % Model2 22.2656 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0644 +Epoch [137/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0691, Loss2: 0.0638 +Epoch [137/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0476, Loss2: 0.0460 +Epoch [137/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 53.1250, Loss1: 0.0505, Loss2: 0.0536 +Epoch [137/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0787, Loss2: 0.0738 +Epoch [137/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0540, Loss2: 0.0546 +Epoch [137/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0694, Loss2: 0.0716 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 22.5962 % Model2 22.5160 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0594, Loss2: 0.0572 +Epoch [138/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0704, Loss2: 0.0662 +Epoch [138/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0790, Loss2: 0.0762 +Epoch [138/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0657, Loss2: 0.0619 +Epoch [138/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0544, Loss2: 0.0518 +Epoch [138/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0610, Loss2: 0.0662 +Epoch [138/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0678, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 22.4659 % Model2 21.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0553, Loss2: 0.0549 +Epoch [139/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0715, Loss2: 0.0709 +Epoch [139/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0775, Loss2: 0.0733 +Epoch [139/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0663, Loss2: 0.0680 +Epoch [139/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0596, Loss2: 0.0566 +Epoch [139/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 50.7812, Loss1: 0.0487, Loss2: 0.0516 +Epoch [139/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0683, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 22.3558 % Model2 21.7849 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0712, Loss2: 0.0643 +Epoch [140/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0596, Loss2: 0.0582 +Epoch [140/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0827, Loss2: 0.0839 +Epoch [140/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0927, Loss2: 0.0919 +Epoch [140/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0891, Loss2: 0.0879 +Epoch [140/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0763, Loss2: 0.0862 +Epoch [140/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0493, Loss2: 0.0487 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.0453 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0588, Loss2: 0.0577 +Epoch [141/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0633, Loss2: 0.0639 +Epoch [141/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0547, Loss2: 0.0593 +Epoch [141/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0743, Loss2: 0.0714 +Epoch [141/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0598, Loss2: 0.0550 +Epoch [141/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0641, Loss2: 0.0638 +Epoch [141/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0701, Loss2: 0.0770 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 22.0252 % Model2 22.0753 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0588, Loss2: 0.0578 +Epoch [142/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0718, Loss2: 0.0706 +Epoch [142/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0650, Loss2: 0.0601 +Epoch [142/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0715, Loss2: 0.0683 +Epoch [142/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0675, Loss2: 0.0700 +Epoch [142/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0565, Loss2: 0.0592 +Epoch [142/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0546, Loss2: 0.0557 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.3758 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0606, Loss2: 0.0645 +Epoch [143/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0544, Loss2: 0.0566 +Epoch [143/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0680, Loss2: 0.0685 +Epoch [143/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0555, Loss2: 0.0551 +Epoch [143/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0802, Loss2: 0.0865 +Epoch [143/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 55.4688, Loss1: 0.0556, Loss2: 0.0651 +Epoch [143/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0783, Loss2: 0.0780 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 22.0954 % Model2 21.6647 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0736, Loss2: 0.0751 +Epoch [144/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0645 +Epoch [144/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0662, Loss2: 0.0678 +Epoch [144/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0595, Loss2: 0.0612 +Epoch [144/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0576, Loss2: 0.0586 +Epoch [144/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0616, Loss2: 0.0626 +Epoch [144/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0713, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 22.2055 % Model2 22.0954 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0891, Loss2: 0.0870 +Epoch [145/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0687, Loss2: 0.0671 +Epoch [145/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0800, Loss2: 0.0791 +Epoch [145/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0629, Loss2: 0.0643 +Epoch [145/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0770, Loss2: 0.0745 +Epoch [145/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0832, Loss2: 0.0793 +Epoch [145/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0649, Loss2: 0.0668 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 22.7564 % Model2 22.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0722, Loss2: 0.0691 +Epoch [146/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 59.3750, Loss1: 0.0543, Loss2: 0.0596 +Epoch [146/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0565, Loss2: 0.0566 +Epoch [146/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0655, Loss2: 0.0685 +Epoch [146/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0688, Loss2: 0.0676 +Epoch [146/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 50.7812, Loss1: 0.0587, Loss2: 0.0652 +Epoch [146/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0713, Loss2: 0.0703 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 22.3758 % Model2 22.1554 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0768, Loss2: 0.0812 +Epoch [147/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0729, Loss2: 0.0714 +Epoch [147/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0716, Loss2: 0.0732 +Epoch [147/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0860, Loss2: 0.0813 +Epoch [147/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0582, Loss2: 0.0550 +Epoch [147/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0531, Loss2: 0.0532 +Epoch [147/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0683, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 22.6963 % Model2 22.2656 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0666, Loss2: 0.0663 +Epoch [148/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0616, Loss2: 0.0585 +Epoch [148/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 58.5938, Loss1: 0.0591, Loss2: 0.0623 +Epoch [148/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0737, Loss2: 0.0732 +Epoch [148/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0580, Loss2: 0.0593 +Epoch [148/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0697, Loss2: 0.0685 +Epoch [148/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0888, Loss2: 0.0905 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 22.5761 % Model2 21.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0641, Loss2: 0.0652 +Epoch [149/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0714, Loss2: 0.0766 +Epoch [149/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 44.5312, Loss1: 0.0596, Loss2: 0.0641 +Epoch [149/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0687, Loss2: 0.0661 +Epoch [149/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0605, Loss2: 0.0604 +Epoch [149/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0621, Loss2: 0.0605 +Epoch [149/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0776, Loss2: 0.0804 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 22.0553 % Model2 22.1755 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0720, Loss2: 0.0720 +Epoch [150/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0607, Loss2: 0.0618 +Epoch [150/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0784, Loss2: 0.0738 +Epoch [150/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0660, Loss2: 0.0657 +Epoch [150/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0888, Loss2: 0.0906 +Epoch [150/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0634, Loss2: 0.0618 +Epoch [150/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0621, Loss2: 0.0602 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 22.2656 % Model2 21.8149 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0780, Loss2: 0.0743 +Epoch [151/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0686, Loss2: 0.0678 +Epoch [151/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0772, Loss2: 0.0769 +Epoch [151/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0635, Loss2: 0.0611 +Epoch [151/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0511, Loss2: 0.0534 +Epoch [151/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0783, Loss2: 0.0773 +Epoch [151/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0694, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 22.0653 % Model2 22.1254 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0749, Loss2: 0.0792 +Epoch [152/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0745, Loss2: 0.0692 +Epoch [152/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0746, Loss2: 0.0747 +Epoch [152/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0849, Loss2: 0.0880 +Epoch [152/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0582, Loss2: 0.0548 +Epoch [152/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0725, Loss2: 0.0675 +Epoch [152/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0534, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 22.5060 % Model2 22.0653 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0765, Loss2: 0.0776 +Epoch [153/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 57.8125, Loss1: 0.0673, Loss2: 0.0727 +Epoch [153/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0809, Loss2: 0.0726 +Epoch [153/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0578, Loss2: 0.0591 +Epoch [153/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0793, Loss2: 0.0735 +Epoch [153/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0615, Loss2: 0.0591 +Epoch [153/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0644, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 22.0553 % Model2 22.2756 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0613, Loss2: 0.0644 +Epoch [154/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0756, Loss2: 0.0781 +Epoch [154/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0669, Loss2: 0.0643 +Epoch [154/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0693, Loss2: 0.0666 +Epoch [154/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0715, Loss2: 0.0700 +Epoch [154/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0646, Loss2: 0.0654 +Epoch [154/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0842, Loss2: 0.0847 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 22.0853 % Model2 22.1855 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0808, Loss2: 0.0759 +Epoch [155/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0580, Loss2: 0.0554 +Epoch [155/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0640, Loss2: 0.0607 +Epoch [155/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0574, Loss2: 0.0539 +Epoch [155/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0793, Loss2: 0.0741 +Epoch [155/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0713, Loss2: 0.0706 +Epoch [155/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0865, Loss2: 0.0876 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 22.4459 % Model2 21.7448 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0916, Loss2: 0.0897 +Epoch [156/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0688 +Epoch [156/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0691, Loss2: 0.0709 +Epoch [156/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0777, Loss2: 0.0753 +Epoch [156/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0548, Loss2: 0.0522 +Epoch [156/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0771, Loss2: 0.0797 +Epoch [156/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0671, Loss2: 0.0672 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 22.1254 % Model2 22.0052 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0658, Loss2: 0.0715 +Epoch [157/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0716, Loss2: 0.0756 +Epoch [157/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0619, Loss2: 0.0619 +Epoch [157/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0610, Loss2: 0.0603 +Epoch [157/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0524, Loss2: 0.0522 +Epoch [157/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0659 +Epoch [157/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0673, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 22.6763 % Model2 22.0653 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0597, Loss2: 0.0593 +Epoch [158/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0604, Loss2: 0.0596 +Epoch [158/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0544, Loss2: 0.0569 +Epoch [158/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0603, Loss2: 0.0624 +Epoch [158/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 57.0312, Loss1: 0.0622, Loss2: 0.0545 +Epoch [158/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 63.2812, Loss1: 0.0648, Loss2: 0.0729 +Epoch [158/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0648, Loss2: 0.0623 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 22.3958 % Model2 22.1955 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.0759, Loss2: 0.0684 +Epoch [159/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0720, Loss2: 0.0674 +Epoch [159/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0646, Loss2: 0.0721 +Epoch [159/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0635, Loss2: 0.0619 +Epoch [159/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0616, Loss2: 0.0580 +Epoch [159/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0590, Loss2: 0.0548 +Epoch [159/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0638, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 22.4860 % Model2 22.6262 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0678, Loss2: 0.0678 +Epoch [160/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0609, Loss2: 0.0630 +Epoch [160/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0689, Loss2: 0.0695 +Epoch [160/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0647, Loss2: 0.0679 +Epoch [160/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0787, Loss2: 0.0710 +Epoch [160/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0608, Loss2: 0.0610 +Epoch [160/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0917, Loss2: 0.0837 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 22.4159 % Model2 22.3558 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0637, Loss2: 0.0622 +Epoch [161/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0766, Loss2: 0.0719 +Epoch [161/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0863, Loss2: 0.0874 +Epoch [161/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0724, Loss2: 0.0708 +Epoch [161/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0743, Loss2: 0.0706 +Epoch [161/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0588, Loss2: 0.0534 +Epoch [161/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0972, Loss2: 0.0939 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 22.2656 % Model2 22.1755 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0776, Loss2: 0.0791 +Epoch [162/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0718, Loss2: 0.0738 +Epoch [162/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0547, Loss2: 0.0537 +Epoch [162/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0700, Loss2: 0.0682 +Epoch [162/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0525, Loss2: 0.0541 +Epoch [162/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0859, Loss2: 0.0799 +Epoch [162/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0656, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 22.3558 % Model2 22.3658 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0631, Loss2: 0.0611 +Epoch [163/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0670, Loss2: 0.0665 +Epoch [163/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0634, Loss2: 0.0607 +Epoch [163/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0979, Loss2: 0.0959 +Epoch [163/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0730, Loss2: 0.0694 +Epoch [163/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0581, Loss2: 0.0574 +Epoch [163/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0733, Loss2: 0.0731 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 22.2256 % Model2 22.1154 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0772, Loss2: 0.0764 +Epoch [164/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0639, Loss2: 0.0613 +Epoch [164/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0755, Loss2: 0.0728 +Epoch [164/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0679, Loss2: 0.0690 +Epoch [164/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0673, Loss2: 0.0674 +Epoch [164/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0698, Loss2: 0.0707 +Epoch [164/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0516, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 22.7664 % Model2 21.8249 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0656, Loss2: 0.0670 +Epoch [165/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0835, Loss2: 0.0881 +Epoch [165/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0623, Loss2: 0.0593 +Epoch [165/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0691, Loss2: 0.0767 +Epoch [165/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0697, Loss2: 0.0694 +Epoch [165/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0632, Loss2: 0.0673 +Epoch [165/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0812, Loss2: 0.0798 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 22.3057 % Model2 22.1655 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0719, Loss2: 0.0729 +Epoch [166/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0777, Loss2: 0.0785 +Epoch [166/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0844, Loss2: 0.0832 +Epoch [166/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.1075, Loss2: 0.1118 +Epoch [166/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0690, Loss2: 0.0676 +Epoch [166/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0786, Loss2: 0.0735 +Epoch [166/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0814, Loss2: 0.0775 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 22.2055 % Model2 22.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0841, Loss2: 0.0841 +Epoch [167/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0513, Loss2: 0.0499 +Epoch [167/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0727, Loss2: 0.0789 +Epoch [167/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0746, Loss2: 0.0758 +Epoch [167/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0623, Loss2: 0.0596 +Epoch [167/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0684, Loss2: 0.0716 +Epoch [167/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0725, Loss2: 0.0784 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 21.8850 % Model2 22.4359 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0627, Loss2: 0.0622 +Epoch [168/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0767, Loss2: 0.0797 +Epoch [168/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0683, Loss2: 0.0676 +Epoch [168/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0887, Loss2: 0.0792 +Epoch [168/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0659, Loss2: 0.0716 +Epoch [168/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0868, Loss2: 0.0896 +Epoch [168/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 73.4375, Loss1: 0.0919, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 22.2356 % Model2 21.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0830, Loss2: 0.0781 +Epoch [169/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0633, Loss2: 0.0586 +Epoch [169/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0694, Loss2: 0.0655 +Epoch [169/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0878, Loss2: 0.0924 +Epoch [169/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0659 +Epoch [169/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0634, Loss2: 0.0655 +Epoch [169/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0715, Loss2: 0.0662 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 22.3958 % Model2 22.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0730, Loss2: 0.0734 +Epoch [170/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0673, Loss2: 0.0676 +Epoch [170/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0726, Loss2: 0.0650 +Epoch [170/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0672, Loss2: 0.0640 +Epoch [170/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0729, Loss2: 0.0748 +Epoch [170/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0700, Loss2: 0.0667 +Epoch [170/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0738, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 21.9050 % Model2 21.6947 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0685, Loss2: 0.0754 +Epoch [171/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0868, Loss2: 0.0766 +Epoch [171/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0730, Loss2: 0.0756 +Epoch [171/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0733, Loss2: 0.0658 +Epoch [171/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0669, Loss2: 0.0692 +Epoch [171/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0575, Loss2: 0.0578 +Epoch [171/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0697, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 22.4058 % Model2 22.0353 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0760, Loss2: 0.0756 +Epoch [172/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.1562, Loss1: 0.0653, Loss2: 0.0746 +Epoch [172/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0566, Loss2: 0.0582 +Epoch [172/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0864, Loss2: 0.0743 +Epoch [172/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0825, Loss2: 0.0867 +Epoch [172/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 52.3438, Loss1: 0.0577, Loss2: 0.0642 +Epoch [172/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0660, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 22.4860 % Model2 22.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0856, Loss2: 0.0902 +Epoch [173/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0772, Loss2: 0.0799 +Epoch [173/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0699, Loss2: 0.0722 +Epoch [173/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0739, Loss2: 0.0704 +Epoch [173/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0747, Loss2: 0.0729 +Epoch [173/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0853, Loss2: 0.0855 +Epoch [173/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0511, Loss2: 0.0542 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 22.4760 % Model2 21.9351 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0622, Loss2: 0.0668 +Epoch [174/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0809, Loss2: 0.0863 +Epoch [174/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0843, Loss2: 0.0866 +Epoch [174/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0797, Loss2: 0.0805 +Epoch [174/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0628, Loss2: 0.0658 +Epoch [174/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0689, Loss2: 0.0678 +Epoch [174/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0848, Loss2: 0.0863 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 22.5361 % Model2 21.6747 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0747, Loss2: 0.0729 +Epoch [175/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0686, Loss2: 0.0657 +Epoch [175/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0775, Loss2: 0.0794 +Epoch [175/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0688, Loss2: 0.0678 +Epoch [175/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0578, Loss2: 0.0574 +Epoch [175/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0699, Loss2: 0.0669 +Epoch [175/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0901, Loss2: 0.0796 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 22.2756 % Model2 21.7548 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0696, Loss2: 0.0669 +Epoch [176/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0722, Loss2: 0.0688 +Epoch [176/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0700, Loss2: 0.0678 +Epoch [176/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0800, Loss2: 0.0762 +Epoch [176/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 55.4688, Loss1: 0.0651, Loss2: 0.0701 +Epoch [176/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0709, Loss2: 0.0673 +Epoch [176/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0670, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.2155 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.1196, Loss2: 0.1103 +Epoch [177/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.0625, Loss1: 0.0679, Loss2: 0.0734 +Epoch [177/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0727, Loss2: 0.0731 +Epoch [177/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0791, Loss2: 0.0742 +Epoch [177/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0741, Loss2: 0.0721 +Epoch [177/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 57.0312, Loss1: 0.0567, Loss2: 0.0624 +Epoch [177/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0671, Loss2: 0.0707 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 22.3858 % Model2 22.1655 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0831, Loss2: 0.0837 +Epoch [178/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0654, Loss2: 0.0610 +Epoch [178/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0611, Loss2: 0.0620 +Epoch [178/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0743, Loss2: 0.0665 +Epoch [178/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0699, Loss2: 0.0666 +Epoch [178/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0589, Loss2: 0.0603 +Epoch [178/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0585, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 22.2356 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0706, Loss2: 0.0694 +Epoch [179/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0670, Loss2: 0.0692 +Epoch [179/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0810, Loss2: 0.0788 +Epoch [179/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0787, Loss2: 0.0861 +Epoch [179/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0643, Loss2: 0.0608 +Epoch [179/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0637, Loss2: 0.0641 +Epoch [179/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0743, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 22.3357 % Model2 21.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 64.8438, Loss1: 0.0963, Loss2: 0.0807 +Epoch [180/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0958, Loss2: 0.0883 +Epoch [180/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0931, Loss2: 0.0950 +Epoch [180/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0949, Loss2: 0.0859 +Epoch [180/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0693, Loss2: 0.0667 +Epoch [180/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0739, Loss2: 0.0751 +Epoch [180/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0814, Loss2: 0.0819 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 22.2155 % Model2 21.7849 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0570, Loss2: 0.0538 +Epoch [181/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0696, Loss2: 0.0679 +Epoch [181/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0628, Loss2: 0.0642 +Epoch [181/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0734, Loss2: 0.0761 +Epoch [181/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.1034, Loss2: 0.0911 +Epoch [181/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0680, Loss2: 0.0716 +Epoch [181/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0761, Loss2: 0.0822 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 22.1554 % Model2 22.1354 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0736, Loss2: 0.0730 +Epoch [182/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0710, Loss2: 0.0692 +Epoch [182/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 67.1875, Loss1: 0.0817, Loss2: 0.0711 +Epoch [182/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0782, Loss2: 0.0696 +Epoch [182/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0596, Loss2: 0.0609 +Epoch [182/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0736, Loss2: 0.0795 +Epoch [182/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0637, Loss2: 0.0656 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 22.2356 % Model2 22.1054 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 59.3750, Loss1: 0.0816, Loss2: 0.0739 +Epoch [183/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0774, Loss2: 0.0767 +Epoch [183/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0814, Loss2: 0.0786 +Epoch [183/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0812, Loss2: 0.0795 +Epoch [183/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0791, Loss2: 0.0774 +Epoch [183/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0867, Loss2: 0.0831 +Epoch [183/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0700, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 22.2456 % Model2 21.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0657, Loss2: 0.0663 +Epoch [184/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0691, Loss2: 0.0721 +Epoch [184/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0998, Loss2: 0.0939 +Epoch [184/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0689, Loss2: 0.0662 +Epoch [184/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0736, Loss2: 0.0688 +Epoch [184/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0800, Loss2: 0.0832 +Epoch [184/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0734, Loss2: 0.0692 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 22.1554 % Model2 22.1154 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0758, Loss2: 0.0728 +Epoch [185/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0743, Loss2: 0.0762 +Epoch [185/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0565, Loss2: 0.0568 +Epoch [185/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0798, Loss2: 0.0795 +Epoch [185/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0751, Loss2: 0.0727 +Epoch [185/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 58.5938, Loss1: 0.0662, Loss2: 0.0610 +Epoch [185/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0742, Loss2: 0.0751 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 22.4459 % Model2 22.0152 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0781, Loss2: 0.0775 +Epoch [186/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0759, Loss2: 0.0750 +Epoch [186/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0735, Loss2: 0.0721 +Epoch [186/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0690, Loss2: 0.0682 +Epoch [186/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0817, Loss2: 0.0739 +Epoch [186/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0882, Loss2: 0.0845 +Epoch [186/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0795, Loss2: 0.0785 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 22.3357 % Model2 22.0853 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0795, Loss2: 0.0811 +Epoch [187/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0656, Loss2: 0.0612 +Epoch [187/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0697, Loss2: 0.0659 +Epoch [187/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0837, Loss2: 0.0815 +Epoch [187/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0705, Loss2: 0.0676 +Epoch [187/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0950, Loss2: 0.0880 +Epoch [187/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0817, Loss2: 0.0800 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 22.2256 % Model2 21.7147 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0811, Loss2: 0.0746 +Epoch [188/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0879, Loss2: 0.0927 +Epoch [188/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0739, Loss2: 0.0786 +Epoch [188/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0594, Loss2: 0.0579 +Epoch [188/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0791, Loss2: 0.0824 +Epoch [188/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0738, Loss2: 0.0781 +Epoch [188/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0562, Loss2: 0.0530 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 22.2957 % Model2 21.7949 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0771, Loss2: 0.0763 +Epoch [189/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0651, Loss2: 0.0668 +Epoch [189/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0709, Loss2: 0.0675 +Epoch [189/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0702, Loss2: 0.0679 +Epoch [189/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0592, Loss2: 0.0579 +Epoch [189/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0657, Loss2: 0.0666 +Epoch [189/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0703, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 22.2756 % Model2 21.8850 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0826, Loss2: 0.0812 +Epoch [190/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0626, Loss2: 0.0618 +Epoch [190/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0826, Loss2: 0.0750 +Epoch [190/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0754, Loss2: 0.0743 +Epoch [190/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0818, Loss2: 0.0774 +Epoch [190/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0763, Loss2: 0.0746 +Epoch [190/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0971, Loss2: 0.0981 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 22.4159 % Model2 21.7849 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0915, Loss2: 0.0881 +Epoch [191/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0668, Loss2: 0.0713 +Epoch [191/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0837, Loss2: 0.0836 +Epoch [191/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0751, Loss2: 0.0775 +Epoch [191/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0599, Loss2: 0.0608 +Epoch [191/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0764, Loss2: 0.0705 +Epoch [191/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.0882, Loss2: 0.1008 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 22.4159 % Model2 21.9451 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0750, Loss2: 0.0742 +Epoch [192/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.8438, Loss1: 0.0665, Loss2: 0.0714 +Epoch [192/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.1032, Loss2: 0.1074 +Epoch [192/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0832, Loss2: 0.0809 +Epoch [192/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0837, Loss2: 0.0798 +Epoch [192/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0714, Loss2: 0.0675 +Epoch [192/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0672, Loss2: 0.0684 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 22.2857 % Model2 22.0052 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0626, Loss2: 0.0618 +Epoch [193/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0902, Loss2: 0.0918 +Epoch [193/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0870, Loss2: 0.0851 +Epoch [193/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0792, Loss2: 0.0778 +Epoch [193/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0696, Loss2: 0.0705 +Epoch [193/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0811, Loss2: 0.0793 +Epoch [193/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0508, Loss2: 0.0516 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 22.3157 % Model2 21.8049 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0701, Loss2: 0.0668 +Epoch [194/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0661, Loss2: 0.0628 +Epoch [194/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.1009, Loss2: 0.1018 +Epoch [194/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.1033, Loss2: 0.1085 +Epoch [194/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0861, Loss2: 0.0824 +Epoch [194/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0802, Loss2: 0.0730 +Epoch [194/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 22.4159 % Model2 22.0252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1052, Loss2: 0.1076 +Epoch [195/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0728, Loss2: 0.0770 +Epoch [195/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0749, Loss2: 0.0710 +Epoch [195/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0781, Loss2: 0.0775 +Epoch [195/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0970, Loss2: 0.1007 +Epoch [195/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0754, Loss2: 0.0747 +Epoch [195/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0581, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 22.1955 % Model2 21.9651 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0661, Loss2: 0.0668 +Epoch [196/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0804, Loss2: 0.0748 +Epoch [196/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0797, Loss2: 0.0777 +Epoch [196/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0638, Loss2: 0.0588 +Epoch [196/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0785, Loss2: 0.0739 +Epoch [196/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0783, Loss2: 0.0753 +Epoch [196/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0748, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 22.3458 % Model2 21.8450 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0773, Loss2: 0.0785 +Epoch [197/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0957, Loss2: 0.0923 +Epoch [197/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0795, Loss2: 0.0715 +Epoch [197/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0728 +Epoch [197/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0656, Loss2: 0.0668 +Epoch [197/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0915, Loss2: 0.0962 +Epoch [197/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0635, Loss2: 0.0651 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 22.2556 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0881, Loss2: 0.0911 +Epoch [198/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0933, Loss2: 0.0878 +Epoch [198/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0791, Loss2: 0.0811 +Epoch [198/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0726, Loss2: 0.0686 +Epoch [198/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0758, Loss2: 0.0793 +Epoch [198/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0784, Loss2: 0.0809 +Epoch [198/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0806, Loss2: 0.0714 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 22.3257 % Model2 21.9251 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0680, Loss2: 0.0696 +Epoch [199/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0622, Loss2: 0.0568 +Epoch [199/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0695, Loss2: 0.0761 +Epoch [199/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0791, Loss2: 0.0761 +Epoch [199/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 57.8125, Loss1: 0.0890, Loss2: 0.1004 +Epoch [199/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0706, Loss2: 0.0726 +Epoch [199/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0847, Loss2: 0.0787 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 22.3157 % Model2 21.8850 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0726, Loss2: 0.0717 +Epoch [200/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0609, Loss2: 0.0620 +Epoch [200/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0867, Loss2: 0.0856 +Epoch [200/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0862, Loss2: 0.0768 +Epoch [200/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0866, Loss2: 0.0880 +Epoch [200/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0723, Loss2: 0.0676 +Epoch [200/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0763, Loss2: 0.0798 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 22.2756 % Model2 21.8750 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_6_2.log b/other_methods/coteaching_plus/coteaching_plus_results/out_6_2.log new file mode 100644 index 0000000..0fe4bab --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_6_2.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.20 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 15.6250, Training Accuracy2: 15.6250, Loss1: 0.0181, Loss2: 0.0181 +Epoch [2/200], Iter [100/390] Training Accuracy1: 18.7500, Training Accuracy2: 25.0000, Loss1: 0.0170, Loss2: 0.0165 +Epoch [2/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 30.4688, Loss1: 0.0155, Loss2: 0.0153 +Epoch [2/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0159, Loss2: 0.0158 +Epoch [2/200], Iter [250/390] Training Accuracy1: 33.5938, Training Accuracy2: 32.0312, Loss1: 0.0156, Loss2: 0.0154 +Epoch [2/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0147, Loss2: 0.0145 +Epoch [2/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 27.3438, Loss1: 0.0150, Loss2: 0.0154 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 33.3433 % Model2 34.4852 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 29.6875, Loss1: 0.0152, Loss2: 0.0152 +Epoch [3/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 33.5938, Loss1: 0.0141, Loss2: 0.0152 +Epoch [3/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 28.9062, Loss1: 0.0146, Loss2: 0.0147 +Epoch [3/200], Iter [200/390] Training Accuracy1: 30.4688, Training Accuracy2: 32.8125, Loss1: 0.0144, Loss2: 0.0146 +Epoch [3/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0139, Loss2: 0.0136 +Epoch [3/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0143, Loss2: 0.0142 +Epoch [3/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.8438, Loss1: 0.0138, Loss2: 0.0133 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 41.7468 % Model2 39.8438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0146, Loss2: 0.0139 +Epoch [4/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0117, Loss2: 0.0121 +Epoch [4/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 32.0312, Loss1: 0.0145, Loss2: 0.0152 +Epoch [4/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0125, Loss2: 0.0126 +Epoch [4/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 34.3750, Loss1: 0.0130, Loss2: 0.0129 +Epoch [4/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.0625, Loss1: 0.0128, Loss2: 0.0131 +Epoch [4/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.1562, Loss1: 0.0157, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 44.0304 % Model2 45.4627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0129, Loss2: 0.0125 +Epoch [5/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0133, Loss2: 0.0128 +Epoch [5/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0117, Loss2: 0.0113 +Epoch [5/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0135 +Epoch [5/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0139 +Epoch [5/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.8438, Loss1: 0.0140, Loss2: 0.0135 +Epoch [5/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0138, Loss2: 0.0142 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 45.7833 % Model2 47.8666 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0111, Loss2: 0.0110 +Epoch [6/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0110, Loss2: 0.0115 +Epoch [6/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0125, Loss2: 0.0124 +Epoch [6/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0132, Loss2: 0.0127 +Epoch [6/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0126, Loss2: 0.0124 +Epoch [6/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0115, Loss2: 0.0116 +Epoch [6/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0127, Loss2: 0.0122 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 47.6262 % Model2 50.3706 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0111, Loss2: 0.0106 +Epoch [7/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0117, Loss2: 0.0113 +Epoch [7/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0123, Loss2: 0.0120 +Epoch [7/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0125, Loss2: 0.0118 +Epoch [7/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0130, Loss2: 0.0130 +Epoch [7/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0112, Loss2: 0.0116 +Epoch [7/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.9688, Loss1: 0.0127, Loss2: 0.0119 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 50.6310 % Model2 51.0317 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 52.3438, Loss1: 0.0106, Loss2: 0.0093 +Epoch [8/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.9688, Loss1: 0.0132, Loss2: 0.0128 +Epoch [8/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0103, Loss2: 0.0095 +Epoch [8/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 54.6875, Loss1: 0.0109, Loss2: 0.0099 +Epoch [8/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0097, Loss2: 0.0089 +Epoch [8/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0124, Loss2: 0.0118 +Epoch [8/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0122, Loss2: 0.0119 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 51.6827 % Model2 53.1450 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0106, Loss2: 0.0096 +Epoch [9/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0115, Loss2: 0.0109 +Epoch [9/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0112, Loss2: 0.0107 +Epoch [9/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0111, Loss2: 0.0110 +Epoch [9/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0112, Loss2: 0.0117 +Epoch [9/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0118, Loss2: 0.0108 +Epoch [9/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0112, Loss2: 0.0102 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 53.2853 % Model2 53.9563 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0111, Loss2: 0.0105 +Epoch [10/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0100, Loss2: 0.0094 +Epoch [10/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 54.6875, Loss1: 0.0109, Loss2: 0.0091 +Epoch [10/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0098, Loss2: 0.0095 +Epoch [10/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0094, Loss2: 0.0086 +Epoch [10/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0115, Loss2: 0.0110 +Epoch [10/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0101, Loss2: 0.0098 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 53.0749 % Model2 54.1767 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0112, Loss2: 0.0097 +Epoch [11/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0103, Loss2: 0.0096 +Epoch [11/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0125, Loss2: 0.0112 +Epoch [11/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0072, Loss2: 0.0072 +Epoch [11/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0102, Loss2: 0.0098 +Epoch [11/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0087, Loss2: 0.0090 +Epoch [11/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0099, Loss2: 0.0098 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 53.1550 % Model2 55.3285 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0090, Loss2: 0.0091 +Epoch [12/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0126, Loss2: 0.0129 +Epoch [12/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0108, Loss2: 0.0099 +Epoch [12/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0111, Loss2: 0.0118 +Epoch [12/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0107, Loss2: 0.0095 +Epoch [12/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0090, Loss2: 0.0078 +Epoch [12/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0090, Loss2: 0.0096 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 53.7760 % Model2 54.5773 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0091, Loss2: 0.0086 +Epoch [13/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.8438, Loss1: 0.0116, Loss2: 0.0113 +Epoch [13/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0106, Loss2: 0.0099 +Epoch [13/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0108, Loss2: 0.0108 +Epoch [13/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0085, Loss2: 0.0084 +Epoch [13/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0074, Loss2: 0.0070 +Epoch [13/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0090, Loss2: 0.0082 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 50.3906 % Model2 53.8762 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0098, Loss2: 0.0094 +Epoch [14/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0124, Loss2: 0.0112 +Epoch [14/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0083, Loss2: 0.0079 +Epoch [14/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0103, Loss2: 0.0098 +Epoch [14/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0092, Loss2: 0.0098 +Epoch [14/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0096, Loss2: 0.0078 +Epoch [14/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0101, Loss2: 0.0115 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 54.0865 % Model2 56.9611 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0099, Loss2: 0.0087 +Epoch [15/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0091, Loss2: 0.0086 +Epoch [15/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0100, Loss2: 0.0104 +Epoch [15/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 43.7500, Loss1: 0.0116, Loss2: 0.0109 +Epoch [15/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0095, Loss2: 0.0088 +Epoch [15/200], Iter [300/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0115, Loss2: 0.0107 +Epoch [15/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0110, Loss2: 0.0102 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 55.5088 % Model2 57.8325 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0096, Loss2: 0.0090 +Epoch [16/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0086, Loss2: 0.0085 +Epoch [16/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0091, Loss2: 0.0075 +Epoch [16/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0082, Loss2: 0.0069 +Epoch [16/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 54.6875, Loss1: 0.0098, Loss2: 0.0088 +Epoch [16/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 52.3438, Loss1: 0.0106, Loss2: 0.0096 +Epoch [16/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0091, Loss2: 0.0091 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 55.6591 % Model2 57.0613 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0096, Loss2: 0.0092 +Epoch [17/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0082, Loss2: 0.0079 +Epoch [17/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0100, Loss2: 0.0092 +Epoch [17/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0114, Loss2: 0.0100 +Epoch [17/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0086, Loss2: 0.0085 +Epoch [17/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0083, Loss2: 0.0078 +Epoch [17/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0084, Loss2: 0.0086 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 56.5905 % Model2 58.5537 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0091, Loss2: 0.0089 +Epoch [18/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0097, Loss2: 0.0093 +Epoch [18/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0082, Loss2: 0.0076 +Epoch [18/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0095, Loss2: 0.0100 +Epoch [18/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0094, Loss2: 0.0076 +Epoch [18/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 50.7812, Loss1: 0.0089, Loss2: 0.0093 +Epoch [18/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0078, Loss2: 0.0072 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 55.0581 % Model2 57.7424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0100, Loss2: 0.0090 +Epoch [19/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0095, Loss2: 0.0089 +Epoch [19/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0078, Loss2: 0.0069 +Epoch [19/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.8750, Loss1: 0.0090, Loss2: 0.0100 +Epoch [19/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0066, Loss2: 0.0069 +Epoch [19/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0090, Loss2: 0.0115 +Epoch [19/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0078, Loss2: 0.0083 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 55.7492 % Model2 58.0729 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0080, Loss2: 0.0082 +Epoch [20/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0091, Loss2: 0.0096 +Epoch [20/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0058, Loss2: 0.0052 +Epoch [20/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0096, Loss2: 0.0092 +Epoch [20/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0074, Loss2: 0.0073 +Epoch [20/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0078, Loss2: 0.0069 +Epoch [20/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0079, Loss2: 0.0070 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 57.0012 % Model2 57.8125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0683, Loss2: 0.0680 +Epoch [21/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0775, Loss2: 0.0745 +Epoch [21/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0820, Loss2: 0.0783 +Epoch [21/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 41.4062, Loss1: 0.0496, Loss2: 0.0562 +Epoch [21/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0482, Loss2: 0.0477 +Epoch [21/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0777, Loss2: 0.0703 +Epoch [21/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0675, Loss2: 0.0658 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 52.5140 % Model2 49.7596 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0664, Loss2: 0.0642 +Epoch [22/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0522, Loss2: 0.0528 +Epoch [22/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 55.4688, Loss1: 0.0723, Loss2: 0.0667 +Epoch [22/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0565, Loss2: 0.0585 +Epoch [22/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0583, Loss2: 0.0582 +Epoch [22/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0537, Loss2: 0.0543 +Epoch [22/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0578, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 53.6158 % Model2 57.6022 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0787, Loss2: 0.0731 +Epoch [23/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0526, Loss2: 0.0533 +Epoch [23/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0646, Loss2: 0.0649 +Epoch [23/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0627, Loss2: 0.0581 +Epoch [23/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0582, Loss2: 0.0579 +Epoch [23/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0643, Loss2: 0.0632 +Epoch [23/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0652, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 54.3970 % Model2 54.4371 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0769, Loss2: 0.0764 +Epoch [24/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0541, Loss2: 0.0552 +Epoch [24/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0591, Loss2: 0.0607 +Epoch [24/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0607, Loss2: 0.0604 +Epoch [24/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0906, Loss2: 0.0900 +Epoch [24/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0541, Loss2: 0.0519 +Epoch [24/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0527, Loss2: 0.0543 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 55.2284 % Model2 55.7993 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0701, Loss2: 0.0691 +Epoch [25/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0523, Loss2: 0.0510 +Epoch [25/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0766, Loss2: 0.0728 +Epoch [25/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.9062, Loss1: 0.0524, Loss2: 0.0557 +Epoch [25/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0747, Loss2: 0.0727 +Epoch [25/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0556, Loss2: 0.0543 +Epoch [25/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0652, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 55.9295 % Model2 56.6006 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0704, Loss2: 0.0710 +Epoch [26/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0540, Loss2: 0.0518 +Epoch [26/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0670, Loss2: 0.0665 +Epoch [26/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0707, Loss2: 0.0677 +Epoch [26/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0859, Loss2: 0.0792 +Epoch [26/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0601, Loss2: 0.0576 +Epoch [26/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0591, Loss2: 0.0593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 56.6106 % Model2 56.1599 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0682 +Epoch [27/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 47.6562, Loss1: 0.0640, Loss2: 0.0679 +Epoch [27/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 51.5625, Loss1: 0.0780, Loss2: 0.0865 +Epoch [27/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0671, Loss2: 0.0634 +Epoch [27/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0706, Loss2: 0.0678 +Epoch [27/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0519, Loss2: 0.0519 +Epoch [27/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0704, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 56.9411 % Model2 55.4087 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 54.6875, Loss1: 0.0777, Loss2: 0.0746 +Epoch [28/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0535, Loss2: 0.0511 +Epoch [28/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0893, Loss2: 0.0847 +Epoch [28/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0648, Loss2: 0.0643 +Epoch [28/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0684, Loss2: 0.0639 +Epoch [28/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0561, Loss2: 0.0573 +Epoch [28/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0691, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 56.9211 % Model2 57.5220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0639, Loss2: 0.0643 +Epoch [29/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0753, Loss2: 0.0714 +Epoch [29/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0671, Loss2: 0.0614 +Epoch [29/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 65.6250, Loss1: 0.0669, Loss2: 0.0583 +Epoch [29/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0652, Loss2: 0.0620 +Epoch [29/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.0000, Loss1: 0.0561, Loss2: 0.0533 +Epoch [29/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0651, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 55.8193 % Model2 58.2332 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0700, Loss2: 0.0705 +Epoch [30/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 51.5625, Loss1: 0.0742, Loss2: 0.0778 +Epoch [30/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0653, Loss2: 0.0649 +Epoch [30/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0759, Loss2: 0.0752 +Epoch [30/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0594, Loss2: 0.0553 +Epoch [30/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0693, Loss2: 0.0665 +Epoch [30/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0578, Loss2: 0.0549 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 55.4187 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0534, Loss2: 0.0500 +Epoch [31/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0721, Loss2: 0.0656 +Epoch [31/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0584, Loss2: 0.0580 +Epoch [31/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0507, Loss2: 0.0495 +Epoch [31/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0503, Loss2: 0.0482 +Epoch [31/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0605, Loss2: 0.0579 +Epoch [31/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.0312, Loss1: 0.0521, Loss2: 0.0493 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 56.6306 % Model2 58.2833 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0574, Loss2: 0.0563 +Epoch [32/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0565, Loss2: 0.0547 +Epoch [32/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0658, Loss2: 0.0639 +Epoch [32/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0638, Loss2: 0.0612 +Epoch [32/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0711, Loss2: 0.0713 +Epoch [32/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0758, Loss2: 0.0758 +Epoch [32/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0635, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 56.8109 % Model2 57.7224 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0606, Loss2: 0.0594 +Epoch [33/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0809, Loss2: 0.0763 +Epoch [33/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0694, Loss2: 0.0688 +Epoch [33/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0874, Loss2: 0.0896 +Epoch [33/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0755, Loss2: 0.0754 +Epoch [33/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0666, Loss2: 0.0639 +Epoch [33/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0711, Loss2: 0.0694 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 56.0096 % Model2 58.6338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0590, Loss2: 0.0558 +Epoch [34/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 58.5938, Loss1: 0.0783, Loss2: 0.0723 +Epoch [34/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0667, Loss2: 0.0649 +Epoch [34/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0700, Loss2: 0.0718 +Epoch [34/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0596, Loss2: 0.0599 +Epoch [34/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0694, Loss2: 0.0682 +Epoch [34/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0609, Loss2: 0.0596 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 56.9411 % Model2 58.2031 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0768, Loss2: 0.0730 +Epoch [35/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0665, Loss2: 0.0714 +Epoch [35/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0589, Loss2: 0.0578 +Epoch [35/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0581, Loss2: 0.0577 +Epoch [35/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0673, Loss2: 0.0627 +Epoch [35/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0725, Loss2: 0.0691 +Epoch [35/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0705, Loss2: 0.0740 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 57.2216 % Model2 58.0629 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0678, Loss2: 0.0649 +Epoch [36/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0684, Loss2: 0.0626 +Epoch [36/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0697, Loss2: 0.0688 +Epoch [36/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0661, Loss2: 0.0639 +Epoch [36/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0573, Loss2: 0.0587 +Epoch [36/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0636, Loss2: 0.0612 +Epoch [36/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0622, Loss2: 0.0608 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 55.8694 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0738, Loss2: 0.0694 +Epoch [37/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0773, Loss2: 0.0722 +Epoch [37/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 52.3438, Loss1: 0.0735, Loss2: 0.0778 +Epoch [37/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0593, Loss2: 0.0615 +Epoch [37/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 63.2812, Loss1: 0.0567, Loss2: 0.0520 +Epoch [37/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0664 +Epoch [37/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0607, Loss2: 0.0597 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 55.5990 % Model2 57.1214 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0658, Loss2: 0.0643 +Epoch [38/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0641, Loss2: 0.0644 +Epoch [38/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.8438, Loss1: 0.0628, Loss2: 0.0592 +Epoch [38/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0499, Loss2: 0.0470 +Epoch [38/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.1250, Loss1: 0.0656, Loss2: 0.0649 +Epoch [38/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0860, Loss2: 0.0830 +Epoch [38/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0775, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 55.8994 % Model2 59.4551 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0623, Loss2: 0.0585 +Epoch [39/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0702, Loss2: 0.0716 +Epoch [39/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0749, Loss2: 0.0768 +Epoch [39/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0681, Loss2: 0.0689 +Epoch [39/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 50.0000, Loss1: 0.0532, Loss2: 0.0553 +Epoch [39/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0746, Loss2: 0.0718 +Epoch [39/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0874, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 58.0429 % Model2 58.7440 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0708, Loss2: 0.0667 +Epoch [40/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0751, Loss2: 0.0718 +Epoch [40/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0658, Loss2: 0.0655 +Epoch [40/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0697, Loss2: 0.0682 +Epoch [40/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0566, Loss2: 0.0584 +Epoch [40/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 66.4062, Loss1: 0.0637, Loss2: 0.0588 +Epoch [40/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0756, Loss2: 0.0733 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 56.5705 % Model2 57.8726 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 56.2500, Loss1: 0.0750, Loss2: 0.0826 +Epoch [41/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0729, Loss2: 0.0678 +Epoch [41/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0635, Loss2: 0.0683 +Epoch [41/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0688, Loss2: 0.0661 +Epoch [41/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0544, Loss2: 0.0504 +Epoch [41/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0828, Loss2: 0.0790 +Epoch [41/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0559, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 57.0613 % Model2 59.1346 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.9688, Loss1: 0.0732, Loss2: 0.0645 +Epoch [42/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0743, Loss2: 0.0771 +Epoch [42/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0672, Loss2: 0.0624 +Epoch [42/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 62.5000, Loss1: 0.0747, Loss2: 0.0815 +Epoch [42/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0826, Loss2: 0.0774 +Epoch [42/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0530, Loss2: 0.0517 +Epoch [42/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0637, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 54.4772 % Model2 58.3834 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0698 +Epoch [43/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0735, Loss2: 0.0744 +Epoch [43/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0655, Loss2: 0.0635 +Epoch [43/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 53.1250, Loss1: 0.0594, Loss2: 0.0651 +Epoch [43/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0619, Loss2: 0.0585 +Epoch [43/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0646, Loss2: 0.0616 +Epoch [43/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0659, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 55.8193 % Model2 56.7007 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0702, Loss2: 0.0708 +Epoch [44/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0682, Loss2: 0.0686 +Epoch [44/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0621, Loss2: 0.0608 +Epoch [44/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0592, Loss2: 0.0580 +Epoch [44/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0818, Loss2: 0.0812 +Epoch [44/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0761, Loss2: 0.0749 +Epoch [44/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0705, Loss2: 0.0671 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 56.9812 % Model2 56.4704 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0698, Loss2: 0.0666 +Epoch [45/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 55.4688, Loss1: 0.0625, Loss2: 0.0687 +Epoch [45/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0591, Loss2: 0.0579 +Epoch [45/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0824, Loss2: 0.0758 +Epoch [45/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0692, Loss2: 0.0705 +Epoch [45/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0520, Loss2: 0.0522 +Epoch [45/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0518, Loss2: 0.0507 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 55.6891 % Model2 56.7007 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0621, Loss2: 0.0618 +Epoch [46/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0718, Loss2: 0.0693 +Epoch [46/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0640, Loss2: 0.0630 +Epoch [46/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0587, Loss2: 0.0595 +Epoch [46/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0769, Loss2: 0.0712 +Epoch [46/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0708, Loss2: 0.0674 +Epoch [46/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0662, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 56.7208 % Model2 58.1530 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 64.0625, Loss1: 0.0574, Loss2: 0.0535 +Epoch [47/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0790, Loss2: 0.0753 +Epoch [47/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0757, Loss2: 0.0757 +Epoch [47/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0576, Loss2: 0.0539 +Epoch [47/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0604, Loss2: 0.0608 +Epoch [47/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0774, Loss2: 0.0698 +Epoch [47/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0752, Loss2: 0.0743 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 56.4904 % Model2 59.0845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0774, Loss2: 0.0753 +Epoch [48/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0757, Loss2: 0.0780 +Epoch [48/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0500, Loss2: 0.0496 +Epoch [48/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0651, Loss2: 0.0686 +Epoch [48/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0564, Loss2: 0.0545 +Epoch [48/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0552, Loss2: 0.0526 +Epoch [48/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0609, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 56.7608 % Model2 58.2131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0664, Loss2: 0.0635 +Epoch [49/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 52.3438, Loss1: 0.0630, Loss2: 0.0701 +Epoch [49/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 52.3438, Loss1: 0.0707, Loss2: 0.0787 +Epoch [49/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0593, Loss2: 0.0565 +Epoch [49/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0585, Loss2: 0.0535 +Epoch [49/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0871, Loss2: 0.0879 +Epoch [49/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0692, Loss2: 0.0682 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 56.0797 % Model2 58.0929 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0721 +Epoch [50/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0620, Loss2: 0.0564 +Epoch [50/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0702, Loss2: 0.0654 +Epoch [50/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0790, Loss2: 0.0783 +Epoch [50/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0637, Loss2: 0.0627 +Epoch [50/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0621, Loss2: 0.0609 +Epoch [50/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0736, Loss2: 0.0744 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 56.0897 % Model2 58.3934 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0665, Loss2: 0.0616 +Epoch [51/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0632, Loss2: 0.0657 +Epoch [51/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0818, Loss2: 0.0842 +Epoch [51/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0702, Loss2: 0.0712 +Epoch [51/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 63.2812, Loss1: 0.0602, Loss2: 0.0531 +Epoch [51/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0650, Loss2: 0.0638 +Epoch [51/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 69.5312, Loss1: 0.0782, Loss2: 0.0687 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 56.3702 % Model2 55.1983 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0611, Loss2: 0.0568 +Epoch [52/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0576, Loss2: 0.0548 +Epoch [52/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0670, Loss2: 0.0650 +Epoch [52/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0631, Loss2: 0.0595 +Epoch [52/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 64.0625, Loss1: 0.0681, Loss2: 0.0613 +Epoch [52/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0653, Loss2: 0.0686 +Epoch [52/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0589, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 55.8794 % Model2 57.9527 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0724, Loss2: 0.0694 +Epoch [53/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0705, Loss2: 0.0640 +Epoch [53/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0562, Loss2: 0.0525 +Epoch [53/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0774, Loss2: 0.0735 +Epoch [53/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0823, Loss2: 0.0808 +Epoch [53/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0595, Loss2: 0.0596 +Epoch [53/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0638, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 56.6506 % Model2 58.0729 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 64.8438, Loss1: 0.0594, Loss2: 0.0551 +Epoch [54/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 48.4375, Loss1: 0.0604, Loss2: 0.0643 +Epoch [54/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0621, Loss2: 0.0623 +Epoch [54/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0898, Loss2: 0.0893 +Epoch [54/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 59.3750, Loss1: 0.0766, Loss2: 0.0838 +Epoch [54/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0602, Loss2: 0.0570 +Epoch [54/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0900, Loss2: 0.0879 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 57.0212 % Model2 57.2216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0700, Loss2: 0.0683 +Epoch [55/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0722, Loss2: 0.0723 +Epoch [55/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0557, Loss2: 0.0571 +Epoch [55/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0681, Loss2: 0.0669 +Epoch [55/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0750, Loss2: 0.0762 +Epoch [55/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 57.8125, Loss1: 0.0654, Loss2: 0.0710 +Epoch [55/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0818, Loss2: 0.0864 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 56.9010 % Model2 57.6522 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0869, Loss2: 0.0868 +Epoch [56/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0649, Loss2: 0.0622 +Epoch [56/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0722, Loss2: 0.0708 +Epoch [56/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0904, Loss2: 0.0822 +Epoch [56/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 51.5625, Loss1: 0.0621, Loss2: 0.0681 +Epoch [56/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0795, Loss2: 0.0803 +Epoch [56/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0572, Loss2: 0.0562 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 57.0413 % Model2 57.6222 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0809, Loss2: 0.0823 +Epoch [57/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0734, Loss2: 0.0660 +Epoch [57/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0633, Loss2: 0.0640 +Epoch [57/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0622, Loss2: 0.0605 +Epoch [57/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0622, Loss2: 0.0628 +Epoch [57/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0556, Loss2: 0.0541 +Epoch [57/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0795, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 57.0713 % Model2 58.2532 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 73.4375, Loss1: 0.0716, Loss2: 0.0645 +Epoch [58/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0746, Loss2: 0.0681 +Epoch [58/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0688, Loss2: 0.0652 +Epoch [58/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 51.5625, Loss1: 0.0535, Loss2: 0.0510 +Epoch [58/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.1562, Loss1: 0.0572, Loss2: 0.0597 +Epoch [58/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0720, Loss2: 0.0680 +Epoch [58/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0610, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 58.7740 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0719, Loss2: 0.0696 +Epoch [59/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0763, Loss2: 0.0702 +Epoch [59/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0789, Loss2: 0.0794 +Epoch [59/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0605, Loss2: 0.0619 +Epoch [59/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0752, Loss2: 0.0744 +Epoch [59/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0580, Loss2: 0.0557 +Epoch [59/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0888, Loss2: 0.0923 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 55.7091 % Model2 58.8241 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 73.4375, Loss1: 0.0816, Loss2: 0.0664 +Epoch [60/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0834 +Epoch [60/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 65.6250, Loss1: 0.0608, Loss2: 0.0533 +Epoch [60/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0690, Loss2: 0.0669 +Epoch [60/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0577, Loss2: 0.0563 +Epoch [60/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0652, Loss2: 0.0646 +Epoch [60/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0679, Loss2: 0.0609 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 55.5188 % Model2 56.9912 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0631, Loss2: 0.0633 +Epoch [61/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0689, Loss2: 0.0674 +Epoch [61/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0693, Loss2: 0.0689 +Epoch [61/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0615, Loss2: 0.0606 +Epoch [61/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0556, Loss2: 0.0551 +Epoch [61/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0669, Loss2: 0.0630 +Epoch [61/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0674, Loss2: 0.0656 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 56.3001 % Model2 57.1214 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0562, Loss2: 0.0570 +Epoch [62/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0664, Loss2: 0.0643 +Epoch [62/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0545, Loss2: 0.0507 +Epoch [62/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0799, Loss2: 0.0733 +Epoch [62/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0633, Loss2: 0.0573 +Epoch [62/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0631, Loss2: 0.0578 +Epoch [62/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0656, Loss2: 0.0655 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 55.2584 % Model2 56.8209 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0677, Loss2: 0.0662 +Epoch [63/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0760, Loss2: 0.0804 +Epoch [63/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0799, Loss2: 0.0782 +Epoch [63/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0636, Loss2: 0.0618 +Epoch [63/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0872, Loss2: 0.0864 +Epoch [63/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0571, Loss2: 0.0577 +Epoch [63/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0702, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 55.8293 % Model2 58.7139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0661, Loss2: 0.0633 +Epoch [64/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 58.5938, Loss1: 0.0639, Loss2: 0.0679 +Epoch [64/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0706, Loss2: 0.0720 +Epoch [64/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 53.9062, Loss1: 0.0575, Loss2: 0.0657 +Epoch [64/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0595, Loss2: 0.0573 +Epoch [64/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0647, Loss2: 0.0610 +Epoch [64/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0503, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 56.6306 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0628, Loss2: 0.0560 +Epoch [65/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0624, Loss2: 0.0607 +Epoch [65/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0841, Loss2: 0.0814 +Epoch [65/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0687, Loss2: 0.0634 +Epoch [65/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0627, Loss2: 0.0588 +Epoch [65/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0754, Loss2: 0.0670 +Epoch [65/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0706, Loss2: 0.0676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 57.8926 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0673, Loss2: 0.0653 +Epoch [66/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0745, Loss2: 0.0729 +Epoch [66/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0636, Loss2: 0.0647 +Epoch [66/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0728, Loss2: 0.0651 +Epoch [66/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0739, Loss2: 0.0708 +Epoch [66/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.0312, Loss1: 0.0567, Loss2: 0.0619 +Epoch [66/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1045, Loss2: 0.0995 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 56.4804 % Model2 58.4435 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0714, Loss2: 0.0669 +Epoch [67/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0956, Loss2: 0.0969 +Epoch [67/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0644, Loss2: 0.0689 +Epoch [67/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0763, Loss2: 0.0746 +Epoch [67/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0555, Loss2: 0.0558 +Epoch [67/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0526, Loss2: 0.0491 +Epoch [67/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0682, Loss2: 0.0623 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 55.4888 % Model2 58.8442 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0736, Loss2: 0.0690 +Epoch [68/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0665, Loss2: 0.0663 +Epoch [68/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0810, Loss2: 0.0894 +Epoch [68/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 59.3750, Loss1: 0.0586, Loss2: 0.0662 +Epoch [68/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0693, Loss2: 0.0749 +Epoch [68/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0872, Loss2: 0.0847 +Epoch [68/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.0940, Loss2: 0.0963 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 54.9179 % Model2 58.1030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0814, Loss2: 0.0804 +Epoch [69/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0850, Loss2: 0.0815 +Epoch [69/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 60.9375, Loss1: 0.0644, Loss2: 0.0671 +Epoch [69/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0784, Loss2: 0.0741 +Epoch [69/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0619, Loss2: 0.0635 +Epoch [69/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0729, Loss2: 0.0750 +Epoch [69/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0588, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 56.0998 % Model2 57.4319 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0825, Loss2: 0.0802 +Epoch [70/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0694, Loss2: 0.0644 +Epoch [70/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0871, Loss2: 0.0839 +Epoch [70/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.9688, Loss1: 0.0820, Loss2: 0.0764 +Epoch [70/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0775, Loss2: 0.0741 +Epoch [70/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0720, Loss2: 0.0706 +Epoch [70/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0814, Loss2: 0.0785 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 56.4603 % Model2 58.2232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0558, Loss2: 0.0540 +Epoch [71/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0832, Loss2: 0.0752 +Epoch [71/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0670 +Epoch [71/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0772, Loss2: 0.0736 +Epoch [71/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0749, Loss2: 0.0776 +Epoch [71/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0553, Loss2: 0.0567 +Epoch [71/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 60.1562, Loss1: 0.0475, Loss2: 0.0503 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 56.4904 % Model2 57.3618 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0898, Loss2: 0.0847 +Epoch [72/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0752, Loss2: 0.0769 +Epoch [72/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0715, Loss2: 0.0754 +Epoch [72/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0722, Loss2: 0.0749 +Epoch [72/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0555, Loss2: 0.0508 +Epoch [72/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0665, Loss2: 0.0660 +Epoch [72/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0759, Loss2: 0.0737 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 56.2200 % Model2 57.8826 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0787, Loss2: 0.0718 +Epoch [73/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0680, Loss2: 0.0634 +Epoch [73/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0833, Loss2: 0.0792 +Epoch [73/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0555, Loss2: 0.0506 +Epoch [73/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0813, Loss2: 0.0800 +Epoch [73/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0695 +Epoch [73/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.0719, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 54.7977 % Model2 56.9311 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0764, Loss2: 0.0731 +Epoch [74/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0652, Loss2: 0.0626 +Epoch [74/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0899, Loss2: 0.0897 +Epoch [74/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0702, Loss2: 0.0673 +Epoch [74/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0764, Loss2: 0.0743 +Epoch [74/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0867, Loss2: 0.0835 +Epoch [74/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 58.5938, Loss1: 0.0604, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 56.2300 % Model2 58.0829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0909, Loss2: 0.0847 +Epoch [75/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0730, Loss2: 0.0693 +Epoch [75/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0479, Loss2: 0.0489 +Epoch [75/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.1043, Loss2: 0.1051 +Epoch [75/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0778, Loss2: 0.0687 +Epoch [75/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0715, Loss2: 0.0674 +Epoch [75/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0662, Loss2: 0.0647 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 58.4836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0584, Loss2: 0.0566 +Epoch [76/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0670, Loss2: 0.0621 +Epoch [76/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0766, Loss2: 0.0720 +Epoch [76/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0633, Loss2: 0.0600 +Epoch [76/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0639, Loss2: 0.0644 +Epoch [76/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0622, Loss2: 0.0624 +Epoch [76/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0592, Loss2: 0.0623 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 55.9595 % Model2 57.7524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0500, Loss2: 0.0501 +Epoch [77/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0626, Loss2: 0.0617 +Epoch [77/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0568, Loss2: 0.0555 +Epoch [77/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 68.7500, Loss1: 0.0711, Loss2: 0.0636 +Epoch [77/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0752, Loss2: 0.0788 +Epoch [77/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0729, Loss2: 0.0712 +Epoch [77/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0831, Loss2: 0.0763 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 56.3902 % Model2 57.8926 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0795, Loss2: 0.0778 +Epoch [78/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0849, Loss2: 0.0821 +Epoch [78/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0676, Loss2: 0.0677 +Epoch [78/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0740, Loss2: 0.0696 +Epoch [78/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 72.6562, Loss1: 0.0831, Loss2: 0.0720 +Epoch [78/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0801, Loss2: 0.0821 +Epoch [78/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.0629, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 56.4002 % Model2 58.6038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0810, Loss2: 0.0794 +Epoch [79/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0732, Loss2: 0.0729 +Epoch [79/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0615, Loss2: 0.0590 +Epoch [79/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0754, Loss2: 0.0776 +Epoch [79/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0695, Loss2: 0.0677 +Epoch [79/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0679, Loss2: 0.0666 +Epoch [79/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0824, Loss2: 0.0808 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 56.9712 % Model2 57.5721 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0792, Loss2: 0.0788 +Epoch [80/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0621, Loss2: 0.0657 +Epoch [80/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.1058, Loss2: 0.1115 +Epoch [80/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.9062, Loss1: 0.0727, Loss2: 0.0798 +Epoch [80/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0678, Loss2: 0.0677 +Epoch [80/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0705, Loss2: 0.0667 +Epoch [80/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0654, Loss2: 0.0674 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 56.4303 % Model2 57.6823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0813, Loss2: 0.0813 +Epoch [81/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0768, Loss2: 0.0752 +Epoch [81/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0583, Loss2: 0.0536 +Epoch [81/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0673, Loss2: 0.0641 +Epoch [81/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0627, Loss2: 0.0624 +Epoch [81/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.0692, Loss2: 0.0634 +Epoch [81/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 57.0312, Loss1: 0.0688, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 56.6707 % Model2 57.9427 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0587, Loss2: 0.0566 +Epoch [82/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0880, Loss2: 0.0836 +Epoch [82/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0581, Loss2: 0.0535 +Epoch [82/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0731, Loss2: 0.0748 +Epoch [82/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0581, Loss2: 0.0553 +Epoch [82/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0718, Loss2: 0.0695 +Epoch [82/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0676, Loss2: 0.0624 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 56.0597 % Model2 56.6707 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0646, Loss2: 0.0656 +Epoch [83/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0717, Loss2: 0.0724 +Epoch [83/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0603, Loss2: 0.0578 +Epoch [83/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0682, Loss2: 0.0647 +Epoch [83/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0847, Loss2: 0.0810 +Epoch [83/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0559, Loss2: 0.0568 +Epoch [83/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0540, Loss2: 0.0562 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 55.1983 % Model2 57.9527 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 73.4375, Loss1: 0.0710, Loss2: 0.0651 +Epoch [84/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0729, Loss2: 0.0706 +Epoch [84/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0740, Loss2: 0.0733 +Epoch [84/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 61.7188, Loss1: 0.0784, Loss2: 0.0708 +Epoch [84/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0766, Loss2: 0.0708 +Epoch [84/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0637, Loss2: 0.0665 +Epoch [84/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0670, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 56.7808 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0590, Loss2: 0.0587 +Epoch [85/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0761, Loss2: 0.0714 +Epoch [85/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0704, Loss2: 0.0692 +Epoch [85/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0693, Loss2: 0.0736 +Epoch [85/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0689, Loss2: 0.0597 +Epoch [85/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0805, Loss2: 0.0745 +Epoch [85/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0651, Loss2: 0.0644 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 56.2600 % Model2 57.4119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0714, Loss2: 0.0721 +Epoch [86/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0786, Loss2: 0.0770 +Epoch [86/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0582, Loss2: 0.0596 +Epoch [86/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0620, Loss2: 0.0598 +Epoch [86/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0645, Loss2: 0.0594 +Epoch [86/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0680, Loss2: 0.0717 +Epoch [86/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0685, Loss2: 0.0723 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 56.8510 % Model2 58.0028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.1875, Loss1: 0.0776, Loss2: 0.0696 +Epoch [87/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.1228, Loss2: 0.1098 +Epoch [87/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0897, Loss2: 0.0879 +Epoch [87/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0740, Loss2: 0.0737 +Epoch [87/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0765, Loss2: 0.0774 +Epoch [87/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0819, Loss2: 0.0775 +Epoch [87/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.0813, Loss2: 0.0911 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 56.1098 % Model2 57.3117 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0600, Loss2: 0.0588 +Epoch [88/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1103, Loss2: 0.1144 +Epoch [88/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0629, Loss2: 0.0623 +Epoch [88/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0607, Loss2: 0.0598 +Epoch [88/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0654, Loss2: 0.0628 +Epoch [88/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0864, Loss2: 0.0818 +Epoch [88/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0678, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 55.4287 % Model2 56.9010 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0999, Loss2: 0.0930 +Epoch [89/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0877, Loss2: 0.0829 +Epoch [89/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0909, Loss2: 0.0939 +Epoch [89/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0647, Loss2: 0.0687 +Epoch [89/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0668, Loss2: 0.0643 +Epoch [89/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0787, Loss2: 0.0769 +Epoch [89/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0544, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 56.7308 % Model2 57.5421 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0793, Loss2: 0.0705 +Epoch [90/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0725, Loss2: 0.0691 +Epoch [90/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0674, Loss2: 0.0685 +Epoch [90/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0766, Loss2: 0.0761 +Epoch [90/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 74.2188, Loss1: 0.0974, Loss2: 0.0855 +Epoch [90/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0921, Loss2: 0.0877 +Epoch [90/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0554, Loss2: 0.0590 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 56.1498 % Model2 58.2432 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0748, Loss2: 0.0765 +Epoch [91/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0695, Loss2: 0.0663 +Epoch [91/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0900, Loss2: 0.0884 +Epoch [91/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 70.3125, Loss1: 0.0728, Loss2: 0.0622 +Epoch [91/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1074, Loss2: 0.1124 +Epoch [91/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.1190, Loss2: 0.1057 +Epoch [91/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0612, Loss2: 0.0628 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 56.6406 % Model2 57.9728 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 68.7500, Loss1: 0.0683, Loss2: 0.0765 +Epoch [92/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0789, Loss2: 0.0739 +Epoch [92/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0739, Loss2: 0.0716 +Epoch [92/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0774, Loss2: 0.0712 +Epoch [92/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0777, Loss2: 0.0705 +Epoch [92/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0754, Loss2: 0.0669 +Epoch [92/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 63.2812, Loss1: 0.0544, Loss2: 0.0459 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 56.7007 % Model2 57.7324 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0778, Loss2: 0.0744 +Epoch [93/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0736, Loss2: 0.0742 +Epoch [93/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0723, Loss2: 0.0671 +Epoch [93/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0911, Loss2: 0.0930 +Epoch [93/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0768, Loss2: 0.0794 +Epoch [93/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0916, Loss2: 0.0858 +Epoch [93/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 65.6250, Loss1: 0.0648, Loss2: 0.0553 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 56.8009 % Model2 56.6006 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0825, Loss2: 0.0796 +Epoch [94/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0636, Loss2: 0.0663 +Epoch [94/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0760, Loss2: 0.0785 +Epoch [94/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0670, Loss2: 0.0634 +Epoch [94/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0860, Loss2: 0.0813 +Epoch [94/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0661, Loss2: 0.0638 +Epoch [94/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0720, Loss2: 0.0686 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 56.2300 % Model2 57.0312 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 66.4062, Loss1: 0.0652, Loss2: 0.0567 +Epoch [95/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0731, Loss2: 0.0760 +Epoch [95/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0571, Loss2: 0.0565 +Epoch [95/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0661, Loss2: 0.0677 +Epoch [95/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0837, Loss2: 0.0769 +Epoch [95/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0906, Loss2: 0.0940 +Epoch [95/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0746, Loss2: 0.0765 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 56.7708 % Model2 57.7825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0905, Loss2: 0.0913 +Epoch [96/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 69.5312, Loss1: 0.0700, Loss2: 0.0590 +Epoch [96/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0635, Loss2: 0.0696 +Epoch [96/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0611, Loss2: 0.0570 +Epoch [96/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0655, Loss2: 0.0660 +Epoch [96/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0948, Loss2: 0.0934 +Epoch [96/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0705, Loss2: 0.0753 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 55.7091 % Model2 56.5505 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0897, Loss2: 0.0870 +Epoch [97/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 59.3750, Loss1: 0.0562, Loss2: 0.0613 +Epoch [97/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 64.0625, Loss1: 0.0626, Loss2: 0.0682 +Epoch [97/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0714, Loss2: 0.0754 +Epoch [97/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0601, Loss2: 0.0547 +Epoch [97/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0749, Loss2: 0.0724 +Epoch [97/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0631, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 56.1699 % Model2 56.9712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 68.7500, Loss1: 0.0904, Loss2: 0.0791 +Epoch [98/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0778, Loss2: 0.0692 +Epoch [98/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0959, Loss2: 0.0991 +Epoch [98/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0596, Loss2: 0.0601 +Epoch [98/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0604, Loss2: 0.0579 +Epoch [98/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0564, Loss2: 0.0569 +Epoch [98/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 60.1562, Loss1: 0.0541, Loss2: 0.0479 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 54.9780 % Model2 57.4119 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0652, Loss2: 0.0613 +Epoch [99/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0770, Loss2: 0.0764 +Epoch [99/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0653, Loss2: 0.0693 +Epoch [99/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0756, Loss2: 0.0800 +Epoch [99/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0588, Loss2: 0.0594 +Epoch [99/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 60.1562, Loss1: 0.0571, Loss2: 0.0635 +Epoch [99/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0826, Loss2: 0.0856 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 56.2700 % Model2 56.9712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0757, Loss2: 0.0741 +Epoch [100/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0976, Loss2: 0.0866 +Epoch [100/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0987, Loss2: 0.1002 +Epoch [100/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0740, Loss2: 0.0785 +Epoch [100/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 70.3125, Loss1: 0.0892, Loss2: 0.0797 +Epoch [100/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0700, Loss2: 0.0681 +Epoch [100/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0928, Loss2: 0.0921 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 55.8093 % Model2 56.4603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0895, Loss2: 0.0896 +Epoch [101/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0741, Loss2: 0.0724 +Epoch [101/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0553, Loss2: 0.0586 +Epoch [101/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0844, Loss2: 0.0839 +Epoch [101/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0690, Loss2: 0.0643 +Epoch [101/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0643, Loss2: 0.0597 +Epoch [101/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0698, Loss2: 0.0652 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 56.1799 % Model2 56.7408 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0692, Loss2: 0.0693 +Epoch [102/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 72.6562, Loss1: 0.0902, Loss2: 0.0858 +Epoch [102/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0602, Loss2: 0.0592 +Epoch [102/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.0766, Loss2: 0.0812 +Epoch [102/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0695, Loss2: 0.0703 +Epoch [102/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0796, Loss2: 0.0726 +Epoch [102/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.0938, Loss1: 0.0683, Loss2: 0.0639 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 55.2985 % Model2 56.6106 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0621, Loss2: 0.0588 +Epoch [103/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0930, Loss2: 0.0883 +Epoch [103/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0660, Loss2: 0.0658 +Epoch [103/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0763, Loss2: 0.0759 +Epoch [103/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0916, Loss2: 0.0862 +Epoch [103/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0806, Loss2: 0.0773 +Epoch [103/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0755, Loss2: 0.0730 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 56.4303 % Model2 57.9026 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0752, Loss2: 0.0715 +Epoch [104/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1049, Loss2: 0.1063 +Epoch [104/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0920, Loss2: 0.0904 +Epoch [104/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0642, Loss2: 0.0654 +Epoch [104/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0719, Loss2: 0.0682 +Epoch [104/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0657, Loss2: 0.0650 +Epoch [104/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0809, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 55.9996 % Model2 56.6206 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1067, Loss2: 0.1020 +Epoch [105/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0804, Loss2: 0.0764 +Epoch [105/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.8750, Loss1: 0.0968, Loss2: 0.0841 +Epoch [105/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0776, Loss2: 0.0754 +Epoch [105/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 72.6562, Loss1: 0.0767, Loss2: 0.0633 +Epoch [105/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0731, Loss2: 0.0761 +Epoch [105/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 63.2812, Loss1: 0.0822, Loss2: 0.0891 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 55.9595 % Model2 57.3417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.0957, Loss2: 0.0917 +Epoch [106/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0823, Loss2: 0.0814 +Epoch [106/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0753, Loss2: 0.0771 +Epoch [106/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0848, Loss2: 0.0887 +Epoch [106/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.1241, Loss2: 0.1016 +Epoch [106/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.1103, Loss2: 0.1040 +Epoch [106/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0735, Loss2: 0.0768 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 56.3902 % Model2 56.3101 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0908, Loss2: 0.0838 +Epoch [107/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0628, Loss2: 0.0664 +Epoch [107/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0701, Loss2: 0.0689 +Epoch [107/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0600, Loss2: 0.0625 +Epoch [107/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0877, Loss2: 0.0860 +Epoch [107/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.1284, Loss2: 0.1415 +Epoch [107/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0913, Loss2: 0.0871 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 55.3686 % Model2 57.3017 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1057, Loss2: 0.1020 +Epoch [108/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.0898, Loss2: 0.0844 +Epoch [108/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 58.5938, Loss1: 0.0794, Loss2: 0.0899 +Epoch [108/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0840, Loss2: 0.0832 +Epoch [108/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1037, Loss2: 0.1045 +Epoch [108/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.0840, Loss2: 0.0870 +Epoch [108/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 71.0938, Loss1: 0.0837, Loss2: 0.0725 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 55.5889 % Model2 57.5521 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0630, Loss2: 0.0652 +Epoch [109/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1003, Loss2: 0.0976 +Epoch [109/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0747, Loss2: 0.0759 +Epoch [109/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0614, Loss2: 0.0624 +Epoch [109/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.0849, Loss2: 0.0788 +Epoch [109/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0852, Loss2: 0.0907 +Epoch [109/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1077, Loss2: 0.1148 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 55.9996 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0751, Loss2: 0.0701 +Epoch [110/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0872, Loss2: 0.0877 +Epoch [110/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0884, Loss2: 0.0871 +Epoch [110/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1083, Loss2: 0.1010 +Epoch [110/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0897, Loss2: 0.0832 +Epoch [110/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.0937, Loss2: 0.0899 +Epoch [110/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1065, Loss2: 0.1102 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 55.8293 % Model2 57.4419 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.0924, Loss2: 0.1045 +Epoch [111/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0774, Loss2: 0.0749 +Epoch [111/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.1180, Loss2: 0.1094 +Epoch [111/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0729, Loss2: 0.0715 +Epoch [111/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.0878, Loss2: 0.0898 +Epoch [111/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0923, Loss2: 0.0866 +Epoch [111/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 66.4062, Loss1: 0.0756, Loss2: 0.0867 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 55.9395 % Model2 57.0413 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1060, Loss2: 0.1069 +Epoch [112/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0786, Loss2: 0.0713 +Epoch [112/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.0897, Loss2: 0.0901 +Epoch [112/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0658, Loss2: 0.0625 +Epoch [112/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.1097, Loss2: 0.1136 +Epoch [112/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0863, Loss2: 0.0770 +Epoch [112/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1246, Loss2: 0.1132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 55.1783 % Model2 57.4519 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.0938, Loss1: 0.0926, Loss2: 0.0996 +Epoch [113/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1371, Loss2: 0.1289 +Epoch [113/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.1096, Loss2: 0.0979 +Epoch [113/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0908, Loss2: 0.0880 +Epoch [113/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0716, Loss2: 0.0728 +Epoch [113/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1048, Loss2: 0.1074 +Epoch [113/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 67.1875, Loss1: 0.0747, Loss2: 0.0669 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 55.4087 % Model2 57.6823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0772, Loss2: 0.0797 +Epoch [114/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1033, Loss2: 0.1012 +Epoch [114/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0872, Loss2: 0.0868 +Epoch [114/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0756, Loss2: 0.0684 +Epoch [114/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0717, Loss2: 0.0743 +Epoch [114/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0834, Loss2: 0.0778 +Epoch [114/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0971, Loss2: 0.0970 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 55.8994 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0732, Loss2: 0.0758 +Epoch [115/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0821, Loss2: 0.0754 +Epoch [115/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1011, Loss2: 0.1028 +Epoch [115/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0828, Loss2: 0.0780 +Epoch [115/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1403, Loss2: 0.1398 +Epoch [115/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0995, Loss2: 0.1032 +Epoch [115/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0900, Loss2: 0.0966 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 55.1382 % Model2 57.3918 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.1208, Loss2: 0.1240 +Epoch [116/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0864, Loss2: 0.0790 +Epoch [116/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0838, Loss2: 0.0813 +Epoch [116/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0688, Loss2: 0.0629 +Epoch [116/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0870, Loss2: 0.0837 +Epoch [116/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0618, Loss2: 0.0626 +Epoch [116/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0627, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 55.9595 % Model2 56.8009 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0652, Loss2: 0.0631 +Epoch [117/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0908, Loss2: 0.0915 +Epoch [117/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 73.4375, Loss1: 0.0965, Loss2: 0.0856 +Epoch [117/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0642, Loss2: 0.0617 +Epoch [117/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.0796, Loss2: 0.0784 +Epoch [117/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 70.3125, Loss1: 0.0808, Loss2: 0.0714 +Epoch [117/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.1017, Loss2: 0.0924 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 55.7893 % Model2 56.8109 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 74.2188, Loss1: 0.0953, Loss2: 0.0978 +Epoch [118/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.0882, Loss2: 0.0830 +Epoch [118/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0786, Loss2: 0.0829 +Epoch [118/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0892, Loss2: 0.0813 +Epoch [118/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0769, Loss2: 0.0767 +Epoch [118/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0862, Loss2: 0.0792 +Epoch [118/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0739, Loss2: 0.0695 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 55.7893 % Model2 56.8910 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0796, Loss2: 0.0741 +Epoch [119/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.0986, Loss2: 0.1042 +Epoch [119/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1091, Loss2: 0.1012 +Epoch [119/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 72.6562, Loss1: 0.1039, Loss2: 0.0852 +Epoch [119/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.0984, Loss2: 0.0915 +Epoch [119/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0759, Loss2: 0.0767 +Epoch [119/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0738, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 55.7392 % Model2 57.0813 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0751, Loss2: 0.0745 +Epoch [120/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 76.5625, Loss1: 0.1125, Loss2: 0.0982 +Epoch [120/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0736, Loss2: 0.0798 +Epoch [120/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0985, Loss2: 0.0955 +Epoch [120/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0586, Loss2: 0.0587 +Epoch [120/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1039, Loss2: 0.1055 +Epoch [120/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0872, Loss2: 0.0855 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 55.7091 % Model2 56.7909 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 69.5312, Loss1: 0.0821, Loss2: 0.0752 +Epoch [121/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0783, Loss2: 0.0752 +Epoch [121/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0828, Loss2: 0.0799 +Epoch [121/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1041, Loss2: 0.0986 +Epoch [121/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1109, Loss2: 0.1038 +Epoch [121/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0903, Loss2: 0.0885 +Epoch [121/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0878, Loss2: 0.0938 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 55.7292 % Model2 56.7508 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 72.6562, Loss1: 0.0908, Loss2: 0.0833 +Epoch [122/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0975, Loss2: 0.0958 +Epoch [122/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0701, Loss2: 0.0738 +Epoch [122/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0942, Loss2: 0.0879 +Epoch [122/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1021, Loss2: 0.1077 +Epoch [122/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.0822, Loss2: 0.0863 +Epoch [122/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1102, Loss2: 0.1087 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 56.2400 % Model2 58.0329 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0858, Loss2: 0.0809 +Epoch [123/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1064, Loss2: 0.1007 +Epoch [123/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0886, Loss2: 0.0870 +Epoch [123/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.1043, Loss2: 0.0896 +Epoch [123/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0896, Loss2: 0.0907 +Epoch [123/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0774, Loss2: 0.0731 +Epoch [123/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0715, Loss2: 0.0693 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 55.9295 % Model2 56.4904 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 71.8750, Loss1: 0.0903, Loss2: 0.1093 +Epoch [124/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0894, Loss2: 0.0833 +Epoch [124/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0852, Loss2: 0.0849 +Epoch [124/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1102, Loss2: 0.1031 +Epoch [124/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.1131, Loss2: 0.1043 +Epoch [124/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0992, Loss2: 0.1024 +Epoch [124/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0823, Loss2: 0.0794 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 55.4087 % Model2 57.0012 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1033, Loss2: 0.1034 +Epoch [125/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.0625, Loss1: 0.0766, Loss2: 0.0817 +Epoch [125/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0703, Loss2: 0.0729 +Epoch [125/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0862, Loss2: 0.0786 +Epoch [125/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.0808, Loss2: 0.0747 +Epoch [125/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0927, Loss2: 0.0895 +Epoch [125/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.1065, Loss2: 0.0985 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 55.6791 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0758, Loss2: 0.0709 +Epoch [126/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0789, Loss2: 0.0794 +Epoch [126/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0793, Loss2: 0.0764 +Epoch [126/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0818, Loss2: 0.0789 +Epoch [126/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0845, Loss2: 0.0863 +Epoch [126/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0988, Loss2: 0.0973 +Epoch [126/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.0856, Loss2: 0.0881 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 55.6991 % Model2 56.8610 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0805, Loss2: 0.0812 +Epoch [127/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0841, Loss2: 0.0876 +Epoch [127/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 63.2812, Loss1: 0.0699, Loss2: 0.0775 +Epoch [127/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0886, Loss2: 0.0855 +Epoch [127/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1063, Loss2: 0.0999 +Epoch [127/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.1080, Loss2: 0.0959 +Epoch [127/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 68.7500, Loss1: 0.1137, Loss2: 0.1322 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 54.7776 % Model2 56.0697 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 68.7500, Loss1: 0.0861, Loss2: 0.0843 +Epoch [128/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 75.0000, Loss1: 0.0793, Loss2: 0.0715 +Epoch [128/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0667, Loss2: 0.0642 +Epoch [128/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0761, Loss2: 0.0739 +Epoch [128/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.1211, Loss2: 0.1108 +Epoch [128/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0726, Loss2: 0.0761 +Epoch [128/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1150, Loss2: 0.1089 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 55.5088 % Model2 57.2416 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0832, Loss2: 0.0786 +Epoch [129/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 76.5625, Loss1: 0.1090, Loss2: 0.0997 +Epoch [129/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0916, Loss2: 0.0886 +Epoch [129/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1554, Loss2: 0.1434 +Epoch [129/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.8750, Loss1: 0.0872, Loss2: 0.0928 +Epoch [129/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0851, Loss2: 0.0807 +Epoch [129/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0963, Loss2: 0.0967 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 55.6290 % Model2 57.2917 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.0888, Loss2: 0.0821 +Epoch [130/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 59.3750, Loss1: 0.0601, Loss2: 0.0638 +Epoch [130/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1030, Loss2: 0.1075 +Epoch [130/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.1043, Loss2: 0.1103 +Epoch [130/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0913, Loss2: 0.0839 +Epoch [130/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.0914, Loss2: 0.0820 +Epoch [130/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.0992, Loss2: 0.0910 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 55.6090 % Model2 57.0613 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0995, Loss2: 0.0997 +Epoch [131/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0956, Loss2: 0.0936 +Epoch [131/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.0710, Loss2: 0.0776 +Epoch [131/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 66.4062, Loss1: 0.0901, Loss2: 0.1011 +Epoch [131/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1192, Loss2: 0.1213 +Epoch [131/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1106, Loss2: 0.1069 +Epoch [131/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0836, Loss2: 0.0837 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 55.3586 % Model2 56.0397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1246, Loss2: 0.1310 +Epoch [132/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0954, Loss2: 0.0940 +Epoch [132/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0951, Loss2: 0.0850 +Epoch [132/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.8438, Loss1: 0.0791, Loss2: 0.0700 +Epoch [132/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 77.3438, Loss1: 0.1124, Loss2: 0.1046 +Epoch [132/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0856, Loss2: 0.0846 +Epoch [132/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1308, Loss2: 0.1209 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 54.6074 % Model2 56.8910 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0798, Loss2: 0.0818 +Epoch [133/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0890, Loss2: 0.0846 +Epoch [133/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1012, Loss2: 0.1033 +Epoch [133/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1109, Loss2: 0.1121 +Epoch [133/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1059, Loss2: 0.0972 +Epoch [133/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0883, Loss2: 0.0875 +Epoch [133/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0900, Loss2: 0.0841 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 55.5889 % Model2 57.1715 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1137, Loss2: 0.1121 +Epoch [134/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1066, Loss2: 0.0978 +Epoch [134/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0712, Loss2: 0.0716 +Epoch [134/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.0912, Loss2: 0.0960 +Epoch [134/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0726, Loss2: 0.0742 +Epoch [134/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0963, Loss2: 0.0940 +Epoch [134/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.0821, Loss2: 0.0844 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 54.8978 % Model2 57.0112 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0780, Loss2: 0.0783 +Epoch [135/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0966, Loss2: 0.0898 +Epoch [135/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1114, Loss2: 0.1071 +Epoch [135/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1405, Loss2: 0.1246 +Epoch [135/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0823, Loss2: 0.0782 +Epoch [135/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0904, Loss2: 0.0866 +Epoch [135/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0799, Loss2: 0.0778 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 55.1282 % Model2 56.7608 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0744, Loss2: 0.0805 +Epoch [136/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1545, Loss2: 0.1508 +Epoch [136/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0967, Loss2: 0.0848 +Epoch [136/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.0936, Loss2: 0.0911 +Epoch [136/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0713, Loss2: 0.0706 +Epoch [136/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.0868, Loss2: 0.0846 +Epoch [136/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0769, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 55.4287 % Model2 56.7508 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0766, Loss2: 0.0746 +Epoch [137/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0862, Loss2: 0.0875 +Epoch [137/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0871, Loss2: 0.0816 +Epoch [137/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0828, Loss2: 0.0753 +Epoch [137/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1301, Loss2: 0.1385 +Epoch [137/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0830, Loss2: 0.0795 +Epoch [137/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0904, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 55.8393 % Model2 57.2216 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.1039, Loss2: 0.1113 +Epoch [138/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.1081, Loss2: 0.1144 +Epoch [138/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1102, Loss2: 0.1171 +Epoch [138/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1563, Loss2: 0.1540 +Epoch [138/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0925, Loss2: 0.0937 +Epoch [138/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1071, Loss2: 0.1055 +Epoch [138/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1221, Loss2: 0.1238 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 55.0381 % Model2 55.7392 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0841, Loss2: 0.0888 +Epoch [139/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1217, Loss2: 0.1174 +Epoch [139/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1197, Loss2: 0.1331 +Epoch [139/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0901, Loss2: 0.0929 +Epoch [139/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1005, Loss2: 0.1056 +Epoch [139/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.0938, Loss1: 0.1058, Loss2: 0.0945 +Epoch [139/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 68.7500, Loss1: 0.0735, Loss2: 0.0802 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 55.4387 % Model2 56.9812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 73.4375, Loss1: 0.0857, Loss2: 0.0964 +Epoch [140/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0917, Loss2: 0.0856 +Epoch [140/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 77.3438, Loss1: 0.1124, Loss2: 0.0965 +Epoch [140/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1204, Loss2: 0.1186 +Epoch [140/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0878, Loss2: 0.0874 +Epoch [140/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1019, Loss2: 0.1071 +Epoch [140/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0927, Loss2: 0.0911 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 54.5473 % Model2 56.6306 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1159, Loss2: 0.1092 +Epoch [141/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1105, Loss2: 0.1181 +Epoch [141/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1267, Loss2: 0.1390 +Epoch [141/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1330, Loss2: 0.1234 +Epoch [141/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0897, Loss2: 0.0862 +Epoch [141/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1453, Loss2: 0.1339 +Epoch [141/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1364, Loss2: 0.1518 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 54.9980 % Model2 56.5705 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0944, Loss2: 0.0944 +Epoch [142/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0885, Loss2: 0.0855 +Epoch [142/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.0938, Loss1: 0.0907, Loss2: 0.1036 +Epoch [142/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.0902, Loss2: 0.0863 +Epoch [142/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.1250, Loss1: 0.0951, Loss2: 0.0876 +Epoch [142/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0992, Loss2: 0.1008 +Epoch [142/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.8750, Loss1: 0.1096, Loss2: 0.1032 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 55.4187 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0864, Loss2: 0.0806 +Epoch [143/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0724, Loss2: 0.0708 +Epoch [143/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1050, Loss2: 0.1002 +Epoch [143/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 69.5312, Loss1: 0.0888, Loss2: 0.0986 +Epoch [143/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.1284, Loss2: 0.1340 +Epoch [143/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0933, Loss2: 0.0935 +Epoch [143/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0746, Loss2: 0.0729 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 54.6274 % Model2 57.1114 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.1043, Loss2: 0.0956 +Epoch [144/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1321, Loss2: 0.1289 +Epoch [144/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.0974, Loss2: 0.0971 +Epoch [144/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0893, Loss2: 0.0885 +Epoch [144/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1145, Loss2: 0.1138 +Epoch [144/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 77.3438, Loss1: 0.1314, Loss2: 0.1161 +Epoch [144/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 77.3438, Loss1: 0.1146, Loss2: 0.1107 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 55.2484 % Model2 56.8009 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.0891, Loss2: 0.0967 +Epoch [145/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1044, Loss2: 0.1084 +Epoch [145/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1316, Loss2: 0.1386 +Epoch [145/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0985, Loss2: 0.0927 +Epoch [145/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0953, Loss2: 0.0963 +Epoch [145/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.7812, Loss1: 0.1202, Loss2: 0.1037 +Epoch [145/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0920, Loss2: 0.0914 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 55.1783 % Model2 56.6907 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 79.6875, Training Accuracy2: 73.4375, Loss1: 0.0823, Loss2: 0.0948 +Epoch [146/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1650, Loss2: 0.1384 +Epoch [146/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0860, Loss2: 0.0840 +Epoch [146/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.1081, Loss2: 0.0945 +Epoch [146/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 73.4375, Loss1: 0.0880, Loss2: 0.0886 +Epoch [146/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 79.6875, Loss1: 0.1118, Loss2: 0.0935 +Epoch [146/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0970, Loss2: 0.0949 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 55.5288 % Model2 57.0813 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1279, Loss2: 0.1163 +Epoch [147/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1046, Loss2: 0.1095 +Epoch [147/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.0885, Loss2: 0.0805 +Epoch [147/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.0904, Loss2: 0.0951 +Epoch [147/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 74.2188, Loss1: 0.1157, Loss2: 0.0925 +Epoch [147/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.0868, Loss2: 0.0909 +Epoch [147/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1029, Loss2: 0.1105 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 54.2468 % Model2 56.5004 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 61.7188, Loss1: 0.0716, Loss2: 0.0827 +Epoch [148/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1108, Loss2: 0.1066 +Epoch [148/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.8750, Loss1: 0.1083, Loss2: 0.1181 +Epoch [148/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1204, Loss2: 0.1121 +Epoch [148/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 69.5312, Loss1: 0.0859, Loss2: 0.0933 +Epoch [148/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1364, Loss2: 0.1237 +Epoch [148/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.0874, Loss2: 0.0959 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 54.9379 % Model2 56.4203 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.1229, Loss2: 0.1377 +Epoch [149/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1278, Loss2: 0.1285 +Epoch [149/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.1034, Loss2: 0.1047 +Epoch [149/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1069, Loss2: 0.1038 +Epoch [149/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1047, Loss2: 0.1047 +Epoch [149/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1220, Loss2: 0.1271 +Epoch [149/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0894, Loss2: 0.0858 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 55.2684 % Model2 56.8710 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0804, Loss2: 0.0837 +Epoch [150/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0740, Loss2: 0.0764 +Epoch [150/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 76.5625, Loss1: 0.1289, Loss2: 0.1366 +Epoch [150/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.0938, Loss1: 0.1035, Loss2: 0.1148 +Epoch [150/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0895, Loss2: 0.0900 +Epoch [150/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1156, Loss2: 0.1204 +Epoch [150/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.0985, Loss2: 0.0943 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 54.6575 % Model2 56.8209 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 74.2188, Loss1: 0.0812, Loss2: 0.0895 +Epoch [151/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0790, Loss2: 0.0754 +Epoch [151/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1676, Loss2: 0.1629 +Epoch [151/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.1405, Loss2: 0.1401 +Epoch [151/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1392, Loss2: 0.1304 +Epoch [151/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 71.0938, Loss1: 0.1233, Loss2: 0.1390 +Epoch [151/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.1057, Loss2: 0.1100 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 54.9279 % Model2 56.7508 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0891, Loss2: 0.0933 +Epoch [152/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1296, Loss2: 0.1195 +Epoch [152/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.0901, Loss2: 0.0897 +Epoch [152/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1161, Loss2: 0.1014 +Epoch [152/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.1275, Loss2: 0.1264 +Epoch [152/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.1035, Loss2: 0.1113 +Epoch [152/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 74.2188, Loss1: 0.0764, Loss2: 0.0692 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 55.2584 % Model2 56.6306 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.1131, Loss2: 0.1059 +Epoch [153/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1191, Loss2: 0.1245 +Epoch [153/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.1282, Loss2: 0.1348 +Epoch [153/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0961, Loss2: 0.0898 +Epoch [153/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1084, Loss2: 0.1059 +Epoch [153/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 69.5312, Loss1: 0.0913, Loss2: 0.0997 +Epoch [153/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.0964, Loss2: 0.0995 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 55.5389 % Model2 56.8409 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 74.2188, Loss1: 0.0933, Loss2: 0.0817 +Epoch [154/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1693, Loss2: 0.1778 +Epoch [154/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1114, Loss2: 0.1126 +Epoch [154/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.0843, Loss2: 0.0812 +Epoch [154/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1023, Loss2: 0.1053 +Epoch [154/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0967, Loss2: 0.1022 +Epoch [154/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 70.3125, Loss1: 0.1329, Loss2: 0.1223 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 54.7776 % Model2 56.1198 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.0000, Loss1: 0.1007, Loss2: 0.1065 +Epoch [155/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0834, Loss2: 0.0787 +Epoch [155/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1049, Loss2: 0.0994 +Epoch [155/200], Iter [200/390] Training Accuracy1: 81.2500, Training Accuracy2: 83.5938, Loss1: 0.1310, Loss2: 0.1253 +Epoch [155/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0919, Loss2: 0.0912 +Epoch [155/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.0000, Loss1: 0.1354, Loss2: 0.1352 +Epoch [155/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1458, Loss2: 0.1464 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 54.9179 % Model2 56.6306 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0899, Loss2: 0.0921 +Epoch [156/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.0910, Loss2: 0.0935 +Epoch [156/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 79.6875, Loss1: 0.1315, Loss2: 0.1401 +Epoch [156/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0949, Loss2: 0.0900 +Epoch [156/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0922, Loss2: 0.0925 +Epoch [156/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0864, Loss2: 0.0903 +Epoch [156/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1687, Loss2: 0.1640 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 55.4788 % Model2 56.5405 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1057, Loss2: 0.1098 +Epoch [157/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1003, Loss2: 0.0972 +Epoch [157/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1111, Loss2: 0.1014 +Epoch [157/200], Iter [200/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.1365, Loss2: 0.1319 +Epoch [157/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 74.2188, Loss1: 0.1123, Loss2: 0.0992 +Epoch [157/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1009, Loss2: 0.0990 +Epoch [157/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1115, Loss2: 0.1128 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 55.4487 % Model2 56.6907 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1088, Loss2: 0.1093 +Epoch [158/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1402, Loss2: 0.1327 +Epoch [158/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1144, Loss2: 0.1152 +Epoch [158/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1257, Loss2: 0.1167 +Epoch [158/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1086, Loss2: 0.1119 +Epoch [158/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1206, Loss2: 0.1169 +Epoch [158/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 82.8125, Loss1: 0.1269, Loss2: 0.1049 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 54.9780 % Model2 56.8910 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.0875, Loss2: 0.0877 +Epoch [159/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.1398, Loss2: 0.1629 +Epoch [159/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1041, Loss2: 0.0971 +Epoch [159/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1232, Loss2: 0.1262 +Epoch [159/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.1755, Loss2: 0.1629 +Epoch [159/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0873, Loss2: 0.0838 +Epoch [159/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1376, Loss2: 0.1395 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 55.2384 % Model2 57.0613 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 79.6875, Loss1: 0.1039, Loss2: 0.1019 +Epoch [160/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.0964, Loss2: 0.0990 +Epoch [160/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0904, Loss2: 0.0879 +Epoch [160/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1379, Loss2: 0.1342 +Epoch [160/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.0873, Loss2: 0.0913 +Epoch [160/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 73.4375, Loss1: 0.1037, Loss2: 0.1079 +Epoch [160/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1482, Loss2: 0.1360 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 55.0581 % Model2 56.7708 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0880, Loss2: 0.0866 +Epoch [161/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 73.4375, Loss1: 0.0785, Loss2: 0.0708 +Epoch [161/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.1136, Loss2: 0.1062 +Epoch [161/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1079, Loss2: 0.1104 +Epoch [161/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1138, Loss2: 0.1187 +Epoch [161/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0974, Loss2: 0.1007 +Epoch [161/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1106, Loss2: 0.1070 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 54.9780 % Model2 56.5004 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1340, Loss2: 0.1458 +Epoch [162/200], Iter [100/390] Training Accuracy1: 83.5938, Training Accuracy2: 85.1562, Loss1: 0.1882, Loss2: 0.1638 +Epoch [162/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.0934, Loss2: 0.0895 +Epoch [162/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.0938, Loss2: 0.0847 +Epoch [162/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1146, Loss2: 0.1077 +Epoch [162/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 77.3438, Loss1: 0.1186, Loss2: 0.1075 +Epoch [162/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0898, Loss2: 0.0967 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 54.5172 % Model2 56.3902 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1105, Loss2: 0.1005 +Epoch [163/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 80.4688, Loss1: 0.1789, Loss2: 0.1531 +Epoch [163/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0884, Loss2: 0.0851 +Epoch [163/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.8750, Loss1: 0.1357, Loss2: 0.1556 +Epoch [163/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1201, Loss2: 0.1195 +Epoch [163/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0842, Loss2: 0.0882 +Epoch [163/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1062, Loss2: 0.1034 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 54.9579 % Model2 56.8209 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1222, Loss2: 0.1270 +Epoch [164/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1085, Loss2: 0.1061 +Epoch [164/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 76.5625, Loss1: 0.1653, Loss2: 0.1752 +Epoch [164/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0843, Loss2: 0.0845 +Epoch [164/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 76.5625, Loss1: 0.0931, Loss2: 0.0888 +Epoch [164/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1179, Loss2: 0.1316 +Epoch [164/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1240, Loss2: 0.1231 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 54.6474 % Model2 56.4904 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1094, Loss2: 0.1113 +Epoch [165/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.1363, Loss2: 0.1360 +Epoch [165/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1440, Loss2: 0.1444 +Epoch [165/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 73.4375, Loss1: 0.0929, Loss2: 0.0946 +Epoch [165/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.1485, Loss2: 0.1621 +Epoch [165/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1083, Loss2: 0.1081 +Epoch [165/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1198, Loss2: 0.1072 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 54.5873 % Model2 56.3802 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1224, Loss2: 0.1139 +Epoch [166/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1170, Loss2: 0.1125 +Epoch [166/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.1453, Loss2: 0.1248 +Epoch [166/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1170, Loss2: 0.1184 +Epoch [166/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1258, Loss2: 0.1187 +Epoch [166/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.1093, Loss2: 0.0950 +Epoch [166/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1668, Loss2: 0.1697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 55.3085 % Model2 56.1098 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0969, Loss2: 0.0873 +Epoch [167/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 74.2188, Loss1: 0.1160, Loss2: 0.1051 +Epoch [167/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1316, Loss2: 0.1260 +Epoch [167/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 83.5938, Loss1: 0.1759, Loss2: 0.1568 +Epoch [167/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0989, Loss2: 0.0912 +Epoch [167/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1351, Loss2: 0.1227 +Epoch [167/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.1034, Loss2: 0.0931 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 54.5773 % Model2 56.6206 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1316, Loss2: 0.1401 +Epoch [168/200], Iter [100/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1029, Loss2: 0.0961 +Epoch [168/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1038, Loss2: 0.1073 +Epoch [168/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1233, Loss2: 0.1240 +Epoch [168/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.0938, Loss1: 0.0858, Loss2: 0.0866 +Epoch [168/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.0000, Loss1: 0.0936, Loss2: 0.0960 +Epoch [168/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 76.5625, Loss1: 0.1134, Loss2: 0.1039 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 54.6274 % Model2 56.3101 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1069, Loss2: 0.1063 +Epoch [169/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1190, Loss2: 0.1249 +Epoch [169/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1775, Loss2: 0.1884 +Epoch [169/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 76.5625, Loss1: 0.0964, Loss2: 0.0872 +Epoch [169/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 71.0938, Loss1: 0.1738, Loss2: 0.1537 +Epoch [169/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1067, Loss2: 0.1075 +Epoch [169/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.0894, Loss2: 0.0931 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 54.5072 % Model2 56.6206 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1458, Loss2: 0.1603 +Epoch [170/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.1581, Loss2: 0.1536 +Epoch [170/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1361, Loss2: 0.1205 +Epoch [170/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 69.5312, Loss1: 0.1405, Loss2: 0.1445 +Epoch [170/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1097, Loss2: 0.1166 +Epoch [170/200], Iter [300/390] Training Accuracy1: 82.8125, Training Accuracy2: 84.3750, Loss1: 0.4369, Loss2: 0.4022 +Epoch [170/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1019, Loss2: 0.1019 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 54.7075 % Model2 56.7007 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1190, Loss2: 0.1137 +Epoch [171/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 71.8750, Loss1: 0.0891, Loss2: 0.0986 +Epoch [171/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1231, Loss2: 0.1142 +Epoch [171/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.0000, Loss1: 0.1297, Loss2: 0.1152 +Epoch [171/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1092, Loss2: 0.1177 +Epoch [171/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 78.1250, Loss1: 0.1847, Loss2: 0.1656 +Epoch [171/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1359, Loss2: 0.1325 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 54.7576 % Model2 56.4603 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1070, Loss2: 0.1098 +Epoch [172/200], Iter [100/390] Training Accuracy1: 81.2500, Training Accuracy2: 82.8125, Loss1: 0.1792, Loss2: 0.1691 +Epoch [172/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0920, Loss2: 0.0933 +Epoch [172/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1071, Loss2: 0.1069 +Epoch [172/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.0965, Loss2: 0.0895 +Epoch [172/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1448, Loss2: 0.1375 +Epoch [172/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 68.7500, Loss1: 0.1071, Loss2: 0.1170 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 54.7576 % Model2 56.5304 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1554, Loss2: 0.1417 +Epoch [173/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1068, Loss2: 0.1055 +Epoch [173/200], Iter [150/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.0843, Loss2: 0.0935 +Epoch [173/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1136, Loss2: 0.1181 +Epoch [173/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1052, Loss2: 0.1086 +Epoch [173/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1160, Loss2: 0.1130 +Epoch [173/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1281, Loss2: 0.1173 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 54.2869 % Model2 56.5705 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.1340, Loss2: 0.1458 +Epoch [174/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1081, Loss2: 0.1145 +Epoch [174/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1106, Loss2: 0.1101 +Epoch [174/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1440, Loss2: 0.1301 +Epoch [174/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1051, Loss2: 0.1049 +Epoch [174/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 76.5625, Loss1: 0.1612, Loss2: 0.1474 +Epoch [174/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0868, Loss2: 0.0892 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 54.1266 % Model2 55.6891 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1124, Loss2: 0.1255 +Epoch [175/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.1166, Loss2: 0.1121 +Epoch [175/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1600, Loss2: 0.1657 +Epoch [175/200], Iter [200/390] Training Accuracy1: 80.4688, Training Accuracy2: 76.5625, Loss1: 0.1278, Loss2: 0.1416 +Epoch [175/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.1252, Loss2: 0.1411 +Epoch [175/200], Iter [300/390] Training Accuracy1: 78.9062, Training Accuracy2: 75.7812, Loss1: 0.1229, Loss2: 0.1381 +Epoch [175/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.0312, Loss1: 0.1927, Loss2: 0.1632 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 54.4271 % Model2 56.1699 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.1157, Loss2: 0.1156 +Epoch [176/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1213, Loss2: 0.1063 +Epoch [176/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0951, Loss2: 0.0953 +Epoch [176/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1777, Loss2: 0.1597 +Epoch [176/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1383, Loss2: 0.1459 +Epoch [176/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 72.6562, Loss1: 0.1103, Loss2: 0.1205 +Epoch [176/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 66.4062, Loss1: 0.1263, Loss2: 0.1454 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 54.3970 % Model2 56.3001 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1490, Loss2: 0.1388 +Epoch [177/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1802, Loss2: 0.1771 +Epoch [177/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.0000, Loss1: 0.1421, Loss2: 0.1309 +Epoch [177/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 85.1562, Loss1: 0.2902, Loss2: 0.2443 +Epoch [177/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 75.7812, Loss1: 0.1187, Loss2: 0.1075 +Epoch [177/200], Iter [300/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.8125, Loss1: 0.1072, Loss2: 0.0982 +Epoch [177/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1357, Loss2: 0.1276 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 54.4571 % Model2 56.1298 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0992, Loss2: 0.1004 +Epoch [178/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1087, Loss2: 0.1189 +Epoch [178/200], Iter [150/390] Training Accuracy1: 78.9062, Training Accuracy2: 80.4688, Loss1: 0.1374, Loss2: 0.1301 +Epoch [178/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.9062, Loss1: 0.1622, Loss2: 0.1545 +Epoch [178/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1092, Loss2: 0.1051 +Epoch [178/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1278, Loss2: 0.1141 +Epoch [178/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 74.2188, Loss1: 0.1043, Loss2: 0.1011 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 54.7376 % Model2 56.2700 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1880, Loss2: 0.2021 +Epoch [179/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.0915, Loss2: 0.0882 +Epoch [179/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1192, Loss2: 0.1296 +Epoch [179/200], Iter [200/390] Training Accuracy1: 82.8125, Training Accuracy2: 81.2500, Loss1: 0.1731, Loss2: 0.1765 +Epoch [179/200], Iter [250/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1604, Loss2: 0.1648 +Epoch [179/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 67.9688, Loss1: 0.0746, Loss2: 0.0799 +Epoch [179/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1061, Loss2: 0.0962 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 54.7977 % Model2 56.1599 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1180, Loss2: 0.1222 +Epoch [180/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 82.8125, Loss1: 0.1636, Loss2: 0.1396 +Epoch [180/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 73.4375, Loss1: 0.1231, Loss2: 0.1378 +Epoch [180/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1395, Loss2: 0.1451 +Epoch [180/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 78.1250, Loss1: 0.1402, Loss2: 0.1171 +Epoch [180/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1028, Loss2: 0.1051 +Epoch [180/200], Iter [350/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1899, Loss2: 0.1766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 54.4571 % Model2 56.1999 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.0965, Loss2: 0.0899 +Epoch [181/200], Iter [100/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.9062, Loss1: 0.1855, Loss2: 0.1847 +Epoch [181/200], Iter [150/390] Training Accuracy1: 76.5625, Training Accuracy2: 75.7812, Loss1: 0.1423, Loss2: 0.1490 +Epoch [181/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.0000, Loss1: 0.1235, Loss2: 0.1183 +Epoch [181/200], Iter [250/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.1899, Loss2: 0.1901 +Epoch [181/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1285, Loss2: 0.1299 +Epoch [181/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1709, Loss2: 0.1767 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 54.3870 % Model2 56.6406 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1174, Loss2: 0.1306 +Epoch [182/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 77.3438, Loss1: 0.1890, Loss2: 0.1574 +Epoch [182/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.1011, Loss2: 0.1057 +Epoch [182/200], Iter [200/390] Training Accuracy1: 79.6875, Training Accuracy2: 78.9062, Loss1: 0.2097, Loss2: 0.2155 +Epoch [182/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.0892, Loss2: 0.0883 +Epoch [182/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 76.5625, Loss1: 0.1200, Loss2: 0.1375 +Epoch [182/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 75.7812, Loss1: 0.1155, Loss2: 0.0979 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 54.3870 % Model2 56.3702 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 75.7812, Loss1: 0.1179, Loss2: 0.1239 +Epoch [183/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1287, Loss2: 0.1265 +Epoch [183/200], Iter [150/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1333, Loss2: 0.1440 +Epoch [183/200], Iter [200/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.1394, Loss2: 0.1423 +Epoch [183/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1795, Loss2: 0.1778 +Epoch [183/200], Iter [300/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.1638, Loss2: 0.1587 +Epoch [183/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.1340, Loss2: 0.1335 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 54.6575 % Model2 56.4904 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1498, Loss2: 0.1391 +Epoch [184/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1363, Loss2: 0.1259 +Epoch [184/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 76.5625, Loss1: 0.1288, Loss2: 0.1237 +Epoch [184/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.1420, Loss2: 0.1405 +Epoch [184/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.1079, Loss2: 0.1147 +Epoch [184/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.0886, Loss2: 0.0894 +Epoch [184/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1065, Loss2: 0.1069 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 54.4872 % Model2 56.0497 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 78.1250, Training Accuracy2: 69.5312, Loss1: 0.1067, Loss2: 0.1353 +Epoch [185/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1245, Loss2: 0.1259 +Epoch [185/200], Iter [150/390] Training Accuracy1: 81.2500, Training Accuracy2: 81.2500, Loss1: 0.2730, Loss2: 0.2896 +Epoch [185/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0909, Loss2: 0.0914 +Epoch [185/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.0000, Loss1: 0.1074, Loss2: 0.0977 +Epoch [185/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1454, Loss2: 0.1415 +Epoch [185/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1318, Loss2: 0.1381 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 54.1867 % Model2 56.1799 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.0000, Loss1: 0.2872, Loss2: 0.2594 +Epoch [186/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 74.2188, Loss1: 0.1739, Loss2: 0.1531 +Epoch [186/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1483, Loss2: 0.1412 +Epoch [186/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0938, Loss2: 0.0941 +Epoch [186/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 73.4375, Loss1: 0.1240, Loss2: 0.1417 +Epoch [186/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 77.3438, Loss1: 0.1204, Loss2: 0.1171 +Epoch [186/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1267, Loss2: 0.1187 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 54.4471 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1471, Loss2: 0.1524 +Epoch [187/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.8750, Loss1: 0.1309, Loss2: 0.1535 +Epoch [187/200], Iter [150/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.7812, Loss1: 0.1583, Loss2: 0.1460 +Epoch [187/200], Iter [200/390] Training Accuracy1: 74.2188, Training Accuracy2: 74.2188, Loss1: 0.1486, Loss2: 0.1478 +Epoch [187/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 78.1250, Loss1: 0.1428, Loss2: 0.1313 +Epoch [187/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.7812, Loss1: 0.1689, Loss2: 0.1546 +Epoch [187/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.1385, Loss2: 0.1284 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 54.1667 % Model2 56.1198 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1029, Loss2: 0.1099 +Epoch [188/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1899, Loss2: 0.1763 +Epoch [188/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1347, Loss2: 0.1267 +Epoch [188/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.0950, Loss2: 0.0877 +Epoch [188/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1491, Loss2: 0.1572 +Epoch [188/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 72.6562, Loss1: 0.1140, Loss2: 0.0987 +Epoch [188/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 81.2500, Loss1: 0.1374, Loss2: 0.1156 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 54.0264 % Model2 56.2200 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.0966, Loss2: 0.1007 +Epoch [189/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.1451, Loss2: 0.1302 +Epoch [189/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1234, Loss2: 0.1231 +Epoch [189/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.0000, Loss1: 0.1987, Loss2: 0.2158 +Epoch [189/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1172, Loss2: 0.1103 +Epoch [189/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.1450, Loss2: 0.1336 +Epoch [189/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1899, Loss2: 0.1985 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 54.3369 % Model2 56.2300 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 81.2500, Loss1: 0.1418, Loss2: 0.1255 +Epoch [190/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1343, Loss2: 0.1303 +Epoch [190/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1649, Loss2: 0.1736 +Epoch [190/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 81.2500, Loss1: 0.1135, Loss2: 0.1001 +Epoch [190/200], Iter [250/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1257, Loss2: 0.1265 +Epoch [190/200], Iter [300/390] Training Accuracy1: 76.5625, Training Accuracy2: 74.2188, Loss1: 0.1451, Loss2: 0.1649 +Epoch [190/200], Iter [350/390] Training Accuracy1: 82.8125, Training Accuracy2: 82.8125, Loss1: 0.1682, Loss2: 0.1723 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 54.2869 % Model2 56.0397 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 78.9062, Training Accuracy2: 78.1250, Loss1: 0.1345, Loss2: 0.1399 +Epoch [191/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 73.4375, Loss1: 0.1430, Loss2: 0.1237 +Epoch [191/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1199, Loss2: 0.1401 +Epoch [191/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1323, Loss2: 0.1210 +Epoch [191/200], Iter [250/390] Training Accuracy1: 75.0000, Training Accuracy2: 70.3125, Loss1: 0.0910, Loss2: 0.1017 +Epoch [191/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1170, Loss2: 0.1179 +Epoch [191/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 74.2188, Loss1: 0.1502, Loss2: 0.1309 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 54.3069 % Model2 56.0797 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 74.2188, Loss1: 0.1864, Loss2: 0.1771 +Epoch [192/200], Iter [100/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.1446, Loss2: 0.1406 +Epoch [192/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1257, Loss2: 0.1137 +Epoch [192/200], Iter [200/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.2180, Loss2: 0.2347 +Epoch [192/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1349, Loss2: 0.1520 +Epoch [192/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 71.8750, Loss1: 0.1546, Loss2: 0.1622 +Epoch [192/200], Iter [350/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.2063, Loss2: 0.1947 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 54.0565 % Model2 56.1799 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1180, Loss2: 0.1251 +Epoch [193/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1229, Loss2: 0.1178 +Epoch [193/200], Iter [150/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.7812, Loss1: 0.1246, Loss2: 0.1337 +Epoch [193/200], Iter [200/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.1745, Loss2: 0.1532 +Epoch [193/200], Iter [250/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.1826, Loss2: 0.1972 +Epoch [193/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 75.7812, Loss1: 0.1746, Loss2: 0.1644 +Epoch [193/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 74.2188, Loss1: 0.1336, Loss2: 0.1148 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 53.9864 % Model2 56.0497 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.1336, Loss2: 0.1369 +Epoch [194/200], Iter [100/390] Training Accuracy1: 84.3750, Training Accuracy2: 78.9062, Loss1: 0.1448, Loss2: 0.1852 +Epoch [194/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.1079, Loss2: 0.1180 +Epoch [194/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 76.5625, Loss1: 0.2309, Loss2: 0.2287 +Epoch [194/200], Iter [250/390] Training Accuracy1: 78.1250, Training Accuracy2: 78.1250, Loss1: 0.1237, Loss2: 0.1249 +Epoch [194/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 76.5625, Loss1: 0.1648, Loss2: 0.1597 +Epoch [194/200], Iter [350/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.2072, Loss2: 0.2066 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 54.1466 % Model2 56.0697 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.1250, Loss1: 0.1973, Loss2: 0.1909 +Epoch [195/200], Iter [100/390] Training Accuracy1: 77.3438, Training Accuracy2: 77.3438, Loss1: 0.1334, Loss2: 0.1359 +Epoch [195/200], Iter [150/390] Training Accuracy1: 80.4688, Training Accuracy2: 80.4688, Loss1: 0.2480, Loss2: 0.2644 +Epoch [195/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 78.9062, Loss1: 0.1693, Loss2: 0.1560 +Epoch [195/200], Iter [250/390] Training Accuracy1: 77.3438, Training Accuracy2: 78.1250, Loss1: 0.2063, Loss2: 0.1938 +Epoch [195/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1128, Loss2: 0.1065 +Epoch [195/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.1261, Loss2: 0.1180 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 54.0565 % Model2 56.2099 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 74.2188, Loss1: 0.1251, Loss2: 0.1239 +Epoch [196/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1740, Loss2: 0.1673 +Epoch [196/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.1346, Loss2: 0.1356 +Epoch [196/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 69.5312, Loss1: 0.1032, Loss2: 0.1112 +Epoch [196/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.1097, Loss2: 0.1052 +Epoch [196/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1302, Loss2: 0.1225 +Epoch [196/200], Iter [350/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1701, Loss2: 0.1826 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 54.1266 % Model2 55.9696 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1411, Loss2: 0.1380 +Epoch [197/200], Iter [100/390] Training Accuracy1: 73.4375, Training Accuracy2: 78.1250, Loss1: 0.1703, Loss2: 0.1477 +Epoch [197/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1235, Loss2: 0.1220 +Epoch [197/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1160, Loss2: 0.1126 +Epoch [197/200], Iter [250/390] Training Accuracy1: 78.9062, Training Accuracy2: 79.6875, Loss1: 0.1831, Loss2: 0.1797 +Epoch [197/200], Iter [300/390] Training Accuracy1: 73.4375, Training Accuracy2: 76.5625, Loss1: 0.1159, Loss2: 0.1076 +Epoch [197/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1248, Loss2: 0.1255 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 53.9363 % Model2 55.9495 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 77.3438, Training Accuracy2: 75.0000, Loss1: 0.1410, Loss2: 0.1545 +Epoch [198/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 75.7812, Loss1: 0.1170, Loss2: 0.1051 +Epoch [198/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1380, Loss2: 0.1494 +Epoch [198/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 78.1250, Loss1: 0.1491, Loss2: 0.1204 +Epoch [198/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 75.0000, Loss1: 0.1246, Loss2: 0.1478 +Epoch [198/200], Iter [300/390] Training Accuracy1: 82.0312, Training Accuracy2: 82.0312, Loss1: 0.1982, Loss2: 0.1906 +Epoch [198/200], Iter [350/390] Training Accuracy1: 74.2188, Training Accuracy2: 75.0000, Loss1: 0.1512, Loss2: 0.1453 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 54.0665 % Model2 55.9395 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 74.2188, Training Accuracy2: 77.3438, Loss1: 0.1821, Loss2: 0.1677 +Epoch [199/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.1394, Loss2: 0.1387 +Epoch [199/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1415, Loss2: 0.1389 +Epoch [199/200], Iter [200/390] Training Accuracy1: 76.5625, Training Accuracy2: 77.3438, Loss1: 0.1274, Loss2: 0.1248 +Epoch [199/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1599, Loss2: 0.1598 +Epoch [199/200], Iter [300/390] Training Accuracy1: 75.7812, Training Accuracy2: 73.4375, Loss1: 0.1251, Loss2: 0.1342 +Epoch [199/200], Iter [350/390] Training Accuracy1: 80.4688, Training Accuracy2: 82.0312, Loss1: 0.2461, Loss2: 0.2414 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 54.1567 % Model2 55.9295 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 75.0000, Training Accuracy2: 75.0000, Loss1: 0.1380, Loss2: 0.1370 +Epoch [200/200], Iter [100/390] Training Accuracy1: 78.1250, Training Accuracy2: 79.6875, Loss1: 0.3037, Loss2: 0.2742 +Epoch [200/200], Iter [150/390] Training Accuracy1: 75.0000, Training Accuracy2: 77.3438, Loss1: 0.1449, Loss2: 0.1314 +Epoch [200/200], Iter [200/390] Training Accuracy1: 75.0000, Training Accuracy2: 71.0938, Loss1: 0.1306, Loss2: 0.1533 +Epoch [200/200], Iter [250/390] Training Accuracy1: 80.4688, Training Accuracy2: 78.9062, Loss1: 0.1352, Loss2: 0.1386 +Epoch [200/200], Iter [300/390] Training Accuracy1: 75.0000, Training Accuracy2: 79.6875, Loss1: 0.1143, Loss2: 0.1024 +Epoch [200/200], Iter [350/390] Training Accuracy1: 78.1250, Training Accuracy2: 74.2188, Loss1: 0.1289, Loss2: 0.1496 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 54.0264 % Model2 55.9696 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_6_4.log b/other_methods/coteaching_plus/coteaching_plus_results/out_6_4.log new file mode 100644 index 0000000..a1f8267 --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_6_4.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.40 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 18.7500, Training Accuracy2: 18.7500, Loss1: 0.0169, Loss2: 0.0168 +Epoch [2/200], Iter [100/390] Training Accuracy1: 20.3125, Training Accuracy2: 23.4375, Loss1: 0.0171, Loss2: 0.0169 +Epoch [2/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 24.2188, Loss1: 0.0156, Loss2: 0.0154 +Epoch [2/200], Iter [200/390] Training Accuracy1: 25.7812, Training Accuracy2: 24.2188, Loss1: 0.0163, Loss2: 0.0163 +Epoch [2/200], Iter [250/390] Training Accuracy1: 26.5625, Training Accuracy2: 27.3438, Loss1: 0.0152, Loss2: 0.0152 +Epoch [2/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 32.0312, Loss1: 0.0152, Loss2: 0.0150 +Epoch [2/200], Iter [350/390] Training Accuracy1: 17.9688, Training Accuracy2: 21.8750, Loss1: 0.0166, Loss2: 0.0164 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 21.0938 % Model2 20.0020 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 36.7188, Loss1: 0.0143, Loss2: 0.0140 +Epoch [3/200], Iter [100/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0155, Loss2: 0.0153 +Epoch [3/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 29.6875, Loss1: 0.0154, Loss2: 0.0153 +Epoch [3/200], Iter [200/390] Training Accuracy1: 34.3750, Training Accuracy2: 31.2500, Loss1: 0.0151, Loss2: 0.0149 +Epoch [3/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 32.0312, Loss1: 0.0141, Loss2: 0.0137 +Epoch [3/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.1562, Loss1: 0.0147, Loss2: 0.0138 +Epoch [3/200], Iter [350/390] Training Accuracy1: 28.9062, Training Accuracy2: 31.2500, Loss1: 0.0155, Loss2: 0.0151 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 30.1382 % Model2 31.6607 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 35.9375, Loss1: 0.0139, Loss2: 0.0137 +Epoch [4/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0132, Loss2: 0.0128 +Epoch [4/200], Iter [150/390] Training Accuracy1: 30.4688, Training Accuracy2: 25.0000, Loss1: 0.0155, Loss2: 0.0157 +Epoch [4/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0136 +Epoch [4/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0138, Loss2: 0.0134 +Epoch [4/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 33.5938, Loss1: 0.0136, Loss2: 0.0132 +Epoch [4/200], Iter [350/390] Training Accuracy1: 23.4375, Training Accuracy2: 24.2188, Loss1: 0.0149, Loss2: 0.0150 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 31.4103 % Model2 32.3117 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0134, Loss2: 0.0132 +Epoch [5/200], Iter [100/390] Training Accuracy1: 26.5625, Training Accuracy2: 29.6875, Loss1: 0.0144, Loss2: 0.0143 +Epoch [5/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0133, Loss2: 0.0131 +Epoch [5/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 34.3750, Loss1: 0.0157, Loss2: 0.0154 +Epoch [5/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 35.9375, Loss1: 0.0133, Loss2: 0.0130 +Epoch [5/200], Iter [300/390] Training Accuracy1: 25.7812, Training Accuracy2: 28.9062, Loss1: 0.0147, Loss2: 0.0144 +Epoch [5/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 29.6875, Loss1: 0.0146, Loss2: 0.0141 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 29.3269 % Model2 31.1599 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 39.8438, Loss1: 0.0121, Loss2: 0.0120 +Epoch [6/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0128, Loss2: 0.0126 +Epoch [6/200], Iter [150/390] Training Accuracy1: 33.5938, Training Accuracy2: 33.5938, Loss1: 0.0133, Loss2: 0.0131 +Epoch [6/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0139, Loss2: 0.0136 +Epoch [6/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 29.6875, Loss1: 0.0135, Loss2: 0.0138 +Epoch [6/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 26.5625, Loss1: 0.0146, Loss2: 0.0148 +Epoch [6/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 46.8750, Loss1: 0.0137, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 33.3934 % Model2 32.3017 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0111 +Epoch [7/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0132, Loss2: 0.0134 +Epoch [7/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0124, Loss2: 0.0121 +Epoch [7/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0135, Loss2: 0.0133 +Epoch [7/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0139, Loss2: 0.0134 +Epoch [7/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 45.3125, Loss1: 0.0123, Loss2: 0.0121 +Epoch [7/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 35.9375, Loss1: 0.0129, Loss2: 0.0125 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 36.7889 % Model2 39.0224 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 41.4062, Loss1: 0.0113, Loss2: 0.0115 +Epoch [8/200], Iter [100/390] Training Accuracy1: 30.4688, Training Accuracy2: 31.2500, Loss1: 0.0143, Loss2: 0.0145 +Epoch [8/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0121, Loss2: 0.0116 +Epoch [8/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0115, Loss2: 0.0113 +Epoch [8/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 40.6250, Loss1: 0.0134, Loss2: 0.0127 +Epoch [8/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 38.2812, Loss1: 0.0133, Loss2: 0.0131 +Epoch [8/200], Iter [350/390] Training Accuracy1: 32.0312, Training Accuracy2: 37.5000, Loss1: 0.0139, Loss2: 0.0131 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 37.4800 % Model2 40.4948 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0122, Loss2: 0.0113 +Epoch [9/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0117, Loss2: 0.0109 +Epoch [9/200], Iter [150/390] Training Accuracy1: 32.8125, Training Accuracy2: 42.1875, Loss1: 0.0132, Loss2: 0.0122 +Epoch [9/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0127, Loss2: 0.0120 +Epoch [9/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0118, Loss2: 0.0116 +Epoch [9/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 38.2812, Loss1: 0.0118, Loss2: 0.0121 +Epoch [9/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 45.3125, Loss1: 0.0118, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 37.2596 % Model2 40.0641 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0102, Loss2: 0.0100 +Epoch [10/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0130, Loss2: 0.0134 +Epoch [10/200], Iter [150/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0127, Loss2: 0.0121 +Epoch [10/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0111, Loss2: 0.0109 +Epoch [10/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0126 +Epoch [10/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.9375, Loss1: 0.0109, Loss2: 0.0109 +Epoch [10/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0110, Loss2: 0.0104 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 36.9892 % Model2 39.7937 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 43.7500, Loss1: 0.0111, Loss2: 0.0101 +Epoch [11/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0117, Loss2: 0.0108 +Epoch [11/200], Iter [150/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0132, Loss2: 0.0127 +Epoch [11/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0104, Loss2: 0.0100 +Epoch [11/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0119, Loss2: 0.0113 +Epoch [11/200], Iter [300/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0134, Loss2: 0.0126 +Epoch [11/200], Iter [350/390] Training Accuracy1: 33.5938, Training Accuracy2: 37.5000, Loss1: 0.0135, Loss2: 0.0120 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 39.2328 % Model2 39.6034 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0111, Loss2: 0.0108 +Epoch [12/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.9375, Loss1: 0.0136, Loss2: 0.0126 +Epoch [12/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0113, Loss2: 0.0106 +Epoch [12/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 31.2500, Loss1: 0.0128, Loss2: 0.0133 +Epoch [12/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0115, Loss2: 0.0109 +Epoch [12/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0102, Loss2: 0.0101 +Epoch [12/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0120, Loss2: 0.0111 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 39.9840 % Model2 41.6266 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0114, Loss2: 0.0102 +Epoch [13/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0120, Loss2: 0.0129 +Epoch [13/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0122, Loss2: 0.0123 +Epoch [13/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0122, Loss2: 0.0115 +Epoch [13/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0110, Loss2: 0.0105 +Epoch [13/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0103, Loss2: 0.0100 +Epoch [13/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0115, Loss2: 0.0112 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 39.8638 % Model2 42.4379 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0114, Loss2: 0.0105 +Epoch [14/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.9375, Loss1: 0.0134, Loss2: 0.0131 +Epoch [14/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 46.0938, Loss1: 0.0108, Loss2: 0.0100 +Epoch [14/200], Iter [200/390] Training Accuracy1: 33.5938, Training Accuracy2: 38.2812, Loss1: 0.0108, Loss2: 0.0116 +Epoch [14/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 37.5000, Loss1: 0.0119, Loss2: 0.0122 +Epoch [14/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0106, Loss2: 0.0100 +Epoch [14/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0118, Loss2: 0.0110 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 40.0841 % Model2 41.5465 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 36.7188, Loss1: 0.0104, Loss2: 0.0108 +Epoch [15/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 47.6562, Loss1: 0.0108, Loss2: 0.0099 +Epoch [15/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 39.0625, Loss1: 0.0120, Loss2: 0.0119 +Epoch [15/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0139, Loss2: 0.0118 +Epoch [15/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0096, Loss2: 0.0100 +Epoch [15/200], Iter [300/390] Training Accuracy1: 33.5938, Training Accuracy2: 40.6250, Loss1: 0.0128, Loss2: 0.0127 +Epoch [15/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 49.2188, Loss1: 0.0113, Loss2: 0.0104 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 43.2692 % Model2 43.3093 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0113, Loss2: 0.0105 +Epoch [16/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0103, Loss2: 0.0103 +Epoch [16/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0104, Loss2: 0.0100 +Epoch [16/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0124, Loss2: 0.0116 +Epoch [16/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 45.3125, Loss1: 0.0119, Loss2: 0.0111 +Epoch [16/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 46.0938, Loss1: 0.0105, Loss2: 0.0096 +Epoch [16/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0098, Loss2: 0.0098 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 40.2444 % Model2 41.0457 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.0000, Loss1: 0.0096, Loss2: 0.0093 +Epoch [17/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0120, Loss2: 0.0110 +Epoch [17/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0105, Loss2: 0.0101 +Epoch [17/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0111, Loss2: 0.0109 +Epoch [17/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0102, Loss2: 0.0094 +Epoch [17/200], Iter [300/390] Training Accuracy1: 39.0625, Training Accuracy2: 44.5312, Loss1: 0.0124, Loss2: 0.0111 +Epoch [17/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 40.6250, Loss1: 0.0107, Loss2: 0.0112 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 39.4231 % Model2 39.3429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0115, Loss2: 0.0109 +Epoch [18/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 39.0625, Loss1: 0.0113, Loss2: 0.0112 +Epoch [18/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0097, Loss2: 0.0093 +Epoch [18/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0115, Loss2: 0.0105 +Epoch [18/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 44.5312, Loss1: 0.0106, Loss2: 0.0100 +Epoch [18/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0098, Loss2: 0.0083 +Epoch [18/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0108, Loss2: 0.0094 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 39.3229 % Model2 40.6751 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0109, Loss2: 0.0104 +Epoch [19/200], Iter [100/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0116, Loss2: 0.0108 +Epoch [19/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0102, Loss2: 0.0101 +Epoch [19/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0116, Loss2: 0.0115 +Epoch [19/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0106, Loss2: 0.0091 +Epoch [19/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0106, Loss2: 0.0100 +Epoch [19/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0117, Loss2: 0.0116 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 41.4263 % Model2 43.8301 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0107, Loss2: 0.0097 +Epoch [20/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 38.2812, Loss1: 0.0121, Loss2: 0.0121 +Epoch [20/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0102, Loss2: 0.0110 +Epoch [20/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0123, Loss2: 0.0114 +Epoch [20/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0112, Loss2: 0.0113 +Epoch [20/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0099, Loss2: 0.0098 +Epoch [20/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0111, Loss2: 0.0099 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 44.8818 % Model2 46.0938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0537, Loss2: 0.0526 +Epoch [21/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 42.1875, Loss1: 0.0512, Loss2: 0.0535 +Epoch [21/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0501, Loss2: 0.0500 +Epoch [21/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0584, Loss2: 0.0565 +Epoch [21/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0602, Loss2: 0.0597 +Epoch [21/200], Iter [300/390] Training Accuracy1: 27.3438, Training Accuracy2: 34.3750, Loss1: 0.0478, Loss2: 0.0460 +Epoch [21/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0668, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 33.7139 % Model2 36.8490 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.8438, Loss1: 0.0888, Loss2: 0.0905 +Epoch [22/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 47.6562, Loss1: 0.0593, Loss2: 0.0588 +Epoch [22/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0732, Loss2: 0.0723 +Epoch [22/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 41.4062, Loss1: 0.0492, Loss2: 0.0478 +Epoch [22/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0798, Loss2: 0.0828 +Epoch [22/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 36.7188, Loss1: 0.0725, Loss2: 0.0677 +Epoch [22/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 41.4062, Loss1: 0.0567, Loss2: 0.0587 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 36.1679 % Model2 34.7256 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0859, Loss2: 0.0853 +Epoch [23/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 35.9375, Loss1: 0.0475, Loss2: 0.0494 +Epoch [23/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 39.0625, Loss1: 0.0519, Loss2: 0.0563 +Epoch [23/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0935, Loss2: 0.0888 +Epoch [23/200], Iter [250/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0684, Loss2: 0.0643 +Epoch [23/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 34.3750, Loss1: 0.0557, Loss2: 0.0555 +Epoch [23/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0873, Loss2: 0.0850 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 37.3397 % Model2 39.7736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0717, Loss2: 0.0699 +Epoch [24/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0599, Loss2: 0.0594 +Epoch [24/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0868, Loss2: 0.0855 +Epoch [24/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0511, Loss2: 0.0504 +Epoch [24/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0768, Loss2: 0.0764 +Epoch [24/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0521, Loss2: 0.0511 +Epoch [24/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0556, Loss2: 0.0578 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 41.0056 % Model2 41.4764 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 37.5000, Loss1: 0.0542, Loss2: 0.0537 +Epoch [25/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0730, Loss2: 0.0728 +Epoch [25/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0593, Loss2: 0.0608 +Epoch [25/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.1875, Loss1: 0.0734, Loss2: 0.0711 +Epoch [25/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0474, Loss2: 0.0473 +Epoch [25/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0661, Loss2: 0.0647 +Epoch [25/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0576, Loss2: 0.0569 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 40.9455 % Model2 43.2492 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0693, Loss2: 0.0707 +Epoch [26/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 40.6250, Loss1: 0.0550, Loss2: 0.0575 +Epoch [26/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0642, Loss2: 0.0638 +Epoch [26/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 40.6250, Loss1: 0.0563, Loss2: 0.0569 +Epoch [26/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 48.4375, Loss1: 0.0687, Loss2: 0.0648 +Epoch [26/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0559, Loss2: 0.0555 +Epoch [26/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0684, Loss2: 0.0689 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 39.6234 % Model2 39.8538 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0578, Loss2: 0.0577 +Epoch [27/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0736, Loss2: 0.0728 +Epoch [27/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0698, Loss2: 0.0679 +Epoch [27/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0533, Loss2: 0.0525 +Epoch [27/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0567, Loss2: 0.0557 +Epoch [27/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0769, Loss2: 0.0786 +Epoch [27/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0594, Loss2: 0.0545 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 37.0192 % Model2 38.2712 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0684, Loss2: 0.0671 +Epoch [28/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 39.0625, Loss1: 0.0582, Loss2: 0.0585 +Epoch [28/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0611, Loss2: 0.0603 +Epoch [28/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0715, Loss2: 0.0695 +Epoch [28/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0784, Loss2: 0.0748 +Epoch [28/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0635, Loss2: 0.0643 +Epoch [28/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0664, Loss2: 0.0670 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 38.9824 % Model2 38.9523 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0697, Loss2: 0.0681 +Epoch [29/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0809, Loss2: 0.0742 +Epoch [29/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0489, Loss2: 0.0493 +Epoch [29/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0569, Loss2: 0.0575 +Epoch [29/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0602, Loss2: 0.0593 +Epoch [29/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0674, Loss2: 0.0674 +Epoch [29/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0570, Loss2: 0.0550 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 38.9123 % Model2 40.9455 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0529, Loss2: 0.0526 +Epoch [30/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0542, Loss2: 0.0528 +Epoch [30/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 40.6250, Loss1: 0.0533, Loss2: 0.0533 +Epoch [30/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0803, Loss2: 0.0822 +Epoch [30/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0716, Loss2: 0.0704 +Epoch [30/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 50.7812, Loss1: 0.0510, Loss2: 0.0475 +Epoch [30/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0626, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 36.1478 % Model2 36.9692 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0445, Loss2: 0.0446 +Epoch [31/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0530, Loss2: 0.0531 +Epoch [31/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0779, Loss2: 0.0721 +Epoch [31/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 48.4375, Loss1: 0.0561, Loss2: 0.0533 +Epoch [31/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 42.9688, Loss1: 0.0499, Loss2: 0.0467 +Epoch [31/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 48.4375, Loss1: 0.0607, Loss2: 0.0577 +Epoch [31/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0557, Loss2: 0.0555 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 41.0056 % Model2 41.9471 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 44.5312, Loss1: 0.0655, Loss2: 0.0629 +Epoch [32/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0685, Loss2: 0.0654 +Epoch [32/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0692, Loss2: 0.0617 +Epoch [32/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0492, Loss2: 0.0498 +Epoch [32/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0703, Loss2: 0.0650 +Epoch [32/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0650, Loss2: 0.0675 +Epoch [32/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0483, Loss2: 0.0475 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 38.4215 % Model2 41.9471 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0613, Loss2: 0.0607 +Epoch [33/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.9688, Loss1: 0.0572, Loss2: 0.0584 +Epoch [33/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 41.4062, Loss1: 0.0485, Loss2: 0.0497 +Epoch [33/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 47.6562, Loss1: 0.0770, Loss2: 0.0802 +Epoch [33/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 38.2812, Loss1: 0.0567, Loss2: 0.0610 +Epoch [33/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0629, Loss2: 0.0636 +Epoch [33/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0449, Loss2: 0.0450 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 37.5000 % Model2 38.9123 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0575, Loss2: 0.0549 +Epoch [34/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0486, Loss2: 0.0477 +Epoch [34/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 48.4375, Loss1: 0.0648, Loss2: 0.0607 +Epoch [34/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0676, Loss2: 0.0634 +Epoch [34/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0539, Loss2: 0.0555 +Epoch [34/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0643, Loss2: 0.0628 +Epoch [34/200], Iter [350/390] Training Accuracy1: 37.5000, Training Accuracy2: 43.7500, Loss1: 0.0491, Loss2: 0.0480 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 42.3277 % Model2 43.6799 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0606, Loss2: 0.0573 +Epoch [35/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0594, Loss2: 0.0585 +Epoch [35/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 53.1250, Loss1: 0.0506, Loss2: 0.0451 +Epoch [35/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.0938, Loss1: 0.0610, Loss2: 0.0636 +Epoch [35/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0573, Loss2: 0.0578 +Epoch [35/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0538, Loss2: 0.0547 +Epoch [35/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0593, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 40.7752 % Model2 40.2344 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 41.4062, Loss1: 0.0528, Loss2: 0.0550 +Epoch [36/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0571, Loss2: 0.0552 +Epoch [36/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.8125, Loss1: 0.0572, Loss2: 0.0537 +Epoch [36/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0620, Loss2: 0.0649 +Epoch [36/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0609, Loss2: 0.0596 +Epoch [36/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0757, Loss2: 0.0733 +Epoch [36/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0571, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 42.7284 % Model2 42.8786 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0776, Loss2: 0.0720 +Epoch [37/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0648 +Epoch [37/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0574, Loss2: 0.0558 +Epoch [37/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 42.9688, Loss1: 0.0516, Loss2: 0.0538 +Epoch [37/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 41.4062, Loss1: 0.0575, Loss2: 0.0608 +Epoch [37/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0766, Loss2: 0.0680 +Epoch [37/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0521, Loss2: 0.0498 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 39.8638 % Model2 40.5048 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 47.6562, Loss1: 0.0665, Loss2: 0.0681 +Epoch [38/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0687, Loss2: 0.0686 +Epoch [38/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0537, Loss2: 0.0538 +Epoch [38/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0568, Loss2: 0.0539 +Epoch [38/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.8750, Loss1: 0.0644, Loss2: 0.0702 +Epoch [38/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0712, Loss2: 0.0728 +Epoch [38/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0622, Loss2: 0.0583 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 41.3562 % Model2 39.8838 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0592, Loss2: 0.0575 +Epoch [39/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0690, Loss2: 0.0639 +Epoch [39/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0637, Loss2: 0.0651 +Epoch [39/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0576, Loss2: 0.0566 +Epoch [39/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0531, Loss2: 0.0534 +Epoch [39/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0576, Loss2: 0.0575 +Epoch [39/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0584, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 39.6635 % Model2 41.9571 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 52.3438, Loss1: 0.0573, Loss2: 0.0520 +Epoch [40/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0732, Loss2: 0.0700 +Epoch [40/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0574, Loss2: 0.0589 +Epoch [40/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.0938, Loss1: 0.0528, Loss2: 0.0498 +Epoch [40/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0492, Loss2: 0.0484 +Epoch [40/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0541, Loss2: 0.0522 +Epoch [40/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0664, Loss2: 0.0611 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 40.5449 % Model2 41.3662 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0637, Loss2: 0.0623 +Epoch [41/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0586, Loss2: 0.0586 +Epoch [41/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0662, Loss2: 0.0677 +Epoch [41/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0727, Loss2: 0.0759 +Epoch [41/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 49.2188, Loss1: 0.0665, Loss2: 0.0644 +Epoch [41/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0476, Loss2: 0.0461 +Epoch [41/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0586, Loss2: 0.0583 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 38.0809 % Model2 39.2228 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0655, Loss2: 0.0614 +Epoch [42/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0847, Loss2: 0.0882 +Epoch [42/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0625, Loss2: 0.0654 +Epoch [42/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0523, Loss2: 0.0509 +Epoch [42/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0640, Loss2: 0.0665 +Epoch [42/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0449, Loss2: 0.0440 +Epoch [42/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0601, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 41.1558 % Model2 40.4447 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0566, Loss2: 0.0570 +Epoch [43/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0730, Loss2: 0.0689 +Epoch [43/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0633, Loss2: 0.0625 +Epoch [43/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0473, Loss2: 0.0467 +Epoch [43/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0501, Loss2: 0.0484 +Epoch [43/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0576, Loss2: 0.0571 +Epoch [43/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 47.6562, Loss1: 0.0509, Loss2: 0.0498 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 41.4563 % Model2 41.6466 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 43.7500, Loss1: 0.0519, Loss2: 0.0513 +Epoch [44/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 56.2500, Loss1: 0.0583, Loss2: 0.0542 +Epoch [44/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 51.5625, Loss1: 0.0564, Loss2: 0.0603 +Epoch [44/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0541, Loss2: 0.0534 +Epoch [44/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0572, Loss2: 0.0563 +Epoch [44/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0675, Loss2: 0.0646 +Epoch [44/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0573, Loss2: 0.0579 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 41.1158 % Model2 40.6450 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0668, Loss2: 0.0644 +Epoch [45/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0631, Loss2: 0.0647 +Epoch [45/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0539, Loss2: 0.0537 +Epoch [45/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 54.6875, Loss1: 0.0589, Loss2: 0.0545 +Epoch [45/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 45.3125, Loss1: 0.0664, Loss2: 0.0713 +Epoch [45/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0620, Loss2: 0.0618 +Epoch [45/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0493, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 39.3429 % Model2 39.5333 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.7812, Loss1: 0.0579, Loss2: 0.0539 +Epoch [46/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0538, Loss2: 0.0522 +Epoch [46/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0607, Loss2: 0.0577 +Epoch [46/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0667, Loss2: 0.0651 +Epoch [46/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0589, Loss2: 0.0567 +Epoch [46/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0603, Loss2: 0.0607 +Epoch [46/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0689, Loss2: 0.0678 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 39.3129 % Model2 41.3762 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0622, Loss2: 0.0587 +Epoch [47/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0674, Loss2: 0.0642 +Epoch [47/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0671, Loss2: 0.0665 +Epoch [47/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0514, Loss2: 0.0529 +Epoch [47/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0521, Loss2: 0.0504 +Epoch [47/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 44.5312, Loss1: 0.0577, Loss2: 0.0611 +Epoch [47/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0704, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 40.3446 % Model2 40.1042 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0751, Loss2: 0.0721 +Epoch [48/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 44.5312, Loss1: 0.0555, Loss2: 0.0582 +Epoch [48/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0652, Loss2: 0.0617 +Epoch [48/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0513, Loss2: 0.0473 +Epoch [48/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0636, Loss2: 0.0594 +Epoch [48/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.1875, Loss1: 0.0490, Loss2: 0.0501 +Epoch [48/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0684, Loss2: 0.0691 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 39.8037 % Model2 41.9872 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0573, Loss2: 0.0570 +Epoch [49/200], Iter [100/390] Training Accuracy1: 42.1875, Training Accuracy2: 48.4375, Loss1: 0.0539, Loss2: 0.0509 +Epoch [49/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0661, Loss2: 0.0658 +Epoch [49/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0673, Loss2: 0.0648 +Epoch [49/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0618, Loss2: 0.0582 +Epoch [49/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0680, Loss2: 0.0712 +Epoch [49/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0580, Loss2: 0.0574 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 38.8822 % Model2 40.1342 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0470, Loss2: 0.0461 +Epoch [50/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0579, Loss2: 0.0575 +Epoch [50/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0550, Loss2: 0.0569 +Epoch [50/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0567, Loss2: 0.0537 +Epoch [50/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0642, Loss2: 0.0607 +Epoch [50/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0524, Loss2: 0.0524 +Epoch [50/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0652, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 39.2428 % Model2 39.7436 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0513, Loss2: 0.0523 +Epoch [51/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0671, Loss2: 0.0624 +Epoch [51/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0668, Loss2: 0.0677 +Epoch [51/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0622, Loss2: 0.0595 +Epoch [51/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0653, Loss2: 0.0656 +Epoch [51/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0505, Loss2: 0.0478 +Epoch [51/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0560, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 39.4231 % Model2 38.8121 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0580, Loss2: 0.0564 +Epoch [52/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0798, Loss2: 0.0742 +Epoch [52/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0596, Loss2: 0.0600 +Epoch [52/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0678, Loss2: 0.0662 +Epoch [52/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0668, Loss2: 0.0618 +Epoch [52/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 54.6875, Loss1: 0.0638, Loss2: 0.0585 +Epoch [52/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0707, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 41.6967 % Model2 42.1374 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0607, Loss2: 0.0593 +Epoch [53/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0582, Loss2: 0.0590 +Epoch [53/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.8750, Loss1: 0.0646, Loss2: 0.0639 +Epoch [53/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0621, Loss2: 0.0621 +Epoch [53/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0779, Loss2: 0.0740 +Epoch [53/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0694, Loss2: 0.0645 +Epoch [53/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0494, Loss2: 0.0499 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 38.8421 % Model2 40.9856 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0595, Loss2: 0.0609 +Epoch [54/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0535, Loss2: 0.0519 +Epoch [54/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0666, Loss2: 0.0678 +Epoch [54/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0666, Loss2: 0.0625 +Epoch [54/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 46.0938, Loss1: 0.0577, Loss2: 0.0575 +Epoch [54/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0558, Loss2: 0.0558 +Epoch [54/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0624, Loss2: 0.0612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 40.8353 % Model2 40.9555 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0712, Loss2: 0.0693 +Epoch [55/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0624 +Epoch [55/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0570, Loss2: 0.0567 +Epoch [55/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0750, Loss2: 0.0753 +Epoch [55/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0582, Loss2: 0.0549 +Epoch [55/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0791, Loss2: 0.0753 +Epoch [55/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0577, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 39.5733 % Model2 41.0156 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0531, Loss2: 0.0489 +Epoch [56/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0577, Loss2: 0.0576 +Epoch [56/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0620, Loss2: 0.0584 +Epoch [56/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0827, Loss2: 0.0777 +Epoch [56/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0589, Loss2: 0.0577 +Epoch [56/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0635, Loss2: 0.0612 +Epoch [56/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0558, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 40.3145 % Model2 39.1526 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0782, Loss2: 0.0766 +Epoch [57/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0428, Loss2: 0.0438 +Epoch [57/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0584, Loss2: 0.0582 +Epoch [57/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0599, Loss2: 0.0590 +Epoch [57/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0613, Loss2: 0.0576 +Epoch [57/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0483, Loss2: 0.0467 +Epoch [57/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0557, Loss2: 0.0543 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 40.1042 % Model2 40.1643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0531, Loss2: 0.0537 +Epoch [58/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0575, Loss2: 0.0556 +Epoch [58/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0563, Loss2: 0.0551 +Epoch [58/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 42.9688, Loss1: 0.0440, Loss2: 0.0449 +Epoch [58/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0613, Loss2: 0.0614 +Epoch [58/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.9062, Loss1: 0.0647, Loss2: 0.0582 +Epoch [58/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0594, Loss2: 0.0564 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 42.4479 % Model2 40.7752 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 45.3125, Loss1: 0.0639, Loss2: 0.0653 +Epoch [59/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0560, Loss2: 0.0564 +Epoch [59/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0641, Loss2: 0.0610 +Epoch [59/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0516, Loss2: 0.0531 +Epoch [59/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0656, Loss2: 0.0609 +Epoch [59/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 53.9062, Loss1: 0.0623, Loss2: 0.0573 +Epoch [59/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0596, Loss2: 0.0585 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 39.9639 % Model2 41.1859 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0661, Loss2: 0.0646 +Epoch [60/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0537, Loss2: 0.0533 +Epoch [60/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0524, Loss2: 0.0503 +Epoch [60/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 51.5625, Loss1: 0.0571, Loss2: 0.0549 +Epoch [60/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 59.3750, Loss1: 0.0598, Loss2: 0.0550 +Epoch [60/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0540, Loss2: 0.0547 +Epoch [60/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0687, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 41.1258 % Model2 41.9471 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0711, Loss2: 0.0648 +Epoch [61/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0608, Loss2: 0.0631 +Epoch [61/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0478, Loss2: 0.0461 +Epoch [61/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0615, Loss2: 0.0574 +Epoch [61/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0467, Loss2: 0.0474 +Epoch [61/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0588, Loss2: 0.0578 +Epoch [61/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0539, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 41.2260 % Model2 41.1158 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0508, Loss2: 0.0469 +Epoch [62/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0606, Loss2: 0.0569 +Epoch [62/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0663, Loss2: 0.0653 +Epoch [62/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0540, Loss2: 0.0517 +Epoch [62/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0651, Loss2: 0.0616 +Epoch [62/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0509, Loss2: 0.0507 +Epoch [62/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0645, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 39.7135 % Model2 41.8169 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0636, Loss2: 0.0601 +Epoch [63/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0660, Loss2: 0.0613 +Epoch [63/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0477, Loss2: 0.0467 +Epoch [63/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0572, Loss2: 0.0546 +Epoch [63/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0597, Loss2: 0.0565 +Epoch [63/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0760, Loss2: 0.0715 +Epoch [63/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0591, Loss2: 0.0612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 40.9355 % Model2 39.6034 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0548, Loss2: 0.0513 +Epoch [64/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0540, Loss2: 0.0561 +Epoch [64/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0624, Loss2: 0.0597 +Epoch [64/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0672, Loss2: 0.0660 +Epoch [64/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0566, Loss2: 0.0519 +Epoch [64/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0693, Loss2: 0.0687 +Epoch [64/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0658, Loss2: 0.0678 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 40.5849 % Model2 40.5849 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0540, Loss2: 0.0534 +Epoch [65/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0681, Loss2: 0.0624 +Epoch [65/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0774, Loss2: 0.0737 +Epoch [65/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0683, Loss2: 0.0647 +Epoch [65/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0554, Loss2: 0.0523 +Epoch [65/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0667, Loss2: 0.0653 +Epoch [65/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0584, Loss2: 0.0562 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 41.4363 % Model2 41.2660 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0581, Loss2: 0.0560 +Epoch [66/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0619, Loss2: 0.0605 +Epoch [66/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0712, Loss2: 0.0690 +Epoch [66/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 53.1250, Loss1: 0.0459, Loss2: 0.0416 +Epoch [66/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0469, Loss2: 0.0469 +Epoch [66/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0615, Loss2: 0.0579 +Epoch [66/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0632, Loss2: 0.0589 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 39.3129 % Model2 41.2560 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 44.5312, Loss1: 0.0466, Loss2: 0.0506 +Epoch [67/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0786, Loss2: 0.0779 +Epoch [67/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0556, Loss2: 0.0562 +Epoch [67/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0643, Loss2: 0.0644 +Epoch [67/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 48.4375, Loss1: 0.0496, Loss2: 0.0455 +Epoch [67/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0423, Loss2: 0.0421 +Epoch [67/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0538, Loss2: 0.0540 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 38.9223 % Model2 39.9339 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0673, Loss2: 0.0656 +Epoch [68/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0726, Loss2: 0.0745 +Epoch [68/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0550, Loss2: 0.0572 +Epoch [68/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0619, Loss2: 0.0627 +Epoch [68/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.0000, Loss1: 0.0653, Loss2: 0.0671 +Epoch [68/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0579, Loss2: 0.0566 +Epoch [68/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0677, Loss2: 0.0654 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 39.2628 % Model2 40.6050 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0582, Loss2: 0.0560 +Epoch [69/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0728, Loss2: 0.0716 +Epoch [69/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0670, Loss2: 0.0656 +Epoch [69/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 43.7500, Loss1: 0.0468, Loss2: 0.0482 +Epoch [69/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0622, Loss2: 0.0561 +Epoch [69/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0627, Loss2: 0.0634 +Epoch [69/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0558, Loss2: 0.0551 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 41.0256 % Model2 39.4030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0501, Loss2: 0.0519 +Epoch [70/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0548, Loss2: 0.0565 +Epoch [70/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0652, Loss2: 0.0663 +Epoch [70/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0664, Loss2: 0.0634 +Epoch [70/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0568, Loss2: 0.0567 +Epoch [70/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0652, Loss2: 0.0664 +Epoch [70/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0655, Loss2: 0.0677 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 40.3446 % Model2 40.2143 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0436, Loss2: 0.0465 +Epoch [71/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0714, Loss2: 0.0744 +Epoch [71/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0575, Loss2: 0.0556 +Epoch [71/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0642, Loss2: 0.0624 +Epoch [71/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 46.0938, Loss1: 0.0479, Loss2: 0.0512 +Epoch [71/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0512, Loss2: 0.0497 +Epoch [71/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0602, Loss2: 0.0572 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 39.7236 % Model2 41.4964 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0615, Loss2: 0.0582 +Epoch [72/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0593, Loss2: 0.0558 +Epoch [72/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0519, Loss2: 0.0546 +Epoch [72/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 53.9062, Loss1: 0.0661, Loss2: 0.0597 +Epoch [72/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0519, Loss2: 0.0479 +Epoch [72/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0564, Loss2: 0.0575 +Epoch [72/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0608, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 40.2644 % Model2 41.7468 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0677, Loss2: 0.0653 +Epoch [73/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0618, Loss2: 0.0607 +Epoch [73/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0596, Loss2: 0.0593 +Epoch [73/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0520, Loss2: 0.0502 +Epoch [73/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0762 +Epoch [73/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0544, Loss2: 0.0545 +Epoch [73/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0592, Loss2: 0.0566 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 39.3530 % Model2 40.7552 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0752, Loss2: 0.0785 +Epoch [74/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.1250, Loss1: 0.0624, Loss2: 0.0651 +Epoch [74/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0663, Loss2: 0.0646 +Epoch [74/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 60.1562, Loss1: 0.0514, Loss2: 0.0481 +Epoch [74/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.0312, Loss1: 0.0654, Loss2: 0.0589 +Epoch [74/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0655, Loss2: 0.0655 +Epoch [74/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0613, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 39.7035 % Model2 39.4631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 52.3438, Loss1: 0.0600, Loss2: 0.0628 +Epoch [75/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0641, Loss2: 0.0641 +Epoch [75/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0531, Loss2: 0.0547 +Epoch [75/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0493, Loss2: 0.0515 +Epoch [75/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0648, Loss2: 0.0616 +Epoch [75/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0571, Loss2: 0.0578 +Epoch [75/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0631, Loss2: 0.0588 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 41.3662 % Model2 40.3345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 59.3750, Loss1: 0.0765, Loss2: 0.0670 +Epoch [76/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0598, Loss2: 0.0627 +Epoch [76/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0667, Loss2: 0.0654 +Epoch [76/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0579, Loss2: 0.0556 +Epoch [76/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0574, Loss2: 0.0547 +Epoch [76/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 43.7500, Loss1: 0.0513, Loss2: 0.0526 +Epoch [76/200], Iter [350/390] Training Accuracy1: 46.8750, Training Accuracy2: 57.0312, Loss1: 0.0554, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 39.0325 % Model2 40.5248 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0581, Loss2: 0.0546 +Epoch [77/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0762, Loss2: 0.0705 +Epoch [77/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0664, Loss2: 0.0618 +Epoch [77/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0522, Loss2: 0.0503 +Epoch [77/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0713, Loss2: 0.0722 +Epoch [77/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0641, Loss2: 0.0598 +Epoch [77/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0724, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 38.8922 % Model2 39.8337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0646, Loss2: 0.0608 +Epoch [78/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0715, Loss2: 0.0677 +Epoch [78/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0583, Loss2: 0.0572 +Epoch [78/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0628, Loss2: 0.0575 +Epoch [78/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0596, Loss2: 0.0603 +Epoch [78/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0561, Loss2: 0.0544 +Epoch [78/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0573, Loss2: 0.0537 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 38.4816 % Model2 38.2812 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0678, Loss2: 0.0661 +Epoch [79/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0605, Loss2: 0.0588 +Epoch [79/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0597, Loss2: 0.0570 +Epoch [79/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0745, Loss2: 0.0687 +Epoch [79/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0788, Loss2: 0.0750 +Epoch [79/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0578, Loss2: 0.0526 +Epoch [79/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0606, Loss2: 0.0607 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 40.1342 % Model2 41.7668 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0578, Loss2: 0.0562 +Epoch [80/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0665, Loss2: 0.0661 +Epoch [80/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0605, Loss2: 0.0593 +Epoch [80/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0675, Loss2: 0.0655 +Epoch [80/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0617, Loss2: 0.0614 +Epoch [80/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0565, Loss2: 0.0579 +Epoch [80/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0632, Loss2: 0.0630 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 41.3762 % Model2 39.3129 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0725, Loss2: 0.0699 +Epoch [81/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0598, Loss2: 0.0603 +Epoch [81/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0542, Loss2: 0.0542 +Epoch [81/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 54.6875, Loss1: 0.0514, Loss2: 0.0466 +Epoch [81/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 57.8125, Loss1: 0.0552, Loss2: 0.0560 +Epoch [81/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0638, Loss2: 0.0639 +Epoch [81/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0478, Loss2: 0.0443 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 39.2929 % Model2 38.5517 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0595, Loss2: 0.0583 +Epoch [82/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0551, Loss2: 0.0512 +Epoch [82/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0526, Loss2: 0.0506 +Epoch [82/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0530, Loss2: 0.0521 +Epoch [82/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0675, Loss2: 0.0691 +Epoch [82/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0531, Loss2: 0.0528 +Epoch [82/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0610, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 40.3446 % Model2 39.5232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0485, Loss2: 0.0489 +Epoch [83/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0571, Loss2: 0.0554 +Epoch [83/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 56.2500, Loss1: 0.0504, Loss2: 0.0525 +Epoch [83/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0691, Loss2: 0.0643 +Epoch [83/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0581, Loss2: 0.0558 +Epoch [83/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0798, Loss2: 0.0839 +Epoch [83/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0629, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 39.9339 % Model2 40.0541 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0553, Loss2: 0.0523 +Epoch [84/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0584, Loss2: 0.0586 +Epoch [84/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0609, Loss2: 0.0601 +Epoch [84/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0607, Loss2: 0.0612 +Epoch [84/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 54.6875, Loss1: 0.0490, Loss2: 0.0506 +Epoch [84/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0561, Loss2: 0.0538 +Epoch [84/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0542, Loss2: 0.0497 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 40.0541 % Model2 39.7035 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0465, Loss2: 0.0465 +Epoch [85/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0647, Loss2: 0.0623 +Epoch [85/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0566, Loss2: 0.0577 +Epoch [85/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0663, Loss2: 0.0672 +Epoch [85/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0662, Loss2: 0.0634 +Epoch [85/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 56.2500, Loss1: 0.0540, Loss2: 0.0507 +Epoch [85/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0580, Loss2: 0.0541 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 40.0942 % Model2 41.0757 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0738, Loss2: 0.0675 +Epoch [86/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0621, Loss2: 0.0575 +Epoch [86/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0758, Loss2: 0.0743 +Epoch [86/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0467, Loss2: 0.0462 +Epoch [86/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 50.0000, Loss1: 0.0538, Loss2: 0.0597 +Epoch [86/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0672, Loss2: 0.0726 +Epoch [86/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0558, Loss2: 0.0524 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 41.0857 % Model2 40.7752 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0497, Loss2: 0.0498 +Epoch [87/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0501, Loss2: 0.0496 +Epoch [87/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0676, Loss2: 0.0672 +Epoch [87/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0516, Loss2: 0.0518 +Epoch [87/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0588, Loss2: 0.0561 +Epoch [87/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0688, Loss2: 0.0670 +Epoch [87/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0631, Loss2: 0.0626 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 41.1759 % Model2 40.4147 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 55.4688, Loss1: 0.0607, Loss2: 0.0551 +Epoch [88/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0623, Loss2: 0.0652 +Epoch [88/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0687, Loss2: 0.0653 +Epoch [88/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0548, Loss2: 0.0524 +Epoch [88/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0559, Loss2: 0.0540 +Epoch [88/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 63.2812, Loss1: 0.0527, Loss2: 0.0492 +Epoch [88/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0547, Loss2: 0.0525 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 39.1226 % Model2 38.7220 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0744, Loss2: 0.0759 +Epoch [89/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0750, Loss2: 0.0680 +Epoch [89/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0790, Loss2: 0.0727 +Epoch [89/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0578, Loss2: 0.0546 +Epoch [89/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0563, Loss2: 0.0586 +Epoch [89/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0602, Loss2: 0.0584 +Epoch [89/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0600, Loss2: 0.0608 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 37.9607 % Model2 39.2829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0725, Loss2: 0.0690 +Epoch [90/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0553, Loss2: 0.0566 +Epoch [90/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0525, Loss2: 0.0526 +Epoch [90/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0518, Loss2: 0.0507 +Epoch [90/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0681, Loss2: 0.0647 +Epoch [90/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0519, Loss2: 0.0495 +Epoch [90/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0654, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 40.2244 % Model2 39.5933 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0607, Loss2: 0.0586 +Epoch [91/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0740, Loss2: 0.0727 +Epoch [91/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 57.8125, Loss1: 0.0482, Loss2: 0.0441 +Epoch [91/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0576, Loss2: 0.0609 +Epoch [91/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0627, Loss2: 0.0602 +Epoch [91/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.8125, Loss1: 0.0530, Loss2: 0.0507 +Epoch [91/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0713, Loss2: 0.0705 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 40.5849 % Model2 40.0441 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0573, Loss2: 0.0561 +Epoch [92/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0655, Loss2: 0.0649 +Epoch [92/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0660, Loss2: 0.0632 +Epoch [92/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0620, Loss2: 0.0578 +Epoch [92/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0762, Loss2: 0.0701 +Epoch [92/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0543, Loss2: 0.0518 +Epoch [92/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0515, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 40.5449 % Model2 39.9239 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0736, Loss2: 0.0714 +Epoch [93/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0691, Loss2: 0.0637 +Epoch [93/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0479, Loss2: 0.0455 +Epoch [93/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0704, Loss2: 0.0672 +Epoch [93/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0756, Loss2: 0.0790 +Epoch [93/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0574, Loss2: 0.0592 +Epoch [93/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 48.4375, Loss1: 0.0482, Loss2: 0.0495 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 40.5549 % Model2 40.3446 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0676, Loss2: 0.0642 +Epoch [94/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 63.2812, Loss1: 0.0539, Loss2: 0.0505 +Epoch [94/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 56.2500, Loss1: 0.0785, Loss2: 0.0879 +Epoch [94/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0682, Loss2: 0.0629 +Epoch [94/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 55.4688, Loss1: 0.0637, Loss2: 0.0594 +Epoch [94/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0605, Loss2: 0.0545 +Epoch [94/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0617, Loss2: 0.0627 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 39.4131 % Model2 39.2328 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.0312, Loss1: 0.0551, Loss2: 0.0575 +Epoch [95/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0477, Loss2: 0.0496 +Epoch [95/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0624, Loss2: 0.0583 +Epoch [95/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0691, Loss2: 0.0687 +Epoch [95/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0621, Loss2: 0.0619 +Epoch [95/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0595, Loss2: 0.0583 +Epoch [95/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0545, Loss2: 0.0534 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 40.1342 % Model2 40.8253 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0698, Loss2: 0.0649 +Epoch [96/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.8750, Loss1: 0.0659, Loss2: 0.0696 +Epoch [96/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0589, Loss2: 0.0580 +Epoch [96/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0654, Loss2: 0.0605 +Epoch [96/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 64.0625, Loss1: 0.0633, Loss2: 0.0561 +Epoch [96/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0780, Loss2: 0.0757 +Epoch [96/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0615, Loss2: 0.0606 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 38.5617 % Model2 40.7552 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0912, Loss2: 0.0977 +Epoch [97/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0571, Loss2: 0.0553 +Epoch [97/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0688, Loss2: 0.0717 +Epoch [97/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0815, Loss2: 0.0789 +Epoch [97/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0538, Loss2: 0.0534 +Epoch [97/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0602, Loss2: 0.0553 +Epoch [97/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0662, Loss2: 0.0668 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 39.7236 % Model2 39.9539 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0555, Loss2: 0.0538 +Epoch [98/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0593, Loss2: 0.0573 +Epoch [98/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0613, Loss2: 0.0586 +Epoch [98/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0595, Loss2: 0.0610 +Epoch [98/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0528, Loss2: 0.0541 +Epoch [98/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0726, Loss2: 0.0774 +Epoch [98/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0527, Loss2: 0.0489 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 38.7320 % Model2 39.6635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0656, Loss2: 0.0632 +Epoch [99/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 53.9062, Loss1: 0.0675, Loss2: 0.0763 +Epoch [99/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0533, Loss2: 0.0513 +Epoch [99/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0685, Loss2: 0.0625 +Epoch [99/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0683, Loss2: 0.0669 +Epoch [99/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0663, Loss2: 0.0712 +Epoch [99/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0671, Loss2: 0.0631 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 39.1526 % Model2 39.9038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0807, Loss2: 0.0793 +Epoch [100/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0539, Loss2: 0.0520 +Epoch [100/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0584, Loss2: 0.0578 +Epoch [100/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0706, Loss2: 0.0720 +Epoch [100/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0589, Loss2: 0.0565 +Epoch [100/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0497, Loss2: 0.0493 +Epoch [100/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0669, Loss2: 0.0665 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 40.6050 % Model2 40.7252 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 62.5000, Loss1: 0.1085, Loss2: 0.1265 +Epoch [101/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.7812, Loss1: 0.0666, Loss2: 0.0700 +Epoch [101/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0638, Loss2: 0.0610 +Epoch [101/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0496, Loss2: 0.0482 +Epoch [101/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0588, Loss2: 0.0562 +Epoch [101/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0662, Loss2: 0.0642 +Epoch [101/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0783, Loss2: 0.0727 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 40.1042 % Model2 39.7336 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0609, Loss2: 0.0567 +Epoch [102/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0576, Loss2: 0.0596 +Epoch [102/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0649, Loss2: 0.0627 +Epoch [102/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0666, Loss2: 0.0609 +Epoch [102/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0648, Loss2: 0.0640 +Epoch [102/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0600, Loss2: 0.0598 +Epoch [102/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0732, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.3830 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0708, Loss2: 0.0671 +Epoch [103/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0590, Loss2: 0.0575 +Epoch [103/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0700, Loss2: 0.0722 +Epoch [103/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0679, Loss2: 0.0694 +Epoch [103/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0649, Loss2: 0.0610 +Epoch [103/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0573, Loss2: 0.0512 +Epoch [103/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0660, Loss2: 0.0685 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 41.1759 % Model2 39.7336 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0668, Loss2: 0.0674 +Epoch [104/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0769, Loss2: 0.0733 +Epoch [104/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0657, Loss2: 0.0626 +Epoch [104/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0670, Loss2: 0.0647 +Epoch [104/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0685, Loss2: 0.0669 +Epoch [104/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0707, Loss2: 0.0667 +Epoch [104/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0926, Loss2: 0.0871 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 40.1542 % Model2 40.6350 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0574, Loss2: 0.0572 +Epoch [105/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 52.3438, Loss1: 0.0635, Loss2: 0.0712 +Epoch [105/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0647, Loss2: 0.0634 +Epoch [105/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0631, Loss2: 0.0654 +Epoch [105/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 63.2812, Loss1: 0.0649, Loss2: 0.0575 +Epoch [105/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0583, Loss2: 0.0542 +Epoch [105/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0619, Loss2: 0.0630 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 40.2544 % Model2 39.8237 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0546, Loss2: 0.0543 +Epoch [106/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 60.1562, Loss1: 0.0698, Loss2: 0.0627 +Epoch [106/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0671, Loss2: 0.0695 +Epoch [106/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0608, Loss2: 0.0598 +Epoch [106/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0586, Loss2: 0.0565 +Epoch [106/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0585, Loss2: 0.0514 +Epoch [106/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0665, Loss2: 0.0656 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 40.5449 % Model2 40.4948 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0666, Loss2: 0.0693 +Epoch [107/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0494, Loss2: 0.0482 +Epoch [107/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0570, Loss2: 0.0580 +Epoch [107/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0569, Loss2: 0.0556 +Epoch [107/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0768, Loss2: 0.0738 +Epoch [107/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.0929, Loss2: 0.0839 +Epoch [107/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 61.7188, Loss1: 0.0702, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 40.5449 % Model2 39.8337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0896, Loss2: 0.0881 +Epoch [108/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0792, Loss2: 0.0749 +Epoch [108/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0750, Loss2: 0.0707 +Epoch [108/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0783, Loss2: 0.0758 +Epoch [108/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0744, Loss2: 0.0714 +Epoch [108/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 62.5000, Loss1: 0.0639, Loss2: 0.0591 +Epoch [108/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0589, Loss2: 0.0597 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.4832 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0706, Loss2: 0.0651 +Epoch [109/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0745, Loss2: 0.0700 +Epoch [109/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0677, Loss2: 0.0679 +Epoch [109/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0659, Loss2: 0.0699 +Epoch [109/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 53.1250, Loss1: 0.0505, Loss2: 0.0533 +Epoch [109/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0770, Loss2: 0.0757 +Epoch [109/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0690, Loss2: 0.0644 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 39.9840 % Model2 40.1643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0766, Loss2: 0.0712 +Epoch [110/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0669, Loss2: 0.0674 +Epoch [110/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0703, Loss2: 0.0662 +Epoch [110/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0625, Loss2: 0.0585 +Epoch [110/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0611, Loss2: 0.0589 +Epoch [110/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0802, Loss2: 0.0740 +Epoch [110/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0848, Loss2: 0.0810 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 41.4563 % Model2 39.8838 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0587, Loss2: 0.0568 +Epoch [111/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0621, Loss2: 0.0607 +Epoch [111/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0822, Loss2: 0.0748 +Epoch [111/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0559, Loss2: 0.0533 +Epoch [111/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 57.0312, Loss1: 0.0597, Loss2: 0.0556 +Epoch [111/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0614, Loss2: 0.0615 +Epoch [111/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0677, Loss2: 0.0676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 40.2945 % Model2 39.4231 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0660, Loss2: 0.0632 +Epoch [112/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0750, Loss2: 0.0714 +Epoch [112/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0672, Loss2: 0.0624 +Epoch [112/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0590, Loss2: 0.0528 +Epoch [112/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0747, Loss2: 0.0700 +Epoch [112/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0700, Loss2: 0.0674 +Epoch [112/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 56.2500, Loss1: 0.0575, Loss2: 0.0526 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 39.6735 % Model2 40.5349 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0615, Loss2: 0.0568 +Epoch [113/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0712, Loss2: 0.0696 +Epoch [113/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0683, Loss2: 0.0640 +Epoch [113/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0674, Loss2: 0.0672 +Epoch [113/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0677, Loss2: 0.0647 +Epoch [113/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0598, Loss2: 0.0597 +Epoch [113/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0701, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 39.3630 % Model2 40.2845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0752, Loss2: 0.0718 +Epoch [114/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0750, Loss2: 0.0799 +Epoch [114/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 60.1562, Loss1: 0.0576, Loss2: 0.0545 +Epoch [114/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0585, Loss2: 0.0567 +Epoch [114/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.8750, Loss1: 0.0772, Loss2: 0.0710 +Epoch [114/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0677, Loss2: 0.0677 +Epoch [114/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0682, Loss2: 0.0688 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 39.8237 % Model2 39.6534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0680, Loss2: 0.0679 +Epoch [115/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0820, Loss2: 0.0701 +Epoch [115/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0738, Loss2: 0.0775 +Epoch [115/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0692, Loss2: 0.0651 +Epoch [115/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0691, Loss2: 0.0741 +Epoch [115/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0540, Loss2: 0.0502 +Epoch [115/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0908, Loss2: 0.0883 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 40.0441 % Model2 40.0841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0848, Loss2: 0.0871 +Epoch [116/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0736, Loss2: 0.0689 +Epoch [116/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0609, Loss2: 0.0629 +Epoch [116/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0579, Loss2: 0.0577 +Epoch [116/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0557, Loss2: 0.0520 +Epoch [116/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0584, Loss2: 0.0590 +Epoch [116/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0572, Loss2: 0.0583 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 39.8838 % Model2 39.6835 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0755, Loss2: 0.0761 +Epoch [117/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0642, Loss2: 0.0616 +Epoch [117/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0767, Loss2: 0.0709 +Epoch [117/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0598, Loss2: 0.0548 +Epoch [117/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0851, Loss2: 0.0857 +Epoch [117/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0605, Loss2: 0.0594 +Epoch [117/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0585, Loss2: 0.0573 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 39.1927 % Model2 39.4131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0624, Loss2: 0.0613 +Epoch [118/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0714, Loss2: 0.0659 +Epoch [118/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0543, Loss2: 0.0546 +Epoch [118/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0573, Loss2: 0.0554 +Epoch [118/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0859, Loss2: 0.0768 +Epoch [118/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0681, Loss2: 0.0653 +Epoch [118/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.0312, Loss1: 0.0604, Loss2: 0.0597 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 39.7135 % Model2 39.7937 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0552, Loss2: 0.0541 +Epoch [119/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0745, Loss2: 0.0772 +Epoch [119/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0632, Loss2: 0.0604 +Epoch [119/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0762, Loss2: 0.0746 +Epoch [119/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0641, Loss2: 0.0649 +Epoch [119/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0597, Loss2: 0.0575 +Epoch [119/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0472, Loss2: 0.0456 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 40.0541 % Model2 39.1426 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0774, Loss2: 0.0755 +Epoch [120/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.1019, Loss2: 0.0907 +Epoch [120/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0779, Loss2: 0.0752 +Epoch [120/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0680, Loss2: 0.0630 +Epoch [120/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0470, Loss2: 0.0466 +Epoch [120/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0703, Loss2: 0.0687 +Epoch [120/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0720, Loss2: 0.0729 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 38.6619 % Model2 40.2945 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 62.5000, Loss1: 0.0724, Loss2: 0.0770 +Epoch [121/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 59.3750, Loss1: 0.0598, Loss2: 0.0580 +Epoch [121/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0708, Loss2: 0.0678 +Epoch [121/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0734, Loss2: 0.0733 +Epoch [121/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0847, Loss2: 0.0767 +Epoch [121/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0603, Loss2: 0.0573 +Epoch [121/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0765, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 39.9339 % Model2 40.8454 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0554, Loss2: 0.0538 +Epoch [122/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0751, Loss2: 0.0690 +Epoch [122/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0750, Loss2: 0.0741 +Epoch [122/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0666, Loss2: 0.0666 +Epoch [122/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 64.8438, Loss1: 0.0707, Loss2: 0.0623 +Epoch [122/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0664, Loss2: 0.0626 +Epoch [122/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0573, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 39.9439 % Model2 40.1042 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0688, Loss2: 0.0667 +Epoch [123/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0764, Loss2: 0.0724 +Epoch [123/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 67.9688, Loss1: 0.0670, Loss2: 0.0621 +Epoch [123/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0711, Loss2: 0.0714 +Epoch [123/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0649, Loss2: 0.0589 +Epoch [123/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0742, Loss2: 0.0751 +Epoch [123/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0743, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 40.3746 % Model2 39.2929 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0824, Loss2: 0.0784 +Epoch [124/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0936, Loss2: 0.0919 +Epoch [124/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0686, Loss2: 0.0679 +Epoch [124/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0826, Loss2: 0.0851 +Epoch [124/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0602, Loss2: 0.0603 +Epoch [124/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0637, Loss2: 0.0636 +Epoch [124/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0725, Loss2: 0.0744 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 38.6418 % Model2 39.3630 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0814, Loss2: 0.0820 +Epoch [125/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0523, Loss2: 0.0532 +Epoch [125/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0812, Loss2: 0.0785 +Epoch [125/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0884, Loss2: 0.0891 +Epoch [125/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.1093, Loss2: 0.1040 +Epoch [125/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0660, Loss2: 0.0685 +Epoch [125/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0612, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 40.0641 % Model2 40.3245 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0754, Loss2: 0.0745 +Epoch [126/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0632, Loss2: 0.0643 +Epoch [126/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0692, Loss2: 0.0656 +Epoch [126/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0787, Loss2: 0.0739 +Epoch [126/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0762, Loss2: 0.0722 +Epoch [126/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 59.3750, Loss1: 0.0636, Loss2: 0.0567 +Epoch [126/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 66.4062, Loss1: 0.0749, Loss2: 0.0615 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 40.0040 % Model2 39.8438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0709, Loss2: 0.0648 +Epoch [127/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.0625, Loss1: 0.0612, Loss2: 0.0570 +Epoch [127/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0642, Loss2: 0.0622 +Epoch [127/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0538, Loss2: 0.0542 +Epoch [127/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0735, Loss2: 0.0761 +Epoch [127/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0683, Loss2: 0.0690 +Epoch [127/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 59.3750, Loss1: 0.0621, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 40.4147 % Model2 40.0641 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.1562, Loss1: 0.0603, Loss2: 0.0640 +Epoch [128/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0665, Loss2: 0.0663 +Epoch [128/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0724, Loss2: 0.0711 +Epoch [128/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0595, Loss2: 0.0588 +Epoch [128/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0586, Loss2: 0.0548 +Epoch [128/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0863, Loss2: 0.0958 +Epoch [128/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0711, Loss2: 0.0700 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 39.5933 % Model2 39.8938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0786, Loss2: 0.0773 +Epoch [129/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 65.6250, Loss1: 0.0781, Loss2: 0.0731 +Epoch [129/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0725, Loss2: 0.0724 +Epoch [129/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0776, Loss2: 0.0718 +Epoch [129/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0701, Loss2: 0.0741 +Epoch [129/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0617, Loss2: 0.0619 +Epoch [129/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.9375, Loss1: 0.0535, Loss2: 0.0509 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.0825 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0919, Loss2: 0.1002 +Epoch [130/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 72.6562, Loss1: 0.0793, Loss2: 0.0728 +Epoch [130/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0604, Loss2: 0.0583 +Epoch [130/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0777, Loss2: 0.0761 +Epoch [130/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0820, Loss2: 0.0783 +Epoch [130/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 67.9688, Loss1: 0.0832, Loss2: 0.0735 +Epoch [130/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0650, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 40.0541 % Model2 40.3345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1031, Loss2: 0.0995 +Epoch [131/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0598, Loss2: 0.0598 +Epoch [131/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0736, Loss2: 0.0760 +Epoch [131/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 62.5000, Loss1: 0.0572, Loss2: 0.0533 +Epoch [131/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0909, Loss2: 0.0926 +Epoch [131/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0724, Loss2: 0.0815 +Epoch [131/200], Iter [350/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.8750, Loss1: 0.0906, Loss2: 0.0977 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 39.2528 % Model2 40.1542 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0829, Loss2: 0.0757 +Epoch [132/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0745, Loss2: 0.0718 +Epoch [132/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0751, Loss2: 0.0689 +Epoch [132/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 60.1562, Loss1: 0.0634, Loss2: 0.0591 +Epoch [132/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0802, Loss2: 0.0743 +Epoch [132/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0754, Loss2: 0.0779 +Epoch [132/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0939, Loss2: 0.0823 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 39.3329 % Model2 40.1342 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 58.5938, Loss1: 0.0772, Loss2: 0.0921 +Epoch [133/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.8438, Loss1: 0.0832, Loss2: 0.0762 +Epoch [133/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0599, Loss2: 0.0565 +Epoch [133/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0768, Loss2: 0.0781 +Epoch [133/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0652, Loss2: 0.0699 +Epoch [133/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0734, Loss2: 0.0688 +Epoch [133/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 63.2812, Loss1: 0.0757, Loss2: 0.0712 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 39.3930 % Model2 39.7135 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0607, Loss2: 0.0618 +Epoch [134/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0889, Loss2: 0.0936 +Epoch [134/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0594, Loss2: 0.0591 +Epoch [134/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0839, Loss2: 0.0803 +Epoch [134/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0693, Loss2: 0.0727 +Epoch [134/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0631, Loss2: 0.0634 +Epoch [134/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0755, Loss2: 0.0744 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 39.2728 % Model2 39.3930 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0893, Loss2: 0.0917 +Epoch [135/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0643, Loss2: 0.0682 +Epoch [135/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0752, Loss2: 0.0776 +Epoch [135/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0856, Loss2: 0.0902 +Epoch [135/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 69.5312, Loss1: 0.0853, Loss2: 0.0783 +Epoch [135/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0795, Loss2: 0.0781 +Epoch [135/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0881, Loss2: 0.0833 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 38.9323 % Model2 40.0641 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0744, Loss2: 0.0697 +Epoch [136/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 63.2812, Loss1: 0.0805, Loss2: 0.0837 +Epoch [136/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.0938, Loss1: 0.0661, Loss2: 0.0572 +Epoch [136/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0607, Loss2: 0.0580 +Epoch [136/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0702, Loss2: 0.0677 +Epoch [136/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0697, Loss2: 0.0707 +Epoch [136/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.8125, Loss1: 0.0797, Loss2: 0.0722 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 39.5933 % Model2 39.9339 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0780, Loss2: 0.0725 +Epoch [137/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0842, Loss2: 0.0782 +Epoch [137/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0742, Loss2: 0.0775 +Epoch [137/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0691, Loss2: 0.0701 +Epoch [137/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.1562, Loss1: 0.0616, Loss2: 0.0620 +Epoch [137/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.1010, Loss2: 0.0915 +Epoch [137/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.1562, Loss1: 0.0779, Loss2: 0.0860 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 39.0425 % Model2 39.9840 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0804, Loss2: 0.0819 +Epoch [138/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0716, Loss2: 0.0738 +Epoch [138/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0968, Loss2: 0.0872 +Epoch [138/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0923, Loss2: 0.0904 +Epoch [138/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 69.5312, Loss1: 0.1103, Loss2: 0.0965 +Epoch [138/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 73.4375, Loss1: 0.0917, Loss2: 0.0794 +Epoch [138/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0754, Loss2: 0.0762 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 39.6334 % Model2 39.6534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0867, Loss2: 0.0832 +Epoch [139/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0760, Loss2: 0.0712 +Epoch [139/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0749, Loss2: 0.0811 +Epoch [139/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0679, Loss2: 0.0647 +Epoch [139/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.0865, Loss2: 0.0834 +Epoch [139/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0588, Loss2: 0.0575 +Epoch [139/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0798, Loss2: 0.0772 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 40.3345 % Model2 40.1643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0754, Loss2: 0.0703 +Epoch [140/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0739, Loss2: 0.0816 +Epoch [140/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0910, Loss2: 0.0907 +Epoch [140/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 70.3125, Loss1: 0.0769, Loss2: 0.0686 +Epoch [140/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0670, Loss2: 0.0640 +Epoch [140/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0755, Loss2: 0.0764 +Epoch [140/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.9375, Loss1: 0.0676, Loss2: 0.0621 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 40.0441 % Model2 40.1542 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0635, Loss2: 0.0631 +Epoch [141/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0642, Loss2: 0.0651 +Epoch [141/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0735, Loss2: 0.0760 +Epoch [141/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0938, Loss2: 0.0860 +Epoch [141/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0845, Loss2: 0.0746 +Epoch [141/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0731, Loss2: 0.0775 +Epoch [141/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0746, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 39.7636 % Model2 40.5248 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0668, Loss2: 0.0656 +Epoch [142/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0850, Loss2: 0.0857 +Epoch [142/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0808, Loss2: 0.0821 +Epoch [142/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0882, Loss2: 0.0835 +Epoch [142/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0722, Loss2: 0.0733 +Epoch [142/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0763, Loss2: 0.0734 +Epoch [142/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0727, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 39.7636 % Model2 40.2344 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 63.2812, Loss1: 0.0696, Loss2: 0.0610 +Epoch [143/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0614, Loss2: 0.0602 +Epoch [143/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0691, Loss2: 0.0693 +Epoch [143/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1228, Loss2: 0.1321 +Epoch [143/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.0795, Loss2: 0.0847 +Epoch [143/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0798, Loss2: 0.0776 +Epoch [143/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0665, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 39.7937 % Model2 39.7937 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 59.3750, Loss1: 0.0797, Loss2: 0.0849 +Epoch [144/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0798, Loss2: 0.0778 +Epoch [144/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0764, Loss2: 0.0760 +Epoch [144/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0732 +Epoch [144/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0571, Loss2: 0.0589 +Epoch [144/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.0819, Loss2: 0.0793 +Epoch [144/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0605, Loss2: 0.0636 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 39.6935 % Model2 40.5449 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0698, Loss2: 0.0641 +Epoch [145/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.0745, Loss2: 0.0738 +Epoch [145/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0890, Loss2: 0.0854 +Epoch [145/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0822, Loss2: 0.0800 +Epoch [145/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 63.2812, Loss1: 0.0784, Loss2: 0.0720 +Epoch [145/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1049, Loss2: 0.1089 +Epoch [145/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0598, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 40.0541 % Model2 39.7336 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.7812, Loss1: 0.1057, Loss2: 0.0972 +Epoch [146/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0883, Loss2: 0.0910 +Epoch [146/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0654, Loss2: 0.0585 +Epoch [146/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0766, Loss2: 0.0755 +Epoch [146/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.0577, Loss2: 0.0600 +Epoch [146/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0774, Loss2: 0.0763 +Epoch [146/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0721, Loss2: 0.0718 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 39.6735 % Model2 39.4331 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0624, Loss2: 0.0650 +Epoch [147/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0760, Loss2: 0.0794 +Epoch [147/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0754, Loss2: 0.0717 +Epoch [147/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.1034, Loss2: 0.0997 +Epoch [147/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 61.7188, Loss1: 0.0988, Loss2: 0.1101 +Epoch [147/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0620, Loss2: 0.0593 +Epoch [147/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0878, Loss2: 0.0820 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 39.2829 % Model2 40.2444 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1189, Loss2: 0.1062 +Epoch [148/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0693, Loss2: 0.0685 +Epoch [148/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0750, Loss2: 0.0767 +Epoch [148/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0674, Loss2: 0.0663 +Epoch [148/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.0839, Loss2: 0.0816 +Epoch [148/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.0767, Loss2: 0.0779 +Epoch [148/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0762, Loss2: 0.0751 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 39.1526 % Model2 39.7035 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0676, Loss2: 0.0649 +Epoch [149/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.1875, Loss1: 0.1144, Loss2: 0.1248 +Epoch [149/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0914, Loss2: 0.0932 +Epoch [149/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0762, Loss2: 0.0769 +Epoch [149/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0675, Loss2: 0.0694 +Epoch [149/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0649, Loss2: 0.0607 +Epoch [149/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0918, Loss2: 0.0853 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 40.3045 % Model2 39.8037 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0765, Loss2: 0.0767 +Epoch [150/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 59.3750, Loss1: 0.0851, Loss2: 0.0858 +Epoch [150/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0777, Loss2: 0.0766 +Epoch [150/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.0790, Loss2: 0.0724 +Epoch [150/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 65.6250, Loss1: 0.0656, Loss2: 0.0566 +Epoch [150/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0771, Loss2: 0.0771 +Epoch [150/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0659, Loss2: 0.0643 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 40.5048 % Model2 40.4247 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.1034, Loss2: 0.1099 +Epoch [151/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0838, Loss2: 0.0850 +Epoch [151/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0853, Loss2: 0.0869 +Epoch [151/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0822, Loss2: 0.0811 +Epoch [151/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0674, Loss2: 0.0685 +Epoch [151/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0869, Loss2: 0.0907 +Epoch [151/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0746, Loss2: 0.0707 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 39.7236 % Model2 39.6134 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0835, Loss2: 0.0817 +Epoch [152/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.8438, Loss1: 0.0690, Loss2: 0.0659 +Epoch [152/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0842, Loss2: 0.0840 +Epoch [152/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.1066, Loss2: 0.1117 +Epoch [152/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0763, Loss2: 0.0737 +Epoch [152/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.9688, Loss1: 0.0743, Loss2: 0.0653 +Epoch [152/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0771, Loss2: 0.0741 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 39.7436 % Model2 39.7135 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.1154, Loss2: 0.1129 +Epoch [153/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0810, Loss2: 0.0762 +Epoch [153/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0786, Loss2: 0.0745 +Epoch [153/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.1121, Loss2: 0.1038 +Epoch [153/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0694, Loss2: 0.0727 +Epoch [153/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0713, Loss2: 0.0708 +Epoch [153/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.0924, Loss2: 0.0891 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 39.3029 % Model2 38.9223 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0876, Loss2: 0.0833 +Epoch [154/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0711, Loss2: 0.0725 +Epoch [154/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0686, Loss2: 0.0698 +Epoch [154/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 72.6562, Loss1: 0.0898, Loss2: 0.0861 +Epoch [154/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0778, Loss2: 0.0753 +Epoch [154/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0667, Loss2: 0.0677 +Epoch [154/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0876, Loss2: 0.0845 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 39.3129 % Model2 39.3429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0903, Loss2: 0.0911 +Epoch [155/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0886, Loss2: 0.0899 +Epoch [155/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0613, Loss2: 0.0630 +Epoch [155/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 75.7812, Loss1: 0.0951, Loss2: 0.0848 +Epoch [155/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 64.8438, Loss1: 0.0861, Loss2: 0.0919 +Epoch [155/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0841, Loss2: 0.0785 +Epoch [155/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0837, Loss2: 0.0869 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.7736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.1062, Loss2: 0.0994 +Epoch [156/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0872, Loss2: 0.0893 +Epoch [156/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0748, Loss2: 0.0694 +Epoch [156/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0744, Loss2: 0.0730 +Epoch [156/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0731, Loss2: 0.0737 +Epoch [156/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0871, Loss2: 0.0828 +Epoch [156/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 73.4375, Loss1: 0.0993, Loss2: 0.0879 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 39.7937 % Model2 40.0741 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0746, Loss2: 0.0731 +Epoch [157/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 70.3125, Loss1: 0.0836, Loss2: 0.0733 +Epoch [157/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 68.7500, Loss1: 0.1043, Loss2: 0.0951 +Epoch [157/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0918, Loss2: 0.0867 +Epoch [157/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 66.4062, Loss1: 0.0818, Loss2: 0.0740 +Epoch [157/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.0946, Loss2: 0.0904 +Epoch [157/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0875, Loss2: 0.0793 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 39.3329 % Model2 39.9840 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 73.4375, Loss1: 0.1092, Loss2: 0.1082 +Epoch [158/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0800, Loss2: 0.0718 +Epoch [158/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.0859, Loss2: 0.0869 +Epoch [158/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0813, Loss2: 0.0827 +Epoch [158/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0958, Loss2: 0.1004 +Epoch [158/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.0824, Loss2: 0.0784 +Epoch [158/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.8750, Loss1: 0.0922, Loss2: 0.0835 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 39.1827 % Model2 39.7436 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0916, Loss2: 0.0941 +Epoch [159/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0985, Loss2: 0.1015 +Epoch [159/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0948, Loss2: 0.0913 +Epoch [159/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1071, Loss2: 0.1128 +Epoch [159/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 70.3125, Loss1: 0.0841, Loss2: 0.0753 +Epoch [159/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 65.6250, Loss1: 0.0822, Loss2: 0.0746 +Epoch [159/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0914, Loss2: 0.0875 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 39.4732 % Model2 40.0240 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0760, Loss2: 0.0759 +Epoch [160/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0949, Loss2: 0.0952 +Epoch [160/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0930, Loss2: 0.0904 +Epoch [160/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0777, Loss2: 0.0771 +Epoch [160/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0801, Loss2: 0.0801 +Epoch [160/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 72.6562, Loss1: 0.0952, Loss2: 0.0845 +Epoch [160/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 62.5000, Loss1: 0.0716, Loss2: 0.0681 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 39.4231 % Model2 39.7736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 62.5000, Loss1: 0.0904, Loss2: 0.0976 +Epoch [161/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0709, Loss2: 0.0673 +Epoch [161/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1039, Loss2: 0.1042 +Epoch [161/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0745, Loss2: 0.0776 +Epoch [161/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 70.3125, Loss1: 0.1026, Loss2: 0.0831 +Epoch [161/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0690, Loss2: 0.0637 +Epoch [161/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1069, Loss2: 0.1080 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 39.2728 % Model2 39.3429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0914, Loss2: 0.0926 +Epoch [162/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0945, Loss2: 0.0902 +Epoch [162/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0710, Loss2: 0.0740 +Epoch [162/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.8438, Loss1: 0.0849, Loss2: 0.0804 +Epoch [162/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.1562, Loss1: 0.0730, Loss2: 0.0714 +Epoch [162/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 66.4062, Loss1: 0.0852, Loss2: 0.0919 +Epoch [162/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0865, Loss2: 0.0883 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 39.2929 % Model2 40.0240 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0729, Loss2: 0.0746 +Epoch [163/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0818, Loss2: 0.0882 +Epoch [163/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0788, Loss2: 0.0746 +Epoch [163/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1172, Loss2: 0.1146 +Epoch [163/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0693, Loss2: 0.0682 +Epoch [163/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1045, Loss2: 0.1055 +Epoch [163/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0750, Loss2: 0.0711 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 39.7135 % Model2 39.9439 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 77.3438, Loss1: 0.0828, Loss2: 0.0727 +Epoch [164/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0782, Loss2: 0.0789 +Epoch [164/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0902, Loss2: 0.0892 +Epoch [164/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0944, Loss2: 0.0939 +Epoch [164/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0761, Loss2: 0.0775 +Epoch [164/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 68.7500, Loss1: 0.0897, Loss2: 0.0943 +Epoch [164/200], Iter [350/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0841, Loss2: 0.0869 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 39.5333 % Model2 40.0441 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0973, Loss2: 0.0972 +Epoch [165/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0738, Loss2: 0.0751 +Epoch [165/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0912, Loss2: 0.0827 +Epoch [165/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 61.7188, Loss1: 0.0793, Loss2: 0.0865 +Epoch [165/200], Iter [250/390] Training Accuracy1: 71.8750, Training Accuracy2: 76.5625, Loss1: 0.0809, Loss2: 0.0773 +Epoch [165/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1014, Loss2: 0.1058 +Epoch [165/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1037, Loss2: 0.1034 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 39.2428 % Model2 39.3029 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1151, Loss2: 0.1130 +Epoch [166/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0726, Loss2: 0.0753 +Epoch [166/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 62.5000, Loss1: 0.0656, Loss2: 0.0737 +Epoch [166/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 70.3125, Loss1: 0.0863, Loss2: 0.0844 +Epoch [166/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.0948, Loss2: 0.0917 +Epoch [166/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.0898, Loss2: 0.0900 +Epoch [166/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.0961, Loss2: 0.0997 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 39.3029 % Model2 39.5433 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1098, Loss2: 0.1046 +Epoch [167/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0745, Loss2: 0.0762 +Epoch [167/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1399, Loss2: 0.1414 +Epoch [167/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1134, Loss2: 0.1075 +Epoch [167/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0729, Loss2: 0.0710 +Epoch [167/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.0820, Loss2: 0.0793 +Epoch [167/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0805, Loss2: 0.0780 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 39.6835 % Model2 39.6234 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0699, Loss2: 0.0662 +Epoch [168/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0896, Loss2: 0.0945 +Epoch [168/200], Iter [150/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0751, Loss2: 0.0765 +Epoch [168/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0973, Loss2: 0.0911 +Epoch [168/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1059, Loss2: 0.1076 +Epoch [168/200], Iter [300/390] Training Accuracy1: 71.8750, Training Accuracy2: 70.3125, Loss1: 0.0971, Loss2: 0.0993 +Epoch [168/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1036, Loss2: 0.0985 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 39.5733 % Model2 39.7035 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 66.4062, Loss1: 0.0819, Loss2: 0.0798 +Epoch [169/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 69.5312, Loss1: 0.0970, Loss2: 0.0973 +Epoch [169/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 73.4375, Loss1: 0.0830, Loss2: 0.0818 +Epoch [169/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1076, Loss2: 0.1023 +Epoch [169/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0867, Loss2: 0.0856 +Epoch [169/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0665, Loss2: 0.0630 +Epoch [169/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0779, Loss2: 0.0750 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 39.2929 % Model2 40.2945 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1042, Loss2: 0.1055 +Epoch [170/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0924, Loss2: 0.0916 +Epoch [170/200], Iter [150/390] Training Accuracy1: 71.8750, Training Accuracy2: 65.6250, Loss1: 0.0734, Loss2: 0.0809 +Epoch [170/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0866, Loss2: 0.0849 +Epoch [170/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.9688, Loss1: 0.0944, Loss2: 0.0920 +Epoch [170/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0936, Loss2: 0.0936 +Epoch [170/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.1076, Loss2: 0.1052 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 39.3630 % Model2 39.2528 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0815, Loss2: 0.0862 +Epoch [171/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.0870, Loss2: 0.0881 +Epoch [171/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0809, Loss2: 0.0839 +Epoch [171/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 58.5938, Loss1: 0.0940, Loss2: 0.0863 +Epoch [171/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0778, Loss2: 0.0788 +Epoch [171/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0994, Loss2: 0.1019 +Epoch [171/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0897, Loss2: 0.0921 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 39.1727 % Model2 39.7236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 73.4375, Training Accuracy2: 71.8750, Loss1: 0.1472, Loss2: 0.1508 +Epoch [172/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0929, Loss2: 0.0947 +Epoch [172/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0917, Loss2: 0.0931 +Epoch [172/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 71.0938, Loss1: 0.1088, Loss2: 0.0897 +Epoch [172/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0759, Loss2: 0.0747 +Epoch [172/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.0880, Loss2: 0.0862 +Epoch [172/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.0740, Loss2: 0.0787 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 39.0224 % Model2 39.4331 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0813, Loss2: 0.0819 +Epoch [173/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.0798, Loss2: 0.0752 +Epoch [173/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.0938, Loss1: 0.0851, Loss2: 0.0900 +Epoch [173/200], Iter [200/390] Training Accuracy1: 71.8750, Training Accuracy2: 68.7500, Loss1: 0.0842, Loss2: 0.0931 +Epoch [173/200], Iter [250/390] Training Accuracy1: 75.7812, Training Accuracy2: 74.2188, Loss1: 0.0940, Loss2: 0.0988 +Epoch [173/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0940, Loss2: 0.0852 +Epoch [173/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0744, Loss2: 0.0719 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 39.5032 % Model2 39.6534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 70.3125, Loss1: 0.1061, Loss2: 0.1103 +Epoch [174/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0942, Loss2: 0.0852 +Epoch [174/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0808, Loss2: 0.0814 +Epoch [174/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0784, Loss2: 0.0768 +Epoch [174/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.1020, Loss2: 0.1051 +Epoch [174/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.0757, Loss2: 0.0701 +Epoch [174/200], Iter [350/390] Training Accuracy1: 72.6562, Training Accuracy2: 69.5312, Loss1: 0.0951, Loss2: 0.1048 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 38.8722 % Model2 39.4531 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.1350, Loss2: 0.1335 +Epoch [175/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0789, Loss2: 0.0828 +Epoch [175/200], Iter [150/390] Training Accuracy1: 78.1250, Training Accuracy2: 81.2500, Loss1: 0.1364, Loss2: 0.1343 +Epoch [175/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0879, Loss2: 0.0920 +Epoch [175/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1062, Loss2: 0.1113 +Epoch [175/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0748, Loss2: 0.0728 +Epoch [175/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0839, Loss2: 0.0901 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.8838 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0901, Loss2: 0.0907 +Epoch [176/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1111, Loss2: 0.1056 +Epoch [176/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0875, Loss2: 0.0885 +Epoch [176/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1188, Loss2: 0.1122 +Epoch [176/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0896, Loss2: 0.0893 +Epoch [176/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.1875, Loss1: 0.0939, Loss2: 0.0901 +Epoch [176/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0891, Loss2: 0.0891 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 39.5433 % Model2 39.3830 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0819, Loss2: 0.0825 +Epoch [177/200], Iter [100/390] Training Accuracy1: 75.7812, Training Accuracy2: 76.5625, Loss1: 0.1433, Loss2: 0.1412 +Epoch [177/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.0746, Loss2: 0.0713 +Epoch [177/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.0904, Loss2: 0.0951 +Epoch [177/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 68.7500, Loss1: 0.0977, Loss2: 0.0924 +Epoch [177/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1129, Loss2: 0.1149 +Epoch [177/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.1025, Loss2: 0.0993 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 39.0425 % Model2 39.8337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.1095, Loss2: 0.1158 +Epoch [178/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.1014, Loss2: 0.1011 +Epoch [178/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0918, Loss2: 0.0903 +Epoch [178/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.1000, Loss2: 0.1082 +Epoch [178/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0858, Loss2: 0.0806 +Epoch [178/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 65.6250, Loss1: 0.0845, Loss2: 0.0783 +Epoch [178/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0804, Loss2: 0.0806 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 39.5032 % Model2 39.4732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 63.2812, Loss1: 0.0906, Loss2: 0.0963 +Epoch [179/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0799, Loss2: 0.0803 +Epoch [179/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0848, Loss2: 0.0797 +Epoch [179/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0747, Loss2: 0.0809 +Epoch [179/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0891, Loss2: 0.0887 +Epoch [179/200], Iter [300/390] Training Accuracy1: 74.2188, Training Accuracy2: 72.6562, Loss1: 0.1185, Loss2: 0.1268 +Epoch [179/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0782, Loss2: 0.0851 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 39.5433 % Model2 39.4131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.0900, Loss2: 0.0907 +Epoch [180/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0906, Loss2: 0.0967 +Epoch [180/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.8438, Loss1: 0.0816, Loss2: 0.0786 +Epoch [180/200], Iter [200/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1071, Loss2: 0.1142 +Epoch [180/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 64.8438, Loss1: 0.1081, Loss2: 0.1249 +Epoch [180/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0856, Loss2: 0.0865 +Epoch [180/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0895, Loss2: 0.0892 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 39.7837 % Model2 39.4431 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.1063, Loss2: 0.0993 +Epoch [181/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 73.4375, Loss1: 0.1104, Loss2: 0.1038 +Epoch [181/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0936, Loss2: 0.0976 +Epoch [181/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 66.4062, Loss1: 0.0881, Loss2: 0.0899 +Epoch [181/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.0903, Loss2: 0.0943 +Epoch [181/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0904, Loss2: 0.0948 +Epoch [181/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0800, Loss2: 0.0760 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 39.3229 % Model2 39.6835 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0687, Loss2: 0.0702 +Epoch [182/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.1058, Loss2: 0.0983 +Epoch [182/200], Iter [150/390] Training Accuracy1: 72.6562, Training Accuracy2: 73.4375, Loss1: 0.1088, Loss2: 0.1083 +Epoch [182/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 70.3125, Loss1: 0.1056, Loss2: 0.1099 +Epoch [182/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0847, Loss2: 0.0851 +Epoch [182/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1159, Loss2: 0.1267 +Epoch [182/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0972, Loss2: 0.0952 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 39.1126 % Model2 39.3730 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0756, Loss2: 0.0798 +Epoch [183/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.0938, Loss1: 0.1331, Loss2: 0.1271 +Epoch [183/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1342, Loss2: 0.1405 +Epoch [183/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.1051, Loss2: 0.0984 +Epoch [183/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0965, Loss2: 0.1029 +Epoch [183/200], Iter [300/390] Training Accuracy1: 71.0938, Training Accuracy2: 65.6250, Loss1: 0.0921, Loss2: 0.1016 +Epoch [183/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0996, Loss2: 0.0973 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 39.6935 % Model2 39.6134 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 73.4375, Loss1: 0.0911, Loss2: 0.0834 +Epoch [184/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.1160, Loss2: 0.1149 +Epoch [184/200], Iter [150/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0654, Loss2: 0.0684 +Epoch [184/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1144, Loss2: 0.1102 +Epoch [184/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.1030, Loss2: 0.1025 +Epoch [184/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.0794, Loss2: 0.0826 +Epoch [184/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 69.5312, Loss1: 0.1197, Loss2: 0.1271 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 39.4331 % Model2 39.3129 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0954, Loss2: 0.0917 +Epoch [185/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0861, Loss2: 0.0801 +Epoch [185/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 65.6250, Loss1: 0.0959, Loss2: 0.1061 +Epoch [185/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0710, Loss2: 0.0730 +Epoch [185/200], Iter [250/390] Training Accuracy1: 70.3125, Training Accuracy2: 67.9688, Loss1: 0.0968, Loss2: 0.1023 +Epoch [185/200], Iter [300/390] Training Accuracy1: 77.3438, Training Accuracy2: 74.2188, Loss1: 0.1001, Loss2: 0.1127 +Epoch [185/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1254, Loss2: 0.1262 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 39.0825 % Model2 39.5333 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 75.0000, Loss1: 0.1169, Loss2: 0.1030 +Epoch [186/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1032, Loss2: 0.1050 +Epoch [186/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1120, Loss2: 0.1066 +Epoch [186/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1350, Loss2: 0.1340 +Epoch [186/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0843, Loss2: 0.0815 +Epoch [186/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1039, Loss2: 0.0987 +Epoch [186/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.1079, Loss2: 0.1138 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 39.4131 % Model2 39.2829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1049, Loss2: 0.0998 +Epoch [187/200], Iter [100/390] Training Accuracy1: 74.2188, Training Accuracy2: 70.3125, Loss1: 0.1120, Loss2: 0.1302 +Epoch [187/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1097, Loss2: 0.1046 +Epoch [187/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1363, Loss2: 0.1403 +Epoch [187/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.1184, Loss2: 0.1243 +Epoch [187/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0907, Loss2: 0.0942 +Epoch [187/200], Iter [350/390] Training Accuracy1: 73.4375, Training Accuracy2: 75.7812, Loss1: 0.1242, Loss2: 0.1213 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 39.3329 % Model2 39.5533 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0957, Loss2: 0.0964 +Epoch [188/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.0938, Loss1: 0.1010, Loss2: 0.0982 +Epoch [188/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0985, Loss2: 0.0982 +Epoch [188/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0785, Loss2: 0.0800 +Epoch [188/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 72.6562, Loss1: 0.0760, Loss2: 0.0804 +Epoch [188/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0924, Loss2: 0.0931 +Epoch [188/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1499, Loss2: 0.1437 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 39.4531 % Model2 39.4631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0912, Loss2: 0.0878 +Epoch [189/200], Iter [100/390] Training Accuracy1: 72.6562, Training Accuracy2: 67.1875, Loss1: 0.1459, Loss2: 0.1688 +Epoch [189/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1094, Loss2: 0.1023 +Epoch [189/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 72.6562, Loss1: 0.1100, Loss2: 0.1057 +Epoch [189/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 69.5312, Loss1: 0.1356, Loss2: 0.1281 +Epoch [189/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 66.4062, Loss1: 0.1027, Loss2: 0.1013 +Epoch [189/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 68.7500, Loss1: 0.1246, Loss2: 0.1138 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 39.5232 % Model2 39.2929 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.1028, Loss2: 0.1074 +Epoch [190/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0947, Loss2: 0.0929 +Epoch [190/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.0921, Loss2: 0.0950 +Epoch [190/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.1159, Loss2: 0.1144 +Epoch [190/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.9688, Loss1: 0.0861, Loss2: 0.0883 +Epoch [190/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.0850, Loss2: 0.0842 +Epoch [190/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1033, Loss2: 0.1078 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 39.2929 % Model2 39.2628 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 72.6562, Loss1: 0.1008, Loss2: 0.0921 +Epoch [191/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0892, Loss2: 0.0888 +Epoch [191/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0804, Loss2: 0.0849 +Epoch [191/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0981, Loss2: 0.0954 +Epoch [191/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.8750, Loss1: 0.1310, Loss2: 0.1152 +Epoch [191/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.1507, Loss2: 0.1376 +Epoch [191/200], Iter [350/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1085, Loss2: 0.1101 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 39.1627 % Model2 39.1226 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.0939, Loss2: 0.0918 +Epoch [192/200], Iter [100/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1018, Loss2: 0.1035 +Epoch [192/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1064, Loss2: 0.1148 +Epoch [192/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 72.6562, Loss1: 0.0917, Loss2: 0.0948 +Epoch [192/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.1152, Loss2: 0.1203 +Epoch [192/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0807, Loss2: 0.0761 +Epoch [192/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.0938, Loss1: 0.1228, Loss2: 0.1242 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 39.2528 % Model2 39.1927 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0908, Loss2: 0.0915 +Epoch [193/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0857, Loss2: 0.0916 +Epoch [193/200], Iter [150/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.1875, Loss1: 0.0819, Loss2: 0.0839 +Epoch [193/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0860, Loss2: 0.0827 +Epoch [193/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1058, Loss2: 0.1091 +Epoch [193/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0964, Loss2: 0.1050 +Epoch [193/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0954, Loss2: 0.0935 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 39.4431 % Model2 38.9423 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 69.5312, Loss1: 0.1482, Loss2: 0.1349 +Epoch [194/200], Iter [100/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.0967, Loss2: 0.0925 +Epoch [194/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1117, Loss2: 0.1174 +Epoch [194/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1289, Loss2: 0.1306 +Epoch [194/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.1017, Loss2: 0.1075 +Epoch [194/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.1105, Loss2: 0.1097 +Epoch [194/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0977, Loss2: 0.0985 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 39.1126 % Model2 39.2628 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.1023, Loss2: 0.1053 +Epoch [195/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1142, Loss2: 0.1115 +Epoch [195/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1393, Loss2: 0.1339 +Epoch [195/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.0938, Loss1: 0.1215, Loss2: 0.1151 +Epoch [195/200], Iter [250/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.0868, Loss2: 0.0845 +Epoch [195/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 69.5312, Loss1: 0.0977, Loss2: 0.0958 +Epoch [195/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1372, Loss2: 0.1373 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 39.0325 % Model2 39.2628 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.1875, Loss1: 0.0933, Loss2: 0.0899 +Epoch [196/200], Iter [100/390] Training Accuracy1: 75.0000, Training Accuracy2: 72.6562, Loss1: 0.1300, Loss2: 0.1453 +Epoch [196/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 73.4375, Loss1: 0.1149, Loss2: 0.1075 +Epoch [196/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0947, Loss2: 0.0971 +Epoch [196/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0948, Loss2: 0.0969 +Epoch [196/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0851, Loss2: 0.0799 +Epoch [196/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0843, Loss2: 0.0930 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 39.2528 % Model2 39.1927 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.1117, Loss2: 0.1144 +Epoch [197/200], Iter [100/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.9375, Loss1: 0.1004, Loss2: 0.1139 +Epoch [197/200], Iter [150/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1131, Loss2: 0.1127 +Epoch [197/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 69.5312, Loss1: 0.0880, Loss2: 0.0868 +Epoch [197/200], Iter [250/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1268, Loss2: 0.1289 +Epoch [197/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.1106, Loss2: 0.1091 +Epoch [197/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.0960, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 39.3129 % Model2 39.1827 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.9688, Loss1: 0.0858, Loss2: 0.0845 +Epoch [198/200], Iter [100/390] Training Accuracy1: 69.5312, Training Accuracy2: 71.8750, Loss1: 0.1121, Loss2: 0.1076 +Epoch [198/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0946, Loss2: 0.0917 +Epoch [198/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0795, Loss2: 0.0790 +Epoch [198/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0908, Loss2: 0.0904 +Epoch [198/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 71.8750, Loss1: 0.1182, Loss2: 0.1266 +Epoch [198/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0850, Loss2: 0.0800 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 39.1326 % Model2 39.1727 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1010, Loss2: 0.1094 +Epoch [199/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.1154, Loss2: 0.1100 +Epoch [199/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0809, Loss2: 0.0822 +Epoch [199/200], Iter [200/390] Training Accuracy1: 70.3125, Training Accuracy2: 70.3125, Loss1: 0.1516, Loss2: 0.1593 +Epoch [199/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0929, Loss2: 0.0869 +Epoch [199/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0968, Loss2: 0.1005 +Epoch [199/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1120, Loss2: 0.1031 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 39.1426 % Model2 39.2127 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.1028, Loss2: 0.0947 +Epoch [200/200], Iter [100/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1046, Loss2: 0.1080 +Epoch [200/200], Iter [150/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.8750, Loss1: 0.1087, Loss2: 0.1098 +Epoch [200/200], Iter [200/390] Training Accuracy1: 71.0938, Training Accuracy2: 71.0938, Loss1: 0.1102, Loss2: 0.1151 +Epoch [200/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1123, Loss2: 0.1097 +Epoch [200/200], Iter [300/390] Training Accuracy1: 72.6562, Training Accuracy2: 70.3125, Loss1: 0.1118, Loss2: 0.1208 +Epoch [200/200], Iter [350/390] Training Accuracy1: 71.8750, Training Accuracy2: 72.6562, Loss1: 0.1082, Loss2: 0.1075 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 39.1727 % Model2 39.1526 % diff --git a/other_methods/coteaching_plus/coteaching_plus_results/out_6_6.log b/other_methods/coteaching_plus/coteaching_plus_results/out_6_6.log new file mode 100644 index 0000000..0b91bbb --- /dev/null +++ b/other_methods/coteaching_plus/coteaching_plus_results/out_6_6.log @@ -0,0 +1,2015 @@ +Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Actual noise 0.70 +(50000, 32, 32, 3) uint8 (32, 32, 3) uint8 +Files already downloaded and verified +loading dataset... +building model... + + +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [1/200] Test Accuracy on the 10000 test data: Model1 9.9860 % Model2 10.7973 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200], Iter [50/390] Training Accuracy1: 20.3125, Training Accuracy2: 20.3125, Loss1: 0.0172, Loss2: 0.0172 +Epoch [2/200], Iter [100/390] Training Accuracy1: 21.8750, Training Accuracy2: 21.8750, Loss1: 0.0167, Loss2: 0.0167 +Epoch [2/200], Iter [150/390] Training Accuracy1: 18.7500, Training Accuracy2: 17.9688, Loss1: 0.0166, Loss2: 0.0167 +Epoch [2/200], Iter [200/390] Training Accuracy1: 22.6562, Training Accuracy2: 25.7812, Loss1: 0.0159, Loss2: 0.0159 +Epoch [2/200], Iter [250/390] Training Accuracy1: 19.5312, Training Accuracy2: 24.2188, Loss1: 0.0160, Loss2: 0.0156 +Epoch [2/200], Iter [300/390] Training Accuracy1: 26.5625, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0153 +Epoch [2/200], Iter [350/390] Training Accuracy1: 29.6875, Training Accuracy2: 25.7812, Loss1: 0.0154, Loss2: 0.0155 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [2/200] Test Accuracy on the 10000 test data: Model1 14.9639 % Model2 14.4631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200], Iter [50/390] Training Accuracy1: 25.7812, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0155 +Epoch [3/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 25.0000, Loss1: 0.0165, Loss2: 0.0165 +Epoch [3/200], Iter [150/390] Training Accuracy1: 28.1250, Training Accuracy2: 28.1250, Loss1: 0.0149, Loss2: 0.0145 +Epoch [3/200], Iter [200/390] Training Accuracy1: 28.1250, Training Accuracy2: 24.2188, Loss1: 0.0155, Loss2: 0.0161 +Epoch [3/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.1562, Loss1: 0.0138, Loss2: 0.0134 +Epoch [3/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0148, Loss2: 0.0147 +Epoch [3/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 30.4688, Loss1: 0.0146, Loss2: 0.0144 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [3/200] Test Accuracy on the 10000 test data: Model1 18.1591 % Model2 18.8301 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200], Iter [50/390] Training Accuracy1: 25.0000, Training Accuracy2: 29.6875, Loss1: 0.0160, Loss2: 0.0156 +Epoch [4/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 35.9375, Loss1: 0.0138, Loss2: 0.0139 +Epoch [4/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0149, Loss2: 0.0148 +Epoch [4/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 36.7188, Loss1: 0.0135, Loss2: 0.0136 +Epoch [4/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0144, Loss2: 0.0142 +Epoch [4/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 38.2812, Loss1: 0.0143, Loss2: 0.0141 +Epoch [4/200], Iter [350/390] Training Accuracy1: 27.3438, Training Accuracy2: 27.3438, Loss1: 0.0148, Loss2: 0.0145 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [4/200] Test Accuracy on the 10000 test data: Model1 19.9319 % Model2 20.1122 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 31.2500, Loss1: 0.0141, Loss2: 0.0142 +Epoch [5/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0146, Loss2: 0.0140 +Epoch [5/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 40.6250, Loss1: 0.0134, Loss2: 0.0126 +Epoch [5/200], Iter [200/390] Training Accuracy1: 32.8125, Training Accuracy2: 29.6875, Loss1: 0.0156, Loss2: 0.0154 +Epoch [5/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0138, Loss2: 0.0140 +Epoch [5/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 34.3750, Loss1: 0.0157, Loss2: 0.0151 +Epoch [5/200], Iter [350/390] Training Accuracy1: 30.4688, Training Accuracy2: 35.1562, Loss1: 0.0157, Loss2: 0.0156 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [5/200] Test Accuracy on the 10000 test data: Model1 20.2825 % Model2 20.0120 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200], Iter [50/390] Training Accuracy1: 33.5938, Training Accuracy2: 34.3750, Loss1: 0.0140, Loss2: 0.0136 +Epoch [6/200], Iter [100/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0138, Loss2: 0.0133 +Epoch [6/200], Iter [150/390] Training Accuracy1: 35.9375, Training Accuracy2: 38.2812, Loss1: 0.0131, Loss2: 0.0129 +Epoch [6/200], Iter [200/390] Training Accuracy1: 21.8750, Training Accuracy2: 33.5938, Loss1: 0.0165, Loss2: 0.0160 +Epoch [6/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 43.7500, Loss1: 0.0133, Loss2: 0.0126 +Epoch [6/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 38.2812, Loss1: 0.0134, Loss2: 0.0137 +Epoch [6/200], Iter [350/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0149, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [6/200] Test Accuracy on the 10000 test data: Model1 21.9050 % Model2 23.2772 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200], Iter [50/390] Training Accuracy1: 29.6875, Training Accuracy2: 33.5938, Loss1: 0.0142, Loss2: 0.0136 +Epoch [7/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0127, Loss2: 0.0122 +Epoch [7/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 33.5938, Loss1: 0.0144, Loss2: 0.0142 +Epoch [7/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0146, Loss2: 0.0145 +Epoch [7/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0140, Loss2: 0.0132 +Epoch [7/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 35.9375, Loss1: 0.0134, Loss2: 0.0126 +Epoch [7/200], Iter [350/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0135, Loss2: 0.0132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [7/200] Test Accuracy on the 10000 test data: Model1 21.1639 % Model2 21.6847 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200], Iter [50/390] Training Accuracy1: 38.2812, Training Accuracy2: 34.3750, Loss1: 0.0130, Loss2: 0.0130 +Epoch [8/200], Iter [100/390] Training Accuracy1: 36.7188, Training Accuracy2: 37.5000, Loss1: 0.0139, Loss2: 0.0135 +Epoch [8/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0125, Loss2: 0.0119 +Epoch [8/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 37.5000, Loss1: 0.0128, Loss2: 0.0128 +Epoch [8/200], Iter [250/390] Training Accuracy1: 29.6875, Training Accuracy2: 30.4688, Loss1: 0.0134, Loss2: 0.0133 +Epoch [8/200], Iter [300/390] Training Accuracy1: 24.2188, Training Accuracy2: 29.6875, Loss1: 0.0150, Loss2: 0.0145 +Epoch [8/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0145, Loss2: 0.0136 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [8/200] Test Accuracy on the 10000 test data: Model1 19.3009 % Model2 19.7015 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.9688, Loss1: 0.0124, Loss2: 0.0122 +Epoch [9/200], Iter [100/390] Training Accuracy1: 28.9062, Training Accuracy2: 35.1562, Loss1: 0.0132, Loss2: 0.0126 +Epoch [9/200], Iter [150/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0135, Loss2: 0.0136 +Epoch [9/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0127, Loss2: 0.0121 +Epoch [9/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0127, Loss2: 0.0120 +Epoch [9/200], Iter [300/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.0312, Loss1: 0.0150, Loss2: 0.0152 +Epoch [9/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0114, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [9/200] Test Accuracy on the 10000 test data: Model1 20.1222 % Model2 20.4527 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 38.2812, Loss1: 0.0137, Loss2: 0.0128 +Epoch [10/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0123, Loss2: 0.0125 +Epoch [10/200], Iter [150/390] Training Accuracy1: 31.2500, Training Accuracy2: 35.1562, Loss1: 0.0141, Loss2: 0.0126 +Epoch [10/200], Iter [200/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.1562, Loss1: 0.0126, Loss2: 0.0126 +Epoch [10/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0135, Loss2: 0.0124 +Epoch [10/200], Iter [300/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0135, Loss2: 0.0131 +Epoch [10/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.0938, Loss1: 0.0121, Loss2: 0.0112 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [10/200] Test Accuracy on the 10000 test data: Model1 19.2708 % Model2 21.8550 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.0625, Loss1: 0.0119, Loss2: 0.0119 +Epoch [11/200], Iter [100/390] Training Accuracy1: 32.8125, Training Accuracy2: 43.7500, Loss1: 0.0134, Loss2: 0.0128 +Epoch [11/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.8438, Loss1: 0.0146, Loss2: 0.0149 +Epoch [11/200], Iter [200/390] Training Accuracy1: 35.1562, Training Accuracy2: 39.0625, Loss1: 0.0121, Loss2: 0.0115 +Epoch [11/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 40.6250, Loss1: 0.0119, Loss2: 0.0115 +Epoch [11/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 41.4062, Loss1: 0.0122, Loss2: 0.0113 +Epoch [11/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0119, Loss2: 0.0116 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [11/200] Test Accuracy on the 10000 test data: Model1 19.1006 % Model2 19.2308 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0133, Loss2: 0.0130 +Epoch [12/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 35.9375, Loss1: 0.0130, Loss2: 0.0131 +Epoch [12/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 42.9688, Loss1: 0.0130, Loss2: 0.0127 +Epoch [12/200], Iter [200/390] Training Accuracy1: 31.2500, Training Accuracy2: 31.2500, Loss1: 0.0142, Loss2: 0.0137 +Epoch [12/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0114, Loss2: 0.0118 +Epoch [12/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0134, Loss2: 0.0130 +Epoch [12/200], Iter [350/390] Training Accuracy1: 31.2500, Training Accuracy2: 37.5000, Loss1: 0.0141, Loss2: 0.0137 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [12/200] Test Accuracy on the 10000 test data: Model1 20.9635 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200], Iter [50/390] Training Accuracy1: 35.9375, Training Accuracy2: 29.6875, Loss1: 0.0134, Loss2: 0.0128 +Epoch [13/200], Iter [100/390] Training Accuracy1: 32.0312, Training Accuracy2: 37.5000, Loss1: 0.0133, Loss2: 0.0130 +Epoch [13/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.1875, Loss1: 0.0115, Loss2: 0.0109 +Epoch [13/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 33.5938, Loss1: 0.0127, Loss2: 0.0126 +Epoch [13/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 32.8125, Loss1: 0.0141, Loss2: 0.0135 +Epoch [13/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0116, Loss2: 0.0107 +Epoch [13/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 36.7188, Loss1: 0.0132, Loss2: 0.0132 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [13/200] Test Accuracy on the 10000 test data: Model1 20.2023 % Model2 21.7949 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200], Iter [50/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.8125, Loss1: 0.0132, Loss2: 0.0133 +Epoch [14/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 41.4062, Loss1: 0.0124, Loss2: 0.0123 +Epoch [14/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.8750, Loss1: 0.0124, Loss2: 0.0109 +Epoch [14/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.0625, Loss1: 0.0115, Loss2: 0.0119 +Epoch [14/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 38.2812, Loss1: 0.0129, Loss2: 0.0127 +Epoch [14/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 38.2812, Loss1: 0.0124, Loss2: 0.0122 +Epoch [14/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0120, Loss2: 0.0110 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [14/200] Test Accuracy on the 10000 test data: Model1 21.1538 % Model2 22.6562 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0120, Loss2: 0.0119 +Epoch [15/200], Iter [100/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.1562, Loss1: 0.0129, Loss2: 0.0120 +Epoch [15/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.8438, Loss1: 0.0117, Loss2: 0.0116 +Epoch [15/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.9375, Loss1: 0.0134, Loss2: 0.0128 +Epoch [15/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 37.5000, Loss1: 0.0116, Loss2: 0.0112 +Epoch [15/200], Iter [300/390] Training Accuracy1: 38.2812, Training Accuracy2: 39.0625, Loss1: 0.0127, Loss2: 0.0118 +Epoch [15/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.9688, Loss1: 0.0116, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [15/200] Test Accuracy on the 10000 test data: Model1 20.2123 % Model2 21.5244 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200], Iter [50/390] Training Accuracy1: 37.5000, Training Accuracy2: 42.1875, Loss1: 0.0133, Loss2: 0.0120 +Epoch [16/200], Iter [100/390] Training Accuracy1: 33.5938, Training Accuracy2: 46.0938, Loss1: 0.0125, Loss2: 0.0110 +Epoch [16/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 43.7500, Loss1: 0.0119, Loss2: 0.0118 +Epoch [16/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0108, Loss2: 0.0109 +Epoch [16/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0125, Loss2: 0.0116 +Epoch [16/200], Iter [300/390] Training Accuracy1: 35.9375, Training Accuracy2: 42.1875, Loss1: 0.0115, Loss2: 0.0110 +Epoch [16/200], Iter [350/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0114, Loss2: 0.0112 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [16/200] Test Accuracy on the 10000 test data: Model1 21.0537 % Model2 21.5445 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 44.5312, Loss1: 0.0105, Loss2: 0.0106 +Epoch [17/200], Iter [100/390] Training Accuracy1: 42.9688, Training Accuracy2: 36.7188, Loss1: 0.0121, Loss2: 0.0127 +Epoch [17/200], Iter [150/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0124, Loss2: 0.0111 +Epoch [17/200], Iter [200/390] Training Accuracy1: 32.0312, Training Accuracy2: 33.5938, Loss1: 0.0133, Loss2: 0.0133 +Epoch [17/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0119, Loss2: 0.0122 +Epoch [17/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 37.5000, Loss1: 0.0111, Loss2: 0.0120 +Epoch [17/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 37.5000, Loss1: 0.0129, Loss2: 0.0123 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [17/200] Test Accuracy on the 10000 test data: Model1 19.6915 % Model2 18.7800 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0092, Loss2: 0.0084 +Epoch [18/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0113, Loss2: 0.0106 +Epoch [18/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0105, Loss2: 0.0105 +Epoch [18/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0113, Loss2: 0.0105 +Epoch [18/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 38.2812, Loss1: 0.0128, Loss2: 0.0129 +Epoch [18/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 41.4062, Loss1: 0.0130, Loss2: 0.0119 +Epoch [18/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 41.4062, Loss1: 0.0115, Loss2: 0.0113 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [18/200] Test Accuracy on the 10000 test data: Model1 22.2756 % Model2 24.5893 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200], Iter [50/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0122, Loss2: 0.0120 +Epoch [19/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0116, Loss2: 0.0112 +Epoch [19/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0118, Loss2: 0.0102 +Epoch [19/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 36.7188, Loss1: 0.0136, Loss2: 0.0132 +Epoch [19/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.9375, Loss1: 0.0121, Loss2: 0.0121 +Epoch [19/200], Iter [300/390] Training Accuracy1: 35.1562, Training Accuracy2: 35.9375, Loss1: 0.0120, Loss2: 0.0115 +Epoch [19/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 40.6250, Loss1: 0.0127, Loss2: 0.0126 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [19/200] Test Accuracy on the 10000 test data: Model1 18.2993 % Model2 18.5897 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0105, Loss2: 0.0100 +Epoch [20/200], Iter [100/390] Training Accuracy1: 38.2812, Training Accuracy2: 35.9375, Loss1: 0.0129, Loss2: 0.0129 +Epoch [20/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0102, Loss2: 0.0102 +Epoch [20/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 38.2812, Loss1: 0.0127, Loss2: 0.0115 +Epoch [20/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 45.3125, Loss1: 0.0113, Loss2: 0.0100 +Epoch [20/200], Iter [300/390] Training Accuracy1: 36.7188, Training Accuracy2: 50.7812, Loss1: 0.0111, Loss2: 0.0105 +Epoch [20/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 39.0625, Loss1: 0.0131, Loss2: 0.0122 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [20/200] Test Accuracy on the 10000 test data: Model1 21.2841 % Model2 23.5076 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0773, Loss2: 0.0778 +Epoch [21/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 30.4688, Loss1: 0.0684, Loss2: 0.0700 +Epoch [21/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0598, Loss2: 0.0585 +Epoch [21/200], Iter [200/390] Training Accuracy1: 29.6875, Training Accuracy2: 34.3750, Loss1: 0.0556, Loss2: 0.0535 +Epoch [21/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.0625, Loss1: 0.0795, Loss2: 0.0791 +Epoch [21/200], Iter [300/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0502, Loss2: 0.0483 +Epoch [21/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 33.5938, Loss1: 0.0770, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [21/200] Test Accuracy on the 10000 test data: Model1 19.8017 % Model2 20.1823 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200], Iter [50/390] Training Accuracy1: 27.3438, Training Accuracy2: 32.8125, Loss1: 0.0592, Loss2: 0.0573 +Epoch [22/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0566, Loss2: 0.0544 +Epoch [22/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 40.6250, Loss1: 0.0552, Loss2: 0.0560 +Epoch [22/200], Iter [200/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0568, Loss2: 0.0551 +Epoch [22/200], Iter [250/390] Training Accuracy1: 35.1562, Training Accuracy2: 32.0312, Loss1: 0.0619, Loss2: 0.0632 +Epoch [22/200], Iter [300/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.8438, Loss1: 0.0608, Loss2: 0.0582 +Epoch [22/200], Iter [350/390] Training Accuracy1: 40.6250, Training Accuracy2: 38.2812, Loss1: 0.0643, Loss2: 0.0644 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [22/200] Test Accuracy on the 10000 test data: Model1 21.8450 % Model2 20.2624 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0628, Loss2: 0.0603 +Epoch [23/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0834, Loss2: 0.0824 +Epoch [23/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 46.8750, Loss1: 0.0665, Loss2: 0.0630 +Epoch [23/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0733, Loss2: 0.0728 +Epoch [23/200], Iter [250/390] Training Accuracy1: 36.7188, Training Accuracy2: 35.1562, Loss1: 0.0673, Loss2: 0.0668 +Epoch [23/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0741, Loss2: 0.0762 +Epoch [23/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0667, Loss2: 0.0603 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [23/200] Test Accuracy on the 10000 test data: Model1 19.5913 % Model2 21.5345 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.1066, Loss2: 0.1008 +Epoch [24/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.1875, Loss1: 0.0698, Loss2: 0.0689 +Epoch [24/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0820, Loss2: 0.0823 +Epoch [24/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0603, Loss2: 0.0607 +Epoch [24/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0714, Loss2: 0.0655 +Epoch [24/200], Iter [300/390] Training Accuracy1: 28.9062, Training Accuracy2: 29.6875, Loss1: 0.0633, Loss2: 0.0628 +Epoch [24/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 39.8438, Loss1: 0.0580, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [24/200] Test Accuracy on the 10000 test data: Model1 19.2007 % Model2 20.1122 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 40.6250, Loss1: 0.0690, Loss2: 0.0685 +Epoch [25/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0750, Loss2: 0.0716 +Epoch [25/200], Iter [150/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0632, Loss2: 0.0631 +Epoch [25/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0686, Loss2: 0.0681 +Epoch [25/200], Iter [250/390] Training Accuracy1: 32.8125, Training Accuracy2: 40.6250, Loss1: 0.0689, Loss2: 0.0638 +Epoch [25/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 35.1562, Loss1: 0.0547, Loss2: 0.0546 +Epoch [25/200], Iter [350/390] Training Accuracy1: 32.8125, Training Accuracy2: 37.5000, Loss1: 0.0572, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [25/200] Test Accuracy on the 10000 test data: Model1 19.7115 % Model2 20.6631 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0668, Loss2: 0.0649 +Epoch [26/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0529, Loss2: 0.0500 +Epoch [26/200], Iter [150/390] Training Accuracy1: 28.9062, Training Accuracy2: 37.5000, Loss1: 0.0682, Loss2: 0.0624 +Epoch [26/200], Iter [200/390] Training Accuracy1: 36.7188, Training Accuracy2: 33.5938, Loss1: 0.0552, Loss2: 0.0567 +Epoch [26/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0721, Loss2: 0.0682 +Epoch [26/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0774, Loss2: 0.0727 +Epoch [26/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 38.2812, Loss1: 0.0715, Loss2: 0.0733 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [26/200] Test Accuracy on the 10000 test data: Model1 23.5276 % Model2 21.8349 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0803, Loss2: 0.0786 +Epoch [27/200], Iter [100/390] Training Accuracy1: 31.2500, Training Accuracy2: 32.8125, Loss1: 0.0516, Loss2: 0.0514 +Epoch [27/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.8750, Loss1: 0.0684, Loss2: 0.0647 +Epoch [27/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 37.5000, Loss1: 0.0589, Loss2: 0.0608 +Epoch [27/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0804, Loss2: 0.0784 +Epoch [27/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0634, Loss2: 0.0626 +Epoch [27/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 36.7188, Loss1: 0.0524, Loss2: 0.0528 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [27/200] Test Accuracy on the 10000 test data: Model1 21.4543 % Model2 20.1222 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 41.4062, Loss1: 0.0555, Loss2: 0.0528 +Epoch [28/200], Iter [100/390] Training Accuracy1: 39.0625, Training Accuracy2: 42.1875, Loss1: 0.0665, Loss2: 0.0631 +Epoch [28/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0761, Loss2: 0.0775 +Epoch [28/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0625, Loss2: 0.0592 +Epoch [28/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 53.1250, Loss1: 0.0594, Loss2: 0.0529 +Epoch [28/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0959, Loss2: 0.0867 +Epoch [28/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 42.1875, Loss1: 0.0643, Loss2: 0.0650 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [28/200] Test Accuracy on the 10000 test data: Model1 19.9920 % Model2 19.5813 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0655, Loss2: 0.0658 +Epoch [29/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0594, Loss2: 0.0558 +Epoch [29/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 40.6250, Loss1: 0.0787, Loss2: 0.0775 +Epoch [29/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0654, Loss2: 0.0631 +Epoch [29/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 45.3125, Loss1: 0.0750, Loss2: 0.0729 +Epoch [29/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.8750, Loss1: 0.0690, Loss2: 0.0708 +Epoch [29/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0594, Loss2: 0.0557 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [29/200] Test Accuracy on the 10000 test data: Model1 20.5529 % Model2 20.2624 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0676, Loss2: 0.0631 +Epoch [30/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 42.9688, Loss1: 0.0702, Loss2: 0.0718 +Epoch [30/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0733, Loss2: 0.0716 +Epoch [30/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 38.2812, Loss1: 0.0588, Loss2: 0.0606 +Epoch [30/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0700, Loss2: 0.0681 +Epoch [30/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 43.7500, Loss1: 0.0567, Loss2: 0.0547 +Epoch [30/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0692, Loss2: 0.0653 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [30/200] Test Accuracy on the 10000 test data: Model1 19.0905 % Model2 20.0721 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200], Iter [50/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.1875, Loss1: 0.0534, Loss2: 0.0526 +Epoch [31/200], Iter [100/390] Training Accuracy1: 40.6250, Training Accuracy2: 39.0625, Loss1: 0.0690, Loss2: 0.0700 +Epoch [31/200], Iter [150/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0646, Loss2: 0.0621 +Epoch [31/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 38.2812, Loss1: 0.0577, Loss2: 0.0600 +Epoch [31/200], Iter [250/390] Training Accuracy1: 37.5000, Training Accuracy2: 38.2812, Loss1: 0.0637, Loss2: 0.0636 +Epoch [31/200], Iter [300/390] Training Accuracy1: 37.5000, Training Accuracy2: 39.0625, Loss1: 0.0526, Loss2: 0.0534 +Epoch [31/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 43.7500, Loss1: 0.0584, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [31/200] Test Accuracy on the 10000 test data: Model1 18.0589 % Model2 21.4243 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200], Iter [50/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0566, Loss2: 0.0542 +Epoch [32/200], Iter [100/390] Training Accuracy1: 34.3750, Training Accuracy2: 39.0625, Loss1: 0.0504, Loss2: 0.0493 +Epoch [32/200], Iter [150/390] Training Accuracy1: 43.7500, Training Accuracy2: 43.7500, Loss1: 0.0698, Loss2: 0.0699 +Epoch [32/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 42.9688, Loss1: 0.0691, Loss2: 0.0675 +Epoch [32/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0600, Loss2: 0.0558 +Epoch [32/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 39.8438, Loss1: 0.0527, Loss2: 0.0513 +Epoch [32/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0747, Loss2: 0.0742 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [32/200] Test Accuracy on the 10000 test data: Model1 22.0353 % Model2 21.3041 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200], Iter [50/390] Training Accuracy1: 35.1562, Training Accuracy2: 33.5938, Loss1: 0.0629, Loss2: 0.0654 +Epoch [33/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0665, Loss2: 0.0643 +Epoch [33/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 44.5312, Loss1: 0.0658, Loss2: 0.0673 +Epoch [33/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 43.7500, Loss1: 0.0643, Loss2: 0.0668 +Epoch [33/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0642, Loss2: 0.0646 +Epoch [33/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 44.5312, Loss1: 0.0584, Loss2: 0.0554 +Epoch [33/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0621, Loss2: 0.0604 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [33/200] Test Accuracy on the 10000 test data: Model1 22.1855 % Model2 20.7131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 44.5312, Loss1: 0.0560, Loss2: 0.0551 +Epoch [34/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0605, Loss2: 0.0571 +Epoch [34/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.8750, Loss1: 0.0766, Loss2: 0.0716 +Epoch [34/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0846, Loss2: 0.0791 +Epoch [34/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 43.7500, Loss1: 0.0703, Loss2: 0.0711 +Epoch [34/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 52.3438, Loss1: 0.0715, Loss2: 0.0666 +Epoch [34/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0556, Loss2: 0.0558 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [34/200] Test Accuracy on the 10000 test data: Model1 21.0637 % Model2 21.2640 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0683, Loss2: 0.0658 +Epoch [35/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.0000, Loss1: 0.0653, Loss2: 0.0682 +Epoch [35/200], Iter [150/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0521, Loss2: 0.0512 +Epoch [35/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0719, Loss2: 0.0724 +Epoch [35/200], Iter [250/390] Training Accuracy1: 35.9375, Training Accuracy2: 36.7188, Loss1: 0.0625, Loss2: 0.0617 +Epoch [35/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0633, Loss2: 0.0619 +Epoch [35/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 51.5625, Loss1: 0.0670, Loss2: 0.0610 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [35/200] Test Accuracy on the 10000 test data: Model1 21.3241 % Model2 22.1054 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200], Iter [50/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0633, Loss2: 0.0613 +Epoch [36/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.0000, Loss1: 0.0695, Loss2: 0.0707 +Epoch [36/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0690, Loss2: 0.0648 +Epoch [36/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0538, Loss2: 0.0521 +Epoch [36/200], Iter [250/390] Training Accuracy1: 32.0312, Training Accuracy2: 32.0312, Loss1: 0.0535, Loss2: 0.0533 +Epoch [36/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 40.6250, Loss1: 0.0549, Loss2: 0.0568 +Epoch [36/200], Iter [350/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0574, Loss2: 0.0551 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [36/200] Test Accuracy on the 10000 test data: Model1 20.6230 % Model2 21.1739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 38.2812, Loss1: 0.0708, Loss2: 0.0738 +Epoch [37/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0677, Loss2: 0.0662 +Epoch [37/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0756, Loss2: 0.0720 +Epoch [37/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 43.7500, Loss1: 0.0765, Loss2: 0.0771 +Epoch [37/200], Iter [250/390] Training Accuracy1: 39.8438, Training Accuracy2: 40.6250, Loss1: 0.0510, Loss2: 0.0496 +Epoch [37/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0659, Loss2: 0.0629 +Epoch [37/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 39.8438, Loss1: 0.0602, Loss2: 0.0575 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [37/200] Test Accuracy on the 10000 test data: Model1 21.0637 % Model2 20.6130 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0736, Loss2: 0.0759 +Epoch [38/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0591, Loss2: 0.0563 +Epoch [38/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0566, Loss2: 0.0566 +Epoch [38/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0664, Loss2: 0.0602 +Epoch [38/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 42.1875, Loss1: 0.0649, Loss2: 0.0638 +Epoch [38/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0585, Loss2: 0.0564 +Epoch [38/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 43.7500, Loss1: 0.0695, Loss2: 0.0655 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [38/200] Test Accuracy on the 10000 test data: Model1 20.9135 % Model2 21.2139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0738, Loss2: 0.0702 +Epoch [39/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0601, Loss2: 0.0614 +Epoch [39/200], Iter [150/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0627, Loss2: 0.0597 +Epoch [39/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 49.2188, Loss1: 0.0619, Loss2: 0.0568 +Epoch [39/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0615, Loss2: 0.0617 +Epoch [39/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0515, Loss2: 0.0497 +Epoch [39/200], Iter [350/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0594, Loss2: 0.0593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [39/200] Test Accuracy on the 10000 test data: Model1 20.4227 % Model2 20.6030 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200], Iter [50/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0534, Loss2: 0.0515 +Epoch [40/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0731, Loss2: 0.0697 +Epoch [40/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0541, Loss2: 0.0563 +Epoch [40/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 44.5312, Loss1: 0.0539, Loss2: 0.0518 +Epoch [40/200], Iter [250/390] Training Accuracy1: 39.0625, Training Accuracy2: 37.5000, Loss1: 0.0598, Loss2: 0.0600 +Epoch [40/200], Iter [300/390] Training Accuracy1: 41.4062, Training Accuracy2: 41.4062, Loss1: 0.0532, Loss2: 0.0529 +Epoch [40/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 42.1875, Loss1: 0.0551, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [40/200] Test Accuracy on the 10000 test data: Model1 20.6130 % Model2 20.6831 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0671, Loss2: 0.0647 +Epoch [41/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0570, Loss2: 0.0580 +Epoch [41/200], Iter [150/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0646, Loss2: 0.0608 +Epoch [41/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0547, Loss2: 0.0522 +Epoch [41/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 43.7500, Loss1: 0.0536, Loss2: 0.0547 +Epoch [41/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0767, Loss2: 0.0716 +Epoch [41/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 46.8750, Loss1: 0.0654, Loss2: 0.0612 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [41/200] Test Accuracy on the 10000 test data: Model1 20.5629 % Model2 20.4928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 50.0000, Loss1: 0.0561, Loss2: 0.0521 +Epoch [42/200], Iter [100/390] Training Accuracy1: 41.4062, Training Accuracy2: 48.4375, Loss1: 0.0569, Loss2: 0.0531 +Epoch [42/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0664, Loss2: 0.0630 +Epoch [42/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0617, Loss2: 0.0595 +Epoch [42/200], Iter [250/390] Training Accuracy1: 38.2812, Training Accuracy2: 42.1875, Loss1: 0.0603, Loss2: 0.0572 +Epoch [42/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0658, Loss2: 0.0643 +Epoch [42/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 50.0000, Loss1: 0.0589, Loss2: 0.0520 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [42/200] Test Accuracy on the 10000 test data: Model1 20.6931 % Model2 22.2055 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0676, Loss2: 0.0663 +Epoch [43/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0508, Loss2: 0.0487 +Epoch [43/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0742, Loss2: 0.0736 +Epoch [43/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0695, Loss2: 0.0641 +Epoch [43/200], Iter [250/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0561, Loss2: 0.0541 +Epoch [43/200], Iter [300/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.9688, Loss1: 0.0548, Loss2: 0.0533 +Epoch [43/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0824, Loss2: 0.0825 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [43/200] Test Accuracy on the 10000 test data: Model1 20.4728 % Model2 20.9836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 46.0938, Loss1: 0.0543, Loss2: 0.0560 +Epoch [44/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0555, Loss2: 0.0536 +Epoch [44/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0609, Loss2: 0.0558 +Epoch [44/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0908, Loss2: 0.0817 +Epoch [44/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 45.3125, Loss1: 0.0552, Loss2: 0.0554 +Epoch [44/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0644, Loss2: 0.0607 +Epoch [44/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 49.2188, Loss1: 0.0550, Loss2: 0.0502 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [44/200] Test Accuracy on the 10000 test data: Model1 20.1823 % Model2 21.0737 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.9062, Loss1: 0.0609, Loss2: 0.0585 +Epoch [45/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0546, Loss2: 0.0524 +Epoch [45/200], Iter [150/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.8750, Loss1: 0.0502, Loss2: 0.0470 +Epoch [45/200], Iter [200/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.8750, Loss1: 0.0607, Loss2: 0.0550 +Epoch [45/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 44.5312, Loss1: 0.0520, Loss2: 0.0522 +Epoch [45/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 39.0625, Loss1: 0.0562, Loss2: 0.0576 +Epoch [45/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0606, Loss2: 0.0620 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [45/200] Test Accuracy on the 10000 test data: Model1 21.0837 % Model2 21.9151 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0615, Loss2: 0.0586 +Epoch [46/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0603, Loss2: 0.0600 +Epoch [46/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0775, Loss2: 0.0709 +Epoch [46/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0793, Loss2: 0.0717 +Epoch [46/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0571, Loss2: 0.0564 +Epoch [46/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0635, Loss2: 0.0638 +Epoch [46/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.0000, Loss1: 0.0667, Loss2: 0.0659 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [46/200] Test Accuracy on the 10000 test data: Model1 21.1338 % Model2 21.3642 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 54.6875, Loss1: 0.0690, Loss2: 0.0632 +Epoch [47/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0755, Loss2: 0.0750 +Epoch [47/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 45.3125, Loss1: 0.0678, Loss2: 0.0654 +Epoch [47/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.0000, Loss1: 0.0548, Loss2: 0.0525 +Epoch [47/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 45.3125, Loss1: 0.0683, Loss2: 0.0661 +Epoch [47/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.0938, Loss1: 0.0537, Loss2: 0.0547 +Epoch [47/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 44.5312, Loss1: 0.0603, Loss2: 0.0607 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [47/200] Test Accuracy on the 10000 test data: Model1 19.8117 % Model2 20.9936 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0611, Loss2: 0.0590 +Epoch [48/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0742, Loss2: 0.0700 +Epoch [48/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 47.6562, Loss1: 0.0632, Loss2: 0.0614 +Epoch [48/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 52.3438, Loss1: 0.0583, Loss2: 0.0566 +Epoch [48/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0685, Loss2: 0.0667 +Epoch [48/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0555, Loss2: 0.0544 +Epoch [48/200], Iter [350/390] Training Accuracy1: 39.8438, Training Accuracy2: 42.1875, Loss1: 0.0541, Loss2: 0.0536 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [48/200] Test Accuracy on the 10000 test data: Model1 19.1206 % Model2 20.5028 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0668, Loss2: 0.0640 +Epoch [49/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0504, Loss2: 0.0507 +Epoch [49/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0784, Loss2: 0.0755 +Epoch [49/200], Iter [200/390] Training Accuracy1: 41.4062, Training Accuracy2: 45.3125, Loss1: 0.0647, Loss2: 0.0621 +Epoch [49/200], Iter [250/390] Training Accuracy1: 41.4062, Training Accuracy2: 42.9688, Loss1: 0.0728, Loss2: 0.0729 +Epoch [49/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0668, Loss2: 0.0682 +Epoch [49/200], Iter [350/390] Training Accuracy1: 36.7188, Training Accuracy2: 45.3125, Loss1: 0.0627, Loss2: 0.0576 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [49/200] Test Accuracy on the 10000 test data: Model1 20.2624 % Model2 21.8049 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0652, Loss2: 0.0636 +Epoch [50/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 39.8438, Loss1: 0.0544, Loss2: 0.0544 +Epoch [50/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0627, Loss2: 0.0589 +Epoch [50/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0657, Loss2: 0.0633 +Epoch [50/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0694, Loss2: 0.0703 +Epoch [50/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 46.0938, Loss1: 0.0574, Loss2: 0.0566 +Epoch [50/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0648, Loss2: 0.0609 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [50/200] Test Accuracy on the 10000 test data: Model1 20.1122 % Model2 22.1454 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0710, Loss2: 0.0680 +Epoch [51/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0619, Loss2: 0.0614 +Epoch [51/200], Iter [150/390] Training Accuracy1: 42.1875, Training Accuracy2: 41.4062, Loss1: 0.0624, Loss2: 0.0613 +Epoch [51/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0709, Loss2: 0.0670 +Epoch [51/200], Iter [250/390] Training Accuracy1: 42.9688, Training Accuracy2: 47.6562, Loss1: 0.0774, Loss2: 0.0704 +Epoch [51/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.8750, Loss1: 0.0513, Loss2: 0.0515 +Epoch [51/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 40.6250, Loss1: 0.0629, Loss2: 0.0696 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [51/200] Test Accuracy on the 10000 test data: Model1 19.9319 % Model2 19.5112 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200], Iter [50/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0576, Loss2: 0.0538 +Epoch [52/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 51.5625, Loss1: 0.0551, Loss2: 0.0514 +Epoch [52/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0656, Loss2: 0.0668 +Epoch [52/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 48.4375, Loss1: 0.0588, Loss2: 0.0586 +Epoch [52/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0622, Loss2: 0.0659 +Epoch [52/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 57.0312, Loss1: 0.0729, Loss2: 0.0661 +Epoch [52/200], Iter [350/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.0938, Loss1: 0.0650, Loss2: 0.0671 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [52/200] Test Accuracy on the 10000 test data: Model1 21.4643 % Model2 20.5228 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0580, Loss2: 0.0587 +Epoch [53/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 42.1875, Loss1: 0.0655, Loss2: 0.0696 +Epoch [53/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0577, Loss2: 0.0571 +Epoch [53/200], Iter [200/390] Training Accuracy1: 37.5000, Training Accuracy2: 44.5312, Loss1: 0.0498, Loss2: 0.0479 +Epoch [53/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0655, Loss2: 0.0622 +Epoch [53/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0564, Loss2: 0.0552 +Epoch [53/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.1250, Loss1: 0.0887, Loss2: 0.0840 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [53/200] Test Accuracy on the 10000 test data: Model1 20.5529 % Model2 20.3125 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.7812, Loss1: 0.0609, Loss2: 0.0565 +Epoch [54/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0648, Loss2: 0.0603 +Epoch [54/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0647, Loss2: 0.0617 +Epoch [54/200], Iter [200/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0674, Loss2: 0.0658 +Epoch [54/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0598, Loss2: 0.0621 +Epoch [54/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0792, Loss2: 0.0778 +Epoch [54/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0512, Loss2: 0.0502 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [54/200] Test Accuracy on the 10000 test data: Model1 20.6230 % Model2 21.8950 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 44.5312, Loss1: 0.0586, Loss2: 0.0588 +Epoch [55/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 52.3438, Loss1: 0.0632, Loss2: 0.0576 +Epoch [55/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0722, Loss2: 0.0731 +Epoch [55/200], Iter [200/390] Training Accuracy1: 40.6250, Training Accuracy2: 46.0938, Loss1: 0.0535, Loss2: 0.0504 +Epoch [55/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0706, Loss2: 0.0640 +Epoch [55/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0757, Loss2: 0.0753 +Epoch [55/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0784, Loss2: 0.0801 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [55/200] Test Accuracy on the 10000 test data: Model1 20.9135 % Model2 20.3025 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0809, Loss2: 0.0742 +Epoch [56/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0789, Loss2: 0.0780 +Epoch [56/200], Iter [150/390] Training Accuracy1: 44.5312, Training Accuracy2: 42.1875, Loss1: 0.0582, Loss2: 0.0575 +Epoch [56/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 47.6562, Loss1: 0.0459, Loss2: 0.0491 +Epoch [56/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.0938, Loss1: 0.0695, Loss2: 0.0681 +Epoch [56/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 57.0312, Loss1: 0.0780, Loss2: 0.0692 +Epoch [56/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0717, Loss2: 0.0713 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [56/200] Test Accuracy on the 10000 test data: Model1 19.9018 % Model2 21.4443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0689, Loss2: 0.0673 +Epoch [57/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0665, Loss2: 0.0638 +Epoch [57/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0818, Loss2: 0.0772 +Epoch [57/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 46.0938, Loss1: 0.0730, Loss2: 0.0724 +Epoch [57/200], Iter [250/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0625, Loss2: 0.0575 +Epoch [57/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0668, Loss2: 0.0621 +Epoch [57/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 49.2188, Loss1: 0.0642, Loss2: 0.0642 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [57/200] Test Accuracy on the 10000 test data: Model1 19.7917 % Model2 20.6831 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0516, Loss2: 0.0528 +Epoch [58/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0765, Loss2: 0.0764 +Epoch [58/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0877, Loss2: 0.0806 +Epoch [58/200], Iter [200/390] Training Accuracy1: 35.9375, Training Accuracy2: 43.7500, Loss1: 0.0554, Loss2: 0.0513 +Epoch [58/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 46.8750, Loss1: 0.0631, Loss2: 0.0615 +Epoch [58/200], Iter [300/390] Training Accuracy1: 42.9688, Training Accuracy2: 42.9688, Loss1: 0.0736, Loss2: 0.0748 +Epoch [58/200], Iter [350/390] Training Accuracy1: 38.2812, Training Accuracy2: 44.5312, Loss1: 0.0552, Loss2: 0.0523 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [58/200] Test Accuracy on the 10000 test data: Model1 20.7131 % Model2 21.8249 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200], Iter [50/390] Training Accuracy1: 39.0625, Training Accuracy2: 41.4062, Loss1: 0.0695, Loss2: 0.0680 +Epoch [59/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0678, Loss2: 0.0634 +Epoch [59/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0739, Loss2: 0.0734 +Epoch [59/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 55.4688, Loss1: 0.0556, Loss2: 0.0513 +Epoch [59/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0592, Loss2: 0.0611 +Epoch [59/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0680, Loss2: 0.0682 +Epoch [59/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0582, Loss2: 0.0571 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [59/200] Test Accuracy on the 10000 test data: Model1 18.9704 % Model2 20.6430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 54.6875, Loss1: 0.0727, Loss2: 0.0659 +Epoch [60/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0682, Loss2: 0.0638 +Epoch [60/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0779, Loss2: 0.0775 +Epoch [60/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0687, Loss2: 0.0670 +Epoch [60/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0654, Loss2: 0.0661 +Epoch [60/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0781, Loss2: 0.0800 +Epoch [60/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0656 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [60/200] Test Accuracy on the 10000 test data: Model1 20.1422 % Model2 20.2424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0681, Loss2: 0.0694 +Epoch [61/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.0000, Loss1: 0.0544, Loss2: 0.0534 +Epoch [61/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0663, Loss2: 0.0675 +Epoch [61/200], Iter [200/390] Training Accuracy1: 42.9688, Training Accuracy2: 48.4375, Loss1: 0.0664, Loss2: 0.0629 +Epoch [61/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.1103, Loss2: 0.0975 +Epoch [61/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0730, Loss2: 0.0721 +Epoch [61/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0675, Loss2: 0.0633 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [61/200] Test Accuracy on the 10000 test data: Model1 20.1222 % Model2 20.7232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0666, Loss2: 0.0605 +Epoch [62/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 43.7500, Loss1: 0.0586, Loss2: 0.0588 +Epoch [62/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0629, Loss2: 0.0654 +Epoch [62/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0744, Loss2: 0.0755 +Epoch [62/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 50.7812, Loss1: 0.0712, Loss2: 0.0780 +Epoch [62/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0640, Loss2: 0.0582 +Epoch [62/200], Iter [350/390] Training Accuracy1: 42.1875, Training Accuracy2: 50.7812, Loss1: 0.0623, Loss2: 0.0568 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [62/200] Test Accuracy on the 10000 test data: Model1 20.1923 % Model2 20.4627 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0536, Loss2: 0.0504 +Epoch [63/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0795, Loss2: 0.0753 +Epoch [63/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0675, Loss2: 0.0699 +Epoch [63/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 64.0625, Loss1: 0.0917, Loss2: 0.0781 +Epoch [63/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 56.2500, Loss1: 0.0682, Loss2: 0.0626 +Epoch [63/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 49.2188, Loss1: 0.0644, Loss2: 0.0624 +Epoch [63/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 49.2188, Loss1: 0.0575, Loss2: 0.0538 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [63/200] Test Accuracy on the 10000 test data: Model1 19.4111 % Model2 20.6430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0614, Loss2: 0.0643 +Epoch [64/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 54.6875, Loss1: 0.0605, Loss2: 0.0596 +Epoch [64/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0728, Loss2: 0.0721 +Epoch [64/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0686, Loss2: 0.0640 +Epoch [64/200], Iter [250/390] Training Accuracy1: 45.3125, Training Accuracy2: 48.4375, Loss1: 0.0491, Loss2: 0.0486 +Epoch [64/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0595, Loss2: 0.0578 +Epoch [64/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0750, Loss2: 0.0767 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [64/200] Test Accuracy on the 10000 test data: Model1 19.3810 % Model2 20.8934 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0672, Loss2: 0.0680 +Epoch [65/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0672, Loss2: 0.0664 +Epoch [65/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0600, Loss2: 0.0592 +Epoch [65/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0741, Loss2: 0.0771 +Epoch [65/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0759, Loss2: 0.0778 +Epoch [65/200], Iter [300/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.0938, Loss1: 0.0596, Loss2: 0.0588 +Epoch [65/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0819, Loss2: 0.0797 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [65/200] Test Accuracy on the 10000 test data: Model1 19.4912 % Model2 20.9435 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200], Iter [50/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0546, Loss2: 0.0523 +Epoch [66/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0796, Loss2: 0.0825 +Epoch [66/200], Iter [150/390] Training Accuracy1: 34.3750, Training Accuracy2: 41.4062, Loss1: 0.0630, Loss2: 0.0587 +Epoch [66/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0595, Loss2: 0.0607 +Epoch [66/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 36.7188, Loss1: 0.0600, Loss2: 0.0620 +Epoch [66/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0520, Loss2: 0.0531 +Epoch [66/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 50.7812, Loss1: 0.0606, Loss2: 0.0601 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [66/200] Test Accuracy on the 10000 test data: Model1 19.8618 % Model2 20.4327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0478, Loss2: 0.0469 +Epoch [67/200], Iter [100/390] Training Accuracy1: 49.2188, Training Accuracy2: 45.3125, Loss1: 0.0641, Loss2: 0.0660 +Epoch [67/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0513, Loss2: 0.0514 +Epoch [67/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0747, Loss2: 0.0734 +Epoch [67/200], Iter [250/390] Training Accuracy1: 40.6250, Training Accuracy2: 47.6562, Loss1: 0.0686, Loss2: 0.0641 +Epoch [67/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.7812, Loss1: 0.0684, Loss2: 0.0663 +Epoch [67/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0602, Loss2: 0.0577 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [67/200] Test Accuracy on the 10000 test data: Model1 20.2925 % Model2 20.9535 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0623, Loss2: 0.0605 +Epoch [68/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0655, Loss2: 0.0643 +Epoch [68/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0686, Loss2: 0.0651 +Epoch [68/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 55.4688, Loss1: 0.0770, Loss2: 0.0692 +Epoch [68/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 46.8750, Loss1: 0.0531, Loss2: 0.0534 +Epoch [68/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0921, Loss2: 0.0787 +Epoch [68/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0690, Loss2: 0.0694 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [68/200] Test Accuracy on the 10000 test data: Model1 20.0821 % Model2 20.8834 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0625, Loss2: 0.0607 +Epoch [69/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0870, Loss2: 0.0877 +Epoch [69/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 49.2188, Loss1: 0.0793, Loss2: 0.0801 +Epoch [69/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0692, Loss2: 0.0711 +Epoch [69/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0689, Loss2: 0.0686 +Epoch [69/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0702, Loss2: 0.0671 +Epoch [69/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0891, Loss2: 0.0872 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [69/200] Test Accuracy on the 10000 test data: Model1 20.3626 % Model2 21.3141 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0842, Loss2: 0.0788 +Epoch [70/200], Iter [100/390] Training Accuracy1: 46.8750, Training Accuracy2: 49.2188, Loss1: 0.0571, Loss2: 0.0558 +Epoch [70/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0585, Loss2: 0.0559 +Epoch [70/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0663, Loss2: 0.0666 +Epoch [70/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 41.4062, Loss1: 0.0624, Loss2: 0.0673 +Epoch [70/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 64.8438, Loss1: 0.0668, Loss2: 0.0582 +Epoch [70/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0583, Loss2: 0.0595 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [70/200] Test Accuracy on the 10000 test data: Model1 20.1823 % Model2 21.0437 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 47.6562, Loss1: 0.0623, Loss2: 0.0606 +Epoch [71/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0617, Loss2: 0.0591 +Epoch [71/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 53.1250, Loss1: 0.0700, Loss2: 0.0643 +Epoch [71/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 50.7812, Loss1: 0.0779, Loss2: 0.0769 +Epoch [71/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0848, Loss2: 0.0800 +Epoch [71/200], Iter [300/390] Training Accuracy1: 42.1875, Training Accuracy2: 44.5312, Loss1: 0.0585, Loss2: 0.0583 +Epoch [71/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0645, Loss2: 0.0623 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [71/200] Test Accuracy on the 10000 test data: Model1 19.1607 % Model2 20.9535 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0851, Loss2: 0.0796 +Epoch [72/200], Iter [100/390] Training Accuracy1: 39.8438, Training Accuracy2: 46.8750, Loss1: 0.0697, Loss2: 0.0651 +Epoch [72/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 50.0000, Loss1: 0.0568, Loss2: 0.0595 +Epoch [72/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0868, Loss2: 0.0853 +Epoch [72/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 42.9688, Loss1: 0.0630, Loss2: 0.0642 +Epoch [72/200], Iter [300/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0624, Loss2: 0.0593 +Epoch [72/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 56.2500, Loss1: 0.0558, Loss2: 0.0516 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [72/200] Test Accuracy on the 10000 test data: Model1 19.4712 % Model2 20.6430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200], Iter [50/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0782, Loss2: 0.0731 +Epoch [73/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.7812, Loss1: 0.0640, Loss2: 0.0592 +Epoch [73/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0631, Loss2: 0.0615 +Epoch [73/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 51.5625, Loss1: 0.0621, Loss2: 0.0568 +Epoch [73/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0700, Loss2: 0.0698 +Epoch [73/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 51.5625, Loss1: 0.0758, Loss2: 0.0715 +Epoch [73/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0692, Loss2: 0.0718 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [73/200] Test Accuracy on the 10000 test data: Model1 19.6114 % Model2 20.6230 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0657, Loss2: 0.0651 +Epoch [74/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0626, Loss2: 0.0642 +Epoch [74/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0776, Loss2: 0.0772 +Epoch [74/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0642, Loss2: 0.0637 +Epoch [74/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0645, Loss2: 0.0640 +Epoch [74/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 47.6562, Loss1: 0.0604, Loss2: 0.0588 +Epoch [74/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0587, Loss2: 0.0588 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [74/200] Test Accuracy on the 10000 test data: Model1 20.4828 % Model2 21.7047 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0936, Loss2: 0.0818 +Epoch [75/200], Iter [100/390] Training Accuracy1: 44.5312, Training Accuracy2: 41.4062, Loss1: 0.0680, Loss2: 0.0698 +Epoch [75/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0623, Loss2: 0.0633 +Epoch [75/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0677, Loss2: 0.0664 +Epoch [75/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0784, Loss2: 0.0710 +Epoch [75/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 53.1250, Loss1: 0.0795, Loss2: 0.0699 +Epoch [75/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0609, Loss2: 0.0614 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [75/200] Test Accuracy on the 10000 test data: Model1 19.4010 % Model2 20.7833 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0770, Loss2: 0.0735 +Epoch [76/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0695, Loss2: 0.0686 +Epoch [76/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 50.7812, Loss1: 0.0614, Loss2: 0.0595 +Epoch [76/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0664, Loss2: 0.0662 +Epoch [76/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0631, Loss2: 0.0613 +Epoch [76/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 51.5625, Loss1: 0.0691, Loss2: 0.0685 +Epoch [76/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0659, Loss2: 0.0649 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [76/200] Test Accuracy on the 10000 test data: Model1 20.4026 % Model2 20.8634 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 49.2188, Loss1: 0.0686, Loss2: 0.0668 +Epoch [77/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0698, Loss2: 0.0663 +Epoch [77/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 47.6562, Loss1: 0.0491, Loss2: 0.0499 +Epoch [77/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0606, Loss2: 0.0582 +Epoch [77/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 50.7812, Loss1: 0.0627, Loss2: 0.0592 +Epoch [77/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 44.5312, Loss1: 0.0601, Loss2: 0.0602 +Epoch [77/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0891, Loss2: 0.0888 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [77/200] Test Accuracy on the 10000 test data: Model1 20.1923 % Model2 20.8433 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200], Iter [50/390] Training Accuracy1: 46.8750, Training Accuracy2: 54.6875, Loss1: 0.0547, Loss2: 0.0491 +Epoch [78/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0798, Loss2: 0.0828 +Epoch [78/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0627, Loss2: 0.0610 +Epoch [78/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0761, Loss2: 0.0739 +Epoch [78/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0698, Loss2: 0.0689 +Epoch [78/200], Iter [300/390] Training Accuracy1: 43.7500, Training Accuracy2: 46.8750, Loss1: 0.0525, Loss2: 0.0520 +Epoch [78/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0896, Loss2: 0.0906 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [78/200] Test Accuracy on the 10000 test data: Model1 20.1823 % Model2 20.0821 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0542, Loss2: 0.0526 +Epoch [79/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0740, Loss2: 0.0738 +Epoch [79/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0806, Loss2: 0.0764 +Epoch [79/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.9062, Loss1: 0.0550, Loss2: 0.0536 +Epoch [79/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0705, Loss2: 0.0645 +Epoch [79/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0880, Loss2: 0.0853 +Epoch [79/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0696, Loss2: 0.0673 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [79/200] Test Accuracy on the 10000 test data: Model1 19.5913 % Model2 21.1338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0684, Loss2: 0.0696 +Epoch [80/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0711, Loss2: 0.0702 +Epoch [80/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0597, Loss2: 0.0564 +Epoch [80/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 45.3125, Loss1: 0.0544, Loss2: 0.0539 +Epoch [80/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0665, Loss2: 0.0622 +Epoch [80/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0576, Loss2: 0.0601 +Epoch [80/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 48.4375, Loss1: 0.0641, Loss2: 0.0626 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [80/200] Test Accuracy on the 10000 test data: Model1 20.5529 % Model2 20.1022 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0664, Loss2: 0.0667 +Epoch [81/200], Iter [100/390] Training Accuracy1: 50.7812, Training Accuracy2: 46.8750, Loss1: 0.0562, Loss2: 0.0587 +Epoch [81/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0865, Loss2: 0.0863 +Epoch [81/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0563, Loss2: 0.0573 +Epoch [81/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0722, Loss2: 0.0737 +Epoch [81/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0634, Loss2: 0.0601 +Epoch [81/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0780, Loss2: 0.0747 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [81/200] Test Accuracy on the 10000 test data: Model1 20.3025 % Model2 20.1923 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0660, Loss2: 0.0698 +Epoch [82/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0660, Loss2: 0.0668 +Epoch [82/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0676, Loss2: 0.0638 +Epoch [82/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.1250, Loss1: 0.0588, Loss2: 0.0541 +Epoch [82/200], Iter [250/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0734, Loss2: 0.0705 +Epoch [82/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0871, Loss2: 0.0836 +Epoch [82/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 53.1250, Loss1: 0.0743, Loss2: 0.0704 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [82/200] Test Accuracy on the 10000 test data: Model1 19.7015 % Model2 20.7131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0742, Loss2: 0.0774 +Epoch [83/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0718, Loss2: 0.0700 +Epoch [83/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0644, Loss2: 0.0646 +Epoch [83/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 46.8750, Loss1: 0.0756, Loss2: 0.0780 +Epoch [83/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0949, Loss2: 0.0880 +Epoch [83/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0641, Loss2: 0.0617 +Epoch [83/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0770, Loss2: 0.0703 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [83/200] Test Accuracy on the 10000 test data: Model1 21.0637 % Model2 20.8033 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0942, Loss2: 0.0968 +Epoch [84/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0800, Loss2: 0.0808 +Epoch [84/200], Iter [150/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0743, Loss2: 0.0732 +Epoch [84/200], Iter [200/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0634, Loss2: 0.0614 +Epoch [84/200], Iter [250/390] Training Accuracy1: 44.5312, Training Accuracy2: 57.8125, Loss1: 0.0712, Loss2: 0.0601 +Epoch [84/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0892, Loss2: 0.0841 +Epoch [84/200], Iter [350/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0618, Loss2: 0.0578 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [84/200] Test Accuracy on the 10000 test data: Model1 18.7600 % Model2 19.8417 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200], Iter [50/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0623, Loss2: 0.0597 +Epoch [85/200], Iter [100/390] Training Accuracy1: 43.7500, Training Accuracy2: 49.2188, Loss1: 0.0634, Loss2: 0.0591 +Epoch [85/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0677, Loss2: 0.0685 +Epoch [85/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0618, Loss2: 0.0610 +Epoch [85/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0761, Loss2: 0.0704 +Epoch [85/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0751, Loss2: 0.0751 +Epoch [85/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.9062, Loss1: 0.0727, Loss2: 0.0761 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [85/200] Test Accuracy on the 10000 test data: Model1 20.0521 % Model2 20.7933 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0645, Loss2: 0.0644 +Epoch [86/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.8125, Loss1: 0.0954, Loss2: 0.0974 +Epoch [86/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0788, Loss2: 0.0732 +Epoch [86/200], Iter [200/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.7812, Loss1: 0.0656, Loss2: 0.0652 +Epoch [86/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0867, Loss2: 0.0931 +Epoch [86/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0780, Loss2: 0.0805 +Epoch [86/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0598, Loss2: 0.0588 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [86/200] Test Accuracy on the 10000 test data: Model1 20.9135 % Model2 20.8333 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200], Iter [50/390] Training Accuracy1: 48.4375, Training Accuracy2: 49.2188, Loss1: 0.0779, Loss2: 0.0754 +Epoch [87/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0651, Loss2: 0.0640 +Epoch [87/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 50.7812, Loss1: 0.0754, Loss2: 0.0780 +Epoch [87/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.0875, Loss2: 0.0887 +Epoch [87/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0912, Loss2: 0.0934 +Epoch [87/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0671, Loss2: 0.0672 +Epoch [87/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0612, Loss2: 0.0622 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [87/200] Test Accuracy on the 10000 test data: Model1 20.5629 % Model2 21.1038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.9062, Loss1: 0.0698, Loss2: 0.0639 +Epoch [88/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0624, Loss2: 0.0614 +Epoch [88/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 51.5625, Loss1: 0.0527, Loss2: 0.0518 +Epoch [88/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.8125, Loss1: 0.0922, Loss2: 0.0838 +Epoch [88/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 55.4688, Loss1: 0.0888, Loss2: 0.0886 +Epoch [88/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 57.8125, Loss1: 0.0621, Loss2: 0.0562 +Epoch [88/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0668, Loss2: 0.0679 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [88/200] Test Accuracy on the 10000 test data: Model1 19.7216 % Model2 20.6330 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.7812, Loss1: 0.0526, Loss2: 0.0526 +Epoch [89/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0676, Loss2: 0.0668 +Epoch [89/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0664, Loss2: 0.0670 +Epoch [89/200], Iter [200/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0578, Loss2: 0.0571 +Epoch [89/200], Iter [250/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0771, Loss2: 0.0775 +Epoch [89/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0670, Loss2: 0.0675 +Epoch [89/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0543, Loss2: 0.0510 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [89/200] Test Accuracy on the 10000 test data: Model1 19.4611 % Model2 20.9235 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 47.6562, Loss1: 0.0757, Loss2: 0.0774 +Epoch [90/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 66.4062, Loss1: 0.0764, Loss2: 0.0716 +Epoch [90/200], Iter [150/390] Training Accuracy1: 46.8750, Training Accuracy2: 45.3125, Loss1: 0.0670, Loss2: 0.0690 +Epoch [90/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 46.0938, Loss1: 0.0612, Loss2: 0.0650 +Epoch [90/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0670, Loss2: 0.0724 +Epoch [90/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0645, Loss2: 0.0636 +Epoch [90/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 59.3750, Loss1: 0.0633, Loss2: 0.0661 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [90/200] Test Accuracy on the 10000 test data: Model1 20.7532 % Model2 20.9335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0815, Loss2: 0.0788 +Epoch [91/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0770, Loss2: 0.0771 +Epoch [91/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 56.2500, Loss1: 0.0657, Loss2: 0.0603 +Epoch [91/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0665, Loss2: 0.0656 +Epoch [91/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 53.1250, Loss1: 0.0628, Loss2: 0.0663 +Epoch [91/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0791, Loss2: 0.0775 +Epoch [91/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0708, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [91/200] Test Accuracy on the 10000 test data: Model1 20.6631 % Model2 20.3826 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0621, Loss2: 0.0662 +Epoch [92/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 56.2500, Loss1: 0.0700, Loss2: 0.0619 +Epoch [92/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0763, Loss2: 0.0710 +Epoch [92/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0555, Loss2: 0.0590 +Epoch [92/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0764, Loss2: 0.0716 +Epoch [92/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0647, Loss2: 0.0589 +Epoch [92/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0634, Loss2: 0.0629 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [92/200] Test Accuracy on the 10000 test data: Model1 19.4111 % Model2 20.6430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 54.6875, Loss1: 0.0539, Loss2: 0.0519 +Epoch [93/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0617, Loss2: 0.0585 +Epoch [93/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0676, Loss2: 0.0679 +Epoch [93/200], Iter [200/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.1017, Loss2: 0.1023 +Epoch [93/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0813, Loss2: 0.0786 +Epoch [93/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 50.7812, Loss1: 0.0702, Loss2: 0.0737 +Epoch [93/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 43.7500, Loss1: 0.0666, Loss2: 0.0697 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [93/200] Test Accuracy on the 10000 test data: Model1 20.1723 % Model2 20.5429 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 61.7188, Loss1: 0.0731, Loss2: 0.0677 +Epoch [94/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0771, Loss2: 0.0777 +Epoch [94/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0718, Loss2: 0.0740 +Epoch [94/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0823, Loss2: 0.0837 +Epoch [94/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0793, Loss2: 0.0734 +Epoch [94/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0673, Loss2: 0.0662 +Epoch [94/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 51.5625, Loss1: 0.0669, Loss2: 0.0634 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [94/200] Test Accuracy on the 10000 test data: Model1 19.7416 % Model2 20.2424 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0867, Loss2: 0.0862 +Epoch [95/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0746, Loss2: 0.0741 +Epoch [95/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0669, Loss2: 0.0673 +Epoch [95/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 51.5625, Loss1: 0.0654, Loss2: 0.0693 +Epoch [95/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0792, Loss2: 0.0774 +Epoch [95/200], Iter [300/390] Training Accuracy1: 46.8750, Training Accuracy2: 48.4375, Loss1: 0.0832, Loss2: 0.0848 +Epoch [95/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0704, Loss2: 0.0698 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [95/200] Test Accuracy on the 10000 test data: Model1 19.7316 % Model2 21.1038 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0783, Loss2: 0.0701 +Epoch [96/200], Iter [100/390] Training Accuracy1: 45.3125, Training Accuracy2: 53.9062, Loss1: 0.0580, Loss2: 0.0507 +Epoch [96/200], Iter [150/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.8750, Loss1: 0.0580, Loss2: 0.0571 +Epoch [96/200], Iter [200/390] Training Accuracy1: 45.3125, Training Accuracy2: 47.6562, Loss1: 0.0689, Loss2: 0.0687 +Epoch [96/200], Iter [250/390] Training Accuracy1: 46.8750, Training Accuracy2: 50.0000, Loss1: 0.0762, Loss2: 0.0714 +Epoch [96/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.0000, Loss1: 0.0734, Loss2: 0.0770 +Epoch [96/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0676, Loss2: 0.0653 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [96/200] Test Accuracy on the 10000 test data: Model1 20.7833 % Model2 20.8934 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 60.1562, Loss1: 0.0625, Loss2: 0.0541 +Epoch [97/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0575, Loss2: 0.0616 +Epoch [97/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0657, Loss2: 0.0655 +Epoch [97/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0678, Loss2: 0.0634 +Epoch [97/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0664, Loss2: 0.0676 +Epoch [97/200], Iter [300/390] Training Accuracy1: 45.3125, Training Accuracy2: 50.0000, Loss1: 0.0598, Loss2: 0.0571 +Epoch [97/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0805, Loss2: 0.0835 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [97/200] Test Accuracy on the 10000 test data: Model1 20.1923 % Model2 20.9635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0771, Loss2: 0.0731 +Epoch [98/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0747, Loss2: 0.0795 +Epoch [98/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0639, Loss2: 0.0595 +Epoch [98/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0923, Loss2: 0.0819 +Epoch [98/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0618, Loss2: 0.0593 +Epoch [98/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0968, Loss2: 0.0880 +Epoch [98/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0664, Loss2: 0.0659 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [98/200] Test Accuracy on the 10000 test data: Model1 20.3425 % Model2 21.1338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0723, Loss2: 0.0704 +Epoch [99/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 49.2188, Loss1: 0.0685, Loss2: 0.0705 +Epoch [99/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 50.7812, Loss1: 0.0708, Loss2: 0.0692 +Epoch [99/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0665, Loss2: 0.0637 +Epoch [99/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 46.0938, Loss1: 0.0693, Loss2: 0.0725 +Epoch [99/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 62.5000, Loss1: 0.0851, Loss2: 0.0727 +Epoch [99/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 51.5625, Loss1: 0.0773, Loss2: 0.0715 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [99/200] Test Accuracy on the 10000 test data: Model1 20.3125 % Model2 21.3041 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0512, Loss2: 0.0525 +Epoch [100/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 61.7188, Loss1: 0.0786, Loss2: 0.0733 +Epoch [100/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0637, Loss2: 0.0616 +Epoch [100/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 49.2188, Loss1: 0.0630, Loss2: 0.0644 +Epoch [100/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0785, Loss2: 0.0790 +Epoch [100/200], Iter [300/390] Training Accuracy1: 44.5312, Training Accuracy2: 50.0000, Loss1: 0.0659, Loss2: 0.0614 +Epoch [100/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0784, Loss2: 0.0759 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [100/200] Test Accuracy on the 10000 test data: Model1 20.4427 % Model2 20.5829 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200], Iter [50/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0617, Loss2: 0.0572 +Epoch [101/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0724, Loss2: 0.0746 +Epoch [101/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 55.4688, Loss1: 0.0602, Loss2: 0.0558 +Epoch [101/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0959, Loss2: 0.0982 +Epoch [101/200], Iter [250/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0516, Loss2: 0.0511 +Epoch [101/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0832, Loss2: 0.0797 +Epoch [101/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0585, Loss2: 0.0584 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [101/200] Test Accuracy on the 10000 test data: Model1 20.4728 % Model2 21.3241 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0617, Loss2: 0.0614 +Epoch [102/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0664, Loss2: 0.0682 +Epoch [102/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0598, Loss2: 0.0579 +Epoch [102/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0708, Loss2: 0.0710 +Epoch [102/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 51.5625, Loss1: 0.0620, Loss2: 0.0665 +Epoch [102/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0677, Loss2: 0.0653 +Epoch [102/200], Iter [350/390] Training Accuracy1: 42.9688, Training Accuracy2: 45.3125, Loss1: 0.0554, Loss2: 0.0532 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [102/200] Test Accuracy on the 10000 test data: Model1 19.8618 % Model2 20.7131 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 62.5000, Loss1: 0.0741, Loss2: 0.0673 +Epoch [103/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0649, Loss2: 0.0618 +Epoch [103/200], Iter [150/390] Training Accuracy1: 54.6875, Training Accuracy2: 49.2188, Loss1: 0.0527, Loss2: 0.0562 +Epoch [103/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0654, Loss2: 0.0643 +Epoch [103/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0608, Loss2: 0.0591 +Epoch [103/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 46.8750, Loss1: 0.0655, Loss2: 0.0674 +Epoch [103/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 52.3438, Loss1: 0.0671, Loss2: 0.0707 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [103/200] Test Accuracy on the 10000 test data: Model1 20.3325 % Model2 20.7632 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0607, Loss2: 0.0623 +Epoch [104/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0760, Loss2: 0.0747 +Epoch [104/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0756, Loss2: 0.0803 +Epoch [104/200], Iter [200/390] Training Accuracy1: 43.7500, Training Accuracy2: 48.4375, Loss1: 0.0540, Loss2: 0.0520 +Epoch [104/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0682, Loss2: 0.0671 +Epoch [104/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0736, Loss2: 0.0707 +Epoch [104/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0774, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [104/200] Test Accuracy on the 10000 test data: Model1 19.9119 % Model2 21.3341 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0763, Loss2: 0.0736 +Epoch [105/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0750, Loss2: 0.0685 +Epoch [105/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0664, Loss2: 0.0662 +Epoch [105/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 57.0312, Loss1: 0.0630, Loss2: 0.0590 +Epoch [105/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0572, Loss2: 0.0549 +Epoch [105/200], Iter [300/390] Training Accuracy1: 49.2188, Training Accuracy2: 47.6562, Loss1: 0.0532, Loss2: 0.0561 +Epoch [105/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 54.6875, Loss1: 0.0658, Loss2: 0.0616 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [105/200] Test Accuracy on the 10000 test data: Model1 20.2224 % Model2 20.7632 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200], Iter [50/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0831, Loss2: 0.0791 +Epoch [106/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0801, Loss2: 0.0811 +Epoch [106/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.0312, Loss1: 0.0685, Loss2: 0.0670 +Epoch [106/200], Iter [200/390] Training Accuracy1: 46.0938, Training Accuracy2: 56.2500, Loss1: 0.0639, Loss2: 0.0575 +Epoch [106/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 55.4688, Loss1: 0.0680, Loss2: 0.0709 +Epoch [106/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0629, Loss2: 0.0626 +Epoch [106/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0704, Loss2: 0.0729 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [106/200] Test Accuracy on the 10000 test data: Model1 21.0637 % Model2 20.4928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0709, Loss2: 0.0699 +Epoch [107/200], Iter [100/390] Training Accuracy1: 47.6562, Training Accuracy2: 52.3438, Loss1: 0.0585, Loss2: 0.0566 +Epoch [107/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0788, Loss2: 0.0721 +Epoch [107/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 50.0000, Loss1: 0.0612, Loss2: 0.0664 +Epoch [107/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0572, Loss2: 0.0539 +Epoch [107/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0744, Loss2: 0.0710 +Epoch [107/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0605, Loss2: 0.0594 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [107/200] Test Accuracy on the 10000 test data: Model1 20.4327 % Model2 20.2524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0746, Loss2: 0.0679 +Epoch [108/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0811, Loss2: 0.0820 +Epoch [108/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 60.1562, Loss1: 0.0817, Loss2: 0.0748 +Epoch [108/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0775, Loss2: 0.0739 +Epoch [108/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.1122, Loss2: 0.1031 +Epoch [108/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0605, Loss2: 0.0580 +Epoch [108/200], Iter [350/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0616, Loss2: 0.0593 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [108/200] Test Accuracy on the 10000 test data: Model1 21.0337 % Model2 20.9736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0658, Loss2: 0.0679 +Epoch [109/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0761, Loss2: 0.0719 +Epoch [109/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 48.4375, Loss1: 0.0697, Loss2: 0.0720 +Epoch [109/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0836, Loss2: 0.0817 +Epoch [109/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0776, Loss2: 0.0769 +Epoch [109/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0920, Loss2: 0.0926 +Epoch [109/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0791, Loss2: 0.0788 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [109/200] Test Accuracy on the 10000 test data: Model1 20.4026 % Model2 20.6430 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0740, Loss2: 0.0722 +Epoch [110/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0738, Loss2: 0.0771 +Epoch [110/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0942, Loss2: 0.0916 +Epoch [110/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0863, Loss2: 0.0859 +Epoch [110/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 47.6562, Loss1: 0.0724, Loss2: 0.0811 +Epoch [110/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0703, Loss2: 0.0701 +Epoch [110/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0635, Loss2: 0.0646 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [110/200] Test Accuracy on the 10000 test data: Model1 21.2640 % Model2 20.2524 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0588, Loss2: 0.0597 +Epoch [111/200], Iter [100/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0687, Loss2: 0.0647 +Epoch [111/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0853, Loss2: 0.0840 +Epoch [111/200], Iter [200/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.1250, Loss1: 0.0695, Loss2: 0.0712 +Epoch [111/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0724, Loss2: 0.0712 +Epoch [111/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0735, Loss2: 0.0685 +Epoch [111/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0703, Loss2: 0.0674 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [111/200] Test Accuracy on the 10000 test data: Model1 20.2424 % Model2 21.0337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0725, Loss2: 0.0679 +Epoch [112/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0709, Loss2: 0.0663 +Epoch [112/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0670, Loss2: 0.0667 +Epoch [112/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 53.1250, Loss1: 0.0623, Loss2: 0.0607 +Epoch [112/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.1011, Loss2: 0.0976 +Epoch [112/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0686, Loss2: 0.0680 +Epoch [112/200], Iter [350/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0648, Loss2: 0.0619 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [112/200] Test Accuracy on the 10000 test data: Model1 20.5529 % Model2 20.9235 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0798, Loss2: 0.0770 +Epoch [113/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 56.2500, Loss1: 0.0748, Loss2: 0.0736 +Epoch [113/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0564, Loss2: 0.0586 +Epoch [113/200], Iter [200/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0735, Loss2: 0.0732 +Epoch [113/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0590, Loss2: 0.0591 +Epoch [113/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0851, Loss2: 0.0761 +Epoch [113/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 50.7812, Loss1: 0.0609, Loss2: 0.0607 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [113/200] Test Accuracy on the 10000 test data: Model1 20.6530 % Model2 21.3842 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0624, Loss2: 0.0581 +Epoch [114/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 57.0312, Loss1: 0.0801, Loss2: 0.0751 +Epoch [114/200], Iter [150/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0688, Loss2: 0.0691 +Epoch [114/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 52.3438, Loss1: 0.0565, Loss2: 0.0581 +Epoch [114/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0711, Loss2: 0.0689 +Epoch [114/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0704, Loss2: 0.0650 +Epoch [114/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0759, Loss2: 0.0776 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [114/200] Test Accuracy on the 10000 test data: Model1 19.5212 % Model2 21.3542 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0674, Loss2: 0.0679 +Epoch [115/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 50.7812, Loss1: 0.0647, Loss2: 0.0626 +Epoch [115/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0796, Loss2: 0.0758 +Epoch [115/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0661, Loss2: 0.0641 +Epoch [115/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0569, Loss2: 0.0551 +Epoch [115/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.9062, Loss1: 0.0837, Loss2: 0.0764 +Epoch [115/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0694, Loss2: 0.0653 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [115/200] Test Accuracy on the 10000 test data: Model1 20.5228 % Model2 20.8133 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0639, Loss2: 0.0637 +Epoch [116/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0828, Loss2: 0.0807 +Epoch [116/200], Iter [150/390] Training Accuracy1: 47.6562, Training Accuracy2: 53.1250, Loss1: 0.0703, Loss2: 0.0660 +Epoch [116/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0682, Loss2: 0.0618 +Epoch [116/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0682, Loss2: 0.0663 +Epoch [116/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 52.3438, Loss1: 0.0754, Loss2: 0.0726 +Epoch [116/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0613, Loss2: 0.0586 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [116/200] Test Accuracy on the 10000 test data: Model1 20.2324 % Model2 21.3842 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 48.4375, Loss1: 0.0772, Loss2: 0.0759 +Epoch [117/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 52.3438, Loss1: 0.0644, Loss2: 0.0648 +Epoch [117/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0678, Loss2: 0.0680 +Epoch [117/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0621, Loss2: 0.0658 +Epoch [117/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0820, Loss2: 0.0775 +Epoch [117/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0672, Loss2: 0.0658 +Epoch [117/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0635, Loss2: 0.0630 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [117/200] Test Accuracy on the 10000 test data: Model1 20.5329 % Model2 21.0236 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0694, Loss2: 0.0704 +Epoch [118/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0979, Loss2: 0.0960 +Epoch [118/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0901, Loss2: 0.0832 +Epoch [118/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0643, Loss2: 0.0615 +Epoch [118/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 50.7812, Loss1: 0.0636, Loss2: 0.0634 +Epoch [118/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0698, Loss2: 0.0690 +Epoch [118/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 55.4688, Loss1: 0.0649, Loss2: 0.0664 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [118/200] Test Accuracy on the 10000 test data: Model1 20.9936 % Model2 20.9335 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0632, Loss2: 0.0614 +Epoch [119/200], Iter [100/390] Training Accuracy1: 46.0938, Training Accuracy2: 46.0938, Loss1: 0.0635, Loss2: 0.0643 +Epoch [119/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0666, Loss2: 0.0644 +Epoch [119/200], Iter [200/390] Training Accuracy1: 44.5312, Training Accuracy2: 48.4375, Loss1: 0.0622, Loss2: 0.0597 +Epoch [119/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0635, Loss2: 0.0643 +Epoch [119/200], Iter [300/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0652, Loss2: 0.0693 +Epoch [119/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 54.6875, Loss1: 0.0761, Loss2: 0.0723 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [119/200] Test Accuracy on the 10000 test data: Model1 20.3926 % Model2 21.0136 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200], Iter [50/390] Training Accuracy1: 50.7812, Training Accuracy2: 60.9375, Loss1: 0.0647, Loss2: 0.0557 +Epoch [120/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 54.6875, Loss1: 0.0744, Loss2: 0.0830 +Epoch [120/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0677, Loss2: 0.0644 +Epoch [120/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 58.5938, Loss1: 0.0682, Loss2: 0.0647 +Epoch [120/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0730, Loss2: 0.0740 +Epoch [120/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0896, Loss2: 0.0922 +Epoch [120/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 54.6875, Loss1: 0.0500, Loss2: 0.0508 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [120/200] Test Accuracy on the 10000 test data: Model1 20.6430 % Model2 20.4928 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0771, Loss2: 0.0735 +Epoch [121/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0624, Loss2: 0.0606 +Epoch [121/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0603, Loss2: 0.0591 +Epoch [121/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0767, Loss2: 0.0699 +Epoch [121/200], Iter [250/390] Training Accuracy1: 47.6562, Training Accuracy2: 57.0312, Loss1: 0.0740, Loss2: 0.0655 +Epoch [121/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 54.6875, Loss1: 0.0897, Loss2: 0.1005 +Epoch [121/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0712, Loss2: 0.0717 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [121/200] Test Accuracy on the 10000 test data: Model1 20.8033 % Model2 21.0837 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0695, Loss2: 0.0653 +Epoch [122/200], Iter [100/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0701, Loss2: 0.0676 +Epoch [122/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0588, Loss2: 0.0573 +Epoch [122/200], Iter [200/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.9062, Loss1: 0.0705, Loss2: 0.0681 +Epoch [122/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0525, Loss2: 0.0522 +Epoch [122/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.1250, Loss1: 0.0700, Loss2: 0.0728 +Epoch [122/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 64.0625, Loss1: 0.0786, Loss2: 0.0676 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [122/200] Test Accuracy on the 10000 test data: Model1 19.9619 % Model2 21.0036 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 54.6875, Loss1: 0.0940, Loss2: 0.0917 +Epoch [123/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0776, Loss2: 0.0744 +Epoch [123/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 55.4688, Loss1: 0.0744, Loss2: 0.0803 +Epoch [123/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0656, Loss2: 0.0640 +Epoch [123/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0526, Loss2: 0.0501 +Epoch [123/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0691, Loss2: 0.0644 +Epoch [123/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0731, Loss2: 0.0727 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [123/200] Test Accuracy on the 10000 test data: Model1 20.8834 % Model2 20.9034 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200], Iter [50/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.0625, Loss1: 0.1000, Loss2: 0.1079 +Epoch [124/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 46.0938, Loss1: 0.0833, Loss2: 0.0907 +Epoch [124/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0677, Loss2: 0.0660 +Epoch [124/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 58.5938, Loss1: 0.0574, Loss2: 0.0625 +Epoch [124/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0689, Loss2: 0.0681 +Epoch [124/200], Iter [300/390] Training Accuracy1: 50.0000, Training Accuracy2: 53.9062, Loss1: 0.0641, Loss2: 0.0618 +Epoch [124/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 58.5938, Loss1: 0.0570, Loss2: 0.0535 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [124/200] Test Accuracy on the 10000 test data: Model1 20.1823 % Model2 21.1438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0805, Loss2: 0.0786 +Epoch [125/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0677, Loss2: 0.0682 +Epoch [125/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 48.4375, Loss1: 0.0699, Loss2: 0.0745 +Epoch [125/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0554, Loss2: 0.0535 +Epoch [125/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1035, Loss2: 0.1096 +Epoch [125/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 51.5625, Loss1: 0.0660, Loss2: 0.0641 +Epoch [125/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0932, Loss2: 0.0912 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [125/200] Test Accuracy on the 10000 test data: Model1 20.6430 % Model2 21.0337 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0742, Loss2: 0.0769 +Epoch [126/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0754, Loss2: 0.0751 +Epoch [126/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 61.7188, Loss1: 0.0780, Loss2: 0.0745 +Epoch [126/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 56.2500, Loss1: 0.0787, Loss2: 0.0752 +Epoch [126/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0700, Loss2: 0.0680 +Epoch [126/200], Iter [300/390] Training Accuracy1: 50.7812, Training Accuracy2: 55.4688, Loss1: 0.0964, Loss2: 0.0874 +Epoch [126/200], Iter [350/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0570, Loss2: 0.0560 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [126/200] Test Accuracy on the 10000 test data: Model1 20.7332 % Model2 20.6330 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0848, Loss2: 0.0861 +Epoch [127/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.9062, Loss1: 0.0780, Loss2: 0.0764 +Epoch [127/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0727, Loss2: 0.0682 +Epoch [127/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0730, Loss2: 0.0679 +Epoch [127/200], Iter [250/390] Training Accuracy1: 50.0000, Training Accuracy2: 52.3438, Loss1: 0.0796, Loss2: 0.0785 +Epoch [127/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0676, Loss2: 0.0664 +Epoch [127/200], Iter [350/390] Training Accuracy1: 46.0938, Training Accuracy2: 48.4375, Loss1: 0.0581, Loss2: 0.0565 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [127/200] Test Accuracy on the 10000 test data: Model1 20.7632 % Model2 20.7332 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0699, Loss2: 0.0715 +Epoch [128/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0835, Loss2: 0.0872 +Epoch [128/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0596, Loss2: 0.0589 +Epoch [128/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0911, Loss2: 0.0920 +Epoch [128/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0657, Loss2: 0.0674 +Epoch [128/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 47.6562, Loss1: 0.0573, Loss2: 0.0577 +Epoch [128/200], Iter [350/390] Training Accuracy1: 47.6562, Training Accuracy2: 45.3125, Loss1: 0.0599, Loss2: 0.0609 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [128/200] Test Accuracy on the 10000 test data: Model1 19.8417 % Model2 20.7532 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0750, Loss2: 0.0748 +Epoch [129/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.0844, Loss2: 0.0834 +Epoch [129/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 61.7188, Loss1: 0.0809, Loss2: 0.0734 +Epoch [129/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0842, Loss2: 0.0893 +Epoch [129/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 53.9062, Loss1: 0.0624, Loss2: 0.0645 +Epoch [129/200], Iter [300/390] Training Accuracy1: 48.4375, Training Accuracy2: 53.1250, Loss1: 0.0854, Loss2: 0.0823 +Epoch [129/200], Iter [350/390] Training Accuracy1: 50.7812, Training Accuracy2: 53.1250, Loss1: 0.0650, Loss2: 0.0624 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [129/200] Test Accuracy on the 10000 test data: Model1 20.8534 % Model2 20.9736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 55.4688, Loss1: 0.0690, Loss2: 0.0640 +Epoch [130/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0946, Loss2: 0.0892 +Epoch [130/200], Iter [150/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0683, Loss2: 0.0681 +Epoch [130/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0626, Loss2: 0.0613 +Epoch [130/200], Iter [250/390] Training Accuracy1: 51.5625, Training Accuracy2: 56.2500, Loss1: 0.0710, Loss2: 0.0654 +Epoch [130/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 49.2188, Loss1: 0.0545, Loss2: 0.0559 +Epoch [130/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.0813, Loss2: 0.0810 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [130/200] Test Accuracy on the 10000 test data: Model1 20.7232 % Model2 20.7632 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0766, Loss2: 0.0735 +Epoch [131/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 50.0000, Loss1: 0.0700, Loss2: 0.0694 +Epoch [131/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.9062, Loss1: 0.0576, Loss2: 0.0599 +Epoch [131/200], Iter [200/390] Training Accuracy1: 51.5625, Training Accuracy2: 52.3438, Loss1: 0.0757, Loss2: 0.0767 +Epoch [131/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0604, Loss2: 0.0602 +Epoch [131/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0893, Loss2: 0.0917 +Epoch [131/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0827, Loss2: 0.0843 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [131/200] Test Accuracy on the 10000 test data: Model1 20.4527 % Model2 21.2440 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 68.7500, Loss1: 0.0771, Loss2: 0.0736 +Epoch [132/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0791, Loss2: 0.0744 +Epoch [132/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 64.0625, Loss1: 0.0719, Loss2: 0.0691 +Epoch [132/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.1562, Loss1: 0.0706, Loss2: 0.0644 +Epoch [132/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.9062, Loss1: 0.0687, Loss2: 0.0699 +Epoch [132/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0772, Loss2: 0.0745 +Epoch [132/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0917, Loss2: 0.0902 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [132/200] Test Accuracy on the 10000 test data: Model1 20.5829 % Model2 20.9635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200], Iter [50/390] Training Accuracy1: 47.6562, Training Accuracy2: 48.4375, Loss1: 0.0650, Loss2: 0.0633 +Epoch [133/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0742, Loss2: 0.0739 +Epoch [133/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0706, Loss2: 0.0688 +Epoch [133/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 54.6875, Loss1: 0.0761, Loss2: 0.0792 +Epoch [133/200], Iter [250/390] Training Accuracy1: 53.1250, Training Accuracy2: 53.1250, Loss1: 0.0728, Loss2: 0.0719 +Epoch [133/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0714, Loss2: 0.0727 +Epoch [133/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1039, Loss2: 0.1009 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [133/200] Test Accuracy on the 10000 test data: Model1 20.2524 % Model2 20.4327 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0832, Loss2: 0.0841 +Epoch [134/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 54.6875, Loss1: 0.0700, Loss2: 0.0696 +Epoch [134/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0736, Loss2: 0.0720 +Epoch [134/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0898, Loss2: 0.0894 +Epoch [134/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0645, Loss2: 0.0618 +Epoch [134/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0801, Loss2: 0.0768 +Epoch [134/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 52.3438, Loss1: 0.0615, Loss2: 0.0645 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [134/200] Test Accuracy on the 10000 test data: Model1 20.1723 % Model2 21.1138 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 53.9062, Loss1: 0.0655, Loss2: 0.0673 +Epoch [135/200], Iter [100/390] Training Accuracy1: 48.4375, Training Accuracy2: 55.4688, Loss1: 0.0760, Loss2: 0.0710 +Epoch [135/200], Iter [150/390] Training Accuracy1: 51.5625, Training Accuracy2: 55.4688, Loss1: 0.0668, Loss2: 0.0652 +Epoch [135/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.1011, Loss2: 0.0972 +Epoch [135/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0969, Loss2: 0.0890 +Epoch [135/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 51.5625, Loss1: 0.0705, Loss2: 0.0727 +Epoch [135/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 67.1875, Loss1: 0.0664, Loss2: 0.0604 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [135/200] Test Accuracy on the 10000 test data: Model1 20.8333 % Model2 21.0537 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200], Iter [50/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.1875, Loss1: 0.0863, Loss2: 0.0878 +Epoch [136/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0784, Loss2: 0.0737 +Epoch [136/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0698, Loss2: 0.0746 +Epoch [136/200], Iter [200/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0766, Loss2: 0.0747 +Epoch [136/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0646, Loss2: 0.0679 +Epoch [136/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0814, Loss2: 0.0821 +Epoch [136/200], Iter [350/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0646, Loss2: 0.0628 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [136/200] Test Accuracy on the 10000 test data: Model1 21.1038 % Model2 21.0036 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0917, Loss2: 0.0871 +Epoch [137/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0687, Loss2: 0.0681 +Epoch [137/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0828, Loss2: 0.0782 +Epoch [137/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0829, Loss2: 0.0814 +Epoch [137/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0730, Loss2: 0.0719 +Epoch [137/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 58.5938, Loss1: 0.0742, Loss2: 0.0724 +Epoch [137/200], Iter [350/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.9375, Loss1: 0.0820, Loss2: 0.0795 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [137/200] Test Accuracy on the 10000 test data: Model1 20.5629 % Model2 21.3942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0694, Loss2: 0.0651 +Epoch [138/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0667, Loss2: 0.0698 +Epoch [138/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0759, Loss2: 0.0773 +Epoch [138/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.8125, Loss1: 0.0696, Loss2: 0.0715 +Epoch [138/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0762, Loss2: 0.0752 +Epoch [138/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0782, Loss2: 0.0768 +Epoch [138/200], Iter [350/390] Training Accuracy1: 52.3438, Training Accuracy2: 48.4375, Loss1: 0.0715, Loss2: 0.0762 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [138/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 20.8534 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.8125, Loss1: 0.0661, Loss2: 0.0656 +Epoch [139/200], Iter [100/390] Training Accuracy1: 53.1250, Training Accuracy2: 60.9375, Loss1: 0.0712, Loss2: 0.0653 +Epoch [139/200], Iter [150/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0581, Loss2: 0.0549 +Epoch [139/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 71.8750, Loss1: 0.0756, Loss2: 0.0668 +Epoch [139/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0843, Loss2: 0.0831 +Epoch [139/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0898, Loss2: 0.0932 +Epoch [139/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.0931, Loss2: 0.0990 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [139/200] Test Accuracy on the 10000 test data: Model1 20.9836 % Model2 20.9736 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0656, Loss2: 0.0652 +Epoch [140/200], Iter [100/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0833, Loss2: 0.0777 +Epoch [140/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0671, Loss2: 0.0662 +Epoch [140/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.8438, Loss1: 0.0799, Loss2: 0.0788 +Epoch [140/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0923, Loss2: 0.0903 +Epoch [140/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 58.5938, Loss1: 0.1052, Loss2: 0.1154 +Epoch [140/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0790, Loss2: 0.0774 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [140/200] Test Accuracy on the 10000 test data: Model1 21.0737 % Model2 21.1438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 55.4688, Loss1: 0.0657, Loss2: 0.0695 +Epoch [141/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0796, Loss2: 0.0764 +Epoch [141/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0759, Loss2: 0.0785 +Epoch [141/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.1562, Loss1: 0.0683, Loss2: 0.0670 +Epoch [141/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0729, Loss2: 0.0674 +Epoch [141/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 60.1562, Loss1: 0.0751, Loss2: 0.0737 +Epoch [141/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 59.3750, Loss1: 0.0663, Loss2: 0.0716 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [141/200] Test Accuracy on the 10000 test data: Model1 20.5729 % Model2 20.7332 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0736, Loss2: 0.0736 +Epoch [142/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0867, Loss2: 0.0886 +Epoch [142/200], Iter [150/390] Training Accuracy1: 53.9062, Training Accuracy2: 51.5625, Loss1: 0.0731, Loss2: 0.0754 +Epoch [142/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0703, Loss2: 0.0717 +Epoch [142/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0752, Loss2: 0.0713 +Epoch [142/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 61.7188, Loss1: 0.0754, Loss2: 0.0697 +Epoch [142/200], Iter [350/390] Training Accuracy1: 49.2188, Training Accuracy2: 49.2188, Loss1: 0.0766, Loss2: 0.0771 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [142/200] Test Accuracy on the 10000 test data: Model1 21.2039 % Model2 20.9836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200], Iter [50/390] Training Accuracy1: 53.1250, Training Accuracy2: 51.5625, Loss1: 0.0653, Loss2: 0.0664 +Epoch [143/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0675, Loss2: 0.0682 +Epoch [143/200], Iter [150/390] Training Accuracy1: 57.0312, Training Accuracy2: 64.8438, Loss1: 0.0956, Loss2: 0.0819 +Epoch [143/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0678, Loss2: 0.0654 +Epoch [143/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0781, Loss2: 0.0738 +Epoch [143/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0779, Loss2: 0.0779 +Epoch [143/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0782, Loss2: 0.0814 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [143/200] Test Accuracy on the 10000 test data: Model1 20.3325 % Model2 20.9235 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0671, Loss2: 0.0692 +Epoch [144/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0746, Loss2: 0.0794 +Epoch [144/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.1562, Loss1: 0.0666, Loss2: 0.0663 +Epoch [144/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0783, Loss2: 0.0803 +Epoch [144/200], Iter [250/390] Training Accuracy1: 67.9688, Training Accuracy2: 71.8750, Loss1: 0.0947, Loss2: 0.0864 +Epoch [144/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0848, Loss2: 0.0867 +Epoch [144/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0874, Loss2: 0.0874 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [144/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 21.0537 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 69.5312, Loss1: 0.0840, Loss2: 0.0762 +Epoch [145/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.1163, Loss2: 0.1072 +Epoch [145/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0976, Loss2: 0.1043 +Epoch [145/200], Iter [200/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0671, Loss2: 0.0640 +Epoch [145/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0911, Loss2: 0.0915 +Epoch [145/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0745, Loss2: 0.0703 +Epoch [145/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0879, Loss2: 0.0810 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [145/200] Test Accuracy on the 10000 test data: Model1 21.0136 % Model2 20.8033 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200], Iter [50/390] Training Accuracy1: 67.9688, Training Accuracy2: 60.1562, Loss1: 0.0809, Loss2: 0.0888 +Epoch [146/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0797, Loss2: 0.0791 +Epoch [146/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.1250, Loss1: 0.0634, Loss2: 0.0620 +Epoch [146/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0868, Loss2: 0.0837 +Epoch [146/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.0968, Loss2: 0.0988 +Epoch [146/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0793, Loss2: 0.0811 +Epoch [146/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 63.2812, Loss1: 0.0809, Loss2: 0.0838 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [146/200] Test Accuracy on the 10000 test data: Model1 20.4026 % Model2 21.0036 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 58.5938, Loss1: 0.0790, Loss2: 0.0742 +Epoch [147/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0766, Loss2: 0.0775 +Epoch [147/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 61.7188, Loss1: 0.0693, Loss2: 0.0622 +Epoch [147/200], Iter [200/390] Training Accuracy1: 54.6875, Training Accuracy2: 55.4688, Loss1: 0.0675, Loss2: 0.0660 +Epoch [147/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0960, Loss2: 0.0918 +Epoch [147/200], Iter [300/390] Training Accuracy1: 51.5625, Training Accuracy2: 54.6875, Loss1: 0.0629, Loss2: 0.0612 +Epoch [147/200], Iter [350/390] Training Accuracy1: 50.0000, Training Accuracy2: 51.5625, Loss1: 0.0572, Loss2: 0.0581 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [147/200] Test Accuracy on the 10000 test data: Model1 20.4427 % Model2 20.7232 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0869, Loss2: 0.0811 +Epoch [148/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 58.5938, Loss1: 0.1002, Loss2: 0.1025 +Epoch [148/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0693, Loss2: 0.0710 +Epoch [148/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 66.4062, Loss1: 0.0851, Loss2: 0.0734 +Epoch [148/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0671, Loss2: 0.0651 +Epoch [148/200], Iter [300/390] Training Accuracy1: 54.6875, Training Accuracy2: 56.2500, Loss1: 0.0551, Loss2: 0.0539 +Epoch [148/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0920, Loss2: 0.0957 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [148/200] Test Accuracy on the 10000 test data: Model1 20.4227 % Model2 21.1739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0782, Loss2: 0.0783 +Epoch [149/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0670, Loss2: 0.0682 +Epoch [149/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0604, Loss2: 0.0620 +Epoch [149/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0807, Loss2: 0.0838 +Epoch [149/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0693, Loss2: 0.0680 +Epoch [149/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 61.7188, Loss1: 0.0630, Loss2: 0.0579 +Epoch [149/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0819, Loss2: 0.0828 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [149/200] Test Accuracy on the 10000 test data: Model1 20.3626 % Model2 21.3241 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 56.2500, Loss1: 0.0697, Loss2: 0.0706 +Epoch [150/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0758, Loss2: 0.0693 +Epoch [150/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0873, Loss2: 0.0851 +Epoch [150/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 63.2812, Loss1: 0.0833, Loss2: 0.0818 +Epoch [150/200], Iter [250/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0679, Loss2: 0.0702 +Epoch [150/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0784, Loss2: 0.0715 +Epoch [150/200], Iter [350/390] Training Accuracy1: 67.9688, Training Accuracy2: 65.6250, Loss1: 0.0725, Loss2: 0.0740 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [150/200] Test Accuracy on the 10000 test data: Model1 20.2224 % Model2 21.3241 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0692, Loss2: 0.0725 +Epoch [151/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0870, Loss2: 0.0877 +Epoch [151/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0733, Loss2: 0.0739 +Epoch [151/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0664, Loss2: 0.0664 +Epoch [151/200], Iter [250/390] Training Accuracy1: 50.7812, Training Accuracy2: 54.6875, Loss1: 0.0752, Loss2: 0.0748 +Epoch [151/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0769, Loss2: 0.0763 +Epoch [151/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.8125, Loss1: 0.0806, Loss2: 0.0830 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [151/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 20.7632 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 75.7812, Loss1: 0.1115, Loss2: 0.1048 +Epoch [152/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 53.9062, Loss1: 0.0686, Loss2: 0.0699 +Epoch [152/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 65.6250, Loss1: 0.0923, Loss2: 0.0890 +Epoch [152/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0805, Loss2: 0.0852 +Epoch [152/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 58.5938, Loss1: 0.0748, Loss2: 0.0681 +Epoch [152/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0664, Loss2: 0.0666 +Epoch [152/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 61.7188, Loss1: 0.0844, Loss2: 0.0821 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [152/200] Test Accuracy on the 10000 test data: Model1 20.3826 % Model2 21.3742 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.1006, Loss2: 0.0927 +Epoch [153/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0813, Loss2: 0.0907 +Epoch [153/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0698, Loss2: 0.0688 +Epoch [153/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 56.2500, Loss1: 0.0779, Loss2: 0.0808 +Epoch [153/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0760, Loss2: 0.0735 +Epoch [153/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 55.4688, Loss1: 0.0618, Loss2: 0.0591 +Epoch [153/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0929, Loss2: 0.0868 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [153/200] Test Accuracy on the 10000 test data: Model1 20.5729 % Model2 21.1939 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0879, Loss2: 0.0884 +Epoch [154/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0898, Loss2: 0.0852 +Epoch [154/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0847, Loss2: 0.0886 +Epoch [154/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.1139, Loss2: 0.1088 +Epoch [154/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 50.0000, Loss1: 0.0590, Loss2: 0.0624 +Epoch [154/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 67.9688, Loss1: 0.1121, Loss2: 0.1013 +Epoch [154/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.1000, Loss2: 0.0919 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [154/200] Test Accuracy on the 10000 test data: Model1 20.0921 % Model2 21.2540 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1228, Loss2: 0.1144 +Epoch [155/200], Iter [100/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0829, Loss2: 0.0740 +Epoch [155/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0937, Loss2: 0.0963 +Epoch [155/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0667, Loss2: 0.0689 +Epoch [155/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0896, Loss2: 0.0875 +Epoch [155/200], Iter [300/390] Training Accuracy1: 53.9062, Training Accuracy2: 55.4688, Loss1: 0.0703, Loss2: 0.0675 +Epoch [155/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0821, Loss2: 0.0785 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [155/200] Test Accuracy on the 10000 test data: Model1 20.6230 % Model2 21.3041 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0780, Loss2: 0.0781 +Epoch [156/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.0680, Loss2: 0.0684 +Epoch [156/200], Iter [150/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0706, Loss2: 0.0648 +Epoch [156/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0625, Loss2: 0.0596 +Epoch [156/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 56.2500, Loss1: 0.0581, Loss2: 0.0593 +Epoch [156/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0718, Loss2: 0.0715 +Epoch [156/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0794, Loss2: 0.0818 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [156/200] Test Accuracy on the 10000 test data: Model1 20.7232 % Model2 20.9836 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200], Iter [50/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0788, Loss2: 0.0741 +Epoch [157/200], Iter [100/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.0658, Loss2: 0.0675 +Epoch [157/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 60.9375, Loss1: 0.0983, Loss2: 0.0998 +Epoch [157/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0968, Loss2: 0.0957 +Epoch [157/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.9375, Loss1: 0.0909, Loss2: 0.0807 +Epoch [157/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 65.6250, Loss1: 0.1080, Loss2: 0.1124 +Epoch [157/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 57.8125, Loss1: 0.0707, Loss2: 0.0749 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [157/200] Test Accuracy on the 10000 test data: Model1 20.5829 % Model2 21.8249 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1033, Loss2: 0.1027 +Epoch [158/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 67.1875, Loss1: 0.0720, Loss2: 0.0677 +Epoch [158/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.8125, Loss1: 0.0755, Loss2: 0.0790 +Epoch [158/200], Iter [200/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.1379, Loss2: 0.1364 +Epoch [158/200], Iter [250/390] Training Accuracy1: 52.3438, Training Accuracy2: 56.2500, Loss1: 0.0737, Loss2: 0.0704 +Epoch [158/200], Iter [300/390] Training Accuracy1: 69.5312, Training Accuracy2: 65.6250, Loss1: 0.0825, Loss2: 0.0899 +Epoch [158/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.0925, Loss2: 0.0939 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [158/200] Test Accuracy on the 10000 test data: Model1 20.7031 % Model2 21.0036 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200], Iter [50/390] Training Accuracy1: 69.5312, Training Accuracy2: 67.9688, Loss1: 0.1299, Loss2: 0.1375 +Epoch [159/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0788, Loss2: 0.0848 +Epoch [159/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0886, Loss2: 0.0844 +Epoch [159/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0722, Loss2: 0.0728 +Epoch [159/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 66.4062, Loss1: 0.0855, Loss2: 0.0857 +Epoch [159/200], Iter [300/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.1113, Loss2: 0.1099 +Epoch [159/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 62.5000, Loss1: 0.0794, Loss2: 0.0842 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [159/200] Test Accuracy on the 10000 test data: Model1 20.9335 % Model2 21.5745 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 59.3750, Loss1: 0.0999, Loss2: 0.0920 +Epoch [160/200], Iter [100/390] Training Accuracy1: 57.0312, Training Accuracy2: 62.5000, Loss1: 0.0866, Loss2: 0.0806 +Epoch [160/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 52.3438, Loss1: 0.0769, Loss2: 0.0814 +Epoch [160/200], Iter [200/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0743, Loss2: 0.0746 +Epoch [160/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0756, Loss2: 0.0746 +Epoch [160/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.0967, Loss2: 0.0982 +Epoch [160/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 71.0938, Loss1: 0.0798, Loss2: 0.0761 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [160/200] Test Accuracy on the 10000 test data: Model1 21.0737 % Model2 21.3842 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200], Iter [50/390] Training Accuracy1: 55.4688, Training Accuracy2: 54.6875, Loss1: 0.0816, Loss2: 0.0829 +Epoch [161/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0783, Loss2: 0.0822 +Epoch [161/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0794, Loss2: 0.0740 +Epoch [161/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0703, Loss2: 0.0703 +Epoch [161/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 64.8438, Loss1: 0.1151, Loss2: 0.1165 +Epoch [161/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 57.0312, Loss1: 0.0677, Loss2: 0.0665 +Epoch [161/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 71.0938, Loss1: 0.0785, Loss2: 0.0730 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [161/200] Test Accuracy on the 10000 test data: Model1 20.7632 % Model2 21.1238 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.0945, Loss2: 0.0943 +Epoch [162/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0955, Loss2: 0.0862 +Epoch [162/200], Iter [150/390] Training Accuracy1: 50.0000, Training Accuracy2: 56.2500, Loss1: 0.0779, Loss2: 0.0736 +Epoch [162/200], Iter [200/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.0312, Loss1: 0.0717, Loss2: 0.0740 +Epoch [162/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0690, Loss2: 0.0684 +Epoch [162/200], Iter [300/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.0760, Loss2: 0.0731 +Epoch [162/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0888, Loss2: 0.0864 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [162/200] Test Accuracy on the 10000 test data: Model1 20.3025 % Model2 20.9635 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0895, Loss2: 0.0880 +Epoch [163/200], Iter [100/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.8125, Loss1: 0.0714, Loss2: 0.0694 +Epoch [163/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 56.2500, Loss1: 0.0684, Loss2: 0.0711 +Epoch [163/200], Iter [200/390] Training Accuracy1: 73.4375, Training Accuracy2: 70.3125, Loss1: 0.1098, Loss2: 0.1177 +Epoch [163/200], Iter [250/390] Training Accuracy1: 55.4688, Training Accuracy2: 63.2812, Loss1: 0.0743, Loss2: 0.0662 +Epoch [163/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 60.1562, Loss1: 0.0752, Loss2: 0.0813 +Epoch [163/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0875, Loss2: 0.0827 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [163/200] Test Accuracy on the 10000 test data: Model1 20.4828 % Model2 21.0938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0908, Loss2: 0.0901 +Epoch [164/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.1067, Loss2: 0.1077 +Epoch [164/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0924, Loss2: 0.0914 +Epoch [164/200], Iter [200/390] Training Accuracy1: 69.5312, Training Accuracy2: 72.6562, Loss1: 0.0973, Loss2: 0.0928 +Epoch [164/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 64.0625, Loss1: 0.0905, Loss2: 0.0819 +Epoch [164/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0705, Loss2: 0.0663 +Epoch [164/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0880, Loss2: 0.0929 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [164/200] Test Accuracy on the 10000 test data: Model1 21.0036 % Model2 20.7732 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200], Iter [50/390] Training Accuracy1: 52.3438, Training Accuracy2: 57.8125, Loss1: 0.0800, Loss2: 0.0730 +Epoch [165/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 59.3750, Loss1: 0.0753, Loss2: 0.0823 +Epoch [165/200], Iter [150/390] Training Accuracy1: 56.2500, Training Accuracy2: 60.1562, Loss1: 0.0602, Loss2: 0.0586 +Epoch [165/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0755, Loss2: 0.0789 +Epoch [165/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 58.5938, Loss1: 0.0839, Loss2: 0.0760 +Epoch [165/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0940, Loss2: 0.0913 +Epoch [165/200], Iter [350/390] Training Accuracy1: 56.2500, Training Accuracy2: 53.1250, Loss1: 0.0655, Loss2: 0.0691 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [165/200] Test Accuracy on the 10000 test data: Model1 20.5329 % Model2 21.5845 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0960, Loss2: 0.0939 +Epoch [166/200], Iter [100/390] Training Accuracy1: 52.3438, Training Accuracy2: 53.9062, Loss1: 0.0635, Loss2: 0.0630 +Epoch [166/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 66.4062, Loss1: 0.0835, Loss2: 0.0800 +Epoch [166/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0901, Loss2: 0.0895 +Epoch [166/200], Iter [250/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.0852, Loss2: 0.0866 +Epoch [166/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0786, Loss2: 0.0768 +Epoch [166/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 66.4062, Loss1: 0.0800, Loss2: 0.0800 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [166/200] Test Accuracy on the 10000 test data: Model1 21.0938 % Model2 21.1138 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200], Iter [50/390] Training Accuracy1: 57.0312, Training Accuracy2: 57.8125, Loss1: 0.0843, Loss2: 0.0824 +Epoch [167/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 65.6250, Loss1: 0.0726, Loss2: 0.0708 +Epoch [167/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0700, Loss2: 0.0671 +Epoch [167/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.0920, Loss2: 0.0912 +Epoch [167/200], Iter [250/390] Training Accuracy1: 72.6562, Training Accuracy2: 75.7812, Loss1: 0.0904, Loss2: 0.0900 +Epoch [167/200], Iter [300/390] Training Accuracy1: 52.3438, Training Accuracy2: 55.4688, Loss1: 0.0794, Loss2: 0.0747 +Epoch [167/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0953, Loss2: 0.1017 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [167/200] Test Accuracy on the 10000 test data: Model1 20.4026 % Model2 21.2139 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200], Iter [50/390] Training Accuracy1: 57.8125, Training Accuracy2: 59.3750, Loss1: 0.0722, Loss2: 0.0720 +Epoch [168/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0683, Loss2: 0.0672 +Epoch [168/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.1057, Loss2: 0.1084 +Epoch [168/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1025, Loss2: 0.1009 +Epoch [168/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 60.9375, Loss1: 0.0831, Loss2: 0.0766 +Epoch [168/200], Iter [300/390] Training Accuracy1: 70.3125, Training Accuracy2: 71.8750, Loss1: 0.0888, Loss2: 0.0877 +Epoch [168/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.1269, Loss2: 0.1181 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [168/200] Test Accuracy on the 10000 test data: Model1 20.3826 % Model2 21.2039 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.1333, Loss2: 0.1240 +Epoch [169/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.8125, Loss1: 0.0692, Loss2: 0.0740 +Epoch [169/200], Iter [150/390] Training Accuracy1: 55.4688, Training Accuracy2: 57.8125, Loss1: 0.0763, Loss2: 0.0731 +Epoch [169/200], Iter [200/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.9375, Loss1: 0.0860, Loss2: 0.0931 +Epoch [169/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1205, Loss2: 0.1199 +Epoch [169/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 62.5000, Loss1: 0.0745, Loss2: 0.0740 +Epoch [169/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 61.7188, Loss1: 0.0776, Loss2: 0.0833 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [169/200] Test Accuracy on the 10000 test data: Model1 20.5629 % Model2 21.0938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0919, Loss2: 0.0899 +Epoch [170/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0679, Loss2: 0.0716 +Epoch [170/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0819, Loss2: 0.0853 +Epoch [170/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 54.6875, Loss1: 0.0846, Loss2: 0.0848 +Epoch [170/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0853, Loss2: 0.0845 +Epoch [170/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 68.7500, Loss1: 0.0911, Loss2: 0.0844 +Epoch [170/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0867, Loss2: 0.0827 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [170/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 21.0938 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0685, Loss2: 0.0693 +Epoch [171/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.1875, Loss1: 0.0990, Loss2: 0.0914 +Epoch [171/200], Iter [150/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0883, Loss2: 0.0871 +Epoch [171/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1232, Loss2: 0.1247 +Epoch [171/200], Iter [250/390] Training Accuracy1: 53.9062, Training Accuracy2: 53.1250, Loss1: 0.0670, Loss2: 0.0695 +Epoch [171/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 56.2500, Loss1: 0.0847, Loss2: 0.0873 +Epoch [171/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 62.5000, Loss1: 0.0818, Loss2: 0.0795 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [171/200] Test Accuracy on the 10000 test data: Model1 20.6731 % Model2 21.0837 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0989, Loss2: 0.1035 +Epoch [172/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0748, Loss2: 0.0748 +Epoch [172/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.0851, Loss2: 0.0909 +Epoch [172/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.8438, Loss1: 0.1313, Loss2: 0.1373 +Epoch [172/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.1259, Loss2: 0.1284 +Epoch [172/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0760, Loss2: 0.0729 +Epoch [172/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 60.9375, Loss1: 0.0762, Loss2: 0.0766 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [172/200] Test Accuracy on the 10000 test data: Model1 20.6330 % Model2 21.0637 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 63.2812, Loss1: 0.0928, Loss2: 0.0836 +Epoch [173/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.1106, Loss2: 0.1149 +Epoch [173/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0875, Loss2: 0.0831 +Epoch [173/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 66.4062, Loss1: 0.0878, Loss2: 0.0845 +Epoch [173/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0859, Loss2: 0.0810 +Epoch [173/200], Iter [300/390] Training Accuracy1: 55.4688, Training Accuracy2: 59.3750, Loss1: 0.0735, Loss2: 0.0709 +Epoch [173/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.1057, Loss2: 0.1051 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [173/200] Test Accuracy on the 10000 test data: Model1 20.6831 % Model2 21.2440 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0872, Loss2: 0.0901 +Epoch [174/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.1080, Loss2: 0.1068 +Epoch [174/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.0888, Loss2: 0.0881 +Epoch [174/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.1085, Loss2: 0.1099 +Epoch [174/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 63.2812, Loss1: 0.1009, Loss2: 0.1007 +Epoch [174/200], Iter [300/390] Training Accuracy1: 57.0312, Training Accuracy2: 58.5938, Loss1: 0.0719, Loss2: 0.0724 +Epoch [174/200], Iter [350/390] Training Accuracy1: 71.0938, Training Accuracy2: 67.9688, Loss1: 0.1037, Loss2: 0.1138 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [174/200] Test Accuracy on the 10000 test data: Model1 20.7632 % Model2 20.9535 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200], Iter [50/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.1031, Loss2: 0.0963 +Epoch [175/200], Iter [100/390] Training Accuracy1: 60.9375, Training Accuracy2: 59.3750, Loss1: 0.0774, Loss2: 0.0798 +Epoch [175/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 65.6250, Loss1: 0.0921, Loss2: 0.0837 +Epoch [175/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 68.7500, Loss1: 0.0869, Loss2: 0.0850 +Epoch [175/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.1067, Loss2: 0.1078 +Epoch [175/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.8438, Loss1: 0.0957, Loss2: 0.1078 +Epoch [175/200], Iter [350/390] Training Accuracy1: 62.5000, Training Accuracy2: 60.1562, Loss1: 0.0919, Loss2: 0.0968 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [175/200] Test Accuracy on the 10000 test data: Model1 20.8233 % Model2 21.1438 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 68.7500, Loss1: 0.1376, Loss2: 0.1338 +Epoch [176/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 65.6250, Loss1: 0.0715, Loss2: 0.0699 +Epoch [176/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0798, Loss2: 0.0840 +Epoch [176/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 56.2500, Loss1: 0.0893, Loss2: 0.0902 +Epoch [176/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.9375, Loss1: 0.0930, Loss2: 0.0935 +Epoch [176/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0921, Loss2: 0.0895 +Epoch [176/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.9688, Loss1: 0.1376, Loss2: 0.1257 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [176/200] Test Accuracy on the 10000 test data: Model1 20.4527 % Model2 21.1338 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 67.1875, Loss1: 0.0724, Loss2: 0.0697 +Epoch [177/200], Iter [100/390] Training Accuracy1: 76.5625, Training Accuracy2: 71.0938, Loss1: 0.1443, Loss2: 0.1725 +Epoch [177/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 70.3125, Loss1: 0.1010, Loss2: 0.0958 +Epoch [177/200], Iter [200/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1032, Loss2: 0.1041 +Epoch [177/200], Iter [250/390] Training Accuracy1: 69.5312, Training Accuracy2: 69.5312, Loss1: 0.1121, Loss2: 0.1126 +Epoch [177/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.9375, Loss1: 0.0877, Loss2: 0.0947 +Epoch [177/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0895, Loss2: 0.0909 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [177/200] Test Accuracy on the 10000 test data: Model1 20.6130 % Model2 21.3141 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200], Iter [50/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0865, Loss2: 0.0850 +Epoch [178/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1205, Loss2: 0.1136 +Epoch [178/200], Iter [150/390] Training Accuracy1: 58.5938, Training Accuracy2: 65.6250, Loss1: 0.0715, Loss2: 0.0661 +Epoch [178/200], Iter [200/390] Training Accuracy1: 56.2500, Training Accuracy2: 54.6875, Loss1: 0.0941, Loss2: 0.0951 +Epoch [178/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0893, Loss2: 0.0905 +Epoch [178/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.1562, Loss1: 0.0780, Loss2: 0.0786 +Epoch [178/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 62.5000, Loss1: 0.0799, Loss2: 0.0757 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [178/200] Test Accuracy on the 10000 test data: Model1 20.9535 % Model2 21.4443 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.0754, Loss2: 0.0792 +Epoch [179/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.0625, Loss1: 0.0789, Loss2: 0.0735 +Epoch [179/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 66.4062, Loss1: 0.1179, Loss2: 0.1179 +Epoch [179/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0850, Loss2: 0.0845 +Epoch [179/200], Iter [250/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0982, Loss2: 0.0948 +Epoch [179/200], Iter [300/390] Training Accuracy1: 60.1562, Training Accuracy2: 64.8438, Loss1: 0.0783, Loss2: 0.0726 +Epoch [179/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 62.5000, Loss1: 0.0798, Loss2: 0.0803 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [179/200] Test Accuracy on the 10000 test data: Model1 20.3926 % Model2 21.5745 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200], Iter [50/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.1022, Loss2: 0.1031 +Epoch [180/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 61.7188, Loss1: 0.1063, Loss2: 0.1070 +Epoch [180/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 71.8750, Loss1: 0.0952, Loss2: 0.0788 +Epoch [180/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 60.9375, Loss1: 0.1098, Loss2: 0.1161 +Epoch [180/200], Iter [250/390] Training Accuracy1: 58.5938, Training Accuracy2: 59.3750, Loss1: 0.0949, Loss2: 0.0936 +Epoch [180/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 65.6250, Loss1: 0.1122, Loss2: 0.1113 +Epoch [180/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.1040, Loss2: 0.1013 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [180/200] Test Accuracy on the 10000 test data: Model1 20.6330 % Model2 21.2841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0959, Loss2: 0.0928 +Epoch [181/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.0846, Loss2: 0.0938 +Epoch [181/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.0625, Loss1: 0.0895, Loss2: 0.0923 +Epoch [181/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0827, Loss2: 0.0826 +Epoch [181/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0975, Loss2: 0.0896 +Epoch [181/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 59.3750, Loss1: 0.0934, Loss2: 0.0979 +Epoch [181/200], Iter [350/390] Training Accuracy1: 67.1875, Training Accuracy2: 63.2812, Loss1: 0.0729, Loss2: 0.0784 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [181/200] Test Accuracy on the 10000 test data: Model1 20.5329 % Model2 21.1739 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200], Iter [50/390] Training Accuracy1: 49.2188, Training Accuracy2: 52.3438, Loss1: 0.0811, Loss2: 0.0786 +Epoch [182/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.1116, Loss2: 0.1121 +Epoch [182/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0936, Loss2: 0.0879 +Epoch [182/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.0917, Loss2: 0.0969 +Epoch [182/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 70.3125, Loss1: 0.1082, Loss2: 0.0980 +Epoch [182/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0786, Loss2: 0.0789 +Epoch [182/200], Iter [350/390] Training Accuracy1: 68.7500, Training Accuracy2: 71.0938, Loss1: 0.1296, Loss2: 0.1176 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [182/200] Test Accuracy on the 10000 test data: Model1 20.5829 % Model2 21.4944 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0899, Loss2: 0.0944 +Epoch [183/200], Iter [100/390] Training Accuracy1: 71.8750, Training Accuracy2: 71.8750, Loss1: 0.1029, Loss2: 0.1000 +Epoch [183/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0832, Loss2: 0.0815 +Epoch [183/200], Iter [200/390] Training Accuracy1: 60.1562, Training Accuracy2: 55.4688, Loss1: 0.0904, Loss2: 0.0982 +Epoch [183/200], Iter [250/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0751, Loss2: 0.0770 +Epoch [183/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 67.1875, Loss1: 0.1138, Loss2: 0.1006 +Epoch [183/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0780, Loss2: 0.0701 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [183/200] Test Accuracy on the 10000 test data: Model1 20.7833 % Model2 21.3842 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 66.4062, Loss1: 0.0806, Loss2: 0.0740 +Epoch [184/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 67.9688, Loss1: 0.0988, Loss2: 0.0885 +Epoch [184/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0739, Loss2: 0.0738 +Epoch [184/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 60.9375, Loss1: 0.0767, Loss2: 0.0723 +Epoch [184/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.0625, Loss1: 0.1072, Loss2: 0.1104 +Epoch [184/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 67.1875, Loss1: 0.0771, Loss2: 0.0786 +Epoch [184/200], Iter [350/390] Training Accuracy1: 59.3750, Training Accuracy2: 58.5938, Loss1: 0.0800, Loss2: 0.0781 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [184/200] Test Accuracy on the 10000 test data: Model1 20.7632 % Model2 21.5044 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200], Iter [50/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.0976, Loss2: 0.0953 +Epoch [185/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 52.3438, Loss1: 0.0804, Loss2: 0.0834 +Epoch [185/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0873, Loss2: 0.0852 +Epoch [185/200], Iter [200/390] Training Accuracy1: 53.9062, Training Accuracy2: 59.3750, Loss1: 0.0581, Loss2: 0.0560 +Epoch [185/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0788, Loss2: 0.0794 +Epoch [185/200], Iter [300/390] Training Accuracy1: 58.5938, Training Accuracy2: 57.0312, Loss1: 0.0841, Loss2: 0.0866 +Epoch [185/200], Iter [350/390] Training Accuracy1: 65.6250, Training Accuracy2: 64.8438, Loss1: 0.0980, Loss2: 0.0971 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [185/200] Test Accuracy on the 10000 test data: Model1 20.5128 % Model2 21.4143 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0935, Loss2: 0.0911 +Epoch [186/200], Iter [100/390] Training Accuracy1: 61.7188, Training Accuracy2: 62.5000, Loss1: 0.0773, Loss2: 0.0787 +Epoch [186/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 66.4062, Loss1: 0.0805, Loss2: 0.0789 +Epoch [186/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.0882, Loss2: 0.0863 +Epoch [186/200], Iter [250/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.9688, Loss1: 0.1493, Loss2: 0.1367 +Epoch [186/200], Iter [300/390] Training Accuracy1: 66.4062, Training Accuracy2: 69.5312, Loss1: 0.1111, Loss2: 0.1080 +Epoch [186/200], Iter [350/390] Training Accuracy1: 53.9062, Training Accuracy2: 57.8125, Loss1: 0.0756, Loss2: 0.0719 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [186/200] Test Accuracy on the 10000 test data: Model1 20.8133 % Model2 21.5645 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200], Iter [50/390] Training Accuracy1: 59.3750, Training Accuracy2: 59.3750, Loss1: 0.0831, Loss2: 0.0819 +Epoch [187/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.1229, Loss2: 0.1211 +Epoch [187/200], Iter [150/390] Training Accuracy1: 62.5000, Training Accuracy2: 64.0625, Loss1: 0.0902, Loss2: 0.0857 +Epoch [187/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.1240, Loss2: 0.1198 +Epoch [187/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0981, Loss2: 0.1030 +Epoch [187/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 71.0938, Loss1: 0.1414, Loss2: 0.1271 +Epoch [187/200], Iter [350/390] Training Accuracy1: 66.4062, Training Accuracy2: 67.1875, Loss1: 0.1002, Loss2: 0.0982 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [187/200] Test Accuracy on the 10000 test data: Model1 20.5829 % Model2 21.1639 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200], Iter [50/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.1149, Loss2: 0.1106 +Epoch [188/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 57.0312, Loss1: 0.0956, Loss2: 0.0981 +Epoch [188/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.0625, Loss1: 0.0897, Loss2: 0.0837 +Epoch [188/200], Iter [200/390] Training Accuracy1: 57.8125, Training Accuracy2: 56.2500, Loss1: 0.0884, Loss2: 0.0932 +Epoch [188/200], Iter [250/390] Training Accuracy1: 57.8125, Training Accuracy2: 57.0312, Loss1: 0.0789, Loss2: 0.0806 +Epoch [188/200], Iter [300/390] Training Accuracy1: 67.1875, Training Accuracy2: 64.0625, Loss1: 0.1228, Loss2: 0.1301 +Epoch [188/200], Iter [350/390] Training Accuracy1: 54.6875, Training Accuracy2: 59.3750, Loss1: 0.0876, Loss2: 0.0826 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [188/200] Test Accuracy on the 10000 test data: Model1 20.7031 % Model2 21.2039 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200], Iter [50/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.0312, Loss1: 0.0849, Loss2: 0.0809 +Epoch [189/200], Iter [100/390] Training Accuracy1: 54.6875, Training Accuracy2: 57.8125, Loss1: 0.1017, Loss2: 0.0972 +Epoch [189/200], Iter [150/390] Training Accuracy1: 61.7188, Training Accuracy2: 59.3750, Loss1: 0.0819, Loss2: 0.0882 +Epoch [189/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 60.1562, Loss1: 0.0878, Loss2: 0.0968 +Epoch [189/200], Iter [250/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.0625, Loss1: 0.0775, Loss2: 0.0795 +Epoch [189/200], Iter [300/390] Training Accuracy1: 62.5000, Training Accuracy2: 57.0312, Loss1: 0.0838, Loss2: 0.0924 +Epoch [189/200], Iter [350/390] Training Accuracy1: 58.5938, Training Accuracy2: 53.9062, Loss1: 0.0694, Loss2: 0.0716 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [189/200] Test Accuracy on the 10000 test data: Model1 20.6330 % Model2 21.2540 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200], Iter [50/390] Training Accuracy1: 61.7188, Training Accuracy2: 63.2812, Loss1: 0.0950, Loss2: 0.0947 +Epoch [190/200], Iter [100/390] Training Accuracy1: 55.4688, Training Accuracy2: 60.1562, Loss1: 0.0827, Loss2: 0.0772 +Epoch [190/200], Iter [150/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.0800, Loss2: 0.0805 +Epoch [190/200], Iter [200/390] Training Accuracy1: 58.5938, Training Accuracy2: 58.5938, Loss1: 0.0765, Loss2: 0.0780 +Epoch [190/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1129, Loss2: 0.1049 +Epoch [190/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 65.6250, Loss1: 0.1107, Loss2: 0.1101 +Epoch [190/200], Iter [350/390] Training Accuracy1: 60.9375, Training Accuracy2: 58.5938, Loss1: 0.0793, Loss2: 0.0829 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [190/200] Test Accuracy on the 10000 test data: Model1 20.7532 % Model2 21.4343 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200], Iter [50/390] Training Accuracy1: 62.5000, Training Accuracy2: 65.6250, Loss1: 0.1161, Loss2: 0.1050 +Epoch [191/200], Iter [100/390] Training Accuracy1: 62.5000, Training Accuracy2: 62.5000, Loss1: 0.1063, Loss2: 0.1085 +Epoch [191/200], Iter [150/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1070, Loss2: 0.1070 +Epoch [191/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.1562, Loss1: 0.0925, Loss2: 0.0950 +Epoch [191/200], Iter [250/390] Training Accuracy1: 54.6875, Training Accuracy2: 61.7188, Loss1: 0.0678, Loss2: 0.0639 +Epoch [191/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 54.6875, Loss1: 0.0745, Loss2: 0.0795 +Epoch [191/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1029, Loss2: 0.0992 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [191/200] Test Accuracy on the 10000 test data: Model1 20.5228 % Model2 21.4643 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200], Iter [50/390] Training Accuracy1: 68.7500, Training Accuracy2: 72.6562, Loss1: 0.1246, Loss2: 0.1152 +Epoch [192/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 65.6250, Loss1: 0.1194, Loss2: 0.1150 +Epoch [192/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 64.8438, Loss1: 0.0818, Loss2: 0.0738 +Epoch [192/200], Iter [200/390] Training Accuracy1: 65.6250, Training Accuracy2: 68.7500, Loss1: 0.1354, Loss2: 0.1263 +Epoch [192/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 61.7188, Loss1: 0.0845, Loss2: 0.0828 +Epoch [192/200], Iter [300/390] Training Accuracy1: 61.7188, Training Accuracy2: 60.9375, Loss1: 0.0820, Loss2: 0.0827 +Epoch [192/200], Iter [350/390] Training Accuracy1: 57.8125, Training Accuracy2: 58.5938, Loss1: 0.0880, Loss2: 0.0899 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [192/200] Test Accuracy on the 10000 test data: Model1 20.7031 % Model2 21.2841 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0964, Loss2: 0.0989 +Epoch [193/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0935, Loss2: 0.0885 +Epoch [193/200], Iter [150/390] Training Accuracy1: 59.3750, Training Accuracy2: 55.4688, Loss1: 0.0773, Loss2: 0.0807 +Epoch [193/200], Iter [200/390] Training Accuracy1: 59.3750, Training Accuracy2: 61.7188, Loss1: 0.0926, Loss2: 0.0869 +Epoch [193/200], Iter [250/390] Training Accuracy1: 60.9375, Training Accuracy2: 64.0625, Loss1: 0.0953, Loss2: 0.0923 +Epoch [193/200], Iter [300/390] Training Accuracy1: 59.3750, Training Accuracy2: 60.1562, Loss1: 0.0786, Loss2: 0.0798 +Epoch [193/200], Iter [350/390] Training Accuracy1: 64.8438, Training Accuracy2: 67.9688, Loss1: 0.1129, Loss2: 0.1019 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [193/200] Test Accuracy on the 10000 test data: Model1 20.6430 % Model2 21.2340 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 61.7188, Loss1: 0.1073, Loss2: 0.1174 +Epoch [194/200], Iter [100/390] Training Accuracy1: 60.1562, Training Accuracy2: 56.2500, Loss1: 0.1058, Loss2: 0.1105 +Epoch [194/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 61.7188, Loss1: 0.0932, Loss2: 0.1008 +Epoch [194/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0915, Loss2: 0.0932 +Epoch [194/200], Iter [250/390] Training Accuracy1: 64.0625, Training Accuracy2: 56.2500, Loss1: 0.0755, Loss2: 0.0847 +Epoch [194/200], Iter [300/390] Training Accuracy1: 56.2500, Training Accuracy2: 55.4688, Loss1: 0.0865, Loss2: 0.0870 +Epoch [194/200], Iter [350/390] Training Accuracy1: 64.0625, Training Accuracy2: 64.8438, Loss1: 0.1112, Loss2: 0.1079 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [194/200] Test Accuracy on the 10000 test data: Model1 20.4527 % Model2 21.4744 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200], Iter [50/390] Training Accuracy1: 64.8438, Training Accuracy2: 63.2812, Loss1: 0.0977, Loss2: 0.0980 +Epoch [195/200], Iter [100/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.0625, Loss1: 0.1055, Loss2: 0.1146 +Epoch [195/200], Iter [150/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0983, Loss2: 0.1013 +Epoch [195/200], Iter [200/390] Training Accuracy1: 64.0625, Training Accuracy2: 62.5000, Loss1: 0.0934, Loss2: 0.0939 +Epoch [195/200], Iter [250/390] Training Accuracy1: 63.2812, Training Accuracy2: 62.5000, Loss1: 0.1090, Loss2: 0.1113 +Epoch [195/200], Iter [300/390] Training Accuracy1: 60.9375, Training Accuracy2: 63.2812, Loss1: 0.0858, Loss2: 0.0818 +Epoch [195/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 57.0312, Loss1: 0.0874, Loss2: 0.0926 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [195/200] Test Accuracy on the 10000 test data: Model1 20.5829 % Model2 21.2440 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200], Iter [50/390] Training Accuracy1: 60.9375, Training Accuracy2: 60.9375, Loss1: 0.0891, Loss2: 0.0888 +Epoch [196/200], Iter [100/390] Training Accuracy1: 59.3750, Training Accuracy2: 63.2812, Loss1: 0.0797, Loss2: 0.0779 +Epoch [196/200], Iter [150/390] Training Accuracy1: 60.9375, Training Accuracy2: 57.0312, Loss1: 0.1105, Loss2: 0.1192 +Epoch [196/200], Iter [200/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.0955, Loss2: 0.0912 +Epoch [196/200], Iter [250/390] Training Accuracy1: 61.7188, Training Accuracy2: 58.5938, Loss1: 0.0880, Loss2: 0.0936 +Epoch [196/200], Iter [300/390] Training Accuracy1: 65.6250, Training Accuracy2: 70.3125, Loss1: 0.1061, Loss2: 0.0971 +Epoch [196/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 63.2812, Loss1: 0.0915, Loss2: 0.0899 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [196/200] Test Accuracy on the 10000 test data: Model1 20.6030 % Model2 21.2240 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1077, Loss2: 0.1060 +Epoch [197/200], Iter [100/390] Training Accuracy1: 56.2500, Training Accuracy2: 58.5938, Loss1: 0.0950, Loss2: 0.0899 +Epoch [197/200], Iter [150/390] Training Accuracy1: 65.6250, Training Accuracy2: 67.1875, Loss1: 0.1153, Loss2: 0.1102 +Epoch [197/200], Iter [200/390] Training Accuracy1: 66.4062, Training Accuracy2: 66.4062, Loss1: 0.0914, Loss2: 0.0936 +Epoch [197/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.9688, Loss1: 0.0795, Loss2: 0.0799 +Epoch [197/200], Iter [300/390] Training Accuracy1: 68.7500, Training Accuracy2: 70.3125, Loss1: 0.1249, Loss2: 0.1179 +Epoch [197/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 62.5000, Loss1: 0.0822, Loss2: 0.0800 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [197/200] Test Accuracy on the 10000 test data: Model1 20.6130 % Model2 21.3942 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200], Iter [50/390] Training Accuracy1: 70.3125, Training Accuracy2: 68.7500, Loss1: 0.1065, Loss2: 0.1110 +Epoch [198/200], Iter [100/390] Training Accuracy1: 64.0625, Training Accuracy2: 65.6250, Loss1: 0.0757, Loss2: 0.0739 +Epoch [198/200], Iter [150/390] Training Accuracy1: 68.7500, Training Accuracy2: 64.0625, Loss1: 0.0970, Loss2: 0.1071 +Epoch [198/200], Iter [200/390] Training Accuracy1: 61.7188, Training Accuracy2: 61.7188, Loss1: 0.0884, Loss2: 0.0867 +Epoch [198/200], Iter [250/390] Training Accuracy1: 59.3750, Training Accuracy2: 62.5000, Loss1: 0.0947, Loss2: 0.0880 +Epoch [198/200], Iter [300/390] Training Accuracy1: 63.2812, Training Accuracy2: 57.8125, Loss1: 0.0796, Loss2: 0.0824 +Epoch [198/200], Iter [350/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.0625, Loss1: 0.1058, Loss2: 0.1068 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [198/200] Test Accuracy on the 10000 test data: Model1 20.5128 % Model2 21.3442 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200], Iter [50/390] Training Accuracy1: 63.2812, Training Accuracy2: 61.7188, Loss1: 0.0908, Loss2: 0.0951 +Epoch [199/200], Iter [100/390] Training Accuracy1: 64.8438, Training Accuracy2: 64.8438, Loss1: 0.1071, Loss2: 0.1070 +Epoch [199/200], Iter [150/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.0935, Loss2: 0.0950 +Epoch [199/200], Iter [200/390] Training Accuracy1: 68.7500, Training Accuracy2: 67.1875, Loss1: 0.1018, Loss2: 0.1031 +Epoch [199/200], Iter [250/390] Training Accuracy1: 67.1875, Training Accuracy2: 67.1875, Loss1: 0.0927, Loss2: 0.0955 +Epoch [199/200], Iter [300/390] Training Accuracy1: 53.1250, Training Accuracy2: 49.2188, Loss1: 0.0774, Loss2: 0.0823 +Epoch [199/200], Iter [350/390] Training Accuracy1: 60.1562, Training Accuracy2: 61.7188, Loss1: 0.0944, Loss2: 0.0916 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [199/200] Test Accuracy on the 10000 test data: Model1 20.6731 % Model2 21.2941 % +Training cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200], Iter [50/390] Training Accuracy1: 60.1562, Training Accuracy2: 59.3750, Loss1: 0.1216, Loss2: 0.1229 +Epoch [200/200], Iter [100/390] Training Accuracy1: 63.2812, Training Accuracy2: 64.8438, Loss1: 0.1160, Loss2: 0.1102 +Epoch [200/200], Iter [150/390] Training Accuracy1: 57.8125, Training Accuracy2: 60.9375, Loss1: 0.0857, Loss2: 0.0824 +Epoch [200/200], Iter [200/390] Training Accuracy1: 64.8438, Training Accuracy2: 71.0938, Loss1: 0.1010, Loss2: 0.0900 +Epoch [200/200], Iter [250/390] Training Accuracy1: 62.5000, Training Accuracy2: 63.2812, Loss1: 0.0976, Loss2: 0.0988 +Epoch [200/200], Iter [300/390] Training Accuracy1: 67.9688, Training Accuracy2: 64.8438, Loss1: 0.0926, Loss2: 0.0987 +Epoch [200/200], Iter [350/390] Training Accuracy1: 61.7188, Training Accuracy2: 64.0625, Loss1: 0.1023, Loss2: 0.0955 +Evaluating cifar10_coteaching_plus_from_file_0.2... +Epoch [200/200] Test Accuracy on the 10000 test data: Model1 20.6130 % Model2 21.3742 % diff --git a/other_methods/sceloss/sceloss_results/out_0_2.log b/other_methods/sceloss/sceloss_results/out_0_2.log new file mode 100644 index 0000000..53a2146 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_0_2.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.0__noise_amount__0.2.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=6.66330 loss_avg=7.23714 acc=0.30469 acc_top1_avg=0.27750 acc_top5_avg=0.71953 lr=0.01000 gn=8.16120 time=63.71it/s +epoch=0 global_step=100 loss=6.13369 loss_avg=6.86517 acc=0.37500 acc_top1_avg=0.31258 acc_top5_avg=0.75727 lr=0.01000 gn=8.29910 time=65.54it/s +epoch=0 global_step=150 loss=5.57940 loss_avg=6.64543 acc=0.41406 acc_top1_avg=0.33380 acc_top5_avg=0.78042 lr=0.01000 gn=6.58758 time=58.18it/s +epoch=0 global_step=200 loss=6.18387 loss_avg=6.48345 acc=0.34375 acc_top1_avg=0.34984 acc_top5_avg=0.79336 lr=0.01000 gn=6.26850 time=65.38it/s +epoch=0 global_step=250 loss=5.63400 loss_avg=6.36069 acc=0.46094 acc_top1_avg=0.36309 acc_top5_avg=0.80137 lr=0.01000 gn=5.80019 time=62.30it/s +epoch=0 global_step=300 loss=5.41309 loss_avg=6.23898 acc=0.46875 acc_top1_avg=0.37586 acc_top5_avg=0.81057 lr=0.01000 gn=5.86286 time=62.30it/s +epoch=0 global_step=350 loss=5.94627 loss_avg=6.15486 acc=0.39062 acc_top1_avg=0.38431 acc_top5_avg=0.81632 lr=0.01000 gn=5.91256 time=66.18it/s +====================Eval==================== +epoch=0 global_step=391 loss=0.39103 test_loss_avg=2.81117 acc=0.92188 test_acc_avg=0.37812 test_acc_top5_avg=0.92031 time=250.53it/s +epoch=0 global_step=391 loss=1.90106 test_loss_avg=2.32057 acc=0.43750 test_acc_avg=0.47814 test_acc_top5_avg=0.91792 time=34.40it/s +curr_acc 0.4781 +BEST_ACC 0.0000 +curr_acc_top5 0.9179 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.05829 loss_avg=5.62918 acc=0.36719 acc_top1_avg=0.44271 acc_top5_avg=0.83333 lr=0.01000 gn=6.21735 time=61.20it/s +epoch=1 global_step=450 loss=5.14490 loss_avg=5.40700 acc=0.50000 acc_top1_avg=0.46292 acc_top5_avg=0.85275 lr=0.01000 gn=6.26045 time=57.53it/s +epoch=1 global_step=500 loss=4.66170 loss_avg=5.29785 acc=0.55469 acc_top1_avg=0.47527 acc_top5_avg=0.85536 lr=0.01000 gn=6.60884 time=64.82it/s +epoch=1 global_step=550 loss=4.96107 loss_avg=5.22990 acc=0.50781 acc_top1_avg=0.48231 acc_top5_avg=0.86291 lr=0.01000 gn=6.68783 time=65.40it/s +epoch=1 global_step=600 loss=5.10841 loss_avg=5.19146 acc=0.50000 acc_top1_avg=0.48617 acc_top5_avg=0.86435 lr=0.01000 gn=5.98695 time=65.82it/s +epoch=1 global_step=650 loss=4.74807 loss_avg=5.11800 acc=0.54688 acc_top1_avg=0.49400 acc_top5_avg=0.86800 lr=0.01000 gn=6.46480 time=65.49it/s +epoch=1 global_step=700 loss=5.30355 loss_avg=5.08508 acc=0.45312 acc_top1_avg=0.49737 acc_top5_avg=0.86911 lr=0.01000 gn=7.17570 time=65.49it/s +epoch=1 global_step=750 loss=4.41514 loss_avg=5.04827 acc=0.56250 acc_top1_avg=0.50159 acc_top5_avg=0.86923 lr=0.01000 gn=6.96176 time=64.11it/s +====================Eval==================== +epoch=1 global_step=782 loss=2.22574 test_loss_avg=1.76245 acc=0.42188 test_acc_avg=0.55580 test_acc_top5_avg=0.96540 time=243.77it/s +epoch=1 global_step=782 loss=0.41767 test_loss_avg=1.61941 acc=0.89844 test_acc_avg=0.59848 test_acc_top5_avg=0.94729 time=249.54it/s +epoch=1 global_step=782 loss=0.31884 test_loss_avg=1.50261 acc=0.93750 test_acc_avg=0.62609 test_acc_top5_avg=0.95204 time=868.21it/s +curr_acc 0.6261 +BEST_ACC 0.4781 +curr_acc_top5 0.9520 +BEST_ACC_top5 0.9179 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=4.49508 loss_avg=4.72163 acc=0.55469 acc_top1_avg=0.53168 acc_top5_avg=0.88672 lr=0.01000 gn=7.42476 time=61.12it/s +epoch=2 global_step=850 loss=3.80041 loss_avg=4.70630 acc=0.61719 acc_top1_avg=0.53435 acc_top5_avg=0.88178 lr=0.01000 gn=6.27363 time=57.50it/s +epoch=2 global_step=900 loss=4.96506 loss_avg=4.64218 acc=0.50000 acc_top1_avg=0.54151 acc_top5_avg=0.88539 lr=0.01000 gn=6.40353 time=58.58it/s +epoch=2 global_step=950 loss=4.94095 loss_avg=4.65638 acc=0.49219 acc_top1_avg=0.54041 acc_top5_avg=0.88170 lr=0.01000 gn=5.58733 time=65.12it/s +epoch=2 global_step=1000 loss=4.92234 loss_avg=4.62372 acc=0.50781 acc_top1_avg=0.54451 acc_top5_avg=0.88170 lr=0.01000 gn=6.47415 time=59.68it/s +epoch=2 global_step=1050 loss=4.00655 loss_avg=4.60228 acc=0.61719 acc_top1_avg=0.54664 acc_top5_avg=0.88211 lr=0.01000 gn=8.62525 time=52.41it/s +epoch=2 global_step=1100 loss=4.78850 loss_avg=4.58444 acc=0.53125 acc_top1_avg=0.54914 acc_top5_avg=0.88225 lr=0.01000 gn=6.46780 time=63.39it/s +epoch=2 global_step=1150 loss=3.57321 loss_avg=4.54989 acc=0.65625 acc_top1_avg=0.55271 acc_top5_avg=0.88249 lr=0.01000 gn=7.30688 time=56.34it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.82737 test_loss_avg=1.46062 acc=0.55469 test_acc_avg=0.64435 test_acc_top5_avg=0.95331 time=246.54it/s +epoch=2 global_step=1173 loss=0.54541 test_loss_avg=1.15348 acc=0.75000 test_acc_avg=0.70797 test_acc_top5_avg=0.96499 time=545.92it/s +curr_acc 0.7080 +BEST_ACC 0.6261 +curr_acc_top5 0.9650 +BEST_ACC_top5 0.9520 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=4.88872 loss_avg=4.44315 acc=0.52344 acc_top1_avg=0.56568 acc_top5_avg=0.88079 lr=0.01000 gn=6.25166 time=57.62it/s +epoch=3 global_step=1250 loss=4.45068 loss_avg=4.34100 acc=0.57031 acc_top1_avg=0.57468 acc_top5_avg=0.88575 lr=0.01000 gn=5.74128 time=63.86it/s +epoch=3 global_step=1300 loss=4.59839 loss_avg=4.32729 acc=0.57031 acc_top1_avg=0.57708 acc_top5_avg=0.88706 lr=0.01000 gn=6.21082 time=56.65it/s +epoch=3 global_step=1350 loss=4.08161 loss_avg=4.29477 acc=0.60156 acc_top1_avg=0.58024 acc_top5_avg=0.88930 lr=0.01000 gn=5.94846 time=63.82it/s +epoch=3 global_step=1400 loss=3.98479 loss_avg=4.29011 acc=0.60938 acc_top1_avg=0.58016 acc_top5_avg=0.88925 lr=0.01000 gn=7.73163 time=62.18it/s +epoch=3 global_step=1450 loss=3.53536 loss_avg=4.27415 acc=0.67969 acc_top1_avg=0.58196 acc_top5_avg=0.89023 lr=0.01000 gn=7.06856 time=56.06it/s +epoch=3 global_step=1500 loss=4.07019 loss_avg=4.28051 acc=0.59375 acc_top1_avg=0.58090 acc_top5_avg=0.89027 lr=0.01000 gn=5.72177 time=58.10it/s +epoch=3 global_step=1550 loss=3.98773 loss_avg=4.27163 acc=0.61719 acc_top1_avg=0.58188 acc_top5_avg=0.88986 lr=0.01000 gn=5.98255 time=55.96it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.10250 test_loss_avg=0.81773 acc=0.95312 test_acc_avg=0.76262 test_acc_top5_avg=0.98678 time=247.74it/s +epoch=3 global_step=1564 loss=0.23488 test_loss_avg=1.26197 acc=0.92969 test_acc_avg=0.69072 test_acc_top5_avg=0.96106 time=247.51it/s +epoch=3 global_step=1564 loss=1.78865 test_loss_avg=1.24463 acc=0.43750 test_acc_avg=0.69264 test_acc_top5_avg=0.96420 time=882.27it/s +curr_acc 0.6926 +BEST_ACC 0.7080 +curr_acc_top5 0.9642 +BEST_ACC_top5 0.9650 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.56917 loss_avg=4.20277 acc=0.54688 acc_top1_avg=0.58442 acc_top5_avg=0.88411 lr=0.01000 gn=6.30033 time=56.93it/s +epoch=4 global_step=1650 loss=4.36874 loss_avg=4.11187 acc=0.53906 acc_top1_avg=0.59720 acc_top5_avg=0.88835 lr=0.01000 gn=8.04805 time=56.37it/s +epoch=4 global_step=1700 loss=3.92377 loss_avg=4.08026 acc=0.59375 acc_top1_avg=0.60053 acc_top5_avg=0.89264 lr=0.01000 gn=5.78547 time=64.18it/s +epoch=4 global_step=1750 loss=4.04252 loss_avg=4.07330 acc=0.59375 acc_top1_avg=0.60202 acc_top5_avg=0.89365 lr=0.01000 gn=6.90483 time=63.75it/s +epoch=4 global_step=1800 loss=4.11550 loss_avg=4.08606 acc=0.61719 acc_top1_avg=0.60083 acc_top5_avg=0.89255 lr=0.01000 gn=5.97436 time=53.43it/s +epoch=4 global_step=1850 loss=4.04052 loss_avg=4.09172 acc=0.60938 acc_top1_avg=0.60082 acc_top5_avg=0.89142 lr=0.01000 gn=6.64977 time=62.66it/s +epoch=4 global_step=1900 loss=3.84348 loss_avg=4.08124 acc=0.62500 acc_top1_avg=0.60221 acc_top5_avg=0.89181 lr=0.01000 gn=6.73525 time=63.31it/s +epoch=4 global_step=1950 loss=3.55807 loss_avg=4.08908 acc=0.65625 acc_top1_avg=0.60122 acc_top5_avg=0.89164 lr=0.01000 gn=6.40201 time=64.07it/s +====================Eval==================== +epoch=4 global_step=1955 loss=2.76937 test_loss_avg=1.47823 acc=0.39844 test_acc_avg=0.66360 test_acc_top5_avg=0.96415 time=249.02it/s +epoch=4 global_step=1955 loss=1.06110 test_loss_avg=1.23403 acc=0.62500 test_acc_avg=0.70629 test_acc_top5_avg=0.97102 time=892.79it/s +curr_acc 0.7063 +BEST_ACC 0.7080 +curr_acc_top5 0.9710 +BEST_ACC_top5 0.9650 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=3.10564 loss_avg=3.89628 acc=0.70312 acc_top1_avg=0.62309 acc_top5_avg=0.89809 lr=0.01000 gn=6.82334 time=62.32it/s +epoch=5 global_step=2050 loss=4.15525 loss_avg=3.95220 acc=0.59375 acc_top1_avg=0.61661 acc_top5_avg=0.89811 lr=0.01000 gn=6.65282 time=56.08it/s +epoch=5 global_step=2100 loss=4.04115 loss_avg=3.95950 acc=0.60156 acc_top1_avg=0.61536 acc_top5_avg=0.89833 lr=0.01000 gn=6.98703 time=64.51it/s +epoch=5 global_step=2150 loss=3.18870 loss_avg=3.95896 acc=0.70312 acc_top1_avg=0.61567 acc_top5_avg=0.89828 lr=0.01000 gn=6.92599 time=63.95it/s +epoch=5 global_step=2200 loss=3.22327 loss_avg=3.96365 acc=0.67969 acc_top1_avg=0.61502 acc_top5_avg=0.89739 lr=0.01000 gn=5.93902 time=53.96it/s +epoch=5 global_step=2250 loss=4.16137 loss_avg=3.97243 acc=0.59375 acc_top1_avg=0.61367 acc_top5_avg=0.89674 lr=0.01000 gn=6.36004 time=60.14it/s +epoch=5 global_step=2300 loss=3.93744 loss_avg=3.96176 acc=0.58594 acc_top1_avg=0.61474 acc_top5_avg=0.89701 lr=0.01000 gn=5.50095 time=64.02it/s +====================Eval==================== +epoch=5 global_step=2346 loss=1.33667 test_loss_avg=1.39975 acc=0.57031 test_acc_avg=0.59375 test_acc_top5_avg=0.97656 time=246.12it/s +epoch=5 global_step=2346 loss=1.19118 test_loss_avg=1.19836 acc=0.61719 test_acc_avg=0.68253 test_acc_top5_avg=0.96619 time=247.80it/s +epoch=5 global_step=2346 loss=0.42056 test_loss_avg=0.95265 acc=0.87500 test_acc_avg=0.74367 test_acc_top5_avg=0.97340 time=885.25it/s +curr_acc 0.7437 +BEST_ACC 0.7080 +curr_acc_top5 0.9734 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=3.90868 loss_avg=3.89546 acc=0.64844 acc_top1_avg=0.63086 acc_top5_avg=0.89453 lr=0.01000 gn=7.39304 time=34.09it/s +epoch=6 global_step=2400 loss=4.02458 loss_avg=3.95140 acc=0.61719 acc_top1_avg=0.61661 acc_top5_avg=0.89482 lr=0.01000 gn=7.54701 time=53.85it/s +epoch=6 global_step=2450 loss=3.80519 loss_avg=3.91525 acc=0.61719 acc_top1_avg=0.61967 acc_top5_avg=0.89911 lr=0.01000 gn=7.13130 time=55.85it/s +epoch=6 global_step=2500 loss=4.16600 loss_avg=3.89283 acc=0.57812 acc_top1_avg=0.62175 acc_top5_avg=0.90138 lr=0.01000 gn=7.95210 time=57.79it/s +epoch=6 global_step=2550 loss=3.82336 loss_avg=3.89260 acc=0.64062 acc_top1_avg=0.62243 acc_top5_avg=0.90062 lr=0.01000 gn=6.15393 time=58.43it/s +epoch=6 global_step=2600 loss=3.51786 loss_avg=3.86758 acc=0.64844 acc_top1_avg=0.62432 acc_top5_avg=0.90145 lr=0.01000 gn=7.75657 time=55.54it/s +epoch=6 global_step=2650 loss=3.75459 loss_avg=3.86667 acc=0.64844 acc_top1_avg=0.62474 acc_top5_avg=0.90044 lr=0.01000 gn=5.36131 time=64.81it/s +epoch=6 global_step=2700 loss=4.73503 loss_avg=3.87613 acc=0.53906 acc_top1_avg=0.62398 acc_top5_avg=0.89989 lr=0.01000 gn=6.97017 time=57.55it/s +====================Eval==================== +epoch=6 global_step=2737 loss=0.75143 test_loss_avg=0.72123 acc=0.78125 test_acc_avg=0.80108 test_acc_top5_avg=0.98798 time=246.64it/s +epoch=6 global_step=2737 loss=1.22190 test_loss_avg=0.95298 acc=0.72656 test_acc_avg=0.74907 test_acc_top5_avg=0.97728 time=261.93it/s +epoch=6 global_step=2737 loss=1.13523 test_loss_avg=0.96050 acc=0.68750 test_acc_avg=0.74713 test_acc_top5_avg=0.97765 time=761.77it/s +curr_acc 0.7471 +BEST_ACC 0.7437 +curr_acc_top5 0.9777 +BEST_ACC_top5 0.9734 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=3.22504 loss_avg=3.69848 acc=0.68750 acc_top1_avg=0.64303 acc_top5_avg=0.90505 lr=0.01000 gn=5.18377 time=59.26it/s +epoch=7 global_step=2800 loss=3.63827 loss_avg=3.70966 acc=0.67969 acc_top1_avg=0.64323 acc_top5_avg=0.89633 lr=0.01000 gn=7.32802 time=59.66it/s +epoch=7 global_step=2850 loss=3.28630 loss_avg=3.75462 acc=0.69531 acc_top1_avg=0.63786 acc_top5_avg=0.89927 lr=0.01000 gn=6.18775 time=57.11it/s +epoch=7 global_step=2900 loss=3.31253 loss_avg=3.75396 acc=0.68750 acc_top1_avg=0.63775 acc_top5_avg=0.90069 lr=0.01000 gn=6.95320 time=51.65it/s +epoch=7 global_step=2950 loss=3.67181 loss_avg=3.78127 acc=0.64844 acc_top1_avg=0.63531 acc_top5_avg=0.90071 lr=0.01000 gn=6.82136 time=61.82it/s +epoch=7 global_step=3000 loss=3.97657 loss_avg=3.78658 acc=0.58594 acc_top1_avg=0.63442 acc_top5_avg=0.89995 lr=0.01000 gn=7.75469 time=56.38it/s +epoch=7 global_step=3050 loss=4.25933 loss_avg=3.79064 acc=0.58594 acc_top1_avg=0.63376 acc_top5_avg=0.89981 lr=0.01000 gn=7.00297 time=54.33it/s +epoch=7 global_step=3100 loss=4.15408 loss_avg=3.77358 acc=0.58594 acc_top1_avg=0.63580 acc_top5_avg=0.90009 lr=0.01000 gn=8.35944 time=60.87it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.18326 test_loss_avg=1.20916 acc=0.67188 test_acc_avg=0.69016 test_acc_top5_avg=0.97706 time=239.94it/s +epoch=7 global_step=3128 loss=0.46367 test_loss_avg=1.02951 acc=0.81250 test_acc_avg=0.73230 test_acc_top5_avg=0.97547 time=875.82it/s +curr_acc 0.7323 +BEST_ACC 0.7471 +curr_acc_top5 0.9755 +BEST_ACC_top5 0.9777 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=3.49372 loss_avg=3.70623 acc=0.65625 acc_top1_avg=0.64134 acc_top5_avg=0.90128 lr=0.01000 gn=8.03784 time=64.70it/s +epoch=8 global_step=3200 loss=4.19302 loss_avg=3.78246 acc=0.57812 acc_top1_avg=0.63303 acc_top5_avg=0.89985 lr=0.01000 gn=8.84306 time=58.75it/s +epoch=8 global_step=3250 loss=4.06326 loss_avg=3.75834 acc=0.60156 acc_top1_avg=0.63525 acc_top5_avg=0.90061 lr=0.01000 gn=7.67260 time=57.68it/s +epoch=8 global_step=3300 loss=3.53332 loss_avg=3.74258 acc=0.67188 acc_top1_avg=0.63713 acc_top5_avg=0.90112 lr=0.01000 gn=7.96732 time=58.83it/s +epoch=8 global_step=3350 loss=4.00725 loss_avg=3.73248 acc=0.64062 acc_top1_avg=0.63904 acc_top5_avg=0.90157 lr=0.01000 gn=7.88219 time=53.91it/s +epoch=8 global_step=3400 loss=4.02930 loss_avg=3.72516 acc=0.61719 acc_top1_avg=0.63985 acc_top5_avg=0.90303 lr=0.01000 gn=7.88402 time=64.31it/s +epoch=8 global_step=3450 loss=4.27668 loss_avg=3.73850 acc=0.58594 acc_top1_avg=0.63825 acc_top5_avg=0.90283 lr=0.01000 gn=7.94656 time=60.48it/s +epoch=8 global_step=3500 loss=3.28698 loss_avg=3.74204 acc=0.68750 acc_top1_avg=0.63773 acc_top5_avg=0.90243 lr=0.01000 gn=10.24762 time=64.96it/s +====================Eval==================== +epoch=8 global_step=3519 loss=0.58383 test_loss_avg=0.90678 acc=0.83594 test_acc_avg=0.77995 test_acc_top5_avg=0.97135 time=238.83it/s +epoch=8 global_step=3519 loss=1.16990 test_loss_avg=1.30093 acc=0.70312 test_acc_avg=0.69773 test_acc_top5_avg=0.97817 time=251.93it/s +epoch=8 global_step=3519 loss=0.75997 test_loss_avg=1.28558 acc=0.75000 test_acc_avg=0.69541 test_acc_top5_avg=0.97271 time=872.00it/s +curr_acc 0.6954 +BEST_ACC 0.7471 +curr_acc_top5 0.9727 +BEST_ACC_top5 0.9777 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=3.61986 loss_avg=3.70903 acc=0.65625 acc_top1_avg=0.63936 acc_top5_avg=0.90146 lr=0.01000 gn=7.81037 time=56.89it/s +epoch=9 global_step=3600 loss=3.34257 loss_avg=3.69006 acc=0.67188 acc_top1_avg=0.64342 acc_top5_avg=0.89979 lr=0.01000 gn=7.95370 time=54.98it/s +epoch=9 global_step=3650 loss=3.72141 loss_avg=3.65916 acc=0.63281 acc_top1_avg=0.64719 acc_top5_avg=0.90297 lr=0.01000 gn=7.66272 time=60.27it/s +epoch=9 global_step=3700 loss=3.19056 loss_avg=3.66298 acc=0.69531 acc_top1_avg=0.64727 acc_top5_avg=0.90288 lr=0.01000 gn=6.51282 time=58.75it/s +epoch=9 global_step=3750 loss=3.21346 loss_avg=3.65139 acc=0.70312 acc_top1_avg=0.64891 acc_top5_avg=0.90283 lr=0.01000 gn=7.45175 time=55.99it/s +epoch=9 global_step=3800 loss=3.85251 loss_avg=3.66788 acc=0.62500 acc_top1_avg=0.64730 acc_top5_avg=0.90266 lr=0.01000 gn=8.34149 time=57.84it/s +epoch=9 global_step=3850 loss=3.64529 loss_avg=3.68090 acc=0.64844 acc_top1_avg=0.64586 acc_top5_avg=0.90200 lr=0.01000 gn=7.73963 time=58.37it/s +epoch=9 global_step=3900 loss=3.68315 loss_avg=3.69740 acc=0.66406 acc_top1_avg=0.64376 acc_top5_avg=0.90192 lr=0.01000 gn=8.65820 time=63.28it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.64262 test_loss_avg=1.01044 acc=0.84375 test_acc_avg=0.73638 test_acc_top5_avg=0.97496 time=245.42it/s +epoch=9 global_step=3910 loss=0.87915 test_loss_avg=0.86511 acc=0.68750 test_acc_avg=0.76661 test_acc_top5_avg=0.98012 time=882.45it/s +curr_acc 0.7666 +BEST_ACC 0.7471 +curr_acc_top5 0.9801 +BEST_ACC_top5 0.9777 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=4.04718 loss_avg=3.63119 acc=0.59375 acc_top1_avg=0.65059 acc_top5_avg=0.91055 lr=0.01000 gn=8.05885 time=57.89it/s +epoch=10 global_step=4000 loss=3.61844 loss_avg=3.60942 acc=0.65625 acc_top1_avg=0.65399 acc_top5_avg=0.90955 lr=0.01000 gn=9.40116 time=63.97it/s +epoch=10 global_step=4050 loss=3.14802 loss_avg=3.61723 acc=0.68750 acc_top1_avg=0.65268 acc_top5_avg=0.90815 lr=0.01000 gn=9.20901 time=56.73it/s +epoch=10 global_step=4100 loss=2.75550 loss_avg=3.62099 acc=0.73438 acc_top1_avg=0.65222 acc_top5_avg=0.90794 lr=0.01000 gn=7.83937 time=56.89it/s +epoch=10 global_step=4150 loss=3.54954 loss_avg=3.62950 acc=0.65625 acc_top1_avg=0.65114 acc_top5_avg=0.90674 lr=0.01000 gn=7.30183 time=59.03it/s +epoch=10 global_step=4200 loss=2.87095 loss_avg=3.62771 acc=0.75000 acc_top1_avg=0.65140 acc_top5_avg=0.90663 lr=0.01000 gn=10.08019 time=56.78it/s +epoch=10 global_step=4250 loss=4.43222 loss_avg=3.63785 acc=0.58594 acc_top1_avg=0.65025 acc_top5_avg=0.90639 lr=0.01000 gn=8.39728 time=62.70it/s +epoch=10 global_step=4300 loss=3.92414 loss_avg=3.64873 acc=0.60156 acc_top1_avg=0.64918 acc_top5_avg=0.90599 lr=0.01000 gn=7.14866 time=56.33it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.10871 test_loss_avg=0.48191 acc=0.95312 test_acc_avg=0.85781 test_acc_top5_avg=0.99297 time=240.80it/s +epoch=10 global_step=4301 loss=0.40420 test_loss_avg=1.15596 acc=0.87500 test_acc_avg=0.72656 test_acc_top5_avg=0.95417 time=248.48it/s +epoch=10 global_step=4301 loss=0.50903 test_loss_avg=1.00486 acc=0.81250 test_acc_avg=0.75178 test_acc_top5_avg=0.96371 time=548.78it/s +curr_acc 0.7518 +BEST_ACC 0.7666 +curr_acc_top5 0.9637 +BEST_ACC_top5 0.9801 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=2.97557 loss_avg=3.63066 acc=0.74219 acc_top1_avg=0.65115 acc_top5_avg=0.90721 lr=0.01000 gn=8.99258 time=55.61it/s +epoch=11 global_step=4400 loss=4.22000 loss_avg=3.62979 acc=0.60156 acc_top1_avg=0.65065 acc_top5_avg=0.90617 lr=0.01000 gn=8.84486 time=56.30it/s +epoch=11 global_step=4450 loss=3.41844 loss_avg=3.62152 acc=0.67188 acc_top1_avg=0.65174 acc_top5_avg=0.90588 lr=0.01000 gn=8.77888 time=55.83it/s +epoch=11 global_step=4500 loss=3.18115 loss_avg=3.62588 acc=0.69531 acc_top1_avg=0.65103 acc_top5_avg=0.90594 lr=0.01000 gn=8.63366 time=54.27it/s +epoch=11 global_step=4550 loss=4.17862 loss_avg=3.62931 acc=0.59375 acc_top1_avg=0.65032 acc_top5_avg=0.90556 lr=0.01000 gn=10.52571 time=62.32it/s +epoch=11 global_step=4600 loss=3.69082 loss_avg=3.63244 acc=0.65625 acc_top1_avg=0.64990 acc_top5_avg=0.90628 lr=0.01000 gn=9.04923 time=57.73it/s +epoch=11 global_step=4650 loss=4.16829 loss_avg=3.63364 acc=0.59375 acc_top1_avg=0.64998 acc_top5_avg=0.90661 lr=0.01000 gn=7.64873 time=54.62it/s +====================Eval==================== +epoch=11 global_step=4692 loss=0.86475 test_loss_avg=1.13140 acc=0.75781 test_acc_avg=0.71421 test_acc_top5_avg=0.98362 time=242.73it/s +epoch=11 global_step=4692 loss=0.02873 test_loss_avg=1.02698 acc=1.00000 test_acc_avg=0.73952 test_acc_top5_avg=0.98042 time=551.37it/s +curr_acc 0.7395 +BEST_ACC 0.7666 +curr_acc_top5 0.9804 +BEST_ACC_top5 0.9801 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=3.32969 loss_avg=3.44112 acc=0.69531 acc_top1_avg=0.66797 acc_top5_avg=0.91406 lr=0.01000 gn=9.70504 time=53.31it/s +epoch=12 global_step=4750 loss=3.44008 loss_avg=3.62102 acc=0.66406 acc_top1_avg=0.64911 acc_top5_avg=0.91231 lr=0.01000 gn=8.31334 time=61.42it/s +epoch=12 global_step=4800 loss=3.60369 loss_avg=3.63702 acc=0.67969 acc_top1_avg=0.64996 acc_top5_avg=0.90705 lr=0.01000 gn=8.96099 time=56.56it/s +epoch=12 global_step=4850 loss=3.05106 loss_avg=3.61341 acc=0.71094 acc_top1_avg=0.65259 acc_top5_avg=0.90541 lr=0.01000 gn=8.46757 time=59.72it/s +epoch=12 global_step=4900 loss=3.48555 loss_avg=3.60644 acc=0.66406 acc_top1_avg=0.65332 acc_top5_avg=0.90572 lr=0.01000 gn=8.94521 time=56.53it/s +epoch=12 global_step=4950 loss=3.16307 loss_avg=3.61022 acc=0.71094 acc_top1_avg=0.65225 acc_top5_avg=0.90643 lr=0.01000 gn=8.78010 time=55.31it/s +epoch=12 global_step=5000 loss=2.96584 loss_avg=3.62736 acc=0.72656 acc_top1_avg=0.65006 acc_top5_avg=0.90592 lr=0.01000 gn=8.64553 time=62.79it/s +epoch=12 global_step=5050 loss=3.52508 loss_avg=3.63394 acc=0.67188 acc_top1_avg=0.64881 acc_top5_avg=0.90614 lr=0.01000 gn=9.87488 time=64.05it/s +====================Eval==================== +epoch=12 global_step=5083 loss=1.13901 test_loss_avg=1.00147 acc=0.75000 test_acc_avg=0.75000 test_acc_top5_avg=0.99609 time=242.22it/s +epoch=12 global_step=5083 loss=0.31270 test_loss_avg=1.31040 acc=0.92188 test_acc_avg=0.66151 test_acc_top5_avg=0.94366 time=241.19it/s +epoch=12 global_step=5083 loss=0.07710 test_loss_avg=0.99814 acc=1.00000 test_acc_avg=0.73586 test_acc_top5_avg=0.96044 time=878.57it/s +curr_acc 0.7359 +BEST_ACC 0.7666 +curr_acc_top5 0.9604 +BEST_ACC_top5 0.9804 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=4.27773 loss_avg=3.55506 acc=0.60938 acc_top1_avg=0.66406 acc_top5_avg=0.91131 lr=0.01000 gn=9.31039 time=63.53it/s +epoch=13 global_step=5150 loss=4.05066 loss_avg=3.55021 acc=0.61719 acc_top1_avg=0.66115 acc_top5_avg=0.91021 lr=0.01000 gn=8.46368 time=53.01it/s +epoch=13 global_step=5200 loss=3.99483 loss_avg=3.55701 acc=0.60156 acc_top1_avg=0.65986 acc_top5_avg=0.90845 lr=0.01000 gn=8.98689 time=53.67it/s +epoch=13 global_step=5250 loss=3.66772 loss_avg=3.57061 acc=0.66406 acc_top1_avg=0.65784 acc_top5_avg=0.90639 lr=0.01000 gn=8.76843 time=59.64it/s +epoch=13 global_step=5300 loss=3.37981 loss_avg=3.58940 acc=0.68750 acc_top1_avg=0.65492 acc_top5_avg=0.90575 lr=0.01000 gn=10.35091 time=48.87it/s +epoch=13 global_step=5350 loss=3.41764 loss_avg=3.58373 acc=0.66406 acc_top1_avg=0.65546 acc_top5_avg=0.90637 lr=0.01000 gn=8.76407 time=47.96it/s +epoch=13 global_step=5400 loss=3.68699 loss_avg=3.58956 acc=0.65625 acc_top1_avg=0.65519 acc_top5_avg=0.90642 lr=0.01000 gn=6.60776 time=46.88it/s +epoch=13 global_step=5450 loss=3.16008 loss_avg=3.59673 acc=0.69531 acc_top1_avg=0.65421 acc_top5_avg=0.90531 lr=0.01000 gn=10.87161 time=59.85it/s +====================Eval==================== +epoch=13 global_step=5474 loss=1.68622 test_loss_avg=1.10081 acc=0.51562 test_acc_avg=0.70958 test_acc_top5_avg=0.98064 time=245.14it/s +epoch=13 global_step=5474 loss=1.14842 test_loss_avg=0.91251 acc=0.69531 test_acc_avg=0.75824 test_acc_top5_avg=0.97314 time=248.43it/s +epoch=13 global_step=5474 loss=0.42488 test_loss_avg=0.90612 acc=0.81250 test_acc_avg=0.75831 test_acc_top5_avg=0.97498 time=535.74it/s +curr_acc 0.7583 +BEST_ACC 0.7666 +curr_acc_top5 0.9750 +BEST_ACC_top5 0.9804 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=3.56003 loss_avg=3.43810 acc=0.66406 acc_top1_avg=0.67548 acc_top5_avg=0.90925 lr=0.01000 gn=6.76272 time=61.95it/s +epoch=14 global_step=5550 loss=4.42604 loss_avg=3.46089 acc=0.57812 acc_top1_avg=0.67116 acc_top5_avg=0.91139 lr=0.01000 gn=8.51486 time=61.92it/s +epoch=14 global_step=5600 loss=3.93989 loss_avg=3.50728 acc=0.62500 acc_top1_avg=0.66450 acc_top5_avg=0.91059 lr=0.01000 gn=9.32595 time=46.17it/s +epoch=14 global_step=5650 loss=3.70173 loss_avg=3.55947 acc=0.65625 acc_top1_avg=0.65887 acc_top5_avg=0.90865 lr=0.01000 gn=9.91100 time=53.36it/s +epoch=14 global_step=5700 loss=3.26437 loss_avg=3.55801 acc=0.70312 acc_top1_avg=0.65898 acc_top5_avg=0.90781 lr=0.01000 gn=8.67961 time=55.81it/s +epoch=14 global_step=5750 loss=2.99990 loss_avg=3.56896 acc=0.71875 acc_top1_avg=0.65803 acc_top5_avg=0.90693 lr=0.01000 gn=9.11555 time=60.95it/s +epoch=14 global_step=5800 loss=3.13415 loss_avg=3.56393 acc=0.71094 acc_top1_avg=0.65800 acc_top5_avg=0.90754 lr=0.01000 gn=9.33431 time=61.98it/s +epoch=14 global_step=5850 loss=3.51822 loss_avg=3.58548 acc=0.66406 acc_top1_avg=0.65640 acc_top5_avg=0.90748 lr=0.01000 gn=9.97724 time=57.41it/s +====================Eval==================== +epoch=14 global_step=5865 loss=1.58748 test_loss_avg=0.91446 acc=0.58594 test_acc_avg=0.74450 test_acc_top5_avg=0.98349 time=248.21it/s +epoch=14 global_step=5865 loss=0.08710 test_loss_avg=0.73690 acc=0.93750 test_acc_avg=0.79490 test_acc_top5_avg=0.98428 time=879.68it/s +curr_acc 0.7949 +BEST_ACC 0.7666 +curr_acc_top5 0.9843 +BEST_ACC_top5 0.9804 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=3.68872 loss_avg=3.49078 acc=0.63281 acc_top1_avg=0.66719 acc_top5_avg=0.90781 lr=0.01000 gn=8.96542 time=63.77it/s +epoch=15 global_step=5950 loss=3.92506 loss_avg=3.54136 acc=0.61719 acc_top1_avg=0.66112 acc_top5_avg=0.90662 lr=0.01000 gn=9.24915 time=51.19it/s +epoch=15 global_step=6000 loss=3.10891 loss_avg=3.54544 acc=0.71875 acc_top1_avg=0.65972 acc_top5_avg=0.90735 lr=0.01000 gn=9.40637 time=61.08it/s +epoch=15 global_step=6050 loss=3.73039 loss_avg=3.54372 acc=0.64062 acc_top1_avg=0.65980 acc_top5_avg=0.90612 lr=0.01000 gn=10.52606 time=62.51it/s +epoch=15 global_step=6100 loss=3.67199 loss_avg=3.55186 acc=0.63281 acc_top1_avg=0.65898 acc_top5_avg=0.90575 lr=0.01000 gn=9.88016 time=60.66it/s +epoch=15 global_step=6150 loss=3.45787 loss_avg=3.56616 acc=0.70312 acc_top1_avg=0.65787 acc_top5_avg=0.90609 lr=0.01000 gn=9.68966 time=56.44it/s +epoch=15 global_step=6200 loss=3.79329 loss_avg=3.56297 acc=0.64844 acc_top1_avg=0.65826 acc_top5_avg=0.90695 lr=0.01000 gn=9.86241 time=56.25it/s +epoch=15 global_step=6250 loss=3.58855 loss_avg=3.56381 acc=0.65625 acc_top1_avg=0.65826 acc_top5_avg=0.90716 lr=0.01000 gn=9.28499 time=63.90it/s +====================Eval==================== +epoch=15 global_step=6256 loss=1.09234 test_loss_avg=1.26259 acc=0.68750 test_acc_avg=0.66042 test_acc_top5_avg=0.94219 time=245.99it/s +epoch=15 global_step=6256 loss=0.86166 test_loss_avg=0.99465 acc=0.75000 test_acc_avg=0.74255 test_acc_top5_avg=0.97680 time=226.69it/s +epoch=15 global_step=6256 loss=0.13777 test_loss_avg=0.90172 acc=0.93750 test_acc_avg=0.76127 test_acc_top5_avg=0.97844 time=449.21it/s +curr_acc 0.7613 +BEST_ACC 0.7949 +curr_acc_top5 0.9784 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=3.76734 loss_avg=3.51390 acc=0.64062 acc_top1_avg=0.66300 acc_top5_avg=0.90412 lr=0.01000 gn=10.25854 time=59.21it/s +epoch=16 global_step=6350 loss=3.40260 loss_avg=3.48999 acc=0.70312 acc_top1_avg=0.66639 acc_top5_avg=0.90550 lr=0.01000 gn=10.25156 time=61.69it/s +epoch=16 global_step=6400 loss=3.07460 loss_avg=3.55399 acc=0.71875 acc_top1_avg=0.66032 acc_top5_avg=0.90511 lr=0.01000 gn=9.10683 time=57.43it/s +epoch=16 global_step=6450 loss=3.67457 loss_avg=3.54855 acc=0.64062 acc_top1_avg=0.66048 acc_top5_avg=0.90520 lr=0.01000 gn=8.09402 time=48.19it/s +epoch=16 global_step=6500 loss=3.53710 loss_avg=3.55028 acc=0.66406 acc_top1_avg=0.66032 acc_top5_avg=0.90686 lr=0.01000 gn=8.49456 time=57.56it/s +epoch=16 global_step=6550 loss=3.77744 loss_avg=3.52509 acc=0.63281 acc_top1_avg=0.66273 acc_top5_avg=0.90851 lr=0.01000 gn=10.70042 time=57.45it/s +epoch=16 global_step=6600 loss=4.36822 loss_avg=3.53611 acc=0.56250 acc_top1_avg=0.66141 acc_top5_avg=0.90836 lr=0.01000 gn=8.83547 time=57.63it/s +====================Eval==================== +epoch=16 global_step=6647 loss=2.28673 test_loss_avg=1.42992 acc=0.56250 test_acc_avg=0.68164 test_acc_top5_avg=0.96897 time=237.33it/s +epoch=16 global_step=6647 loss=0.43590 test_loss_avg=1.28676 acc=0.87500 test_acc_avg=0.71440 test_acc_top5_avg=0.97844 time=754.37it/s +curr_acc 0.7144 +BEST_ACC 0.7949 +curr_acc_top5 0.9784 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=3.64556 loss_avg=3.84141 acc=0.64844 acc_top1_avg=0.63542 acc_top5_avg=0.89583 lr=0.01000 gn=7.46900 time=38.33it/s +epoch=17 global_step=6700 loss=2.34456 loss_avg=3.46060 acc=0.77344 acc_top1_avg=0.66598 acc_top5_avg=0.90625 lr=0.01000 gn=7.01914 time=61.93it/s +epoch=17 global_step=6750 loss=3.06753 loss_avg=3.51122 acc=0.71875 acc_top1_avg=0.66255 acc_top5_avg=0.90595 lr=0.01000 gn=10.42790 time=64.40it/s +epoch=17 global_step=6800 loss=4.40202 loss_avg=3.54294 acc=0.57812 acc_top1_avg=0.65972 acc_top5_avg=0.90630 lr=0.01000 gn=9.43287 time=62.87it/s +epoch=17 global_step=6850 loss=3.15000 loss_avg=3.54099 acc=0.71094 acc_top1_avg=0.65941 acc_top5_avg=0.90833 lr=0.01000 gn=8.67090 time=56.42it/s +epoch=17 global_step=6900 loss=4.40425 loss_avg=3.56334 acc=0.59375 acc_top1_avg=0.65755 acc_top5_avg=0.90912 lr=0.01000 gn=10.43210 time=64.94it/s +epoch=17 global_step=6950 loss=3.22445 loss_avg=3.55356 acc=0.71094 acc_top1_avg=0.65839 acc_top5_avg=0.90885 lr=0.01000 gn=10.23858 time=58.54it/s +epoch=17 global_step=7000 loss=3.43949 loss_avg=3.54948 acc=0.67969 acc_top1_avg=0.65871 acc_top5_avg=0.90904 lr=0.01000 gn=8.54130 time=54.34it/s +====================Eval==================== +epoch=17 global_step=7038 loss=1.64290 test_loss_avg=1.44794 acc=0.56250 test_acc_avg=0.62277 test_acc_top5_avg=0.98661 time=247.64it/s +epoch=17 global_step=7038 loss=1.17036 test_loss_avg=1.08763 acc=0.69531 test_acc_avg=0.73451 test_acc_top5_avg=0.98328 time=241.68it/s +epoch=17 global_step=7038 loss=0.65705 test_loss_avg=1.10111 acc=0.75000 test_acc_avg=0.72488 test_acc_top5_avg=0.97953 time=515.59it/s +curr_acc 0.7249 +BEST_ACC 0.7949 +curr_acc_top5 0.9795 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=3.57130 loss_avg=3.67608 acc=0.67969 acc_top1_avg=0.64844 acc_top5_avg=0.91211 lr=0.01000 gn=10.57763 time=57.02it/s +epoch=18 global_step=7100 loss=4.37947 loss_avg=3.55377 acc=0.57031 acc_top1_avg=0.65927 acc_top5_avg=0.90562 lr=0.01000 gn=8.39455 time=48.33it/s +epoch=18 global_step=7150 loss=3.88624 loss_avg=3.49756 acc=0.61719 acc_top1_avg=0.66441 acc_top5_avg=0.90883 lr=0.01000 gn=9.04463 time=61.27it/s +epoch=18 global_step=7200 loss=3.48831 loss_avg=3.53280 acc=0.66406 acc_top1_avg=0.66098 acc_top5_avg=0.90934 lr=0.01000 gn=10.65176 time=54.10it/s +epoch=18 global_step=7250 loss=3.21203 loss_avg=3.52886 acc=0.69531 acc_top1_avg=0.66211 acc_top5_avg=0.90986 lr=0.01000 gn=8.84797 time=60.72it/s +epoch=18 global_step=7300 loss=3.71924 loss_avg=3.52734 acc=0.64062 acc_top1_avg=0.66221 acc_top5_avg=0.91001 lr=0.01000 gn=10.23035 time=57.47it/s +epoch=18 global_step=7350 loss=3.79457 loss_avg=3.53828 acc=0.63281 acc_top1_avg=0.66066 acc_top5_avg=0.90976 lr=0.01000 gn=10.56871 time=57.67it/s +epoch=18 global_step=7400 loss=3.23336 loss_avg=3.53496 acc=0.71094 acc_top1_avg=0.66078 acc_top5_avg=0.90919 lr=0.01000 gn=10.93448 time=57.12it/s +====================Eval==================== +epoch=18 global_step=7429 loss=3.72804 test_loss_avg=1.44455 acc=0.22656 test_acc_avg=0.64927 test_acc_top5_avg=0.92941 time=247.95it/s +epoch=18 global_step=7429 loss=0.59077 test_loss_avg=1.02609 acc=0.77344 test_acc_avg=0.74058 test_acc_top5_avg=0.95423 time=263.69it/s +epoch=18 global_step=7429 loss=0.32648 test_loss_avg=1.01724 acc=0.81250 test_acc_avg=0.74150 test_acc_top5_avg=0.95481 time=862.49it/s +curr_acc 0.7415 +BEST_ACC 0.7949 +curr_acc_top5 0.9548 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=3.50883 loss_avg=3.42043 acc=0.66406 acc_top1_avg=0.67448 acc_top5_avg=0.91071 lr=0.01000 gn=9.04259 time=49.36it/s +epoch=19 global_step=7500 loss=4.17783 loss_avg=3.47288 acc=0.60156 acc_top1_avg=0.66747 acc_top5_avg=0.91131 lr=0.01000 gn=9.48116 time=59.13it/s +epoch=19 global_step=7550 loss=3.91247 loss_avg=3.46136 acc=0.62500 acc_top1_avg=0.66865 acc_top5_avg=0.91142 lr=0.01000 gn=9.01484 time=56.38it/s +epoch=19 global_step=7600 loss=3.58175 loss_avg=3.46574 acc=0.65625 acc_top1_avg=0.66827 acc_top5_avg=0.91160 lr=0.01000 gn=10.17218 time=55.93it/s +epoch=19 global_step=7650 loss=2.87674 loss_avg=3.48236 acc=0.72656 acc_top1_avg=0.66650 acc_top5_avg=0.91014 lr=0.01000 gn=8.35628 time=60.87it/s +epoch=19 global_step=7700 loss=3.48199 loss_avg=3.48645 acc=0.67188 acc_top1_avg=0.66614 acc_top5_avg=0.91055 lr=0.01000 gn=11.01353 time=62.01it/s +epoch=19 global_step=7750 loss=3.38445 loss_avg=3.47953 acc=0.67188 acc_top1_avg=0.66664 acc_top5_avg=0.91056 lr=0.01000 gn=8.45232 time=64.92it/s +epoch=19 global_step=7800 loss=3.84999 loss_avg=3.50062 acc=0.62500 acc_top1_avg=0.66438 acc_top5_avg=0.90956 lr=0.01000 gn=9.44911 time=64.49it/s +====================Eval==================== +epoch=19 global_step=7820 loss=0.24284 test_loss_avg=1.44848 acc=0.94531 test_acc_avg=0.66980 test_acc_top5_avg=0.95695 time=247.45it/s +epoch=19 global_step=7820 loss=0.81150 test_loss_avg=1.11439 acc=0.81250 test_acc_avg=0.73081 test_acc_top5_avg=0.96806 time=888.81it/s +curr_acc 0.7308 +BEST_ACC 0.7949 +curr_acc_top5 0.9681 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=2.88047 loss_avg=3.41278 acc=0.72656 acc_top1_avg=0.67891 acc_top5_avg=0.91120 lr=0.01000 gn=9.38900 time=64.91it/s +epoch=20 global_step=7900 loss=3.58925 loss_avg=3.38017 acc=0.64844 acc_top1_avg=0.67881 acc_top5_avg=0.91182 lr=0.01000 gn=8.63491 time=56.20it/s +epoch=20 global_step=7950 loss=2.75584 loss_avg=3.40362 acc=0.75781 acc_top1_avg=0.67542 acc_top5_avg=0.91280 lr=0.01000 gn=9.52178 time=59.34it/s +epoch=20 global_step=8000 loss=3.24684 loss_avg=3.44377 acc=0.71094 acc_top1_avg=0.67127 acc_top5_avg=0.91224 lr=0.01000 gn=10.07670 time=53.94it/s +epoch=20 global_step=8050 loss=3.38652 loss_avg=3.45923 acc=0.68750 acc_top1_avg=0.66950 acc_top5_avg=0.91050 lr=0.01000 gn=9.80997 time=50.00it/s +epoch=20 global_step=8100 loss=3.76670 loss_avg=3.45551 acc=0.62500 acc_top1_avg=0.66961 acc_top5_avg=0.91116 lr=0.01000 gn=11.09429 time=64.08it/s +epoch=20 global_step=8150 loss=3.39499 loss_avg=3.46320 acc=0.66406 acc_top1_avg=0.66892 acc_top5_avg=0.91070 lr=0.01000 gn=8.74976 time=57.79it/s +epoch=20 global_step=8200 loss=2.97983 loss_avg=3.47698 acc=0.71875 acc_top1_avg=0.66725 acc_top5_avg=0.90991 lr=0.01000 gn=10.33647 time=64.61it/s +====================Eval==================== +epoch=20 global_step=8211 loss=1.88446 test_loss_avg=0.85558 acc=0.57812 test_acc_avg=0.77344 test_acc_top5_avg=0.98203 time=244.81it/s +epoch=20 global_step=8211 loss=0.35994 test_loss_avg=1.05655 acc=0.89062 test_acc_avg=0.74665 test_acc_top5_avg=0.98158 time=258.08it/s +epoch=20 global_step=8211 loss=1.05074 test_loss_avg=1.09877 acc=0.50000 test_acc_avg=0.73131 test_acc_top5_avg=0.97933 time=878.76it/s +curr_acc 0.7313 +BEST_ACC 0.7949 +curr_acc_top5 0.9793 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=3.32697 loss_avg=3.42604 acc=0.68750 acc_top1_avg=0.67087 acc_top5_avg=0.91126 lr=0.01000 gn=8.57551 time=53.68it/s +epoch=21 global_step=8300 loss=3.54713 loss_avg=3.42541 acc=0.67188 acc_top1_avg=0.67319 acc_top5_avg=0.90976 lr=0.01000 gn=7.38315 time=54.39it/s +epoch=21 global_step=8350 loss=3.26227 loss_avg=3.44909 acc=0.70312 acc_top1_avg=0.67036 acc_top5_avg=0.90788 lr=0.01000 gn=11.35352 time=57.25it/s +epoch=21 global_step=8400 loss=4.17999 loss_avg=3.44945 acc=0.58594 acc_top1_avg=0.66977 acc_top5_avg=0.90923 lr=0.01000 gn=9.03764 time=57.64it/s +epoch=21 global_step=8450 loss=2.95836 loss_avg=3.45695 acc=0.71875 acc_top1_avg=0.66880 acc_top5_avg=0.90916 lr=0.01000 gn=10.96890 time=56.48it/s +epoch=21 global_step=8500 loss=3.67288 loss_avg=3.47789 acc=0.63281 acc_top1_avg=0.66652 acc_top5_avg=0.90841 lr=0.01000 gn=10.13981 time=64.23it/s +epoch=21 global_step=8550 loss=3.99357 loss_avg=3.49250 acc=0.60938 acc_top1_avg=0.66494 acc_top5_avg=0.90906 lr=0.01000 gn=11.52363 time=61.28it/s +epoch=21 global_step=8600 loss=3.17752 loss_avg=3.49823 acc=0.69531 acc_top1_avg=0.66424 acc_top5_avg=0.90954 lr=0.01000 gn=8.57378 time=63.53it/s +====================Eval==================== +epoch=21 global_step=8602 loss=2.36059 test_loss_avg=1.18750 acc=0.46875 test_acc_avg=0.71151 test_acc_top5_avg=0.97618 time=248.43it/s +epoch=21 global_step=8602 loss=1.18486 test_loss_avg=1.05445 acc=0.68750 test_acc_avg=0.74080 test_acc_top5_avg=0.97567 time=872.72it/s +curr_acc 0.7408 +BEST_ACC 0.7949 +curr_acc_top5 0.9757 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=2.78820 loss_avg=3.41010 acc=0.72656 acc_top1_avg=0.67529 acc_top5_avg=0.91227 lr=0.01000 gn=7.81883 time=55.65it/s +epoch=22 global_step=8700 loss=3.34305 loss_avg=3.45257 acc=0.69531 acc_top1_avg=0.66916 acc_top5_avg=0.91048 lr=0.01000 gn=10.41226 time=60.97it/s +epoch=22 global_step=8750 loss=3.43161 loss_avg=3.44608 acc=0.65625 acc_top1_avg=0.67013 acc_top5_avg=0.90952 lr=0.01000 gn=10.30065 time=64.71it/s +epoch=22 global_step=8800 loss=2.76139 loss_avg=3.47260 acc=0.72656 acc_top1_avg=0.66761 acc_top5_avg=0.90976 lr=0.01000 gn=8.58586 time=64.22it/s +epoch=22 global_step=8850 loss=3.53414 loss_avg=3.47466 acc=0.65625 acc_top1_avg=0.66699 acc_top5_avg=0.91009 lr=0.01000 gn=8.79814 time=55.82it/s +epoch=22 global_step=8900 loss=3.32882 loss_avg=3.48965 acc=0.67188 acc_top1_avg=0.66558 acc_top5_avg=0.91003 lr=0.01000 gn=11.40482 time=57.52it/s +epoch=22 global_step=8950 loss=3.09667 loss_avg=3.48178 acc=0.70312 acc_top1_avg=0.66655 acc_top5_avg=0.90953 lr=0.01000 gn=8.66824 time=58.97it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.26259 test_loss_avg=0.49718 acc=0.91406 test_acc_avg=0.87109 test_acc_top5_avg=0.98568 time=258.35it/s +epoch=22 global_step=8993 loss=1.15517 test_loss_avg=1.01026 acc=0.69531 test_acc_avg=0.76878 test_acc_top5_avg=0.98135 time=250.14it/s +epoch=22 global_step=8993 loss=0.56508 test_loss_avg=0.93720 acc=0.81250 test_acc_avg=0.77799 test_acc_top5_avg=0.98259 time=543.37it/s +curr_acc 0.7780 +BEST_ACC 0.7949 +curr_acc_top5 0.9826 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=2.78313 loss_avg=3.19628 acc=0.73438 acc_top1_avg=0.69196 acc_top5_avg=0.92969 lr=0.01000 gn=10.65262 time=65.04it/s +epoch=23 global_step=9050 loss=3.94815 loss_avg=3.43979 acc=0.62500 acc_top1_avg=0.66982 acc_top5_avg=0.91351 lr=0.01000 gn=7.03318 time=65.03it/s +epoch=23 global_step=9100 loss=3.38330 loss_avg=3.45246 acc=0.67969 acc_top1_avg=0.66866 acc_top5_avg=0.91107 lr=0.01000 gn=9.22455 time=58.42it/s +epoch=23 global_step=9150 loss=3.99811 loss_avg=3.44657 acc=0.62500 acc_top1_avg=0.66974 acc_top5_avg=0.91058 lr=0.01000 gn=10.79096 time=51.52it/s +epoch=23 global_step=9200 loss=3.97963 loss_avg=3.44960 acc=0.60156 acc_top1_avg=0.66938 acc_top5_avg=0.91044 lr=0.01000 gn=9.68639 time=62.67it/s +epoch=23 global_step=9250 loss=3.09356 loss_avg=3.47292 acc=0.71094 acc_top1_avg=0.66707 acc_top5_avg=0.91005 lr=0.01000 gn=6.62579 time=55.02it/s +epoch=23 global_step=9300 loss=3.65305 loss_avg=3.47069 acc=0.64062 acc_top1_avg=0.66727 acc_top5_avg=0.91009 lr=0.01000 gn=8.98801 time=61.75it/s +epoch=23 global_step=9350 loss=3.17067 loss_avg=3.48726 acc=0.71094 acc_top1_avg=0.66557 acc_top5_avg=0.90993 lr=0.01000 gn=11.23499 time=61.85it/s +====================Eval==================== +epoch=23 global_step=9384 loss=0.92892 test_loss_avg=0.79849 acc=0.78906 test_acc_avg=0.79332 test_acc_top5_avg=0.98295 time=248.37it/s +epoch=23 global_step=9384 loss=1.38064 test_loss_avg=0.83350 acc=0.56250 test_acc_avg=0.78036 test_acc_top5_avg=0.98299 time=865.16it/s +curr_acc 0.7804 +BEST_ACC 0.7949 +curr_acc_top5 0.9830 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=3.14067 loss_avg=3.45765 acc=0.71094 acc_top1_avg=0.66748 acc_top5_avg=0.91406 lr=0.01000 gn=10.84386 time=48.21it/s +epoch=24 global_step=9450 loss=3.69913 loss_avg=3.46523 acc=0.66406 acc_top1_avg=0.66903 acc_top5_avg=0.90767 lr=0.01000 gn=10.84497 time=60.54it/s +epoch=24 global_step=9500 loss=3.88640 loss_avg=3.44249 acc=0.63281 acc_top1_avg=0.67147 acc_top5_avg=0.91211 lr=0.01000 gn=9.30723 time=55.91it/s +epoch=24 global_step=9550 loss=3.80721 loss_avg=3.42335 acc=0.61719 acc_top1_avg=0.67225 acc_top5_avg=0.91425 lr=0.01000 gn=9.68588 time=62.18it/s +epoch=24 global_step=9600 loss=3.54663 loss_avg=3.43121 acc=0.66406 acc_top1_avg=0.67151 acc_top5_avg=0.91323 lr=0.01000 gn=8.67841 time=59.83it/s +epoch=24 global_step=9650 loss=3.16178 loss_avg=3.44740 acc=0.71094 acc_top1_avg=0.66967 acc_top5_avg=0.91330 lr=0.01000 gn=9.40758 time=60.19it/s +epoch=24 global_step=9700 loss=3.16263 loss_avg=3.45587 acc=0.69531 acc_top1_avg=0.66883 acc_top5_avg=0.91305 lr=0.01000 gn=11.30157 time=58.28it/s +epoch=24 global_step=9750 loss=3.79073 loss_avg=3.46732 acc=0.63281 acc_top1_avg=0.66795 acc_top5_avg=0.91280 lr=0.01000 gn=8.42724 time=55.69it/s +====================Eval==================== +epoch=24 global_step=9775 loss=0.65960 test_loss_avg=0.67293 acc=0.77344 test_acc_avg=0.80664 test_acc_top5_avg=0.98242 time=244.01it/s +epoch=24 global_step=9775 loss=1.82285 test_loss_avg=1.10537 acc=0.53125 test_acc_avg=0.72497 test_acc_top5_avg=0.98119 time=250.06it/s +epoch=24 global_step=9775 loss=0.75148 test_loss_avg=0.94986 acc=0.75000 test_acc_avg=0.75633 test_acc_top5_avg=0.98220 time=770.16it/s +curr_acc 0.7563 +BEST_ACC 0.7949 +curr_acc_top5 0.9822 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=4.00263 loss_avg=3.47917 acc=0.60938 acc_top1_avg=0.66500 acc_top5_avg=0.91594 lr=0.01000 gn=9.41476 time=60.92it/s +epoch=25 global_step=9850 loss=3.17406 loss_avg=3.44737 acc=0.67188 acc_top1_avg=0.66990 acc_top5_avg=0.91250 lr=0.01000 gn=10.72666 time=59.05it/s +epoch=25 global_step=9900 loss=2.67886 loss_avg=3.45464 acc=0.75000 acc_top1_avg=0.66894 acc_top5_avg=0.91063 lr=0.01000 gn=7.25075 time=52.72it/s +epoch=25 global_step=9950 loss=3.63671 loss_avg=3.46358 acc=0.64844 acc_top1_avg=0.66795 acc_top5_avg=0.90982 lr=0.01000 gn=9.48156 time=54.73it/s +epoch=25 global_step=10000 loss=3.12401 loss_avg=3.47792 acc=0.70312 acc_top1_avg=0.66615 acc_top5_avg=0.90962 lr=0.01000 gn=10.46368 time=62.75it/s +epoch=25 global_step=10050 loss=2.85446 loss_avg=3.46498 acc=0.74219 acc_top1_avg=0.66756 acc_top5_avg=0.90997 lr=0.01000 gn=9.99109 time=58.35it/s +epoch=25 global_step=10100 loss=3.09010 loss_avg=3.46059 acc=0.70312 acc_top1_avg=0.66760 acc_top5_avg=0.91034 lr=0.01000 gn=7.92330 time=55.65it/s +epoch=25 global_step=10150 loss=3.74214 loss_avg=3.46002 acc=0.64062 acc_top1_avg=0.66806 acc_top5_avg=0.91075 lr=0.01000 gn=9.32277 time=64.63it/s +====================Eval==================== +epoch=25 global_step=10166 loss=3.59542 test_loss_avg=0.85628 acc=0.37500 test_acc_avg=0.77844 test_acc_top5_avg=0.98219 time=252.30it/s +epoch=25 global_step=10166 loss=0.67314 test_loss_avg=1.03696 acc=0.81250 test_acc_avg=0.75187 test_acc_top5_avg=0.97510 time=262.65it/s +epoch=25 global_step=10166 loss=0.39140 test_loss_avg=1.02017 acc=0.87500 test_acc_avg=0.75465 test_acc_top5_avg=0.97607 time=552.75it/s +curr_acc 0.7546 +BEST_ACC 0.7949 +curr_acc_top5 0.9761 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=3.58086 loss_avg=3.44109 acc=0.64844 acc_top1_avg=0.67509 acc_top5_avg=0.91245 lr=0.01000 gn=10.41319 time=64.18it/s +epoch=26 global_step=10250 loss=3.50552 loss_avg=3.41252 acc=0.65625 acc_top1_avg=0.67550 acc_top5_avg=0.91146 lr=0.01000 gn=8.51978 time=57.23it/s +epoch=26 global_step=10300 loss=3.46044 loss_avg=3.46222 acc=0.65625 acc_top1_avg=0.67048 acc_top5_avg=0.91196 lr=0.01000 gn=7.26350 time=55.03it/s +epoch=26 global_step=10350 loss=4.02201 loss_avg=3.47336 acc=0.59375 acc_top1_avg=0.66784 acc_top5_avg=0.91139 lr=0.01000 gn=9.58628 time=58.77it/s +epoch=26 global_step=10400 loss=3.59621 loss_avg=3.47671 acc=0.64062 acc_top1_avg=0.66723 acc_top5_avg=0.91069 lr=0.01000 gn=7.81750 time=56.71it/s +epoch=26 global_step=10450 loss=2.95322 loss_avg=3.45838 acc=0.71094 acc_top1_avg=0.66915 acc_top5_avg=0.91071 lr=0.01000 gn=7.65473 time=53.95it/s +epoch=26 global_step=10500 loss=2.46227 loss_avg=3.45745 acc=0.76562 acc_top1_avg=0.66916 acc_top5_avg=0.91086 lr=0.01000 gn=10.01031 time=54.43it/s +epoch=26 global_step=10550 loss=3.57667 loss_avg=3.46004 acc=0.66406 acc_top1_avg=0.66878 acc_top5_avg=0.91077 lr=0.01000 gn=9.08486 time=61.57it/s +====================Eval==================== +epoch=26 global_step=10557 loss=0.97816 test_loss_avg=1.00720 acc=0.75781 test_acc_avg=0.75866 test_acc_top5_avg=0.97249 time=239.28it/s +epoch=26 global_step=10557 loss=0.16919 test_loss_avg=0.79263 acc=1.00000 test_acc_avg=0.80261 test_acc_top5_avg=0.98042 time=558.94it/s +curr_acc 0.8026 +BEST_ACC 0.7949 +curr_acc_top5 0.9804 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=2.89483 loss_avg=3.39129 acc=0.76562 acc_top1_avg=0.67733 acc_top5_avg=0.90825 lr=0.01000 gn=12.02126 time=60.43it/s +epoch=27 global_step=10650 loss=2.97319 loss_avg=3.42179 acc=0.71875 acc_top1_avg=0.67322 acc_top5_avg=0.91163 lr=0.01000 gn=8.28720 time=58.53it/s +epoch=27 global_step=10700 loss=3.78223 loss_avg=3.42654 acc=0.63281 acc_top1_avg=0.67357 acc_top5_avg=0.91122 lr=0.01000 gn=8.68082 time=63.36it/s +epoch=27 global_step=10750 loss=3.39167 loss_avg=3.38760 acc=0.66406 acc_top1_avg=0.67742 acc_top5_avg=0.91252 lr=0.01000 gn=8.09545 time=54.94it/s +epoch=27 global_step=10800 loss=3.28652 loss_avg=3.39019 acc=0.67969 acc_top1_avg=0.67724 acc_top5_avg=0.91229 lr=0.01000 gn=9.60202 time=64.38it/s +epoch=27 global_step=10850 loss=3.57129 loss_avg=3.42085 acc=0.68750 acc_top1_avg=0.67406 acc_top5_avg=0.91076 lr=0.01000 gn=10.36606 time=56.02it/s +epoch=27 global_step=10900 loss=2.65066 loss_avg=3.42549 acc=0.75781 acc_top1_avg=0.67342 acc_top5_avg=0.91124 lr=0.01000 gn=9.65130 time=61.85it/s +====================Eval==================== +epoch=27 global_step=10948 loss=2.12223 test_loss_avg=2.07216 acc=0.53125 test_acc_avg=0.51517 test_acc_top5_avg=0.95726 time=241.97it/s +epoch=27 global_step=10948 loss=0.06217 test_loss_avg=1.26793 acc=0.98438 test_acc_avg=0.70009 test_acc_top5_avg=0.97248 time=243.73it/s +epoch=27 global_step=10948 loss=0.79343 test_loss_avg=1.17738 acc=0.81250 test_acc_avg=0.71875 test_acc_top5_avg=0.97528 time=542.04it/s +curr_acc 0.7188 +BEST_ACC 0.8026 +curr_acc_top5 0.9753 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=3.47816 loss_avg=3.44908 acc=0.68750 acc_top1_avg=0.68750 acc_top5_avg=0.90625 lr=0.01000 gn=8.74554 time=41.91it/s +epoch=28 global_step=11000 loss=3.33417 loss_avg=3.45028 acc=0.72656 acc_top1_avg=0.66632 acc_top5_avg=0.90850 lr=0.01000 gn=9.74437 time=55.24it/s +epoch=28 global_step=11050 loss=3.55212 loss_avg=3.44169 acc=0.65625 acc_top1_avg=0.66881 acc_top5_avg=0.90954 lr=0.01000 gn=8.02060 time=64.84it/s +epoch=28 global_step=11100 loss=3.65276 loss_avg=3.44971 acc=0.63281 acc_top1_avg=0.66910 acc_top5_avg=0.91077 lr=0.01000 gn=7.36466 time=58.53it/s +epoch=28 global_step=11150 loss=3.10011 loss_avg=3.43282 acc=0.71875 acc_top1_avg=0.67106 acc_top5_avg=0.91259 lr=0.01000 gn=9.44451 time=63.62it/s +epoch=28 global_step=11200 loss=3.32725 loss_avg=3.43944 acc=0.67969 acc_top1_avg=0.67085 acc_top5_avg=0.91140 lr=0.01000 gn=9.35591 time=57.64it/s +epoch=28 global_step=11250 loss=3.97969 loss_avg=3.43548 acc=0.61719 acc_top1_avg=0.67123 acc_top5_avg=0.91186 lr=0.01000 gn=8.49156 time=57.69it/s +epoch=28 global_step=11300 loss=2.81526 loss_avg=3.43556 acc=0.73438 acc_top1_avg=0.67123 acc_top5_avg=0.91231 lr=0.01000 gn=10.02063 time=51.91it/s +====================Eval==================== +epoch=28 global_step=11339 loss=1.26298 test_loss_avg=1.09679 acc=0.69531 test_acc_avg=0.73191 test_acc_top5_avg=0.98026 time=239.70it/s +epoch=28 global_step=11339 loss=0.78275 test_loss_avg=1.01382 acc=0.75000 test_acc_avg=0.75059 test_acc_top5_avg=0.97963 time=533.08it/s +curr_acc 0.7506 +BEST_ACC 0.8026 +curr_acc_top5 0.9796 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=3.56277 loss_avg=3.49051 acc=0.67188 acc_top1_avg=0.66051 acc_top5_avg=0.91335 lr=0.01000 gn=13.21409 time=63.56it/s +epoch=29 global_step=11400 loss=4.00636 loss_avg=3.41043 acc=0.64062 acc_top1_avg=0.67341 acc_top5_avg=0.91381 lr=0.01000 gn=10.76976 time=65.20it/s +epoch=29 global_step=11450 loss=2.95693 loss_avg=3.40867 acc=0.71875 acc_top1_avg=0.67427 acc_top5_avg=0.91463 lr=0.01000 gn=11.01596 time=60.50it/s +epoch=29 global_step=11500 loss=3.42060 loss_avg=3.41202 acc=0.65625 acc_top1_avg=0.67348 acc_top5_avg=0.91421 lr=0.01000 gn=9.10024 time=58.75it/s +epoch=29 global_step=11550 loss=3.12648 loss_avg=3.41935 acc=0.70312 acc_top1_avg=0.67350 acc_top5_avg=0.91388 lr=0.01000 gn=10.75756 time=55.00it/s +epoch=29 global_step=11600 loss=3.70115 loss_avg=3.42389 acc=0.64062 acc_top1_avg=0.67298 acc_top5_avg=0.91355 lr=0.01000 gn=7.94601 time=58.98it/s +epoch=29 global_step=11650 loss=2.85032 loss_avg=3.43330 acc=0.71875 acc_top1_avg=0.67130 acc_top5_avg=0.91256 lr=0.01000 gn=8.79373 time=56.43it/s +epoch=29 global_step=11700 loss=3.85817 loss_avg=3.43394 acc=0.60938 acc_top1_avg=0.67131 acc_top5_avg=0.91268 lr=0.01000 gn=9.56740 time=64.33it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.21786 test_loss_avg=0.76236 acc=0.92188 test_acc_avg=0.77691 test_acc_top5_avg=0.98351 time=244.37it/s +epoch=29 global_step=11730 loss=0.03621 test_loss_avg=1.15507 acc=0.99219 test_acc_avg=0.71491 test_acc_top5_avg=0.97961 time=244.71it/s +epoch=29 global_step=11730 loss=0.30456 test_loss_avg=0.96307 acc=0.93750 test_acc_avg=0.75663 test_acc_top5_avg=0.98289 time=776.58it/s +curr_acc 0.7566 +BEST_ACC 0.8026 +curr_acc_top5 0.9829 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=3.22923 loss_avg=3.49660 acc=0.67969 acc_top1_avg=0.66406 acc_top5_avg=0.91172 lr=0.01000 gn=11.73381 time=61.30it/s +epoch=30 global_step=11800 loss=3.67938 loss_avg=3.43221 acc=0.64062 acc_top1_avg=0.67165 acc_top5_avg=0.91261 lr=0.01000 gn=8.54171 time=64.32it/s +epoch=30 global_step=11850 loss=3.53454 loss_avg=3.42714 acc=0.67188 acc_top1_avg=0.67155 acc_top5_avg=0.91296 lr=0.01000 gn=11.97789 time=61.49it/s +epoch=30 global_step=11900 loss=4.38405 loss_avg=3.47148 acc=0.57031 acc_top1_avg=0.66668 acc_top5_avg=0.91103 lr=0.01000 gn=9.26131 time=63.75it/s +epoch=30 global_step=11950 loss=3.16275 loss_avg=3.45208 acc=0.72656 acc_top1_avg=0.66871 acc_top5_avg=0.91197 lr=0.01000 gn=9.99451 time=56.69it/s +epoch=30 global_step=12000 loss=3.53035 loss_avg=3.43615 acc=0.67969 acc_top1_avg=0.67072 acc_top5_avg=0.91224 lr=0.01000 gn=8.95328 time=61.77it/s +epoch=30 global_step=12050 loss=3.06179 loss_avg=3.44156 acc=0.70312 acc_top1_avg=0.67056 acc_top5_avg=0.91174 lr=0.01000 gn=7.57654 time=58.15it/s +epoch=30 global_step=12100 loss=3.22827 loss_avg=3.43640 acc=0.70312 acc_top1_avg=0.67092 acc_top5_avg=0.91218 lr=0.01000 gn=9.27062 time=55.56it/s +====================Eval==================== +epoch=30 global_step=12121 loss=0.70468 test_loss_avg=1.09239 acc=0.82031 test_acc_avg=0.73880 test_acc_top5_avg=0.97630 time=238.90it/s +epoch=30 global_step=12121 loss=1.54913 test_loss_avg=1.12307 acc=0.56250 test_acc_avg=0.72745 test_acc_top5_avg=0.98022 time=882.45it/s +curr_acc 0.7275 +BEST_ACC 0.8026 +curr_acc_top5 0.9802 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=3.41048 loss_avg=3.40269 acc=0.67969 acc_top1_avg=0.67295 acc_top5_avg=0.91810 lr=0.01000 gn=9.32595 time=55.95it/s +epoch=31 global_step=12200 loss=3.17094 loss_avg=3.38120 acc=0.69531 acc_top1_avg=0.67623 acc_top5_avg=0.91367 lr=0.01000 gn=9.10308 time=58.48it/s +epoch=31 global_step=12250 loss=4.10791 loss_avg=3.44301 acc=0.59375 acc_top1_avg=0.67042 acc_top5_avg=0.91188 lr=0.01000 gn=9.91974 time=64.97it/s +epoch=31 global_step=12300 loss=3.47600 loss_avg=3.44468 acc=0.67969 acc_top1_avg=0.67026 acc_top5_avg=0.91175 lr=0.01000 gn=9.39612 time=55.54it/s +epoch=31 global_step=12350 loss=3.24336 loss_avg=3.44984 acc=0.71094 acc_top1_avg=0.67010 acc_top5_avg=0.91167 lr=0.01000 gn=10.17045 time=64.89it/s +epoch=31 global_step=12400 loss=3.18736 loss_avg=3.45142 acc=0.70312 acc_top1_avg=0.67008 acc_top5_avg=0.91109 lr=0.01000 gn=8.04237 time=62.53it/s +epoch=31 global_step=12450 loss=3.60194 loss_avg=3.43530 acc=0.64844 acc_top1_avg=0.67138 acc_top5_avg=0.91181 lr=0.01000 gn=12.65710 time=56.08it/s +epoch=31 global_step=12500 loss=3.24918 loss_avg=3.43651 acc=0.67969 acc_top1_avg=0.67109 acc_top5_avg=0.91128 lr=0.01000 gn=8.32494 time=62.77it/s +====================Eval==================== +epoch=31 global_step=12512 loss=0.93437 test_loss_avg=0.93437 acc=0.71094 test_acc_avg=0.71094 test_acc_top5_avg=0.97656 time=216.07it/s +epoch=31 global_step=12512 loss=0.16500 test_loss_avg=1.04803 acc=0.94531 test_acc_avg=0.71844 test_acc_top5_avg=0.96860 time=240.61it/s +epoch=31 global_step=12512 loss=0.04063 test_loss_avg=0.80078 acc=1.00000 test_acc_avg=0.78066 test_acc_top5_avg=0.97656 time=887.68it/s +curr_acc 0.7807 +BEST_ACC 0.8026 +curr_acc_top5 0.9766 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=3.51029 loss_avg=3.31707 acc=0.66406 acc_top1_avg=0.68400 acc_top5_avg=0.91303 lr=0.01000 gn=9.53747 time=62.94it/s +epoch=32 global_step=12600 loss=3.82577 loss_avg=3.32764 acc=0.60938 acc_top1_avg=0.68342 acc_top5_avg=0.91628 lr=0.01000 gn=9.56023 time=61.62it/s +epoch=32 global_step=12650 loss=3.65610 loss_avg=3.35384 acc=0.64844 acc_top1_avg=0.67974 acc_top5_avg=0.91655 lr=0.01000 gn=10.90815 time=64.38it/s +epoch=32 global_step=12700 loss=3.38006 loss_avg=3.34983 acc=0.66406 acc_top1_avg=0.68048 acc_top5_avg=0.91593 lr=0.01000 gn=9.56133 time=63.61it/s +epoch=32 global_step=12750 loss=2.57431 loss_avg=3.39285 acc=0.78125 acc_top1_avg=0.67614 acc_top5_avg=0.91478 lr=0.01000 gn=7.68313 time=54.21it/s +epoch=32 global_step=12800 loss=3.00365 loss_avg=3.40984 acc=0.71875 acc_top1_avg=0.67426 acc_top5_avg=0.91420 lr=0.01000 gn=9.95151 time=64.94it/s +epoch=32 global_step=12850 loss=3.80910 loss_avg=3.41476 acc=0.62500 acc_top1_avg=0.67375 acc_top5_avg=0.91443 lr=0.01000 gn=9.28407 time=63.18it/s +epoch=32 global_step=12900 loss=3.30439 loss_avg=3.42296 acc=0.69531 acc_top1_avg=0.67296 acc_top5_avg=0.91352 lr=0.01000 gn=10.73308 time=55.08it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.39519 test_loss_avg=0.83969 acc=0.61719 test_acc_avg=0.76882 test_acc_top5_avg=0.98295 time=258.73it/s +epoch=32 global_step=12903 loss=1.12009 test_loss_avg=0.92772 acc=0.67969 test_acc_avg=0.75532 test_acc_top5_avg=0.97179 time=251.26it/s +epoch=32 global_step=12903 loss=0.80554 test_loss_avg=0.92033 acc=0.75000 test_acc_avg=0.75623 test_acc_top5_avg=0.97389 time=891.27it/s +curr_acc 0.7562 +BEST_ACC 0.8026 +curr_acc_top5 0.9739 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=3.77459 loss_avg=3.52257 acc=0.63281 acc_top1_avg=0.66240 acc_top5_avg=0.90941 lr=0.01000 gn=11.54510 time=63.81it/s +epoch=33 global_step=13000 loss=3.61534 loss_avg=3.45881 acc=0.66406 acc_top1_avg=0.66962 acc_top5_avg=0.91310 lr=0.01000 gn=10.89909 time=60.23it/s +epoch=33 global_step=13050 loss=3.11242 loss_avg=3.42028 acc=0.71875 acc_top1_avg=0.67342 acc_top5_avg=0.91528 lr=0.01000 gn=9.89338 time=58.27it/s +epoch=33 global_step=13100 loss=2.97795 loss_avg=3.41718 acc=0.71094 acc_top1_avg=0.67402 acc_top5_avg=0.91406 lr=0.01000 gn=11.07973 time=54.28it/s +epoch=33 global_step=13150 loss=3.13876 loss_avg=3.41871 acc=0.70312 acc_top1_avg=0.67403 acc_top5_avg=0.91308 lr=0.01000 gn=9.67692 time=61.68it/s +epoch=33 global_step=13200 loss=3.42653 loss_avg=3.41936 acc=0.68750 acc_top1_avg=0.67361 acc_top5_avg=0.91283 lr=0.01000 gn=8.47066 time=55.63it/s +epoch=33 global_step=13250 loss=3.39045 loss_avg=3.42940 acc=0.67188 acc_top1_avg=0.67255 acc_top5_avg=0.91224 lr=0.01000 gn=11.09251 time=64.53it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.58662 test_loss_avg=1.45123 acc=0.85156 test_acc_avg=0.67605 test_acc_top5_avg=0.94150 time=245.94it/s +epoch=33 global_step=13294 loss=0.61261 test_loss_avg=1.00602 acc=0.81250 test_acc_avg=0.76286 test_acc_top5_avg=0.96499 time=856.85it/s +curr_acc 0.7629 +BEST_ACC 0.8026 +curr_acc_top5 0.9650 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=3.04059 loss_avg=3.25414 acc=0.75000 acc_top1_avg=0.69010 acc_top5_avg=0.90234 lr=0.01000 gn=10.64938 time=61.61it/s +epoch=34 global_step=13350 loss=3.86788 loss_avg=3.39733 acc=0.63281 acc_top1_avg=0.67815 acc_top5_avg=0.90960 lr=0.01000 gn=12.24173 time=61.72it/s +epoch=34 global_step=13400 loss=3.02282 loss_avg=3.36345 acc=0.71094 acc_top1_avg=0.68160 acc_top5_avg=0.91325 lr=0.01000 gn=8.47064 time=52.47it/s +epoch=34 global_step=13450 loss=3.33828 loss_avg=3.38791 acc=0.71875 acc_top1_avg=0.67783 acc_top5_avg=0.91396 lr=0.01000 gn=10.17357 time=56.29it/s +epoch=34 global_step=13500 loss=4.22681 loss_avg=3.37918 acc=0.59375 acc_top1_avg=0.67844 acc_top5_avg=0.91391 lr=0.01000 gn=12.35646 time=62.20it/s +epoch=34 global_step=13550 loss=3.24626 loss_avg=3.38235 acc=0.67969 acc_top1_avg=0.67795 acc_top5_avg=0.91406 lr=0.01000 gn=8.81302 time=55.70it/s +epoch=34 global_step=13600 loss=3.46861 loss_avg=3.41236 acc=0.67969 acc_top1_avg=0.67422 acc_top5_avg=0.91335 lr=0.01000 gn=11.14778 time=52.10it/s +epoch=34 global_step=13650 loss=3.44831 loss_avg=3.43007 acc=0.67969 acc_top1_avg=0.67185 acc_top5_avg=0.91303 lr=0.01000 gn=8.81877 time=63.46it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.50069 test_loss_avg=0.44063 acc=0.87500 test_acc_avg=0.86719 test_acc_top5_avg=0.99721 time=246.65it/s +epoch=34 global_step=13685 loss=0.29256 test_loss_avg=1.11253 acc=0.90625 test_acc_avg=0.73462 test_acc_top5_avg=0.96667 time=246.56it/s +epoch=34 global_step=13685 loss=0.24696 test_loss_avg=0.98226 acc=0.93750 test_acc_avg=0.76177 test_acc_top5_avg=0.97221 time=529.52it/s +curr_acc 0.7618 +BEST_ACC 0.8026 +curr_acc_top5 0.9722 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=3.73810 loss_avg=3.52566 acc=0.64844 acc_top1_avg=0.66406 acc_top5_avg=0.91198 lr=0.01000 gn=10.68056 time=58.34it/s +epoch=35 global_step=13750 loss=3.19720 loss_avg=3.35484 acc=0.70312 acc_top1_avg=0.68077 acc_top5_avg=0.91490 lr=0.01000 gn=8.76765 time=63.95it/s +epoch=35 global_step=13800 loss=3.49618 loss_avg=3.39151 acc=0.64844 acc_top1_avg=0.67683 acc_top5_avg=0.91325 lr=0.01000 gn=8.69616 time=64.94it/s +epoch=35 global_step=13850 loss=2.91305 loss_avg=3.40004 acc=0.74219 acc_top1_avg=0.67609 acc_top5_avg=0.91487 lr=0.01000 gn=9.98516 time=59.62it/s +epoch=35 global_step=13900 loss=3.18391 loss_avg=3.39530 acc=0.70312 acc_top1_avg=0.67569 acc_top5_avg=0.91424 lr=0.01000 gn=9.35418 time=61.73it/s +epoch=35 global_step=13950 loss=2.97661 loss_avg=3.39548 acc=0.72656 acc_top1_avg=0.67529 acc_top5_avg=0.91448 lr=0.01000 gn=7.52691 time=61.73it/s +epoch=35 global_step=14000 loss=3.45224 loss_avg=3.40696 acc=0.67188 acc_top1_avg=0.67431 acc_top5_avg=0.91287 lr=0.01000 gn=12.12351 time=54.48it/s +epoch=35 global_step=14050 loss=3.11603 loss_avg=3.41722 acc=0.71094 acc_top1_avg=0.67286 acc_top5_avg=0.91231 lr=0.01000 gn=10.33334 time=59.15it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.19292 test_loss_avg=1.18094 acc=0.94531 test_acc_avg=0.67879 test_acc_top5_avg=0.97366 time=235.50it/s +epoch=35 global_step=14076 loss=0.30291 test_loss_avg=0.92407 acc=0.87500 test_acc_avg=0.74763 test_acc_top5_avg=0.97656 time=862.32it/s +curr_acc 0.7476 +BEST_ACC 0.8026 +curr_acc_top5 0.9766 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=3.12859 loss_avg=3.38564 acc=0.71094 acc_top1_avg=0.67546 acc_top5_avg=0.90885 lr=0.01000 gn=10.02411 time=58.40it/s +epoch=36 global_step=14150 loss=3.65725 loss_avg=3.40112 acc=0.62500 acc_top1_avg=0.67367 acc_top5_avg=0.91375 lr=0.01000 gn=9.53663 time=63.46it/s +epoch=36 global_step=14200 loss=3.30394 loss_avg=3.36983 acc=0.69531 acc_top1_avg=0.67755 acc_top5_avg=0.91394 lr=0.01000 gn=10.89415 time=64.83it/s +epoch=36 global_step=14250 loss=3.39629 loss_avg=3.37681 acc=0.67188 acc_top1_avg=0.67645 acc_top5_avg=0.91406 lr=0.01000 gn=10.94740 time=63.78it/s +epoch=36 global_step=14300 loss=2.92094 loss_avg=3.39139 acc=0.71094 acc_top1_avg=0.67460 acc_top5_avg=0.91340 lr=0.01000 gn=7.91684 time=62.34it/s +epoch=36 global_step=14350 loss=3.10234 loss_avg=3.40544 acc=0.70312 acc_top1_avg=0.67310 acc_top5_avg=0.91269 lr=0.01000 gn=9.09482 time=58.30it/s +epoch=36 global_step=14400 loss=2.99841 loss_avg=3.40767 acc=0.70312 acc_top1_avg=0.67356 acc_top5_avg=0.91291 lr=0.01000 gn=10.00128 time=62.59it/s +epoch=36 global_step=14450 loss=3.79160 loss_avg=3.41160 acc=0.64062 acc_top1_avg=0.67319 acc_top5_avg=0.91210 lr=0.01000 gn=8.23861 time=59.29it/s +====================Eval==================== +epoch=36 global_step=14467 loss=1.30434 test_loss_avg=1.34346 acc=0.65625 test_acc_avg=0.63021 test_acc_top5_avg=0.95573 time=242.98it/s +epoch=36 global_step=14467 loss=1.49430 test_loss_avg=0.86668 acc=0.69531 test_acc_avg=0.77483 test_acc_top5_avg=0.98061 time=241.11it/s +epoch=36 global_step=14467 loss=0.92629 test_loss_avg=0.80341 acc=0.87500 test_acc_avg=0.79114 test_acc_top5_avg=0.98210 time=524.48it/s +curr_acc 0.7911 +BEST_ACC 0.8026 +curr_acc_top5 0.9821 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=3.88051 loss_avg=3.36375 acc=0.61719 acc_top1_avg=0.67708 acc_top5_avg=0.91170 lr=0.01000 gn=7.42029 time=63.59it/s +epoch=37 global_step=14550 loss=3.49176 loss_avg=3.39854 acc=0.65625 acc_top1_avg=0.67470 acc_top5_avg=0.91218 lr=0.01000 gn=9.68675 time=62.06it/s +epoch=37 global_step=14600 loss=3.18622 loss_avg=3.39727 acc=0.71875 acc_top1_avg=0.67540 acc_top5_avg=0.91224 lr=0.01000 gn=10.16558 time=54.87it/s +epoch=37 global_step=14650 loss=4.28646 loss_avg=3.40435 acc=0.59375 acc_top1_avg=0.67452 acc_top5_avg=0.91321 lr=0.01000 gn=11.07396 time=50.77it/s +epoch=37 global_step=14700 loss=3.23575 loss_avg=3.40662 acc=0.68750 acc_top1_avg=0.67446 acc_top5_avg=0.91326 lr=0.01000 gn=9.22694 time=55.35it/s +epoch=37 global_step=14750 loss=3.63467 loss_avg=3.41170 acc=0.66406 acc_top1_avg=0.67361 acc_top5_avg=0.91293 lr=0.01000 gn=11.16809 time=55.89it/s +epoch=37 global_step=14800 loss=3.73324 loss_avg=3.41671 acc=0.63281 acc_top1_avg=0.67333 acc_top5_avg=0.91282 lr=0.01000 gn=10.92813 time=63.93it/s +epoch=37 global_step=14850 loss=2.90401 loss_avg=3.41372 acc=0.73438 acc_top1_avg=0.67357 acc_top5_avg=0.91359 lr=0.01000 gn=9.50617 time=58.86it/s +====================Eval==================== +epoch=37 global_step=14858 loss=1.12494 test_loss_avg=0.80099 acc=0.77344 test_acc_avg=0.79167 test_acc_top5_avg=0.98727 time=244.65it/s +epoch=37 global_step=14858 loss=0.65220 test_loss_avg=0.83434 acc=0.82031 test_acc_avg=0.78947 test_acc_top5_avg=0.98417 time=246.06it/s +epoch=37 global_step=14858 loss=0.16697 test_loss_avg=0.82199 acc=0.93750 test_acc_avg=0.79213 test_acc_top5_avg=0.98447 time=498.73it/s +curr_acc 0.7921 +BEST_ACC 0.8026 +curr_acc_top5 0.9845 +BEST_ACC_top5 0.9843 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=3.25194 loss_avg=3.33054 acc=0.68750 acc_top1_avg=0.68341 acc_top5_avg=0.90941 lr=0.01000 gn=8.42688 time=59.66it/s +epoch=38 global_step=14950 loss=3.33394 loss_avg=3.34933 acc=0.67969 acc_top1_avg=0.68054 acc_top5_avg=0.91338 lr=0.01000 gn=10.89253 time=64.85it/s +epoch=38 global_step=15000 loss=3.33511 loss_avg=3.35553 acc=0.67969 acc_top1_avg=0.68013 acc_top5_avg=0.91395 lr=0.01000 gn=9.11190 time=58.61it/s +epoch=38 global_step=15050 loss=3.10901 loss_avg=3.37400 acc=0.71875 acc_top1_avg=0.67818 acc_top5_avg=0.91463 lr=0.01000 gn=9.09810 time=56.03it/s +epoch=38 global_step=15100 loss=3.70165 loss_avg=3.38288 acc=0.64062 acc_top1_avg=0.67656 acc_top5_avg=0.91280 lr=0.01000 gn=8.60990 time=58.26it/s +epoch=38 global_step=15150 loss=3.36015 loss_avg=3.38957 acc=0.68750 acc_top1_avg=0.67597 acc_top5_avg=0.91227 lr=0.01000 gn=9.20334 time=63.22it/s +epoch=38 global_step=15200 loss=2.97731 loss_avg=3.39962 acc=0.73438 acc_top1_avg=0.67473 acc_top5_avg=0.91253 lr=0.01000 gn=11.03893 time=54.18it/s +====================Eval==================== +epoch=38 global_step=15249 loss=1.57941 test_loss_avg=1.06035 acc=0.58594 test_acc_avg=0.73112 test_acc_top5_avg=0.97282 time=245.60it/s +epoch=38 global_step=15249 loss=0.55390 test_loss_avg=0.87942 acc=0.81250 test_acc_avg=0.76958 test_acc_top5_avg=0.97854 time=879.12it/s +curr_acc 0.7696 +BEST_ACC 0.8026 +curr_acc_top5 0.9785 +BEST_ACC_top5 0.9845 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=3.33303 loss_avg=3.33303 acc=0.69531 acc_top1_avg=0.69531 acc_top5_avg=0.91406 lr=0.01000 gn=9.11118 time=50.24it/s +epoch=39 global_step=15300 loss=2.77244 loss_avg=3.38263 acc=0.76562 acc_top1_avg=0.67831 acc_top5_avg=0.91575 lr=0.01000 gn=10.95568 time=61.84it/s +epoch=39 global_step=15350 loss=2.89143 loss_avg=3.34497 acc=0.71875 acc_top1_avg=0.68178 acc_top5_avg=0.91762 lr=0.01000 gn=9.78392 time=62.00it/s +epoch=39 global_step=15400 loss=3.35610 loss_avg=3.36490 acc=0.67969 acc_top1_avg=0.67927 acc_top5_avg=0.91608 lr=0.01000 gn=9.18236 time=64.71it/s +epoch=39 global_step=15450 loss=3.16352 loss_avg=3.37052 acc=0.70312 acc_top1_avg=0.67852 acc_top5_avg=0.91519 lr=0.01000 gn=8.93544 time=59.62it/s +epoch=39 global_step=15500 loss=3.36776 loss_avg=3.39403 acc=0.66406 acc_top1_avg=0.67567 acc_top5_avg=0.91381 lr=0.01000 gn=7.47935 time=53.08it/s +epoch=39 global_step=15550 loss=3.46593 loss_avg=3.40920 acc=0.66406 acc_top1_avg=0.67447 acc_top5_avg=0.91357 lr=0.01000 gn=9.72220 time=58.24it/s +epoch=39 global_step=15600 loss=3.72984 loss_avg=3.40302 acc=0.64062 acc_top1_avg=0.67517 acc_top5_avg=0.91326 lr=0.01000 gn=9.70285 time=64.73it/s +====================Eval==================== +epoch=39 global_step=15640 loss=0.51379 test_loss_avg=0.54151 acc=0.89844 test_acc_avg=0.85280 test_acc_top5_avg=0.99301 time=261.59it/s +epoch=39 global_step=15640 loss=0.29395 test_loss_avg=0.94013 acc=0.92969 test_acc_avg=0.76030 test_acc_top5_avg=0.98313 time=261.41it/s +epoch=39 global_step=15640 loss=0.79785 test_loss_avg=0.95746 acc=0.75000 test_acc_avg=0.75465 test_acc_top5_avg=0.98022 time=881.90it/s +curr_acc 0.7546 +BEST_ACC 0.8026 +curr_acc_top5 0.9802 +BEST_ACC_top5 0.9845 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=3.05987 loss_avg=3.37125 acc=0.71094 acc_top1_avg=0.67656 acc_top5_avg=0.91484 lr=0.00100 gn=9.07334 time=61.57it/s +epoch=40 global_step=15700 loss=3.40215 loss_avg=3.26564 acc=0.67969 acc_top1_avg=0.68854 acc_top5_avg=0.91497 lr=0.00100 gn=9.75367 time=61.38it/s +epoch=40 global_step=15750 loss=2.55847 loss_avg=3.17563 acc=0.76562 acc_top1_avg=0.69780 acc_top5_avg=0.91584 lr=0.00100 gn=10.11455 time=56.41it/s +epoch=40 global_step=15800 loss=3.26945 loss_avg=3.10108 acc=0.67188 acc_top1_avg=0.70566 acc_top5_avg=0.91621 lr=0.00100 gn=8.44476 time=62.71it/s +epoch=40 global_step=15850 loss=2.54602 loss_avg=3.05826 acc=0.78125 acc_top1_avg=0.71027 acc_top5_avg=0.91789 lr=0.00100 gn=8.59517 time=64.96it/s +epoch=40 global_step=15900 loss=2.58530 loss_avg=3.03200 acc=0.75000 acc_top1_avg=0.71265 acc_top5_avg=0.91899 lr=0.00100 gn=9.48447 time=57.31it/s +epoch=40 global_step=15950 loss=2.81177 loss_avg=3.02163 acc=0.74219 acc_top1_avg=0.71328 acc_top5_avg=0.91983 lr=0.00100 gn=8.67063 time=58.59it/s +epoch=40 global_step=16000 loss=2.42133 loss_avg=3.01200 acc=0.77344 acc_top1_avg=0.71434 acc_top5_avg=0.91960 lr=0.00100 gn=7.76560 time=62.19it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.70922 test_loss_avg=0.53877 acc=0.79688 test_acc_avg=0.84805 test_acc_top5_avg=0.99062 time=242.78it/s +epoch=40 global_step=16031 loss=0.14759 test_loss_avg=0.44291 acc=0.87500 test_acc_avg=0.87530 test_acc_top5_avg=0.99189 time=874.00it/s +curr_acc 0.8753 +BEST_ACC 0.8026 +curr_acc_top5 0.9919 +BEST_ACC_top5 0.9845 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=3.16137 loss_avg=2.83511 acc=0.68750 acc_top1_avg=0.73355 acc_top5_avg=0.92229 lr=0.00100 gn=7.02586 time=57.78it/s +epoch=41 global_step=16100 loss=2.91802 loss_avg=2.82736 acc=0.72656 acc_top1_avg=0.73392 acc_top5_avg=0.92052 lr=0.00100 gn=11.04788 time=62.85it/s +epoch=41 global_step=16150 loss=1.97783 loss_avg=2.82938 acc=0.82031 acc_top1_avg=0.73424 acc_top5_avg=0.92293 lr=0.00100 gn=6.79598 time=48.02it/s +epoch=41 global_step=16200 loss=2.65920 loss_avg=2.83898 acc=0.75781 acc_top1_avg=0.73253 acc_top5_avg=0.92386 lr=0.00100 gn=7.71265 time=64.16it/s +epoch=41 global_step=16250 loss=2.51149 loss_avg=2.82377 acc=0.78125 acc_top1_avg=0.73430 acc_top5_avg=0.92327 lr=0.00100 gn=8.08258 time=59.15it/s +epoch=41 global_step=16300 loss=2.71735 loss_avg=2.83180 acc=0.74219 acc_top1_avg=0.73295 acc_top5_avg=0.92179 lr=0.00100 gn=8.12726 time=53.15it/s +epoch=41 global_step=16350 loss=2.71006 loss_avg=2.83817 acc=0.74219 acc_top1_avg=0.73197 acc_top5_avg=0.92165 lr=0.00100 gn=9.03491 time=58.01it/s +epoch=41 global_step=16400 loss=3.73404 loss_avg=2.84532 acc=0.63281 acc_top1_avg=0.73131 acc_top5_avg=0.92164 lr=0.00100 gn=8.65390 time=59.17it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.12900 test_loss_avg=0.37113 acc=0.95312 test_acc_avg=0.88210 test_acc_top5_avg=0.99716 time=242.85it/s +epoch=41 global_step=16422 loss=0.13200 test_loss_avg=0.46037 acc=0.95312 test_acc_avg=0.86911 test_acc_top5_avg=0.99168 time=248.58it/s +epoch=41 global_step=16422 loss=0.12437 test_loss_avg=0.40085 acc=0.93750 test_acc_avg=0.88469 test_acc_top5_avg=0.99239 time=810.65it/s +curr_acc 0.8847 +BEST_ACC 0.8753 +curr_acc_top5 0.9924 +BEST_ACC_top5 0.9919 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=2.98140 loss_avg=2.78954 acc=0.72656 acc_top1_avg=0.73828 acc_top5_avg=0.92215 lr=0.00100 gn=10.53718 time=65.40it/s +epoch=42 global_step=16500 loss=2.56641 loss_avg=2.73461 acc=0.75781 acc_top1_avg=0.74289 acc_top5_avg=0.92658 lr=0.00100 gn=7.69579 time=53.09it/s +epoch=42 global_step=16550 loss=3.00284 loss_avg=2.76779 acc=0.71094 acc_top1_avg=0.73871 acc_top5_avg=0.92456 lr=0.00100 gn=9.98463 time=56.78it/s +epoch=42 global_step=16600 loss=3.06871 loss_avg=2.77690 acc=0.71094 acc_top1_avg=0.73723 acc_top5_avg=0.92464 lr=0.00100 gn=8.41564 time=53.50it/s +epoch=42 global_step=16650 loss=2.57222 loss_avg=2.77215 acc=0.76562 acc_top1_avg=0.73749 acc_top5_avg=0.92376 lr=0.00100 gn=10.46379 time=57.43it/s +epoch=42 global_step=16700 loss=3.19474 loss_avg=2.78150 acc=0.68750 acc_top1_avg=0.73645 acc_top5_avg=0.92398 lr=0.00100 gn=9.68067 time=42.64it/s +epoch=42 global_step=16750 loss=3.32502 loss_avg=2.77722 acc=0.67188 acc_top1_avg=0.73719 acc_top5_avg=0.92399 lr=0.00100 gn=8.47131 time=58.28it/s +epoch=42 global_step=16800 loss=2.55029 loss_avg=2.77784 acc=0.75000 acc_top1_avg=0.73723 acc_top5_avg=0.92328 lr=0.00100 gn=10.04028 time=54.91it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.57991 test_loss_avg=0.56588 acc=0.84375 test_acc_avg=0.84326 test_acc_top5_avg=0.98975 time=259.90it/s +epoch=42 global_step=16813 loss=0.40743 test_loss_avg=0.39633 acc=0.87500 test_acc_avg=0.88608 test_acc_top5_avg=0.99268 time=563.52it/s +curr_acc 0.8861 +BEST_ACC 0.8847 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9924 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=2.67752 loss_avg=2.71029 acc=0.73438 acc_top1_avg=0.74261 acc_top5_avg=0.92863 lr=0.00100 gn=9.96982 time=53.37it/s +epoch=43 global_step=16900 loss=2.96719 loss_avg=2.72388 acc=0.71875 acc_top1_avg=0.74327 acc_top5_avg=0.92385 lr=0.00100 gn=9.14914 time=54.53it/s +epoch=43 global_step=16950 loss=2.26657 loss_avg=2.72398 acc=0.79688 acc_top1_avg=0.74293 acc_top5_avg=0.92341 lr=0.00100 gn=8.36808 time=54.98it/s +epoch=43 global_step=17000 loss=3.05033 loss_avg=2.73728 acc=0.70312 acc_top1_avg=0.74152 acc_top5_avg=0.92275 lr=0.00100 gn=9.32173 time=52.46it/s +epoch=43 global_step=17050 loss=2.96896 loss_avg=2.75523 acc=0.72656 acc_top1_avg=0.73958 acc_top5_avg=0.92211 lr=0.00100 gn=11.38461 time=64.74it/s +epoch=43 global_step=17100 loss=2.33633 loss_avg=2.74177 acc=0.78125 acc_top1_avg=0.74113 acc_top5_avg=0.92193 lr=0.00100 gn=10.48690 time=57.22it/s +epoch=43 global_step=17150 loss=2.29164 loss_avg=2.73638 acc=0.79688 acc_top1_avg=0.74182 acc_top5_avg=0.92278 lr=0.00100 gn=9.04130 time=64.67it/s +epoch=43 global_step=17200 loss=2.59485 loss_avg=2.72540 acc=0.75000 acc_top1_avg=0.74279 acc_top5_avg=0.92347 lr=0.00100 gn=8.59719 time=55.75it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.49308 test_loss_avg=0.52116 acc=0.83594 test_acc_avg=0.84896 test_acc_top5_avg=0.99219 time=242.47it/s +epoch=43 global_step=17204 loss=0.18113 test_loss_avg=0.49239 acc=0.95312 test_acc_avg=0.86173 test_acc_top5_avg=0.99278 time=239.62it/s +epoch=43 global_step=17204 loss=0.18318 test_loss_avg=0.39757 acc=0.87500 test_acc_avg=0.88578 test_acc_top5_avg=0.99397 time=542.39it/s +curr_acc 0.8858 +BEST_ACC 0.8861 +curr_acc_top5 0.9940 +BEST_ACC_top5 0.9927 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=2.30962 loss_avg=2.72259 acc=0.78125 acc_top1_avg=0.74202 acc_top5_avg=0.91831 lr=0.00100 gn=7.96569 time=61.25it/s +epoch=44 global_step=17300 loss=2.37058 loss_avg=2.72543 acc=0.77344 acc_top1_avg=0.73958 acc_top5_avg=0.91927 lr=0.00100 gn=7.39226 time=55.71it/s +epoch=44 global_step=17350 loss=3.02586 loss_avg=2.70335 acc=0.72656 acc_top1_avg=0.74219 acc_top5_avg=0.92161 lr=0.00100 gn=12.23132 time=64.94it/s +epoch=44 global_step=17400 loss=1.99206 loss_avg=2.69293 acc=0.82812 acc_top1_avg=0.74414 acc_top5_avg=0.92287 lr=0.00100 gn=8.37920 time=64.11it/s +epoch=44 global_step=17450 loss=2.48210 loss_avg=2.68917 acc=0.78125 acc_top1_avg=0.74508 acc_top5_avg=0.92369 lr=0.00100 gn=11.36393 time=64.85it/s +epoch=44 global_step=17500 loss=2.32244 loss_avg=2.68813 acc=0.79688 acc_top1_avg=0.74567 acc_top5_avg=0.92425 lr=0.00100 gn=11.26011 time=53.46it/s +epoch=44 global_step=17550 loss=3.17213 loss_avg=2.67570 acc=0.68750 acc_top1_avg=0.74706 acc_top5_avg=0.92524 lr=0.00100 gn=11.43344 time=55.95it/s +====================Eval==================== +epoch=44 global_step=17595 loss=0.81641 test_loss_avg=0.47418 acc=0.80469 test_acc_avg=0.86686 test_acc_top5_avg=0.99316 time=245.17it/s +epoch=44 global_step=17595 loss=0.27645 test_loss_avg=0.40196 acc=0.90625 test_acc_avg=0.88746 test_acc_top5_avg=0.99345 time=249.35it/s +epoch=44 global_step=17595 loss=0.20316 test_loss_avg=0.39385 acc=0.87500 test_acc_avg=0.88825 test_acc_top5_avg=0.99367 time=506.31it/s +curr_acc 0.8883 +BEST_ACC 0.8861 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=2.81671 loss_avg=2.32701 acc=0.74219 acc_top1_avg=0.78750 acc_top5_avg=0.92812 lr=0.00100 gn=10.79641 time=58.87it/s +epoch=45 global_step=17650 loss=2.29710 loss_avg=2.67834 acc=0.76562 acc_top1_avg=0.74744 acc_top5_avg=0.92116 lr=0.00100 gn=11.36904 time=61.93it/s +epoch=45 global_step=17700 loss=2.67609 loss_avg=2.62802 acc=0.74219 acc_top1_avg=0.75357 acc_top5_avg=0.92478 lr=0.00100 gn=9.59808 time=63.94it/s +epoch=45 global_step=17750 loss=2.72727 loss_avg=2.63574 acc=0.75000 acc_top1_avg=0.75272 acc_top5_avg=0.92545 lr=0.00100 gn=9.00846 time=54.19it/s +epoch=45 global_step=17800 loss=2.49391 loss_avg=2.65328 acc=0.76562 acc_top1_avg=0.75046 acc_top5_avg=0.92504 lr=0.00100 gn=8.84355 time=63.18it/s +epoch=45 global_step=17850 loss=3.00066 loss_avg=2.65059 acc=0.70312 acc_top1_avg=0.75070 acc_top5_avg=0.92528 lr=0.00100 gn=10.36722 time=64.65it/s +epoch=45 global_step=17900 loss=2.95057 loss_avg=2.64774 acc=0.72656 acc_top1_avg=0.75087 acc_top5_avg=0.92533 lr=0.00100 gn=13.31468 time=54.17it/s +epoch=45 global_step=17950 loss=2.47237 loss_avg=2.65561 acc=0.75781 acc_top1_avg=0.74969 acc_top5_avg=0.92507 lr=0.00100 gn=9.05771 time=63.27it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.87954 test_loss_avg=0.50269 acc=0.74219 test_acc_avg=0.86007 test_acc_top5_avg=0.99253 time=248.86it/s +epoch=45 global_step=17986 loss=0.25931 test_loss_avg=0.40747 acc=0.87500 test_acc_avg=0.88509 test_acc_top5_avg=0.99298 time=898.14it/s +curr_acc 0.8851 +BEST_ACC 0.8883 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=2.42973 loss_avg=2.65771 acc=0.76562 acc_top1_avg=0.75000 acc_top5_avg=0.92188 lr=0.00100 gn=8.27110 time=52.72it/s +epoch=46 global_step=18050 loss=2.66964 loss_avg=2.59987 acc=0.74219 acc_top1_avg=0.75476 acc_top5_avg=0.92493 lr=0.00100 gn=11.02824 time=65.09it/s +epoch=46 global_step=18100 loss=2.71011 loss_avg=2.63297 acc=0.75000 acc_top1_avg=0.75130 acc_top5_avg=0.92407 lr=0.00100 gn=10.55056 time=65.05it/s +epoch=46 global_step=18150 loss=3.07463 loss_avg=2.61901 acc=0.70312 acc_top1_avg=0.75324 acc_top5_avg=0.92492 lr=0.00100 gn=10.31158 time=61.22it/s +epoch=46 global_step=18200 loss=2.53835 loss_avg=2.60655 acc=0.77344 acc_top1_avg=0.75489 acc_top5_avg=0.92483 lr=0.00100 gn=12.11442 time=57.20it/s +epoch=46 global_step=18250 loss=2.76164 loss_avg=2.63768 acc=0.72656 acc_top1_avg=0.75172 acc_top5_avg=0.92341 lr=0.00100 gn=11.01328 time=58.05it/s +epoch=46 global_step=18300 loss=2.14855 loss_avg=2.63490 acc=0.79688 acc_top1_avg=0.75152 acc_top5_avg=0.92369 lr=0.00100 gn=9.41281 time=61.77it/s +epoch=46 global_step=18350 loss=2.99035 loss_avg=2.62364 acc=0.70312 acc_top1_avg=0.75251 acc_top5_avg=0.92445 lr=0.00100 gn=9.04143 time=59.51it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.49419 test_loss_avg=0.33754 acc=0.85938 test_acc_avg=0.89746 test_acc_top5_avg=0.99756 time=253.60it/s +epoch=46 global_step=18377 loss=0.14431 test_loss_avg=0.41311 acc=0.96875 test_acc_avg=0.88080 test_acc_top5_avg=0.99455 time=246.84it/s +epoch=46 global_step=18377 loss=0.18665 test_loss_avg=0.38619 acc=0.87500 test_acc_avg=0.88627 test_acc_top5_avg=0.99496 time=885.25it/s +curr_acc 0.8863 +BEST_ACC 0.8883 +curr_acc_top5 0.9950 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=3.14312 loss_avg=2.61008 acc=0.70312 acc_top1_avg=0.75543 acc_top5_avg=0.92697 lr=0.00100 gn=10.82816 time=63.84it/s +epoch=47 global_step=18450 loss=2.47292 loss_avg=2.60671 acc=0.75781 acc_top1_avg=0.75535 acc_top5_avg=0.92145 lr=0.00100 gn=7.69878 time=57.21it/s +epoch=47 global_step=18500 loss=2.81527 loss_avg=2.60100 acc=0.73438 acc_top1_avg=0.75426 acc_top5_avg=0.92130 lr=0.00100 gn=12.09871 time=59.31it/s +epoch=47 global_step=18550 loss=3.19707 loss_avg=2.61028 acc=0.71875 acc_top1_avg=0.75370 acc_top5_avg=0.92142 lr=0.00100 gn=13.66370 time=55.40it/s +epoch=47 global_step=18600 loss=2.41007 loss_avg=2.60256 acc=0.78906 acc_top1_avg=0.75445 acc_top5_avg=0.92272 lr=0.00100 gn=11.28935 time=59.90it/s +epoch=47 global_step=18650 loss=3.35025 loss_avg=2.60878 acc=0.67969 acc_top1_avg=0.75409 acc_top5_avg=0.92394 lr=0.00100 gn=9.90138 time=52.37it/s +epoch=47 global_step=18700 loss=2.74755 loss_avg=2.60096 acc=0.73438 acc_top1_avg=0.75501 acc_top5_avg=0.92444 lr=0.00100 gn=12.55252 time=64.56it/s +epoch=47 global_step=18750 loss=2.91755 loss_avg=2.60858 acc=0.70312 acc_top1_avg=0.75421 acc_top5_avg=0.92495 lr=0.00100 gn=9.56899 time=50.83it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.21769 test_loss_avg=0.40320 acc=0.90625 test_acc_avg=0.88112 test_acc_top5_avg=0.99345 time=244.65it/s +epoch=47 global_step=18768 loss=0.32180 test_loss_avg=0.37595 acc=0.87500 test_acc_avg=0.88914 test_acc_top5_avg=0.99357 time=879.68it/s +curr_acc 0.8891 +BEST_ACC 0.8883 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9950 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=2.80354 loss_avg=2.46092 acc=0.73438 acc_top1_avg=0.76929 acc_top5_avg=0.93164 lr=0.00100 gn=11.93211 time=64.13it/s +epoch=48 global_step=18850 loss=2.64666 loss_avg=2.53619 acc=0.75000 acc_top1_avg=0.76200 acc_top5_avg=0.92950 lr=0.00100 gn=9.18561 time=55.88it/s +epoch=48 global_step=18900 loss=2.49903 loss_avg=2.58163 acc=0.76562 acc_top1_avg=0.75710 acc_top5_avg=0.92549 lr=0.00100 gn=10.78795 time=61.84it/s +epoch=48 global_step=18950 loss=2.50535 loss_avg=2.56866 acc=0.77344 acc_top1_avg=0.75884 acc_top5_avg=0.92548 lr=0.00100 gn=13.08517 time=57.87it/s +epoch=48 global_step=19000 loss=1.83908 loss_avg=2.56815 acc=0.83594 acc_top1_avg=0.75916 acc_top5_avg=0.92585 lr=0.00100 gn=9.64547 time=57.17it/s +epoch=48 global_step=19050 loss=3.55230 loss_avg=2.57401 acc=0.64062 acc_top1_avg=0.75798 acc_top5_avg=0.92553 lr=0.00100 gn=12.10476 time=55.78it/s +epoch=48 global_step=19100 loss=2.93425 loss_avg=2.56180 acc=0.72656 acc_top1_avg=0.75929 acc_top5_avg=0.92578 lr=0.00100 gn=14.19093 time=54.98it/s +epoch=48 global_step=19150 loss=2.73461 loss_avg=2.56993 acc=0.75781 acc_top1_avg=0.75845 acc_top5_avg=0.92580 lr=0.00100 gn=12.80550 time=57.30it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.38132 test_loss_avg=0.58379 acc=0.85156 test_acc_avg=0.82324 test_acc_top5_avg=0.99023 time=238.96it/s +epoch=48 global_step=19159 loss=0.25965 test_loss_avg=0.46106 acc=0.92969 test_acc_avg=0.86773 test_acc_top5_avg=0.99205 time=233.60it/s +epoch=48 global_step=19159 loss=0.17198 test_loss_avg=0.39067 acc=0.93750 test_acc_avg=0.88588 test_acc_top5_avg=0.99347 time=882.08it/s +curr_acc 0.8859 +BEST_ACC 0.8891 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9950 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=2.35534 loss_avg=2.54505 acc=0.77344 acc_top1_avg=0.76220 acc_top5_avg=0.92912 lr=0.00100 gn=7.86878 time=51.42it/s +epoch=49 global_step=19250 loss=2.43988 loss_avg=2.53402 acc=0.75781 acc_top1_avg=0.76150 acc_top5_avg=0.92771 lr=0.00100 gn=12.24840 time=55.92it/s +epoch=49 global_step=19300 loss=2.95541 loss_avg=2.55091 acc=0.70312 acc_top1_avg=0.75942 acc_top5_avg=0.92620 lr=0.00100 gn=13.07455 time=62.68it/s +epoch=49 global_step=19350 loss=2.78014 loss_avg=2.55302 acc=0.74219 acc_top1_avg=0.75961 acc_top5_avg=0.92756 lr=0.00100 gn=12.21403 time=56.43it/s +epoch=49 global_step=19400 loss=2.27181 loss_avg=2.53894 acc=0.79688 acc_top1_avg=0.76112 acc_top5_avg=0.92823 lr=0.00100 gn=13.00419 time=56.87it/s +epoch=49 global_step=19450 loss=2.87825 loss_avg=2.55049 acc=0.71875 acc_top1_avg=0.75999 acc_top5_avg=0.92824 lr=0.00100 gn=14.29125 time=61.22it/s +epoch=49 global_step=19500 loss=2.77890 loss_avg=2.54759 acc=0.74219 acc_top1_avg=0.76045 acc_top5_avg=0.92788 lr=0.00100 gn=9.47154 time=61.80it/s +epoch=49 global_step=19550 loss=2.39495 loss_avg=2.54900 acc=0.78750 acc_top1_avg=0.76025 acc_top5_avg=0.92759 lr=0.00100 gn=17.05172 time=81.19it/s +====================Eval==================== +epoch=49 global_step=19550 loss=0.63163 test_loss_avg=0.47750 acc=0.79688 test_acc_avg=0.86530 test_acc_top5_avg=0.99273 time=247.17it/s +epoch=49 global_step=19550 loss=0.17845 test_loss_avg=0.37529 acc=0.93750 test_acc_avg=0.89339 test_acc_top5_avg=0.99417 time=640.25it/s +epoch=49 global_step=19550 loss=0.17845 test_loss_avg=0.37529 acc=0.93750 test_acc_avg=0.89339 test_acc_top5_avg=0.99417 time=640.25it/s +curr_acc 0.8934 +BEST_ACC 0.8891 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9950 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=2.73792 loss_avg=2.60027 acc=0.75000 acc_top1_avg=0.75687 acc_top5_avg=0.91875 lr=0.00100 gn=12.36537 time=64.41it/s +epoch=50 global_step=19650 loss=2.83094 loss_avg=2.58816 acc=0.71875 acc_top1_avg=0.75727 acc_top5_avg=0.92281 lr=0.00100 gn=8.90561 time=51.02it/s +epoch=50 global_step=19700 loss=2.19100 loss_avg=2.58602 acc=0.80469 acc_top1_avg=0.75766 acc_top5_avg=0.92453 lr=0.00100 gn=10.92410 time=52.73it/s +epoch=50 global_step=19750 loss=2.65607 loss_avg=2.56613 acc=0.75781 acc_top1_avg=0.75918 acc_top5_avg=0.92574 lr=0.00100 gn=14.96908 time=57.82it/s +epoch=50 global_step=19800 loss=2.77830 loss_avg=2.57552 acc=0.72656 acc_top1_avg=0.75828 acc_top5_avg=0.92572 lr=0.00100 gn=10.89143 time=56.50it/s +epoch=50 global_step=19850 loss=2.52617 loss_avg=2.55498 acc=0.76562 acc_top1_avg=0.76044 acc_top5_avg=0.92591 lr=0.00100 gn=14.01671 time=62.48it/s +epoch=50 global_step=19900 loss=2.49739 loss_avg=2.54022 acc=0.75781 acc_top1_avg=0.76187 acc_top5_avg=0.92656 lr=0.00100 gn=8.64829 time=64.88it/s +====================Eval==================== +epoch=50 global_step=19941 loss=0.11722 test_loss_avg=0.49050 acc=0.97656 test_acc_avg=0.85813 test_acc_top5_avg=0.99344 time=256.77it/s +epoch=50 global_step=19941 loss=0.36014 test_loss_avg=0.40087 acc=0.87500 test_acc_avg=0.88370 test_acc_top5_avg=0.99387 time=874.91it/s +curr_acc 0.8837 +BEST_ACC 0.8934 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9950 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=2.90173 loss_avg=2.73336 acc=0.70312 acc_top1_avg=0.74392 acc_top5_avg=0.92882 lr=0.00100 gn=13.64117 time=57.03it/s +epoch=51 global_step=20000 loss=2.02106 loss_avg=2.47181 acc=0.81250 acc_top1_avg=0.77039 acc_top5_avg=0.93128 lr=0.00100 gn=11.90239 time=64.91it/s +epoch=51 global_step=20050 loss=1.93560 loss_avg=2.51431 acc=0.82031 acc_top1_avg=0.76469 acc_top5_avg=0.93062 lr=0.00100 gn=15.43873 time=62.25it/s +epoch=51 global_step=20100 loss=1.76033 loss_avg=2.51524 acc=0.85156 acc_top1_avg=0.76454 acc_top5_avg=0.92875 lr=0.00100 gn=11.07572 time=50.64it/s +epoch=51 global_step=20150 loss=2.12626 loss_avg=2.53479 acc=0.83594 acc_top1_avg=0.76249 acc_top5_avg=0.92703 lr=0.00100 gn=17.36483 time=55.50it/s +epoch=51 global_step=20200 loss=2.08280 loss_avg=2.51122 acc=0.80469 acc_top1_avg=0.76511 acc_top5_avg=0.92776 lr=0.00100 gn=9.00821 time=40.21it/s +epoch=51 global_step=20250 loss=2.37612 loss_avg=2.52384 acc=0.78125 acc_top1_avg=0.76388 acc_top5_avg=0.92792 lr=0.00100 gn=15.19287 time=53.04it/s +epoch=51 global_step=20300 loss=2.34219 loss_avg=2.51936 acc=0.79688 acc_top1_avg=0.76428 acc_top5_avg=0.92803 lr=0.00100 gn=15.14603 time=65.16it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.66382 test_loss_avg=0.43243 acc=0.80469 test_acc_avg=0.87686 test_acc_top5_avg=0.99479 time=253.19it/s +epoch=51 global_step=20332 loss=0.20008 test_loss_avg=0.39823 acc=0.94531 test_acc_avg=0.88875 test_acc_top5_avg=0.99505 time=236.82it/s +epoch=51 global_step=20332 loss=0.05650 test_loss_avg=0.38275 acc=0.93750 test_acc_avg=0.89161 test_acc_top5_avg=0.99506 time=873.27it/s +curr_acc 0.8916 +BEST_ACC 0.8934 +curr_acc_top5 0.9951 +BEST_ACC_top5 0.9950 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=2.79423 loss_avg=2.53327 acc=0.73438 acc_top1_avg=0.76042 acc_top5_avg=0.92448 lr=0.00100 gn=14.20512 time=64.65it/s +epoch=52 global_step=20400 loss=2.31321 loss_avg=2.53001 acc=0.78125 acc_top1_avg=0.76126 acc_top5_avg=0.92601 lr=0.00100 gn=12.21007 time=62.49it/s +epoch=52 global_step=20450 loss=2.40197 loss_avg=2.53749 acc=0.76562 acc_top1_avg=0.76119 acc_top5_avg=0.92598 lr=0.00100 gn=8.98119 time=57.04it/s +epoch=52 global_step=20500 loss=2.21277 loss_avg=2.50930 acc=0.80469 acc_top1_avg=0.76446 acc_top5_avg=0.92722 lr=0.00100 gn=16.18198 time=55.36it/s +epoch=52 global_step=20550 loss=2.59411 loss_avg=2.50926 acc=0.76562 acc_top1_avg=0.76480 acc_top5_avg=0.92675 lr=0.00100 gn=16.36707 time=59.56it/s +epoch=52 global_step=20600 loss=2.31636 loss_avg=2.50855 acc=0.78906 acc_top1_avg=0.76446 acc_top5_avg=0.92703 lr=0.00100 gn=12.36126 time=61.78it/s +epoch=52 global_step=20650 loss=2.31987 loss_avg=2.50624 acc=0.77344 acc_top1_avg=0.76467 acc_top5_avg=0.92711 lr=0.00100 gn=11.85357 time=57.21it/s +epoch=52 global_step=20700 loss=2.67894 loss_avg=2.49523 acc=0.74219 acc_top1_avg=0.76582 acc_top5_avg=0.92793 lr=0.00100 gn=11.18012 time=53.74it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.62142 test_loss_avg=0.50094 acc=0.85156 test_acc_avg=0.85342 test_acc_top5_avg=0.99107 time=243.73it/s +epoch=52 global_step=20723 loss=0.06943 test_loss_avg=0.37806 acc=0.93750 test_acc_avg=0.88835 test_acc_top5_avg=0.99328 time=875.27it/s +curr_acc 0.8884 +BEST_ACC 0.8934 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=3.05910 loss_avg=2.34778 acc=0.69531 acc_top1_avg=0.78067 acc_top5_avg=0.93605 lr=0.00100 gn=10.92883 time=58.07it/s +epoch=53 global_step=20800 loss=2.37518 loss_avg=2.45566 acc=0.78125 acc_top1_avg=0.76968 acc_top5_avg=0.92888 lr=0.00100 gn=16.95615 time=59.05it/s +epoch=53 global_step=20850 loss=2.73697 loss_avg=2.48056 acc=0.74219 acc_top1_avg=0.76692 acc_top5_avg=0.92735 lr=0.00100 gn=12.91990 time=61.77it/s +epoch=53 global_step=20900 loss=2.38530 loss_avg=2.46949 acc=0.76562 acc_top1_avg=0.76823 acc_top5_avg=0.92805 lr=0.00100 gn=11.86428 time=60.33it/s +epoch=53 global_step=20950 loss=2.54347 loss_avg=2.47380 acc=0.77344 acc_top1_avg=0.76828 acc_top5_avg=0.92838 lr=0.00100 gn=14.81487 time=62.98it/s +epoch=53 global_step=21000 loss=2.72182 loss_avg=2.47012 acc=0.74219 acc_top1_avg=0.76864 acc_top5_avg=0.92901 lr=0.00100 gn=13.13119 time=31.76it/s +epoch=53 global_step=21050 loss=2.57822 loss_avg=2.46784 acc=0.75000 acc_top1_avg=0.76909 acc_top5_avg=0.92888 lr=0.00100 gn=11.28178 time=53.07it/s +epoch=53 global_step=21100 loss=2.22687 loss_avg=2.47245 acc=0.78906 acc_top1_avg=0.76892 acc_top5_avg=0.92836 lr=0.00100 gn=12.42793 time=52.41it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.10658 test_loss_avg=0.37713 acc=0.97656 test_acc_avg=0.88822 test_acc_top5_avg=0.99579 time=243.43it/s +epoch=53 global_step=21114 loss=0.09619 test_loss_avg=0.43032 acc=0.96875 test_acc_avg=0.87959 test_acc_top5_avg=0.99268 time=245.34it/s +epoch=53 global_step=21114 loss=0.12615 test_loss_avg=0.38758 acc=0.93750 test_acc_avg=0.88993 test_acc_top5_avg=0.99357 time=557.90it/s +curr_acc 0.8899 +BEST_ACC 0.8934 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=2.35064 loss_avg=2.39962 acc=0.77344 acc_top1_avg=0.77257 acc_top5_avg=0.93381 lr=0.00100 gn=10.39436 time=60.72it/s +epoch=54 global_step=21200 loss=2.34288 loss_avg=2.42226 acc=0.78125 acc_top1_avg=0.77144 acc_top5_avg=0.92941 lr=0.00100 gn=14.94780 time=62.50it/s +epoch=54 global_step=21250 loss=3.18228 loss_avg=2.46332 acc=0.70312 acc_top1_avg=0.76769 acc_top5_avg=0.92647 lr=0.00100 gn=14.22790 time=55.37it/s +epoch=54 global_step=21300 loss=2.04078 loss_avg=2.47485 acc=0.80469 acc_top1_avg=0.76647 acc_top5_avg=0.92671 lr=0.00100 gn=15.43396 time=52.11it/s +epoch=54 global_step=21350 loss=2.98815 loss_avg=2.46659 acc=0.71875 acc_top1_avg=0.76788 acc_top5_avg=0.92677 lr=0.00100 gn=14.29466 time=65.42it/s +epoch=54 global_step=21400 loss=2.22837 loss_avg=2.45180 acc=0.80469 acc_top1_avg=0.76956 acc_top5_avg=0.92783 lr=0.00100 gn=14.79571 time=61.36it/s +epoch=54 global_step=21450 loss=2.59064 loss_avg=2.46551 acc=0.74219 acc_top1_avg=0.76793 acc_top5_avg=0.92769 lr=0.00100 gn=11.78247 time=55.24it/s +epoch=54 global_step=21500 loss=2.07096 loss_avg=2.46448 acc=0.80469 acc_top1_avg=0.76836 acc_top5_avg=0.92795 lr=0.00100 gn=14.52740 time=55.69it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.16361 test_loss_avg=0.53800 acc=0.93750 test_acc_avg=0.84490 test_acc_top5_avg=0.99127 time=243.74it/s +epoch=54 global_step=21505 loss=0.20493 test_loss_avg=0.39484 acc=0.87500 test_acc_avg=0.88321 test_acc_top5_avg=0.99357 time=859.49it/s +curr_acc 0.8832 +BEST_ACC 0.8934 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=2.65224 loss_avg=2.38244 acc=0.75000 acc_top1_avg=0.77917 acc_top5_avg=0.92969 lr=0.00100 gn=11.36467 time=64.80it/s +epoch=55 global_step=21600 loss=2.25680 loss_avg=2.45467 acc=0.79688 acc_top1_avg=0.77097 acc_top5_avg=0.92961 lr=0.00100 gn=13.38353 time=52.26it/s +epoch=55 global_step=21650 loss=2.24316 loss_avg=2.43132 acc=0.78906 acc_top1_avg=0.77322 acc_top5_avg=0.92829 lr=0.00100 gn=12.08839 time=49.61it/s +epoch=55 global_step=21700 loss=2.38916 loss_avg=2.45756 acc=0.77344 acc_top1_avg=0.77087 acc_top5_avg=0.92897 lr=0.00100 gn=11.10566 time=61.07it/s +epoch=55 global_step=21750 loss=2.59559 loss_avg=2.45718 acc=0.74219 acc_top1_avg=0.77101 acc_top5_avg=0.92886 lr=0.00100 gn=13.88137 time=63.67it/s +epoch=55 global_step=21800 loss=2.30469 loss_avg=2.44672 acc=0.78125 acc_top1_avg=0.77198 acc_top5_avg=0.92879 lr=0.00100 gn=9.96356 time=60.29it/s +epoch=55 global_step=21850 loss=2.01370 loss_avg=2.43334 acc=0.82812 acc_top1_avg=0.77337 acc_top5_avg=0.92912 lr=0.00100 gn=13.66072 time=60.76it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.44868 test_loss_avg=0.58172 acc=0.85156 test_acc_avg=0.81719 test_acc_top5_avg=0.98750 time=247.29it/s +epoch=55 global_step=21896 loss=0.24028 test_loss_avg=0.46411 acc=0.89844 test_acc_avg=0.86562 test_acc_top5_avg=0.99247 time=245.31it/s +epoch=55 global_step=21896 loss=0.12423 test_loss_avg=0.39707 acc=0.93750 test_acc_avg=0.88360 test_acc_top5_avg=0.99318 time=547.77it/s +curr_acc 0.8836 +BEST_ACC 0.8934 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=2.70094 loss_avg=2.41219 acc=0.75000 acc_top1_avg=0.77539 acc_top5_avg=0.91797 lr=0.00100 gn=16.26927 time=57.17it/s +epoch=56 global_step=21950 loss=1.89238 loss_avg=2.36733 acc=0.82812 acc_top1_avg=0.77836 acc_top5_avg=0.92694 lr=0.00100 gn=10.67741 time=57.90it/s +epoch=56 global_step=22000 loss=2.40934 loss_avg=2.43591 acc=0.77344 acc_top1_avg=0.77118 acc_top5_avg=0.92811 lr=0.00100 gn=11.80998 time=54.63it/s +epoch=56 global_step=22050 loss=2.51201 loss_avg=2.44421 acc=0.76562 acc_top1_avg=0.77105 acc_top5_avg=0.92725 lr=0.00100 gn=16.75583 time=60.13it/s +epoch=56 global_step=22100 loss=2.58811 loss_avg=2.44842 acc=0.76562 acc_top1_avg=0.77087 acc_top5_avg=0.92724 lr=0.00100 gn=18.75797 time=53.07it/s +epoch=56 global_step=22150 loss=2.62071 loss_avg=2.44342 acc=0.75781 acc_top1_avg=0.77138 acc_top5_avg=0.92753 lr=0.00100 gn=15.63514 time=64.28it/s +epoch=56 global_step=22200 loss=2.58667 loss_avg=2.43972 acc=0.76562 acc_top1_avg=0.77166 acc_top5_avg=0.92827 lr=0.00100 gn=17.71098 time=53.54it/s +epoch=56 global_step=22250 loss=2.45123 loss_avg=2.43512 acc=0.75781 acc_top1_avg=0.77220 acc_top5_avg=0.92883 lr=0.00100 gn=9.20628 time=52.40it/s +====================Eval==================== +epoch=56 global_step=22287 loss=0.87366 test_loss_avg=0.57823 acc=0.78906 test_acc_avg=0.83293 test_acc_top5_avg=0.99099 time=254.28it/s +epoch=56 global_step=22287 loss=0.29982 test_loss_avg=0.40081 acc=0.91406 test_acc_avg=0.88487 test_acc_top5_avg=0.99332 time=251.47it/s +epoch=56 global_step=22287 loss=0.04512 test_loss_avg=0.39314 acc=1.00000 test_acc_avg=0.88706 test_acc_top5_avg=0.99357 time=861.61it/s +curr_acc 0.8871 +BEST_ACC 0.8934 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=2.89737 loss_avg=2.35820 acc=0.72656 acc_top1_avg=0.78365 acc_top5_avg=0.92909 lr=0.00100 gn=12.82978 time=54.65it/s +epoch=57 global_step=22350 loss=2.10679 loss_avg=2.37644 acc=0.79688 acc_top1_avg=0.78175 acc_top5_avg=0.92684 lr=0.00100 gn=11.65017 time=64.32it/s +epoch=57 global_step=22400 loss=2.51121 loss_avg=2.36016 acc=0.76562 acc_top1_avg=0.78215 acc_top5_avg=0.93024 lr=0.00100 gn=13.65135 time=60.82it/s +epoch=57 global_step=22450 loss=2.28221 loss_avg=2.38110 acc=0.78906 acc_top1_avg=0.77928 acc_top5_avg=0.92835 lr=0.00100 gn=11.11963 time=57.36it/s +epoch=57 global_step=22500 loss=2.49089 loss_avg=2.38412 acc=0.78125 acc_top1_avg=0.77916 acc_top5_avg=0.92873 lr=0.00100 gn=17.50461 time=61.13it/s +epoch=57 global_step=22550 loss=2.71984 loss_avg=2.40203 acc=0.74219 acc_top1_avg=0.77757 acc_top5_avg=0.92802 lr=0.00100 gn=13.54978 time=49.24it/s +epoch=57 global_step=22600 loss=2.61534 loss_avg=2.40546 acc=0.74219 acc_top1_avg=0.77713 acc_top5_avg=0.92799 lr=0.00100 gn=16.53056 time=60.13it/s +epoch=57 global_step=22650 loss=3.36853 loss_avg=2.41497 acc=0.67969 acc_top1_avg=0.77593 acc_top5_avg=0.92816 lr=0.00100 gn=10.73026 time=64.82it/s +====================Eval==================== +epoch=57 global_step=22678 loss=0.42467 test_loss_avg=0.47758 acc=0.85156 test_acc_avg=0.86320 test_acc_top5_avg=0.99385 time=243.63it/s +epoch=57 global_step=22678 loss=0.06867 test_loss_avg=0.37688 acc=1.00000 test_acc_avg=0.89270 test_acc_top5_avg=0.99476 time=843.25it/s +curr_acc 0.8927 +BEST_ACC 0.8934 +curr_acc_top5 0.9948 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=2.42135 loss_avg=2.50164 acc=0.78125 acc_top1_avg=0.76491 acc_top5_avg=0.92401 lr=0.00100 gn=15.41522 time=60.30it/s +epoch=58 global_step=22750 loss=2.75421 loss_avg=2.44178 acc=0.72656 acc_top1_avg=0.77268 acc_top5_avg=0.92643 lr=0.00100 gn=12.37605 time=53.50it/s +epoch=58 global_step=22800 loss=2.53194 loss_avg=2.41028 acc=0.76562 acc_top1_avg=0.77510 acc_top5_avg=0.92975 lr=0.00100 gn=13.20612 time=57.00it/s +epoch=58 global_step=22850 loss=2.36764 loss_avg=2.37806 acc=0.79688 acc_top1_avg=0.77852 acc_top5_avg=0.93055 lr=0.00100 gn=15.54817 time=57.49it/s +epoch=58 global_step=22900 loss=2.56871 loss_avg=2.39892 acc=0.75000 acc_top1_avg=0.77622 acc_top5_avg=0.92983 lr=0.00100 gn=11.91501 time=51.14it/s +epoch=58 global_step=22950 loss=2.68451 loss_avg=2.40433 acc=0.75781 acc_top1_avg=0.77548 acc_top5_avg=0.92937 lr=0.00100 gn=14.53689 time=57.99it/s +epoch=58 global_step=23000 loss=2.13394 loss_avg=2.39479 acc=0.80469 acc_top1_avg=0.77666 acc_top5_avg=0.92983 lr=0.00100 gn=12.43166 time=56.71it/s +epoch=58 global_step=23050 loss=2.29459 loss_avg=2.40001 acc=0.78906 acc_top1_avg=0.77604 acc_top5_avg=0.92990 lr=0.00100 gn=14.96455 time=58.58it/s +====================Eval==================== +epoch=58 global_step=23069 loss=0.72979 test_loss_avg=0.37801 acc=0.79688 test_acc_avg=0.89236 test_acc_top5_avg=0.99523 time=248.52it/s +epoch=58 global_step=23069 loss=0.22588 test_loss_avg=0.45770 acc=0.92969 test_acc_avg=0.87546 test_acc_top5_avg=0.99334 time=239.25it/s +epoch=58 global_step=23069 loss=0.25854 test_loss_avg=0.43505 acc=0.87500 test_acc_avg=0.88054 test_acc_top5_avg=0.99357 time=409.88it/s +curr_acc 0.8805 +BEST_ACC 0.8934 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=2.54546 loss_avg=2.45427 acc=0.77344 acc_top1_avg=0.77117 acc_top5_avg=0.92364 lr=0.00100 gn=18.63759 time=57.63it/s +epoch=59 global_step=23150 loss=2.62690 loss_avg=2.39181 acc=0.75781 acc_top1_avg=0.77758 acc_top5_avg=0.92631 lr=0.00100 gn=14.53865 time=63.82it/s +epoch=59 global_step=23200 loss=2.32020 loss_avg=2.43123 acc=0.78125 acc_top1_avg=0.77332 acc_top5_avg=0.92611 lr=0.00100 gn=12.52671 time=62.12it/s +epoch=59 global_step=23250 loss=2.40148 loss_avg=2.40153 acc=0.78906 acc_top1_avg=0.77633 acc_top5_avg=0.92805 lr=0.00100 gn=16.94537 time=62.99it/s +epoch=59 global_step=23300 loss=2.53332 loss_avg=2.40093 acc=0.78125 acc_top1_avg=0.77655 acc_top5_avg=0.92813 lr=0.00100 gn=18.52247 time=57.93it/s +epoch=59 global_step=23350 loss=1.92952 loss_avg=2.38496 acc=0.82812 acc_top1_avg=0.77839 acc_top5_avg=0.92777 lr=0.00100 gn=17.25436 time=64.94it/s +epoch=59 global_step=23400 loss=2.58296 loss_avg=2.38058 acc=0.75781 acc_top1_avg=0.77901 acc_top5_avg=0.92841 lr=0.00100 gn=21.11944 time=54.11it/s +epoch=59 global_step=23450 loss=2.41077 loss_avg=2.38518 acc=0.76562 acc_top1_avg=0.77846 acc_top5_avg=0.92872 lr=0.00100 gn=15.42952 time=60.28it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.19574 test_loss_avg=0.42253 acc=0.94531 test_acc_avg=0.87680 test_acc_top5_avg=0.99279 time=250.71it/s +epoch=59 global_step=23460 loss=0.21273 test_loss_avg=0.35683 acc=0.87500 test_acc_avg=0.89577 test_acc_top5_avg=0.99278 time=891.08it/s +curr_acc 0.8958 +BEST_ACC 0.8934 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=2.69986 loss_avg=2.36460 acc=0.75000 acc_top1_avg=0.78027 acc_top5_avg=0.93418 lr=0.00100 gn=15.91241 time=56.83it/s +epoch=60 global_step=23550 loss=2.29933 loss_avg=2.34360 acc=0.78125 acc_top1_avg=0.78255 acc_top5_avg=0.93108 lr=0.00100 gn=14.91635 time=46.67it/s +epoch=60 global_step=23600 loss=2.68627 loss_avg=2.34822 acc=0.75000 acc_top1_avg=0.78153 acc_top5_avg=0.93147 lr=0.00100 gn=17.51995 time=63.92it/s +epoch=60 global_step=23650 loss=2.24165 loss_avg=2.36137 acc=0.78125 acc_top1_avg=0.77993 acc_top5_avg=0.93043 lr=0.00100 gn=12.04832 time=56.11it/s +epoch=60 global_step=23700 loss=1.82657 loss_avg=2.37016 acc=0.83594 acc_top1_avg=0.77939 acc_top5_avg=0.92998 lr=0.00100 gn=14.20150 time=65.22it/s +epoch=60 global_step=23750 loss=2.76613 loss_avg=2.37637 acc=0.74219 acc_top1_avg=0.77907 acc_top5_avg=0.93015 lr=0.00100 gn=17.32326 time=58.24it/s +epoch=60 global_step=23800 loss=2.25438 loss_avg=2.37291 acc=0.78906 acc_top1_avg=0.77973 acc_top5_avg=0.92960 lr=0.00100 gn=9.77546 time=58.27it/s +epoch=60 global_step=23850 loss=2.71774 loss_avg=2.36887 acc=0.73438 acc_top1_avg=0.78049 acc_top5_avg=0.92959 lr=0.00100 gn=17.32456 time=60.97it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.19090 test_loss_avg=0.56753 acc=0.96094 test_acc_avg=0.82891 test_acc_top5_avg=0.99375 time=231.67it/s +epoch=60 global_step=23851 loss=0.12773 test_loss_avg=0.46657 acc=0.96094 test_acc_avg=0.86641 test_acc_top5_avg=0.99232 time=222.45it/s +epoch=60 global_step=23851 loss=0.26412 test_loss_avg=0.40028 acc=0.87500 test_acc_avg=0.88321 test_acc_top5_avg=0.99337 time=738.30it/s +curr_acc 0.8832 +BEST_ACC 0.8958 +curr_acc_top5 0.9934 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=1.66407 loss_avg=2.38095 acc=0.84375 acc_top1_avg=0.77710 acc_top5_avg=0.92411 lr=0.00100 gn=13.54820 time=56.23it/s +epoch=61 global_step=23950 loss=1.97340 loss_avg=2.37317 acc=0.81250 acc_top1_avg=0.77841 acc_top5_avg=0.92693 lr=0.00100 gn=17.73392 time=61.07it/s +epoch=61 global_step=24000 loss=2.50109 loss_avg=2.36357 acc=0.77344 acc_top1_avg=0.78036 acc_top5_avg=0.92722 lr=0.00100 gn=20.99265 time=57.26it/s +epoch=61 global_step=24050 loss=2.55379 loss_avg=2.35754 acc=0.76562 acc_top1_avg=0.78113 acc_top5_avg=0.92792 lr=0.00100 gn=15.64013 time=57.68it/s +epoch=61 global_step=24100 loss=3.04798 loss_avg=2.35714 acc=0.70312 acc_top1_avg=0.78125 acc_top5_avg=0.92928 lr=0.00100 gn=14.49902 time=58.57it/s +epoch=61 global_step=24150 loss=2.12216 loss_avg=2.36562 acc=0.82812 acc_top1_avg=0.78028 acc_top5_avg=0.92945 lr=0.00100 gn=19.13377 time=56.87it/s +epoch=61 global_step=24200 loss=2.01455 loss_avg=2.35950 acc=0.82812 acc_top1_avg=0.78096 acc_top5_avg=0.93052 lr=0.00100 gn=15.39346 time=56.21it/s +====================Eval==================== +epoch=61 global_step=24242 loss=0.27868 test_loss_avg=0.53185 acc=0.91406 test_acc_avg=0.84148 test_acc_top5_avg=0.98891 time=251.37it/s +epoch=61 global_step=24242 loss=0.26509 test_loss_avg=0.42397 acc=0.87500 test_acc_avg=0.87757 test_acc_top5_avg=0.99258 time=885.81it/s +curr_acc 0.8776 +BEST_ACC 0.8958 +curr_acc_top5 0.9926 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=2.35052 loss_avg=2.22764 acc=0.77344 acc_top1_avg=0.79492 acc_top5_avg=0.93848 lr=0.00100 gn=13.78122 time=62.65it/s +epoch=62 global_step=24300 loss=3.17031 loss_avg=2.43765 acc=0.71094 acc_top1_avg=0.77169 acc_top5_avg=0.92821 lr=0.00100 gn=18.32519 time=55.27it/s +epoch=62 global_step=24350 loss=2.64761 loss_avg=2.38211 acc=0.75000 acc_top1_avg=0.77720 acc_top5_avg=0.93056 lr=0.00100 gn=23.11309 time=61.28it/s +epoch=62 global_step=24400 loss=1.84642 loss_avg=2.37318 acc=0.84375 acc_top1_avg=0.77858 acc_top5_avg=0.92880 lr=0.00100 gn=18.77200 time=54.88it/s +epoch=62 global_step=24450 loss=2.26089 loss_avg=2.37257 acc=0.79688 acc_top1_avg=0.77870 acc_top5_avg=0.92819 lr=0.00100 gn=18.92299 time=61.69it/s +epoch=62 global_step=24500 loss=2.86183 loss_avg=2.36159 acc=0.72656 acc_top1_avg=0.78025 acc_top5_avg=0.92860 lr=0.00100 gn=17.23184 time=57.26it/s +epoch=62 global_step=24550 loss=2.18687 loss_avg=2.35154 acc=0.79688 acc_top1_avg=0.78138 acc_top5_avg=0.92903 lr=0.00100 gn=16.43499 time=64.61it/s +epoch=62 global_step=24600 loss=2.16266 loss_avg=2.35692 acc=0.79688 acc_top1_avg=0.78099 acc_top5_avg=0.92912 lr=0.00100 gn=13.44967 time=58.10it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.83916 test_loss_avg=0.85527 acc=0.75000 test_acc_avg=0.75781 test_acc_top5_avg=0.98438 time=194.61it/s +epoch=62 global_step=24633 loss=0.22568 test_loss_avg=0.46785 acc=0.89844 test_acc_avg=0.86313 test_acc_top5_avg=0.99354 time=247.28it/s +epoch=62 global_step=24633 loss=0.37444 test_loss_avg=0.41235 acc=0.87500 test_acc_avg=0.87886 test_acc_top5_avg=0.99387 time=808.15it/s +curr_acc 0.8789 +BEST_ACC 0.8958 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=2.40984 loss_avg=2.31309 acc=0.77344 acc_top1_avg=0.78768 acc_top5_avg=0.93290 lr=0.00100 gn=12.74722 time=61.67it/s +epoch=63 global_step=24700 loss=2.23046 loss_avg=2.29822 acc=0.81250 acc_top1_avg=0.78673 acc_top5_avg=0.93120 lr=0.00100 gn=20.39812 time=63.20it/s +epoch=63 global_step=24750 loss=2.77777 loss_avg=2.30520 acc=0.72656 acc_top1_avg=0.78646 acc_top5_avg=0.93102 lr=0.00100 gn=15.00026 time=56.87it/s +epoch=63 global_step=24800 loss=2.61890 loss_avg=2.32561 acc=0.75000 acc_top1_avg=0.78396 acc_top5_avg=0.92959 lr=0.00100 gn=15.36429 time=57.59it/s +epoch=63 global_step=24850 loss=2.30598 loss_avg=2.33959 acc=0.77344 acc_top1_avg=0.78255 acc_top5_avg=0.92969 lr=0.00100 gn=15.89868 time=59.74it/s +epoch=63 global_step=24900 loss=2.99597 loss_avg=2.32362 acc=0.71094 acc_top1_avg=0.78429 acc_top5_avg=0.92942 lr=0.00100 gn=17.28000 time=48.37it/s +epoch=63 global_step=24950 loss=2.24348 loss_avg=2.32218 acc=0.78906 acc_top1_avg=0.78448 acc_top5_avg=0.92998 lr=0.00100 gn=19.05050 time=55.67it/s +epoch=63 global_step=25000 loss=2.24206 loss_avg=2.33918 acc=0.78906 acc_top1_avg=0.78251 acc_top5_avg=0.92969 lr=0.00100 gn=13.14493 time=55.51it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.34279 test_loss_avg=0.40935 acc=0.89844 test_acc_avg=0.87500 test_acc_top5_avg=0.99287 time=244.74it/s +epoch=63 global_step=25024 loss=0.51546 test_loss_avg=0.39818 acc=0.84375 test_acc_avg=0.88303 test_acc_top5_avg=0.99347 time=248.37it/s +epoch=63 global_step=25024 loss=0.34710 test_loss_avg=0.40327 acc=0.87500 test_acc_avg=0.88074 test_acc_top5_avg=0.99367 time=863.91it/s +curr_acc 0.8807 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=2.56515 loss_avg=2.37669 acc=0.75781 acc_top1_avg=0.77764 acc_top5_avg=0.92458 lr=0.00100 gn=22.17371 time=62.38it/s +epoch=64 global_step=25100 loss=2.15391 loss_avg=2.29156 acc=0.80469 acc_top1_avg=0.78721 acc_top5_avg=0.92876 lr=0.00100 gn=18.13035 time=59.00it/s +epoch=64 global_step=25150 loss=2.21957 loss_avg=2.31040 acc=0.78906 acc_top1_avg=0.78534 acc_top5_avg=0.93025 lr=0.00100 gn=15.79356 time=55.95it/s +epoch=64 global_step=25200 loss=2.11535 loss_avg=2.30913 acc=0.80469 acc_top1_avg=0.78604 acc_top5_avg=0.93133 lr=0.00100 gn=18.10744 time=59.62it/s +epoch=64 global_step=25250 loss=1.53398 loss_avg=2.31224 acc=0.85938 acc_top1_avg=0.78619 acc_top5_avg=0.93155 lr=0.00100 gn=16.61295 time=55.24it/s +epoch=64 global_step=25300 loss=1.86381 loss_avg=2.31307 acc=0.82812 acc_top1_avg=0.78640 acc_top5_avg=0.93218 lr=0.00100 gn=20.77986 time=64.18it/s +epoch=64 global_step=25350 loss=2.42552 loss_avg=2.31858 acc=0.77344 acc_top1_avg=0.78578 acc_top5_avg=0.93117 lr=0.00100 gn=18.97675 time=59.87it/s +epoch=64 global_step=25400 loss=1.97900 loss_avg=2.32573 acc=0.82031 acc_top1_avg=0.78491 acc_top5_avg=0.93106 lr=0.00100 gn=13.80906 time=61.20it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.85322 test_loss_avg=0.48831 acc=0.78125 test_acc_avg=0.85778 test_acc_top5_avg=0.99077 time=231.18it/s +epoch=64 global_step=25415 loss=0.08836 test_loss_avg=0.37441 acc=0.93750 test_acc_avg=0.89062 test_acc_top5_avg=0.99318 time=873.81it/s +curr_acc 0.8906 +BEST_ACC 0.8958 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=2.39047 loss_avg=2.30317 acc=0.77344 acc_top1_avg=0.78527 acc_top5_avg=0.92902 lr=0.00100 gn=22.08093 time=53.17it/s +epoch=65 global_step=25500 loss=1.99635 loss_avg=2.26121 acc=0.81250 acc_top1_avg=0.79154 acc_top5_avg=0.93051 lr=0.00100 gn=15.34194 time=60.88it/s +epoch=65 global_step=25550 loss=1.60473 loss_avg=2.30670 acc=0.86719 acc_top1_avg=0.78715 acc_top5_avg=0.92755 lr=0.00100 gn=20.31327 time=56.88it/s +epoch=65 global_step=25600 loss=1.22980 loss_avg=2.28727 acc=0.89062 acc_top1_avg=0.78898 acc_top5_avg=0.92969 lr=0.00100 gn=14.14924 time=58.38it/s +epoch=65 global_step=25650 loss=2.63312 loss_avg=2.30401 acc=0.76562 acc_top1_avg=0.78727 acc_top5_avg=0.92916 lr=0.00100 gn=24.47407 time=61.15it/s +epoch=65 global_step=25700 loss=2.68694 loss_avg=2.29740 acc=0.75781 acc_top1_avg=0.78788 acc_top5_avg=0.93048 lr=0.00100 gn=23.15237 time=61.48it/s +epoch=65 global_step=25750 loss=1.87256 loss_avg=2.30051 acc=0.83594 acc_top1_avg=0.78762 acc_top5_avg=0.93083 lr=0.00100 gn=18.33636 time=56.56it/s +epoch=65 global_step=25800 loss=2.29700 loss_avg=2.31684 acc=0.78125 acc_top1_avg=0.78571 acc_top5_avg=0.92995 lr=0.00100 gn=20.05000 time=63.30it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.21891 test_loss_avg=0.39148 acc=0.92188 test_acc_avg=0.87760 test_acc_top5_avg=0.99479 time=244.98it/s +epoch=65 global_step=25806 loss=0.17355 test_loss_avg=0.41561 acc=0.94531 test_acc_avg=0.87800 test_acc_top5_avg=0.99303 time=262.52it/s +epoch=65 global_step=25806 loss=0.07527 test_loss_avg=0.37927 acc=1.00000 test_acc_avg=0.88776 test_acc_top5_avg=0.99367 time=725.16it/s +curr_acc 0.8878 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=3.55981 loss_avg=2.25842 acc=0.66406 acc_top1_avg=0.79190 acc_top5_avg=0.93200 lr=0.00100 gn=22.51795 time=57.91it/s +epoch=66 global_step=25900 loss=2.70738 loss_avg=2.29505 acc=0.73438 acc_top1_avg=0.78649 acc_top5_avg=0.93160 lr=0.00100 gn=19.76706 time=53.71it/s +epoch=66 global_step=25950 loss=2.71457 loss_avg=2.31131 acc=0.74219 acc_top1_avg=0.78488 acc_top5_avg=0.93159 lr=0.00100 gn=16.38640 time=63.39it/s +epoch=66 global_step=26000 loss=2.29432 loss_avg=2.29226 acc=0.78125 acc_top1_avg=0.78725 acc_top5_avg=0.93154 lr=0.00100 gn=13.25414 time=56.76it/s +epoch=66 global_step=26050 loss=2.92054 loss_avg=2.31143 acc=0.71875 acc_top1_avg=0.78535 acc_top5_avg=0.93145 lr=0.00100 gn=14.12117 time=49.74it/s +epoch=66 global_step=26100 loss=2.60090 loss_avg=2.30797 acc=0.75000 acc_top1_avg=0.78579 acc_top5_avg=0.93115 lr=0.00100 gn=21.65531 time=63.44it/s +epoch=66 global_step=26150 loss=2.19862 loss_avg=2.31098 acc=0.79688 acc_top1_avg=0.78570 acc_top5_avg=0.93103 lr=0.00100 gn=16.90867 time=56.64it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.30790 test_loss_avg=0.49416 acc=0.90625 test_acc_avg=0.85742 test_acc_top5_avg=0.99327 time=253.92it/s +epoch=66 global_step=26197 loss=0.09544 test_loss_avg=0.38503 acc=1.00000 test_acc_avg=0.88845 test_acc_top5_avg=0.99486 time=780.19it/s +curr_acc 0.8884 +BEST_ACC 0.8958 +curr_acc_top5 0.9949 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=2.40533 loss_avg=1.87832 acc=0.77344 acc_top1_avg=0.83073 acc_top5_avg=0.94792 lr=0.00100 gn=24.42472 time=58.06it/s +epoch=67 global_step=26250 loss=1.75722 loss_avg=2.22799 acc=0.83594 acc_top1_avg=0.79466 acc_top5_avg=0.93219 lr=0.00100 gn=19.49371 time=62.50it/s +epoch=67 global_step=26300 loss=1.93725 loss_avg=2.23099 acc=0.82031 acc_top1_avg=0.79430 acc_top5_avg=0.93204 lr=0.00100 gn=19.68094 time=59.27it/s +epoch=67 global_step=26350 loss=1.95224 loss_avg=2.24732 acc=0.82031 acc_top1_avg=0.79233 acc_top5_avg=0.93117 lr=0.00100 gn=17.09209 time=61.42it/s +epoch=67 global_step=26400 loss=1.90729 loss_avg=2.25710 acc=0.82031 acc_top1_avg=0.79195 acc_top5_avg=0.93000 lr=0.00100 gn=17.14956 time=54.53it/s +epoch=67 global_step=26450 loss=2.47390 loss_avg=2.26969 acc=0.77344 acc_top1_avg=0.79017 acc_top5_avg=0.93046 lr=0.00100 gn=23.93714 time=56.74it/s +epoch=67 global_step=26500 loss=2.15483 loss_avg=2.29213 acc=0.80469 acc_top1_avg=0.78808 acc_top5_avg=0.93033 lr=0.00100 gn=18.42805 time=56.48it/s +epoch=67 global_step=26550 loss=2.47545 loss_avg=2.29872 acc=0.75781 acc_top1_avg=0.78738 acc_top5_avg=0.93121 lr=0.00100 gn=24.44824 time=50.41it/s +====================Eval==================== +epoch=67 global_step=26588 loss=0.42632 test_loss_avg=0.53558 acc=0.89062 test_acc_avg=0.84487 test_acc_top5_avg=0.98996 time=243.32it/s +epoch=67 global_step=26588 loss=0.24355 test_loss_avg=0.41829 acc=0.96094 test_acc_avg=0.87993 test_acc_top5_avg=0.99328 time=235.61it/s +epoch=67 global_step=26588 loss=0.13471 test_loss_avg=0.37614 acc=1.00000 test_acc_avg=0.89082 test_acc_top5_avg=0.99347 time=867.13it/s +curr_acc 0.8908 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=2.15145 loss_avg=2.17574 acc=0.79688 acc_top1_avg=0.79753 acc_top5_avg=0.93490 lr=0.00100 gn=13.39149 time=64.41it/s +epoch=68 global_step=26650 loss=2.43579 loss_avg=2.28796 acc=0.77344 acc_top1_avg=0.78931 acc_top5_avg=0.93397 lr=0.00100 gn=12.92963 time=52.85it/s +epoch=68 global_step=26700 loss=1.96876 loss_avg=2.32305 acc=0.82031 acc_top1_avg=0.78592 acc_top5_avg=0.93018 lr=0.00100 gn=18.33367 time=64.90it/s +epoch=68 global_step=26750 loss=1.71454 loss_avg=2.30592 acc=0.84375 acc_top1_avg=0.78762 acc_top5_avg=0.93031 lr=0.00100 gn=18.73224 time=63.28it/s +epoch=68 global_step=26800 loss=2.30960 loss_avg=2.28486 acc=0.78906 acc_top1_avg=0.78950 acc_top5_avg=0.93009 lr=0.00100 gn=21.33640 time=57.67it/s +epoch=68 global_step=26850 loss=2.41124 loss_avg=2.29618 acc=0.78906 acc_top1_avg=0.78844 acc_top5_avg=0.93040 lr=0.00100 gn=22.95593 time=63.41it/s +epoch=68 global_step=26900 loss=2.70319 loss_avg=2.28262 acc=0.75000 acc_top1_avg=0.78969 acc_top5_avg=0.93024 lr=0.00100 gn=17.28228 time=54.83it/s +epoch=68 global_step=26950 loss=2.17329 loss_avg=2.28506 acc=0.82031 acc_top1_avg=0.78943 acc_top5_avg=0.93062 lr=0.00100 gn=26.57639 time=56.71it/s +====================Eval==================== +epoch=68 global_step=26979 loss=0.85741 test_loss_avg=0.54179 acc=0.76562 test_acc_avg=0.84319 test_acc_top5_avg=0.99079 time=249.01it/s +epoch=68 global_step=26979 loss=0.52177 test_loss_avg=0.41395 acc=0.86719 test_acc_avg=0.87901 test_acc_top5_avg=0.99309 time=263.11it/s +epoch=68 global_step=26979 loss=0.47323 test_loss_avg=0.41470 acc=0.75000 test_acc_avg=0.87737 test_acc_top5_avg=0.99318 time=870.91it/s +curr_acc 0.8774 +BEST_ACC 0.8958 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=2.15866 loss_avg=2.37702 acc=0.80469 acc_top1_avg=0.77530 acc_top5_avg=0.93043 lr=0.00100 gn=23.32209 time=64.05it/s +epoch=69 global_step=27050 loss=1.74416 loss_avg=2.22504 acc=0.84375 acc_top1_avg=0.79456 acc_top5_avg=0.93464 lr=0.00100 gn=15.91896 time=59.78it/s +epoch=69 global_step=27100 loss=2.38978 loss_avg=2.23835 acc=0.77344 acc_top1_avg=0.79352 acc_top5_avg=0.93104 lr=0.00100 gn=20.50941 time=62.86it/s +epoch=69 global_step=27150 loss=2.41319 loss_avg=2.25008 acc=0.77344 acc_top1_avg=0.79331 acc_top5_avg=0.93133 lr=0.00100 gn=16.77983 time=62.99it/s +epoch=69 global_step=27200 loss=2.34029 loss_avg=2.27443 acc=0.79688 acc_top1_avg=0.79097 acc_top5_avg=0.93146 lr=0.00100 gn=28.73471 time=54.76it/s +epoch=69 global_step=27250 loss=2.64091 loss_avg=2.28042 acc=0.75781 acc_top1_avg=0.79030 acc_top5_avg=0.93119 lr=0.00100 gn=19.76074 time=56.49it/s +epoch=69 global_step=27300 loss=2.26249 loss_avg=2.28696 acc=0.78906 acc_top1_avg=0.78960 acc_top5_avg=0.93088 lr=0.00100 gn=25.14513 time=51.84it/s +epoch=69 global_step=27350 loss=2.87742 loss_avg=2.28755 acc=0.72656 acc_top1_avg=0.78969 acc_top5_avg=0.93133 lr=0.00100 gn=22.08835 time=64.12it/s +====================Eval==================== +epoch=69 global_step=27370 loss=0.14583 test_loss_avg=0.48929 acc=0.93750 test_acc_avg=0.85810 test_acc_top5_avg=0.99123 time=249.13it/s +epoch=69 global_step=27370 loss=0.04327 test_loss_avg=0.38596 acc=1.00000 test_acc_avg=0.88726 test_acc_top5_avg=0.99347 time=554.58it/s +curr_acc 0.8873 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=2.05530 loss_avg=2.20662 acc=0.82812 acc_top1_avg=0.79766 acc_top5_avg=0.93073 lr=0.00100 gn=23.50631 time=58.82it/s +epoch=70 global_step=27450 loss=1.95410 loss_avg=2.24276 acc=0.82812 acc_top1_avg=0.79355 acc_top5_avg=0.93408 lr=0.00100 gn=16.73275 time=58.02it/s +epoch=70 global_step=27500 loss=2.06631 loss_avg=2.25780 acc=0.82031 acc_top1_avg=0.79261 acc_top5_avg=0.93275 lr=0.00100 gn=22.92754 time=62.32it/s +epoch=70 global_step=27550 loss=2.04143 loss_avg=2.25835 acc=0.82812 acc_top1_avg=0.79275 acc_top5_avg=0.93294 lr=0.00100 gn=26.91925 time=63.08it/s +epoch=70 global_step=27600 loss=2.19928 loss_avg=2.25884 acc=0.81250 acc_top1_avg=0.79273 acc_top5_avg=0.93139 lr=0.00100 gn=20.00995 time=64.75it/s +epoch=70 global_step=27650 loss=2.11180 loss_avg=2.26168 acc=0.80469 acc_top1_avg=0.79308 acc_top5_avg=0.93069 lr=0.00100 gn=22.29716 time=57.80it/s +epoch=70 global_step=27700 loss=2.10102 loss_avg=2.26483 acc=0.80469 acc_top1_avg=0.79219 acc_top5_avg=0.93089 lr=0.00100 gn=19.24790 time=64.82it/s +epoch=70 global_step=27750 loss=2.13169 loss_avg=2.26647 acc=0.81250 acc_top1_avg=0.79217 acc_top5_avg=0.93123 lr=0.00100 gn=17.65065 time=60.24it/s +====================Eval==================== +epoch=70 global_step=27761 loss=0.58963 test_loss_avg=0.38435 acc=0.79688 test_acc_avg=0.88516 test_acc_top5_avg=0.99414 time=238.75it/s +epoch=70 global_step=27761 loss=0.15865 test_loss_avg=0.39752 acc=0.96094 test_acc_avg=0.88728 test_acc_top5_avg=0.99375 time=254.65it/s +epoch=70 global_step=27761 loss=0.10382 test_loss_avg=0.38680 acc=0.93750 test_acc_avg=0.88934 test_acc_top5_avg=0.99367 time=550.36it/s +curr_acc 0.8893 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=2.53562 loss_avg=2.26142 acc=0.76562 acc_top1_avg=0.79127 acc_top5_avg=0.92989 lr=0.00100 gn=23.54461 time=63.88it/s +epoch=71 global_step=27850 loss=2.09171 loss_avg=2.31797 acc=0.82031 acc_top1_avg=0.78573 acc_top5_avg=0.92872 lr=0.00100 gn=21.90943 time=61.41it/s +epoch=71 global_step=27900 loss=1.85659 loss_avg=2.29617 acc=0.84375 acc_top1_avg=0.78816 acc_top5_avg=0.93002 lr=0.00100 gn=22.75456 time=60.18it/s +epoch=71 global_step=27950 loss=2.38884 loss_avg=2.28381 acc=0.77344 acc_top1_avg=0.78997 acc_top5_avg=0.93196 lr=0.00100 gn=20.46955 time=64.35it/s +epoch=71 global_step=28000 loss=2.48624 loss_avg=2.26940 acc=0.76562 acc_top1_avg=0.79135 acc_top5_avg=0.93325 lr=0.00100 gn=21.52873 time=58.83it/s +epoch=71 global_step=28050 loss=2.68149 loss_avg=2.27325 acc=0.73438 acc_top1_avg=0.79117 acc_top5_avg=0.93266 lr=0.00100 gn=24.11437 time=65.15it/s +epoch=71 global_step=28100 loss=1.90789 loss_avg=2.26447 acc=0.82812 acc_top1_avg=0.79231 acc_top5_avg=0.93250 lr=0.00100 gn=18.12755 time=60.12it/s +epoch=71 global_step=28150 loss=1.62929 loss_avg=2.26932 acc=0.85938 acc_top1_avg=0.79181 acc_top5_avg=0.93168 lr=0.00100 gn=20.02382 time=60.97it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.48956 test_loss_avg=0.51929 acc=0.86719 test_acc_avg=0.84432 test_acc_top5_avg=0.99104 time=258.73it/s +epoch=71 global_step=28152 loss=0.05624 test_loss_avg=0.40592 acc=1.00000 test_acc_avg=0.87896 test_acc_top5_avg=0.99288 time=860.02it/s +curr_acc 0.8790 +BEST_ACC 0.8958 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=3.41141 loss_avg=2.23614 acc=0.65625 acc_top1_avg=0.79443 acc_top5_avg=0.93278 lr=0.00100 gn=16.53391 time=53.65it/s +epoch=72 global_step=28250 loss=1.96053 loss_avg=2.20852 acc=0.83594 acc_top1_avg=0.79807 acc_top5_avg=0.93232 lr=0.00100 gn=19.22203 time=63.40it/s +epoch=72 global_step=28300 loss=2.52039 loss_avg=2.21287 acc=0.77344 acc_top1_avg=0.79772 acc_top5_avg=0.93217 lr=0.00100 gn=19.42670 time=63.98it/s +epoch=72 global_step=28350 loss=1.79417 loss_avg=2.25541 acc=0.82812 acc_top1_avg=0.79328 acc_top5_avg=0.93004 lr=0.00100 gn=24.92403 time=53.58it/s +epoch=72 global_step=28400 loss=1.85820 loss_avg=2.24908 acc=0.83594 acc_top1_avg=0.79410 acc_top5_avg=0.93164 lr=0.00100 gn=21.07525 time=60.67it/s +epoch=72 global_step=28450 loss=2.28665 loss_avg=2.25317 acc=0.78906 acc_top1_avg=0.79420 acc_top5_avg=0.93087 lr=0.00100 gn=18.05473 time=60.68it/s +epoch=72 global_step=28500 loss=2.10735 loss_avg=2.25085 acc=0.80469 acc_top1_avg=0.79463 acc_top5_avg=0.93187 lr=0.00100 gn=21.83633 time=64.19it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.14524 test_loss_avg=0.47069 acc=0.96875 test_acc_avg=0.85807 test_acc_top5_avg=0.99479 time=261.25it/s +epoch=72 global_step=28543 loss=0.28886 test_loss_avg=0.43458 acc=0.89844 test_acc_avg=0.87172 test_acc_top5_avg=0.99219 time=243.18it/s +epoch=72 global_step=28543 loss=0.16653 test_loss_avg=0.37939 acc=0.93750 test_acc_avg=0.88677 test_acc_top5_avg=0.99347 time=523.05it/s +curr_acc 0.8868 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=2.23622 loss_avg=2.28981 acc=0.78906 acc_top1_avg=0.78795 acc_top5_avg=0.92522 lr=0.00100 gn=24.93520 time=57.25it/s +epoch=73 global_step=28600 loss=2.41574 loss_avg=2.33874 acc=0.78906 acc_top1_avg=0.78440 acc_top5_avg=0.93010 lr=0.00100 gn=20.30940 time=58.29it/s +epoch=73 global_step=28650 loss=2.40917 loss_avg=2.30422 acc=0.77344 acc_top1_avg=0.78841 acc_top5_avg=0.92823 lr=0.00100 gn=15.34206 time=50.77it/s +epoch=73 global_step=28700 loss=2.61617 loss_avg=2.26451 acc=0.76562 acc_top1_avg=0.79274 acc_top5_avg=0.92984 lr=0.00100 gn=21.44249 time=56.56it/s +epoch=73 global_step=28750 loss=2.34430 loss_avg=2.26986 acc=0.79688 acc_top1_avg=0.79182 acc_top5_avg=0.92980 lr=0.00100 gn=24.57010 time=55.71it/s +epoch=73 global_step=28800 loss=2.27881 loss_avg=2.25306 acc=0.79688 acc_top1_avg=0.79344 acc_top5_avg=0.93002 lr=0.00100 gn=21.80691 time=56.04it/s +epoch=73 global_step=28850 loss=2.77369 loss_avg=2.25422 acc=0.73438 acc_top1_avg=0.79354 acc_top5_avg=0.93139 lr=0.00100 gn=21.48899 time=59.03it/s +epoch=73 global_step=28900 loss=2.38827 loss_avg=2.26251 acc=0.76562 acc_top1_avg=0.79265 acc_top5_avg=0.93137 lr=0.00100 gn=24.86579 time=62.61it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.28435 test_loss_avg=0.50298 acc=0.92969 test_acc_avg=0.84991 test_acc_top5_avg=0.99148 time=244.03it/s +epoch=73 global_step=28934 loss=0.08679 test_loss_avg=0.37478 acc=1.00000 test_acc_avg=0.88914 test_acc_top5_avg=0.99347 time=886.75it/s +curr_acc 0.8891 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=2.33377 loss_avg=2.19741 acc=0.78906 acc_top1_avg=0.80176 acc_top5_avg=0.92773 lr=0.00100 gn=22.57285 time=53.07it/s +epoch=74 global_step=29000 loss=2.01388 loss_avg=2.21355 acc=0.83594 acc_top1_avg=0.79806 acc_top5_avg=0.93016 lr=0.00100 gn=23.32514 time=57.38it/s +epoch=74 global_step=29050 loss=1.90281 loss_avg=2.21660 acc=0.83594 acc_top1_avg=0.79714 acc_top5_avg=0.92989 lr=0.00100 gn=23.04771 time=63.98it/s +epoch=74 global_step=29100 loss=2.68285 loss_avg=2.21047 acc=0.75781 acc_top1_avg=0.79810 acc_top5_avg=0.93002 lr=0.00100 gn=24.06043 time=58.65it/s +epoch=74 global_step=29150 loss=1.48416 loss_avg=2.22309 acc=0.89062 acc_top1_avg=0.79698 acc_top5_avg=0.93056 lr=0.00100 gn=26.17627 time=64.02it/s +epoch=74 global_step=29200 loss=2.17571 loss_avg=2.23295 acc=0.81250 acc_top1_avg=0.79635 acc_top5_avg=0.93042 lr=0.00100 gn=24.10511 time=54.90it/s +epoch=74 global_step=29250 loss=1.99386 loss_avg=2.23832 acc=0.82031 acc_top1_avg=0.79613 acc_top5_avg=0.93082 lr=0.00100 gn=20.83997 time=63.05it/s +epoch=74 global_step=29300 loss=1.82307 loss_avg=2.23789 acc=0.84375 acc_top1_avg=0.79645 acc_top5_avg=0.93105 lr=0.00100 gn=21.34037 time=64.73it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.45961 test_loss_avg=0.66244 acc=0.85938 test_acc_avg=0.80078 test_acc_top5_avg=0.99414 time=243.52it/s +epoch=74 global_step=29325 loss=0.23473 test_loss_avg=0.46753 acc=0.92969 test_acc_avg=0.86372 test_acc_top5_avg=0.99161 time=247.06it/s +epoch=74 global_step=29325 loss=0.12748 test_loss_avg=0.38733 acc=0.87500 test_acc_avg=0.88578 test_acc_top5_avg=0.99318 time=852.85it/s +curr_acc 0.8858 +BEST_ACC 0.8958 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=2.09399 loss_avg=2.20223 acc=0.82031 acc_top1_avg=0.79937 acc_top5_avg=0.93375 lr=0.00100 gn=24.56226 time=55.05it/s +epoch=75 global_step=29400 loss=2.34804 loss_avg=2.19867 acc=0.79688 acc_top1_avg=0.79896 acc_top5_avg=0.93229 lr=0.00100 gn=20.54779 time=58.51it/s +epoch=75 global_step=29450 loss=2.30904 loss_avg=2.18327 acc=0.78906 acc_top1_avg=0.80088 acc_top5_avg=0.93337 lr=0.00100 gn=24.75708 time=49.76it/s +epoch=75 global_step=29500 loss=2.19685 loss_avg=2.20996 acc=0.80469 acc_top1_avg=0.79835 acc_top5_avg=0.93375 lr=0.00100 gn=21.03030 time=56.14it/s +epoch=75 global_step=29550 loss=2.09943 loss_avg=2.22829 acc=0.80469 acc_top1_avg=0.79632 acc_top5_avg=0.93257 lr=0.00100 gn=25.05328 time=63.14it/s +epoch=75 global_step=29600 loss=2.53017 loss_avg=2.22544 acc=0.75781 acc_top1_avg=0.79676 acc_top5_avg=0.93236 lr=0.00100 gn=21.15108 time=55.11it/s +epoch=75 global_step=29650 loss=2.47064 loss_avg=2.23106 acc=0.77344 acc_top1_avg=0.79639 acc_top5_avg=0.93190 lr=0.00100 gn=18.94669 time=63.59it/s +epoch=75 global_step=29700 loss=1.86442 loss_avg=2.22700 acc=0.83594 acc_top1_avg=0.79704 acc_top5_avg=0.93206 lr=0.00100 gn=19.26868 time=61.37it/s +====================Eval==================== +epoch=75 global_step=29716 loss=0.63727 test_loss_avg=0.60128 acc=0.82031 test_acc_avg=0.82688 test_acc_top5_avg=0.98687 time=228.19it/s +epoch=75 global_step=29716 loss=0.45178 test_loss_avg=0.43901 acc=0.82031 test_acc_avg=0.87021 test_acc_top5_avg=0.99187 time=262.37it/s +epoch=75 global_step=29716 loss=0.23459 test_loss_avg=0.43924 acc=0.87500 test_acc_avg=0.86887 test_acc_top5_avg=0.99219 time=886.75it/s +curr_acc 0.8689 +BEST_ACC 0.8958 +curr_acc_top5 0.9922 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=2.20389 loss_avg=2.16033 acc=0.78906 acc_top1_avg=0.80078 acc_top5_avg=0.93520 lr=0.00100 gn=19.24938 time=59.95it/s +epoch=76 global_step=29800 loss=1.57150 loss_avg=2.17271 acc=0.86719 acc_top1_avg=0.80078 acc_top5_avg=0.93238 lr=0.00100 gn=23.90553 time=59.81it/s +epoch=76 global_step=29850 loss=2.64645 loss_avg=2.20207 acc=0.76562 acc_top1_avg=0.79827 acc_top5_avg=0.93289 lr=0.00100 gn=32.74475 time=58.86it/s +epoch=76 global_step=29900 loss=2.28123 loss_avg=2.20977 acc=0.79688 acc_top1_avg=0.79794 acc_top5_avg=0.93334 lr=0.00100 gn=24.65522 time=53.51it/s +epoch=76 global_step=29950 loss=2.53713 loss_avg=2.20714 acc=0.75781 acc_top1_avg=0.79841 acc_top5_avg=0.93363 lr=0.00100 gn=26.45627 time=63.49it/s +epoch=76 global_step=30000 loss=2.60751 loss_avg=2.21203 acc=0.75781 acc_top1_avg=0.79809 acc_top5_avg=0.93373 lr=0.00100 gn=23.13457 time=50.14it/s +epoch=76 global_step=30050 loss=2.32826 loss_avg=2.21150 acc=0.79688 acc_top1_avg=0.79818 acc_top5_avg=0.93329 lr=0.00100 gn=27.73525 time=63.37it/s +epoch=76 global_step=30100 loss=2.73279 loss_avg=2.22268 acc=0.75000 acc_top1_avg=0.79712 acc_top5_avg=0.93282 lr=0.00100 gn=30.12646 time=59.38it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.57124 test_loss_avg=0.58415 acc=0.82812 test_acc_avg=0.83628 test_acc_top5_avg=0.98998 time=256.05it/s +epoch=76 global_step=30107 loss=0.23918 test_loss_avg=0.43780 acc=0.93750 test_acc_avg=0.87441 test_acc_top5_avg=0.99248 time=544.15it/s +curr_acc 0.8744 +BEST_ACC 0.8958 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=2.00611 loss_avg=2.06629 acc=0.82812 acc_top1_avg=0.81250 acc_top5_avg=0.93550 lr=0.00100 gn=19.29265 time=55.86it/s +epoch=77 global_step=30200 loss=2.23559 loss_avg=2.16534 acc=0.78906 acc_top1_avg=0.80150 acc_top5_avg=0.93044 lr=0.00100 gn=22.85908 time=58.12it/s +epoch=77 global_step=30250 loss=2.18321 loss_avg=2.19159 acc=0.80469 acc_top1_avg=0.79944 acc_top5_avg=0.92914 lr=0.00100 gn=27.14384 time=61.41it/s +epoch=77 global_step=30300 loss=2.23147 loss_avg=2.21713 acc=0.79688 acc_top1_avg=0.79728 acc_top5_avg=0.92928 lr=0.00100 gn=17.98419 time=63.90it/s +epoch=77 global_step=30350 loss=2.15772 loss_avg=2.22291 acc=0.82031 acc_top1_avg=0.79697 acc_top5_avg=0.92949 lr=0.00100 gn=30.15441 time=62.03it/s +epoch=77 global_step=30400 loss=1.45344 loss_avg=2.21726 acc=0.86719 acc_top1_avg=0.79762 acc_top5_avg=0.92979 lr=0.00100 gn=17.67655 time=63.93it/s +epoch=77 global_step=30450 loss=1.52518 loss_avg=2.22399 acc=0.86719 acc_top1_avg=0.79713 acc_top5_avg=0.92953 lr=0.00100 gn=21.06767 time=58.57it/s +====================Eval==================== +epoch=77 global_step=30498 loss=0.76361 test_loss_avg=0.41609 acc=0.77344 test_acc_avg=0.87546 test_acc_top5_avg=0.99449 time=251.44it/s +epoch=77 global_step=30498 loss=0.15773 test_loss_avg=0.41415 acc=0.95312 test_acc_avg=0.88001 test_acc_top5_avg=0.99254 time=237.44it/s +epoch=77 global_step=30498 loss=0.29460 test_loss_avg=0.40610 acc=0.87500 test_acc_avg=0.88064 test_acc_top5_avg=0.99268 time=886.37it/s +curr_acc 0.8806 +BEST_ACC 0.8958 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=2.29207 loss_avg=2.47713 acc=0.79688 acc_top1_avg=0.76953 acc_top5_avg=0.93359 lr=0.00100 gn=23.95053 time=61.46it/s +epoch=78 global_step=30550 loss=2.27085 loss_avg=2.16040 acc=0.78906 acc_top1_avg=0.80303 acc_top5_avg=0.93960 lr=0.00100 gn=21.06567 time=48.46it/s +epoch=78 global_step=30600 loss=2.25187 loss_avg=2.20297 acc=0.78906 acc_top1_avg=0.79917 acc_top5_avg=0.93520 lr=0.00100 gn=21.21390 time=57.32it/s +epoch=78 global_step=30650 loss=3.06357 loss_avg=2.22007 acc=0.70312 acc_top1_avg=0.79759 acc_top5_avg=0.93452 lr=0.00100 gn=21.67728 time=57.99it/s +epoch=78 global_step=30700 loss=1.69384 loss_avg=2.22830 acc=0.86719 acc_top1_avg=0.79664 acc_top5_avg=0.93263 lr=0.00100 gn=27.14408 time=64.38it/s +epoch=78 global_step=30750 loss=2.35906 loss_avg=2.22977 acc=0.78906 acc_top1_avg=0.79660 acc_top5_avg=0.93279 lr=0.00100 gn=23.23926 time=57.06it/s +epoch=78 global_step=30800 loss=1.89290 loss_avg=2.21988 acc=0.82812 acc_top1_avg=0.79763 acc_top5_avg=0.93261 lr=0.00100 gn=17.52604 time=64.82it/s +epoch=78 global_step=30850 loss=2.19822 loss_avg=2.21635 acc=0.80469 acc_top1_avg=0.79796 acc_top5_avg=0.93217 lr=0.00100 gn=24.46112 time=54.78it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.25824 test_loss_avg=0.54793 acc=0.90625 test_acc_avg=0.83697 test_acc_top5_avg=0.99054 time=246.80it/s +epoch=78 global_step=30889 loss=0.06242 test_loss_avg=0.41546 acc=1.00000 test_acc_avg=0.87905 test_acc_top5_avg=0.99229 time=865.88it/s +curr_acc 0.8791 +BEST_ACC 0.8958 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=1.99737 loss_avg=2.16264 acc=0.82031 acc_top1_avg=0.80185 acc_top5_avg=0.93679 lr=0.00100 gn=22.63195 time=47.64it/s +epoch=79 global_step=30950 loss=2.64680 loss_avg=2.16523 acc=0.75000 acc_top1_avg=0.80174 acc_top5_avg=0.93353 lr=0.00100 gn=22.12787 time=49.47it/s +epoch=79 global_step=31000 loss=2.55348 loss_avg=2.18260 acc=0.77344 acc_top1_avg=0.80096 acc_top5_avg=0.93194 lr=0.00100 gn=25.85292 time=58.69it/s +epoch=79 global_step=31050 loss=2.42043 loss_avg=2.19886 acc=0.78125 acc_top1_avg=0.79945 acc_top5_avg=0.93143 lr=0.00100 gn=24.27301 time=53.02it/s +epoch=79 global_step=31100 loss=2.28991 loss_avg=2.21475 acc=0.79688 acc_top1_avg=0.79776 acc_top5_avg=0.93054 lr=0.00100 gn=25.84731 time=57.72it/s +epoch=79 global_step=31150 loss=2.02617 loss_avg=2.21397 acc=0.81250 acc_top1_avg=0.79795 acc_top5_avg=0.93115 lr=0.00100 gn=19.58847 time=54.83it/s +epoch=79 global_step=31200 loss=1.89858 loss_avg=2.22313 acc=0.82812 acc_top1_avg=0.79733 acc_top5_avg=0.93142 lr=0.00100 gn=29.38392 time=63.81it/s +epoch=79 global_step=31250 loss=1.77293 loss_avg=2.21868 acc=0.85156 acc_top1_avg=0.79774 acc_top5_avg=0.93155 lr=0.00100 gn=24.69399 time=61.99it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.23670 test_loss_avg=0.61832 acc=0.94531 test_acc_avg=0.81858 test_acc_top5_avg=0.98958 time=241.25it/s +epoch=79 global_step=31280 loss=0.16956 test_loss_avg=0.50152 acc=0.93750 test_acc_avg=0.85845 test_acc_top5_avg=0.99325 time=248.95it/s +epoch=79 global_step=31280 loss=0.19479 test_loss_avg=0.43503 acc=0.93750 test_acc_avg=0.87569 test_acc_top5_avg=0.99407 time=398.43it/s +curr_acc 0.8757 +BEST_ACC 0.8958 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=2.19758 loss_avg=2.15631 acc=0.80469 acc_top1_avg=0.80391 acc_top5_avg=0.93008 lr=0.00010 gn=21.33850 time=60.24it/s +epoch=80 global_step=31350 loss=2.70240 loss_avg=2.18025 acc=0.74219 acc_top1_avg=0.80022 acc_top5_avg=0.93315 lr=0.00010 gn=23.95799 time=63.15it/s +epoch=80 global_step=31400 loss=1.88904 loss_avg=2.16493 acc=0.83594 acc_top1_avg=0.80241 acc_top5_avg=0.93327 lr=0.00010 gn=24.59643 time=61.00it/s +epoch=80 global_step=31450 loss=2.22679 loss_avg=2.15843 acc=0.80469 acc_top1_avg=0.80290 acc_top5_avg=0.93304 lr=0.00010 gn=22.40429 time=54.76it/s +epoch=80 global_step=31500 loss=1.92060 loss_avg=2.14043 acc=0.85156 acc_top1_avg=0.80458 acc_top5_avg=0.93303 lr=0.00010 gn=24.55039 time=64.56it/s +epoch=80 global_step=31550 loss=1.90316 loss_avg=2.12335 acc=0.83594 acc_top1_avg=0.80622 acc_top5_avg=0.93328 lr=0.00010 gn=21.46338 time=50.85it/s +epoch=80 global_step=31600 loss=1.90251 loss_avg=2.11101 acc=0.82812 acc_top1_avg=0.80742 acc_top5_avg=0.93357 lr=0.00010 gn=25.22817 time=59.50it/s +epoch=80 global_step=31650 loss=1.78837 loss_avg=2.11753 acc=0.85156 acc_top1_avg=0.80667 acc_top5_avg=0.93281 lr=0.00010 gn=23.61537 time=64.20it/s +====================Eval==================== +epoch=80 global_step=31671 loss=0.46075 test_loss_avg=0.50085 acc=0.83594 test_acc_avg=0.85078 test_acc_top5_avg=0.99219 time=241.16it/s +epoch=80 global_step=31671 loss=0.09442 test_loss_avg=0.38002 acc=1.00000 test_acc_avg=0.88835 test_acc_top5_avg=0.99407 time=541.06it/s +curr_acc 0.8884 +BEST_ACC 0.8958 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=2.23117 loss_avg=2.07813 acc=0.78125 acc_top1_avg=0.80954 acc_top5_avg=0.93157 lr=0.00010 gn=16.76374 time=61.90it/s +epoch=81 global_step=31750 loss=1.41674 loss_avg=2.10235 acc=0.87500 acc_top1_avg=0.80667 acc_top5_avg=0.92880 lr=0.00010 gn=17.39306 time=60.76it/s +epoch=81 global_step=31800 loss=1.67439 loss_avg=2.11822 acc=0.85156 acc_top1_avg=0.80451 acc_top5_avg=0.93090 lr=0.00010 gn=21.42830 time=57.79it/s +epoch=81 global_step=31850 loss=1.87167 loss_avg=2.09030 acc=0.82031 acc_top1_avg=0.80770 acc_top5_avg=0.93261 lr=0.00010 gn=28.60612 time=57.18it/s +epoch=81 global_step=31900 loss=1.50576 loss_avg=2.08006 acc=0.87500 acc_top1_avg=0.80946 acc_top5_avg=0.93341 lr=0.00010 gn=21.02240 time=61.93it/s +epoch=81 global_step=31950 loss=1.54988 loss_avg=2.08876 acc=0.85938 acc_top1_avg=0.80872 acc_top5_avg=0.93235 lr=0.00010 gn=15.63934 time=64.73it/s +epoch=81 global_step=32000 loss=2.30969 loss_avg=2.07751 acc=0.78125 acc_top1_avg=0.81003 acc_top5_avg=0.93296 lr=0.00010 gn=17.21290 time=56.64it/s +epoch=81 global_step=32050 loss=2.36281 loss_avg=2.08132 acc=0.78125 acc_top1_avg=0.80955 acc_top5_avg=0.93264 lr=0.00010 gn=24.38420 time=64.13it/s +====================Eval==================== +epoch=81 global_step=32062 loss=0.77341 test_loss_avg=0.77341 acc=0.75000 test_acc_avg=0.75000 test_acc_top5_avg=0.99219 time=210.99it/s +epoch=81 global_step=32062 loss=0.15246 test_loss_avg=0.47559 acc=0.94531 test_acc_avg=0.86336 test_acc_top5_avg=0.99203 time=242.19it/s +epoch=81 global_step=32062 loss=0.12688 test_loss_avg=0.38500 acc=0.93750 test_acc_avg=0.88835 test_acc_top5_avg=0.99377 time=881.34it/s +curr_acc 0.8884 +BEST_ACC 0.8958 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=1.53008 loss_avg=2.07035 acc=0.87500 acc_top1_avg=0.81271 acc_top5_avg=0.93709 lr=0.00010 gn=24.08513 time=56.95it/s +epoch=82 global_step=32150 loss=2.37010 loss_avg=2.03931 acc=0.78906 acc_top1_avg=0.81499 acc_top5_avg=0.93608 lr=0.00010 gn=24.37506 time=63.89it/s +epoch=82 global_step=32200 loss=2.16173 loss_avg=2.05805 acc=0.80469 acc_top1_avg=0.81210 acc_top5_avg=0.93535 lr=0.00010 gn=23.91405 time=57.65it/s +epoch=82 global_step=32250 loss=1.76292 loss_avg=2.07933 acc=0.84375 acc_top1_avg=0.80901 acc_top5_avg=0.93310 lr=0.00010 gn=19.37883 time=63.39it/s +epoch=82 global_step=32300 loss=1.28224 loss_avg=2.06606 acc=0.89844 acc_top1_avg=0.81017 acc_top5_avg=0.93376 lr=0.00010 gn=22.84819 time=58.59it/s +epoch=82 global_step=32350 loss=2.13617 loss_avg=2.06538 acc=0.81250 acc_top1_avg=0.81052 acc_top5_avg=0.93392 lr=0.00010 gn=23.72734 time=53.10it/s +epoch=82 global_step=32400 loss=2.05855 loss_avg=2.06528 acc=0.81250 acc_top1_avg=0.81074 acc_top5_avg=0.93336 lr=0.00010 gn=19.93290 time=53.49it/s +epoch=82 global_step=32450 loss=2.25804 loss_avg=2.06289 acc=0.78906 acc_top1_avg=0.81103 acc_top5_avg=0.93323 lr=0.00010 gn=25.11328 time=63.42it/s +====================Eval==================== +epoch=82 global_step=32453 loss=0.64704 test_loss_avg=0.49000 acc=0.82031 test_acc_avg=0.85227 test_acc_top5_avg=0.99290 time=249.19it/s +epoch=82 global_step=32453 loss=0.54025 test_loss_avg=0.39149 acc=0.84375 test_acc_avg=0.88542 test_acc_top5_avg=0.99360 time=246.99it/s +epoch=82 global_step=32453 loss=0.22687 test_loss_avg=0.39514 acc=0.87500 test_acc_avg=0.88311 test_acc_top5_avg=0.99387 time=549.71it/s +curr_acc 0.8831 +BEST_ACC 0.8958 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=1.93556 loss_avg=2.05827 acc=0.81250 acc_top1_avg=0.81017 acc_top5_avg=0.93102 lr=0.00010 gn=24.70511 time=64.90it/s +epoch=83 global_step=32550 loss=2.10426 loss_avg=2.09024 acc=0.81250 acc_top1_avg=0.80710 acc_top5_avg=0.93082 lr=0.00010 gn=27.09686 time=42.73it/s +epoch=83 global_step=32600 loss=2.36045 loss_avg=2.04686 acc=0.78906 acc_top1_avg=0.81229 acc_top5_avg=0.93234 lr=0.00010 gn=25.04224 time=52.33it/s +epoch=83 global_step=32650 loss=2.17240 loss_avg=2.03810 acc=0.80469 acc_top1_avg=0.81309 acc_top5_avg=0.93318 lr=0.00010 gn=21.98184 time=56.45it/s +epoch=83 global_step=32700 loss=1.88982 loss_avg=2.04711 acc=0.82812 acc_top1_avg=0.81225 acc_top5_avg=0.93301 lr=0.00010 gn=18.48321 time=62.06it/s +epoch=83 global_step=32750 loss=1.94777 loss_avg=2.05414 acc=0.81250 acc_top1_avg=0.81126 acc_top5_avg=0.93292 lr=0.00010 gn=15.39256 time=59.95it/s +epoch=83 global_step=32800 loss=2.41359 loss_avg=2.05490 acc=0.77344 acc_top1_avg=0.81115 acc_top5_avg=0.93313 lr=0.00010 gn=25.74121 time=46.92it/s +====================Eval==================== +epoch=83 global_step=32844 loss=0.71519 test_loss_avg=0.46498 acc=0.81250 test_acc_avg=0.86392 test_acc_top5_avg=0.99382 time=259.89it/s +epoch=83 global_step=32844 loss=0.12304 test_loss_avg=0.37139 acc=0.93750 test_acc_avg=0.89102 test_acc_top5_avg=0.99476 time=854.24it/s +curr_acc 0.8910 +BEST_ACC 0.8958 +curr_acc_top5 0.9948 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=1.46146 loss_avg=1.82606 acc=0.86719 acc_top1_avg=0.83073 acc_top5_avg=0.93750 lr=0.00010 gn=18.02945 time=53.31it/s +epoch=84 global_step=32900 loss=1.50643 loss_avg=2.01312 acc=0.86719 acc_top1_avg=0.81487 acc_top5_avg=0.93234 lr=0.00010 gn=21.03107 time=62.26it/s +epoch=84 global_step=32950 loss=2.34431 loss_avg=2.01789 acc=0.78125 acc_top1_avg=0.81390 acc_top5_avg=0.93381 lr=0.00010 gn=20.32183 time=64.32it/s +epoch=84 global_step=33000 loss=1.98856 loss_avg=2.01530 acc=0.82031 acc_top1_avg=0.81390 acc_top5_avg=0.93485 lr=0.00010 gn=16.80538 time=56.89it/s +epoch=84 global_step=33050 loss=2.03005 loss_avg=2.02430 acc=0.82031 acc_top1_avg=0.81314 acc_top5_avg=0.93359 lr=0.00010 gn=25.89736 time=53.16it/s +epoch=84 global_step=33100 loss=2.38870 loss_avg=2.03496 acc=0.76562 acc_top1_avg=0.81235 acc_top5_avg=0.93256 lr=0.00010 gn=16.46461 time=64.18it/s +epoch=84 global_step=33150 loss=1.96509 loss_avg=2.03156 acc=0.81250 acc_top1_avg=0.81270 acc_top5_avg=0.93257 lr=0.00010 gn=19.95938 time=57.23it/s +epoch=84 global_step=33200 loss=1.67932 loss_avg=2.03561 acc=0.85156 acc_top1_avg=0.81257 acc_top5_avg=0.93313 lr=0.00010 gn=21.57970 time=58.33it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.18005 test_loss_avg=0.44323 acc=0.95312 test_acc_avg=0.86775 test_acc_top5_avg=0.99554 time=86.73it/s +epoch=84 global_step=33235 loss=0.18205 test_loss_avg=0.41277 acc=0.95312 test_acc_avg=0.87866 test_acc_top5_avg=0.99402 time=246.88it/s +epoch=84 global_step=33235 loss=0.14111 test_loss_avg=0.38375 acc=0.93750 test_acc_avg=0.88637 test_acc_top5_avg=0.99456 time=889.00it/s +curr_acc 0.8864 +BEST_ACC 0.8958 +curr_acc_top5 0.9946 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=1.85727 loss_avg=1.96005 acc=0.82812 acc_top1_avg=0.81979 acc_top5_avg=0.93906 lr=0.00010 gn=23.71835 time=56.11it/s +epoch=85 global_step=33300 loss=1.92032 loss_avg=1.99644 acc=0.82812 acc_top1_avg=0.81623 acc_top5_avg=0.93401 lr=0.00010 gn=17.34628 time=61.24it/s +epoch=85 global_step=33350 loss=2.48262 loss_avg=2.03496 acc=0.78125 acc_top1_avg=0.81284 acc_top5_avg=0.93098 lr=0.00010 gn=29.20690 time=59.70it/s +epoch=85 global_step=33400 loss=1.63061 loss_avg=2.03712 acc=0.84375 acc_top1_avg=0.81222 acc_top5_avg=0.93187 lr=0.00010 gn=18.19867 time=57.81it/s +epoch=85 global_step=33450 loss=2.16246 loss_avg=2.04537 acc=0.79688 acc_top1_avg=0.81156 acc_top5_avg=0.93176 lr=0.00010 gn=24.11421 time=53.79it/s +epoch=85 global_step=33500 loss=1.96107 loss_avg=2.03693 acc=0.81250 acc_top1_avg=0.81247 acc_top5_avg=0.93275 lr=0.00010 gn=14.53066 time=58.11it/s +epoch=85 global_step=33550 loss=2.25884 loss_avg=2.03314 acc=0.79688 acc_top1_avg=0.81292 acc_top5_avg=0.93366 lr=0.00010 gn=21.20642 time=64.45it/s +epoch=85 global_step=33600 loss=1.86985 loss_avg=2.03023 acc=0.84375 acc_top1_avg=0.81331 acc_top5_avg=0.93423 lr=0.00010 gn=31.10729 time=63.23it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.09865 test_loss_avg=0.46257 acc=0.98438 test_acc_avg=0.86540 test_acc_top5_avg=0.99219 time=253.37it/s +epoch=85 global_step=33626 loss=0.12453 test_loss_avg=0.38416 acc=0.93750 test_acc_avg=0.88855 test_acc_top5_avg=0.99347 time=510.44it/s +curr_acc 0.8885 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=1.56669 loss_avg=2.07401 acc=0.85156 acc_top1_avg=0.81120 acc_top5_avg=0.93685 lr=0.00010 gn=15.07849 time=57.44it/s +epoch=86 global_step=33700 loss=1.51425 loss_avg=2.06533 acc=0.86719 acc_top1_avg=0.81113 acc_top5_avg=0.93433 lr=0.00010 gn=19.51235 time=56.33it/s +epoch=86 global_step=33750 loss=2.14677 loss_avg=2.06121 acc=0.79688 acc_top1_avg=0.81099 acc_top5_avg=0.93315 lr=0.00010 gn=22.70358 time=58.11it/s +epoch=86 global_step=33800 loss=1.68817 loss_avg=2.03673 acc=0.85156 acc_top1_avg=0.81326 acc_top5_avg=0.93292 lr=0.00010 gn=21.65024 time=49.71it/s +epoch=86 global_step=33850 loss=2.24706 loss_avg=2.02793 acc=0.79688 acc_top1_avg=0.81407 acc_top5_avg=0.93356 lr=0.00010 gn=23.91744 time=63.38it/s +epoch=86 global_step=33900 loss=1.77148 loss_avg=2.02655 acc=0.84375 acc_top1_avg=0.81407 acc_top5_avg=0.93331 lr=0.00010 gn=22.27884 time=61.46it/s +epoch=86 global_step=33950 loss=1.82744 loss_avg=2.02031 acc=0.83594 acc_top1_avg=0.81477 acc_top5_avg=0.93376 lr=0.00010 gn=23.62445 time=55.54it/s +epoch=86 global_step=34000 loss=1.83265 loss_avg=2.02887 acc=0.84375 acc_top1_avg=0.81382 acc_top5_avg=0.93328 lr=0.00010 gn=24.61774 time=65.16it/s +====================Eval==================== +epoch=86 global_step=34017 loss=0.82630 test_loss_avg=0.74767 acc=0.78125 test_acc_avg=0.78385 test_acc_top5_avg=0.98828 time=238.60it/s +epoch=86 global_step=34017 loss=0.26713 test_loss_avg=0.45040 acc=0.91406 test_acc_avg=0.86872 test_acc_top5_avg=0.99289 time=255.33it/s +epoch=86 global_step=34017 loss=0.11112 test_loss_avg=0.38531 acc=1.00000 test_acc_avg=0.88766 test_acc_top5_avg=0.99407 time=887.87it/s +curr_acc 0.8877 +BEST_ACC 0.8958 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=2.65752 loss_avg=2.06782 acc=0.74219 acc_top1_avg=0.80895 acc_top5_avg=0.93063 lr=0.00010 gn=27.41282 time=61.63it/s +epoch=87 global_step=34100 loss=1.81211 loss_avg=2.06107 acc=0.83594 acc_top1_avg=0.80977 acc_top5_avg=0.93251 lr=0.00010 gn=17.09824 time=62.28it/s +epoch=87 global_step=34150 loss=2.42677 loss_avg=2.06274 acc=0.76562 acc_top1_avg=0.80986 acc_top5_avg=0.93257 lr=0.00010 gn=18.68269 time=62.31it/s +epoch=87 global_step=34200 loss=1.50837 loss_avg=2.05321 acc=0.87500 acc_top1_avg=0.81105 acc_top5_avg=0.93229 lr=0.00010 gn=21.45393 time=56.29it/s +epoch=87 global_step=34250 loss=2.11171 loss_avg=2.03869 acc=0.81250 acc_top1_avg=0.81260 acc_top5_avg=0.93301 lr=0.00010 gn=23.34079 time=57.42it/s +epoch=87 global_step=34300 loss=1.42229 loss_avg=2.03331 acc=0.88281 acc_top1_avg=0.81308 acc_top5_avg=0.93377 lr=0.00010 gn=25.80824 time=63.35it/s +epoch=87 global_step=34350 loss=1.90492 loss_avg=2.03273 acc=0.82812 acc_top1_avg=0.81304 acc_top5_avg=0.93393 lr=0.00010 gn=22.26297 time=54.09it/s +epoch=87 global_step=34400 loss=1.52683 loss_avg=2.02399 acc=0.86719 acc_top1_avg=0.81407 acc_top5_avg=0.93485 lr=0.00010 gn=16.33351 time=55.20it/s +====================Eval==================== +epoch=87 global_step=34408 loss=0.50688 test_loss_avg=0.50821 acc=0.84375 test_acc_avg=0.85446 test_acc_top5_avg=0.99103 time=239.85it/s +epoch=87 global_step=34408 loss=0.42877 test_loss_avg=0.39174 acc=0.87500 test_acc_avg=0.88657 test_acc_top5_avg=0.99391 time=249.97it/s +epoch=87 global_step=34408 loss=0.09184 test_loss_avg=0.38670 acc=0.93750 test_acc_avg=0.88736 test_acc_top5_avg=0.99407 time=535.60it/s +curr_acc 0.8874 +BEST_ACC 0.8958 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=2.34513 loss_avg=2.05916 acc=0.78125 acc_top1_avg=0.80915 acc_top5_avg=0.92987 lr=0.00010 gn=23.21365 time=52.72it/s +epoch=88 global_step=34500 loss=1.58786 loss_avg=2.02069 acc=0.85938 acc_top1_avg=0.81377 acc_top5_avg=0.93342 lr=0.00010 gn=14.97251 time=53.12it/s +epoch=88 global_step=34550 loss=1.58787 loss_avg=2.02863 acc=0.85938 acc_top1_avg=0.81278 acc_top5_avg=0.93326 lr=0.00010 gn=23.78061 time=60.88it/s +epoch=88 global_step=34600 loss=2.80278 loss_avg=2.03177 acc=0.73438 acc_top1_avg=0.81283 acc_top5_avg=0.93392 lr=0.00010 gn=23.61187 time=58.23it/s +epoch=88 global_step=34650 loss=1.72926 loss_avg=2.02649 acc=0.83594 acc_top1_avg=0.81327 acc_top5_avg=0.93430 lr=0.00010 gn=17.11834 time=64.45it/s +epoch=88 global_step=34700 loss=2.06805 loss_avg=2.02799 acc=0.80469 acc_top1_avg=0.81325 acc_top5_avg=0.93383 lr=0.00010 gn=19.14702 time=54.72it/s +epoch=88 global_step=34750 loss=2.10848 loss_avg=2.02228 acc=0.81250 acc_top1_avg=0.81369 acc_top5_avg=0.93405 lr=0.00010 gn=20.45896 time=56.53it/s +====================Eval==================== +epoch=88 global_step=34799 loss=0.24343 test_loss_avg=0.49827 acc=0.93750 test_acc_avg=0.85710 test_acc_top5_avg=0.99316 time=246.42it/s +epoch=88 global_step=34799 loss=0.11504 test_loss_avg=0.38853 acc=0.93750 test_acc_avg=0.88706 test_acc_top5_avg=0.99456 time=826.79it/s +curr_acc 0.8871 +BEST_ACC 0.8958 +curr_acc_top5 0.9946 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=1.98647 loss_avg=1.98647 acc=0.82031 acc_top1_avg=0.82031 acc_top5_avg=0.95312 lr=0.00010 gn=23.46980 time=53.03it/s +epoch=89 global_step=34850 loss=2.14277 loss_avg=1.97416 acc=0.80469 acc_top1_avg=0.81878 acc_top5_avg=0.93536 lr=0.00010 gn=17.97430 time=64.43it/s +epoch=89 global_step=34900 loss=2.34334 loss_avg=2.02391 acc=0.77344 acc_top1_avg=0.81304 acc_top5_avg=0.93239 lr=0.00010 gn=22.12227 time=59.68it/s +epoch=89 global_step=34950 loss=1.47450 loss_avg=2.01883 acc=0.87500 acc_top1_avg=0.81359 acc_top5_avg=0.93305 lr=0.00010 gn=21.20312 time=54.93it/s +epoch=89 global_step=35000 loss=1.67960 loss_avg=2.02257 acc=0.84375 acc_top1_avg=0.81359 acc_top5_avg=0.93272 lr=0.00010 gn=16.04107 time=61.00it/s +epoch=89 global_step=35050 loss=1.81618 loss_avg=2.02084 acc=0.85156 acc_top1_avg=0.81390 acc_top5_avg=0.93299 lr=0.00010 gn=25.69122 time=53.99it/s +epoch=89 global_step=35100 loss=2.41243 loss_avg=2.01138 acc=0.78906 acc_top1_avg=0.81502 acc_top5_avg=0.93428 lr=0.00010 gn=24.48934 time=64.90it/s +epoch=89 global_step=35150 loss=2.10107 loss_avg=2.01052 acc=0.80469 acc_top1_avg=0.81508 acc_top5_avg=0.93407 lr=0.00010 gn=17.38381 time=51.35it/s +====================Eval==================== +epoch=89 global_step=35190 loss=0.66066 test_loss_avg=0.50492 acc=0.80469 test_acc_avg=0.85567 test_acc_top5_avg=0.99383 time=239.69it/s +epoch=89 global_step=35190 loss=0.13840 test_loss_avg=0.40651 acc=0.96094 test_acc_avg=0.88270 test_acc_top5_avg=0.99400 time=260.45it/s +epoch=89 global_step=35190 loss=0.11408 test_loss_avg=0.39500 acc=1.00000 test_acc_avg=0.88588 test_acc_top5_avg=0.99407 time=888.06it/s +curr_acc 0.8859 +BEST_ACC 0.8958 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=1.75255 loss_avg=2.10734 acc=0.85156 acc_top1_avg=0.80625 acc_top5_avg=0.93047 lr=0.00010 gn=26.51939 time=50.67it/s +epoch=90 global_step=35250 loss=2.68323 loss_avg=2.06385 acc=0.75781 acc_top1_avg=0.80820 acc_top5_avg=0.93294 lr=0.00010 gn=22.20116 time=55.11it/s +epoch=90 global_step=35300 loss=2.30815 loss_avg=2.05127 acc=0.78906 acc_top1_avg=0.80994 acc_top5_avg=0.93352 lr=0.00010 gn=18.23220 time=54.99it/s +epoch=90 global_step=35350 loss=1.66840 loss_avg=2.01539 acc=0.84375 acc_top1_avg=0.81382 acc_top5_avg=0.93472 lr=0.00010 gn=23.06440 time=58.29it/s +epoch=90 global_step=35400 loss=2.06003 loss_avg=2.01244 acc=0.80469 acc_top1_avg=0.81403 acc_top5_avg=0.93575 lr=0.00010 gn=16.72696 time=63.81it/s +epoch=90 global_step=35450 loss=1.72034 loss_avg=2.00855 acc=0.84375 acc_top1_avg=0.81469 acc_top5_avg=0.93501 lr=0.00010 gn=25.51809 time=65.14it/s +epoch=90 global_step=35500 loss=2.05401 loss_avg=2.01731 acc=0.80469 acc_top1_avg=0.81381 acc_top5_avg=0.93369 lr=0.00010 gn=17.64486 time=55.17it/s +epoch=90 global_step=35550 loss=1.76470 loss_avg=2.00725 acc=0.82812 acc_top1_avg=0.81491 acc_top5_avg=0.93437 lr=0.00010 gn=24.49724 time=63.89it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.79483 test_loss_avg=0.44430 acc=0.77344 test_acc_avg=0.87012 test_acc_top5_avg=0.99375 time=246.58it/s +epoch=90 global_step=35581 loss=0.13585 test_loss_avg=0.38483 acc=0.93750 test_acc_avg=0.88805 test_acc_top5_avg=0.99417 time=889.75it/s +curr_acc 0.8881 +BEST_ACC 0.8958 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=1.37613 loss_avg=1.98861 acc=0.87500 acc_top1_avg=0.81743 acc_top5_avg=0.94531 lr=0.00010 gn=14.59568 time=62.36it/s +epoch=91 global_step=35650 loss=2.33554 loss_avg=1.97153 acc=0.78125 acc_top1_avg=0.81850 acc_top5_avg=0.93863 lr=0.00010 gn=19.32271 time=27.62it/s +epoch=91 global_step=35700 loss=2.30855 loss_avg=1.99317 acc=0.78125 acc_top1_avg=0.81657 acc_top5_avg=0.93481 lr=0.00010 gn=13.86470 time=63.78it/s +epoch=91 global_step=35750 loss=2.20712 loss_avg=2.01368 acc=0.78906 acc_top1_avg=0.81430 acc_top5_avg=0.93311 lr=0.00010 gn=22.01750 time=52.46it/s +epoch=91 global_step=35800 loss=1.51553 loss_avg=2.01137 acc=0.86719 acc_top1_avg=0.81436 acc_top5_avg=0.93347 lr=0.00010 gn=26.03953 time=56.79it/s +epoch=91 global_step=35850 loss=2.18307 loss_avg=2.01371 acc=0.79688 acc_top1_avg=0.81418 acc_top5_avg=0.93372 lr=0.00010 gn=24.03612 time=63.86it/s +epoch=91 global_step=35900 loss=1.77858 loss_avg=2.00994 acc=0.83594 acc_top1_avg=0.81475 acc_top5_avg=0.93412 lr=0.00010 gn=20.24697 time=52.19it/s +epoch=91 global_step=35950 loss=1.90844 loss_avg=2.00634 acc=0.82031 acc_top1_avg=0.81496 acc_top5_avg=0.93443 lr=0.00010 gn=20.84067 time=58.55it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.13452 test_loss_avg=0.57080 acc=0.96875 test_acc_avg=0.83452 test_acc_top5_avg=0.99148 time=256.27it/s +epoch=91 global_step=35972 loss=0.12735 test_loss_avg=0.43018 acc=0.96094 test_acc_avg=0.87577 test_acc_top5_avg=0.99321 time=219.94it/s +epoch=91 global_step=35972 loss=0.17080 test_loss_avg=0.38947 acc=0.93750 test_acc_avg=0.88588 test_acc_top5_avg=0.99377 time=548.20it/s +curr_acc 0.8859 +BEST_ACC 0.8958 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=2.83305 loss_avg=2.05783 acc=0.72656 acc_top1_avg=0.81027 acc_top5_avg=0.93331 lr=0.00010 gn=22.54372 time=59.76it/s +epoch=92 global_step=36050 loss=1.77985 loss_avg=2.01419 acc=0.83594 acc_top1_avg=0.81430 acc_top5_avg=0.93249 lr=0.00010 gn=22.79216 time=41.21it/s +epoch=92 global_step=36100 loss=1.55873 loss_avg=2.00515 acc=0.85938 acc_top1_avg=0.81549 acc_top5_avg=0.93500 lr=0.00010 gn=22.42367 time=64.30it/s +epoch=92 global_step=36150 loss=2.01742 loss_avg=1.99949 acc=0.81250 acc_top1_avg=0.81579 acc_top5_avg=0.93566 lr=0.00010 gn=21.90326 time=53.12it/s +epoch=92 global_step=36200 loss=2.22317 loss_avg=2.01549 acc=0.79688 acc_top1_avg=0.81432 acc_top5_avg=0.93431 lr=0.00010 gn=22.55059 time=62.99it/s +epoch=92 global_step=36250 loss=2.17925 loss_avg=2.01609 acc=0.78906 acc_top1_avg=0.81447 acc_top5_avg=0.93430 lr=0.00010 gn=24.04061 time=61.75it/s +epoch=92 global_step=36300 loss=1.96834 loss_avg=2.01479 acc=0.82031 acc_top1_avg=0.81457 acc_top5_avg=0.93424 lr=0.00010 gn=23.51557 time=60.22it/s +epoch=92 global_step=36350 loss=1.68887 loss_avg=2.00724 acc=0.85938 acc_top1_avg=0.81543 acc_top5_avg=0.93485 lr=0.00010 gn=25.21985 time=64.58it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.25255 test_loss_avg=0.48312 acc=0.92188 test_acc_avg=0.85718 test_acc_top5_avg=0.99121 time=260.19it/s +epoch=92 global_step=36363 loss=0.18808 test_loss_avg=0.38654 acc=0.93750 test_acc_avg=0.88548 test_acc_top5_avg=0.99347 time=558.79it/s +curr_acc 0.8855 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=1.54887 loss_avg=1.95828 acc=0.85938 acc_top1_avg=0.81989 acc_top5_avg=0.93666 lr=0.00010 gn=14.99040 time=49.23it/s +epoch=93 global_step=36450 loss=1.92320 loss_avg=1.96833 acc=0.82031 acc_top1_avg=0.81897 acc_top5_avg=0.93669 lr=0.00010 gn=21.03865 time=57.73it/s +epoch=93 global_step=36500 loss=1.72070 loss_avg=1.96958 acc=0.83594 acc_top1_avg=0.81889 acc_top5_avg=0.93727 lr=0.00010 gn=14.90955 time=58.32it/s +epoch=93 global_step=36550 loss=1.80498 loss_avg=1.97092 acc=0.84375 acc_top1_avg=0.81885 acc_top5_avg=0.93671 lr=0.00010 gn=25.46051 time=62.16it/s +epoch=93 global_step=36600 loss=2.36614 loss_avg=1.99130 acc=0.78125 acc_top1_avg=0.81639 acc_top5_avg=0.93562 lr=0.00010 gn=22.16057 time=64.67it/s +epoch=93 global_step=36650 loss=2.31788 loss_avg=2.00692 acc=0.79688 acc_top1_avg=0.81492 acc_top5_avg=0.93461 lr=0.00010 gn=29.55162 time=56.19it/s +epoch=93 global_step=36700 loss=1.84869 loss_avg=1.99041 acc=0.83594 acc_top1_avg=0.81667 acc_top5_avg=0.93497 lr=0.00010 gn=15.80282 time=59.17it/s +epoch=93 global_step=36750 loss=2.04708 loss_avg=1.99156 acc=0.80469 acc_top1_avg=0.81656 acc_top5_avg=0.93475 lr=0.00010 gn=19.71912 time=61.36it/s +====================Eval==================== +epoch=93 global_step=36754 loss=0.83240 test_loss_avg=0.82058 acc=0.76562 test_acc_avg=0.75521 test_acc_top5_avg=0.98177 time=241.12it/s +epoch=93 global_step=36754 loss=0.15398 test_loss_avg=0.46490 acc=0.96875 test_acc_avg=0.86468 test_acc_top5_avg=0.99204 time=247.48it/s +epoch=93 global_step=36754 loss=0.20725 test_loss_avg=0.39499 acc=0.87500 test_acc_avg=0.88331 test_acc_top5_avg=0.99347 time=883.94it/s +curr_acc 0.8833 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=1.21788 loss_avg=1.95932 acc=0.90625 acc_top1_avg=0.82286 acc_top5_avg=0.93529 lr=0.00010 gn=21.45889 time=56.74it/s +epoch=94 global_step=36850 loss=1.89327 loss_avg=1.96314 acc=0.83594 acc_top1_avg=0.82088 acc_top5_avg=0.93416 lr=0.00010 gn=24.78513 time=63.04it/s +epoch=94 global_step=36900 loss=2.67716 loss_avg=1.96251 acc=0.73438 acc_top1_avg=0.82026 acc_top5_avg=0.93429 lr=0.00010 gn=20.30981 time=64.22it/s +epoch=94 global_step=36950 loss=2.09559 loss_avg=1.99744 acc=0.80469 acc_top1_avg=0.81645 acc_top5_avg=0.93335 lr=0.00010 gn=24.10927 time=58.05it/s +epoch=94 global_step=37000 loss=2.33604 loss_avg=2.00260 acc=0.78125 acc_top1_avg=0.81561 acc_top5_avg=0.93382 lr=0.00010 gn=22.37628 time=54.31it/s +epoch=94 global_step=37050 loss=1.34414 loss_avg=2.00650 acc=0.88281 acc_top1_avg=0.81554 acc_top5_avg=0.93338 lr=0.00010 gn=16.33977 time=51.65it/s +epoch=94 global_step=37100 loss=2.30392 loss_avg=2.00114 acc=0.77344 acc_top1_avg=0.81593 acc_top5_avg=0.93416 lr=0.00010 gn=23.28778 time=64.92it/s +====================Eval==================== +epoch=94 global_step=37145 loss=0.78233 test_loss_avg=0.49340 acc=0.80469 test_acc_avg=0.85775 test_acc_top5_avg=0.99251 time=254.66it/s +epoch=94 global_step=37145 loss=0.36672 test_loss_avg=0.39245 acc=0.90625 test_acc_avg=0.88788 test_acc_top5_avg=0.99303 time=246.55it/s +epoch=94 global_step=37145 loss=0.16358 test_loss_avg=0.38858 acc=0.87500 test_acc_avg=0.88697 test_acc_top5_avg=0.99337 time=544.15it/s +curr_acc 0.8870 +BEST_ACC 0.8958 +curr_acc_top5 0.9934 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=2.28354 loss_avg=2.13249 acc=0.78906 acc_top1_avg=0.80625 acc_top5_avg=0.93750 lr=0.00010 gn=25.10026 time=60.96it/s +epoch=95 global_step=37200 loss=1.77779 loss_avg=1.98004 acc=0.83594 acc_top1_avg=0.81747 acc_top5_avg=0.93651 lr=0.00010 gn=23.37120 time=63.23it/s +epoch=95 global_step=37250 loss=2.41530 loss_avg=1.99952 acc=0.78125 acc_top1_avg=0.81607 acc_top5_avg=0.93400 lr=0.00010 gn=26.57649 time=56.89it/s +epoch=95 global_step=37300 loss=2.11503 loss_avg=2.02215 acc=0.81250 acc_top1_avg=0.81406 acc_top5_avg=0.93382 lr=0.00010 gn=21.14279 time=59.71it/s +epoch=95 global_step=37350 loss=2.14835 loss_avg=2.02163 acc=0.78906 acc_top1_avg=0.81399 acc_top5_avg=0.93495 lr=0.00010 gn=23.62257 time=54.29it/s +epoch=95 global_step=37400 loss=1.95059 loss_avg=2.01844 acc=0.82031 acc_top1_avg=0.81428 acc_top5_avg=0.93493 lr=0.00010 gn=25.52661 time=54.76it/s +epoch=95 global_step=37450 loss=2.00169 loss_avg=1.99935 acc=0.82031 acc_top1_avg=0.81603 acc_top5_avg=0.93617 lr=0.00010 gn=26.31881 time=59.98it/s +epoch=95 global_step=37500 loss=2.29454 loss_avg=2.00665 acc=0.78125 acc_top1_avg=0.81503 acc_top5_avg=0.93484 lr=0.00010 gn=23.70997 time=58.62it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.68636 test_loss_avg=0.50767 acc=0.79688 test_acc_avg=0.85382 test_acc_top5_avg=0.99253 time=249.13it/s +epoch=95 global_step=37536 loss=0.13269 test_loss_avg=0.39276 acc=0.87500 test_acc_avg=0.88509 test_acc_top5_avg=0.99397 time=615.99it/s +curr_acc 0.8851 +BEST_ACC 0.8958 +curr_acc_top5 0.9940 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=2.03519 loss_avg=2.00851 acc=0.81250 acc_top1_avg=0.81473 acc_top5_avg=0.93471 lr=0.00010 gn=22.84294 time=60.01it/s +epoch=96 global_step=37600 loss=2.10790 loss_avg=2.02586 acc=0.81250 acc_top1_avg=0.81250 acc_top5_avg=0.93115 lr=0.00010 gn=23.11616 time=60.97it/s +epoch=96 global_step=37650 loss=2.15444 loss_avg=2.03836 acc=0.79688 acc_top1_avg=0.81147 acc_top5_avg=0.93078 lr=0.00010 gn=16.56692 time=60.72it/s +epoch=96 global_step=37700 loss=2.12072 loss_avg=2.02558 acc=0.80469 acc_top1_avg=0.81317 acc_top5_avg=0.93140 lr=0.00010 gn=21.74005 time=50.97it/s +epoch=96 global_step=37750 loss=1.64114 loss_avg=2.00181 acc=0.84375 acc_top1_avg=0.81575 acc_top5_avg=0.93389 lr=0.00010 gn=19.06851 time=63.99it/s +epoch=96 global_step=37800 loss=1.80443 loss_avg=1.99072 acc=0.82812 acc_top1_avg=0.81673 acc_top5_avg=0.93469 lr=0.00010 gn=23.28069 time=53.32it/s +epoch=96 global_step=37850 loss=1.87765 loss_avg=1.98667 acc=0.82812 acc_top1_avg=0.81735 acc_top5_avg=0.93486 lr=0.00010 gn=20.42328 time=61.81it/s +epoch=96 global_step=37900 loss=2.22887 loss_avg=1.99551 acc=0.78906 acc_top1_avg=0.81664 acc_top5_avg=0.93413 lr=0.00010 gn=24.26437 time=55.69it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.48315 test_loss_avg=0.44751 acc=0.85156 test_acc_avg=0.87109 test_acc_top5_avg=0.99414 time=239.41it/s +epoch=96 global_step=37927 loss=0.14048 test_loss_avg=0.40811 acc=0.96094 test_acc_avg=0.88116 test_acc_top5_avg=0.99384 time=247.07it/s +epoch=96 global_step=37927 loss=0.11054 test_loss_avg=0.38460 acc=0.93750 test_acc_avg=0.88716 test_acc_top5_avg=0.99417 time=885.62it/s +curr_acc 0.8872 +BEST_ACC 0.8958 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=2.17695 loss_avg=2.09834 acc=0.80469 acc_top1_avg=0.80503 acc_top5_avg=0.93444 lr=0.00010 gn=23.49302 time=49.57it/s +epoch=97 global_step=38000 loss=2.20172 loss_avg=1.99047 acc=0.80469 acc_top1_avg=0.81635 acc_top5_avg=0.93429 lr=0.00010 gn=26.20517 time=57.35it/s +epoch=97 global_step=38050 loss=2.42158 loss_avg=1.98047 acc=0.77344 acc_top1_avg=0.81752 acc_top5_avg=0.93445 lr=0.00010 gn=21.90394 time=61.57it/s +epoch=97 global_step=38100 loss=1.92312 loss_avg=1.97742 acc=0.82812 acc_top1_avg=0.81747 acc_top5_avg=0.93475 lr=0.00010 gn=18.36823 time=57.42it/s +epoch=97 global_step=38150 loss=2.04632 loss_avg=1.98465 acc=0.80469 acc_top1_avg=0.81674 acc_top5_avg=0.93554 lr=0.00010 gn=16.39951 time=61.44it/s +epoch=97 global_step=38200 loss=1.68722 loss_avg=1.98299 acc=0.85156 acc_top1_avg=0.81725 acc_top5_avg=0.93584 lr=0.00010 gn=21.79276 time=64.68it/s +epoch=97 global_step=38250 loss=2.07047 loss_avg=1.99101 acc=0.81250 acc_top1_avg=0.81647 acc_top5_avg=0.93552 lr=0.00010 gn=20.51056 time=63.38it/s +epoch=97 global_step=38300 loss=1.67990 loss_avg=1.99399 acc=0.85156 acc_top1_avg=0.81614 acc_top5_avg=0.93524 lr=0.00010 gn=22.37357 time=64.39it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.14828 test_loss_avg=0.49441 acc=0.94531 test_acc_avg=0.85241 test_acc_top5_avg=0.99198 time=259.29it/s +epoch=97 global_step=38318 loss=0.20030 test_loss_avg=0.40634 acc=0.87500 test_acc_avg=0.88074 test_acc_top5_avg=0.99377 time=548.20it/s +curr_acc 0.8807 +BEST_ACC 0.8958 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=1.73296 loss_avg=1.91094 acc=0.82812 acc_top1_avg=0.82275 acc_top5_avg=0.93896 lr=0.00010 gn=21.85298 time=51.92it/s +epoch=98 global_step=38400 loss=2.01432 loss_avg=1.91000 acc=0.82031 acc_top1_avg=0.82393 acc_top5_avg=0.93779 lr=0.00010 gn=18.80029 time=60.31it/s +epoch=98 global_step=38450 loss=2.27326 loss_avg=1.93825 acc=0.79688 acc_top1_avg=0.82185 acc_top5_avg=0.93590 lr=0.00010 gn=24.41736 time=61.58it/s +epoch=98 global_step=38500 loss=2.77741 loss_avg=1.94292 acc=0.73438 acc_top1_avg=0.82130 acc_top5_avg=0.93557 lr=0.00010 gn=20.85832 time=64.17it/s +epoch=98 global_step=38550 loss=1.94315 loss_avg=1.97325 acc=0.82031 acc_top1_avg=0.81839 acc_top5_avg=0.93420 lr=0.00010 gn=22.84941 time=59.89it/s +epoch=98 global_step=38600 loss=2.11379 loss_avg=1.96949 acc=0.79688 acc_top1_avg=0.81851 acc_top5_avg=0.93395 lr=0.00010 gn=30.71838 time=55.73it/s +epoch=98 global_step=38650 loss=2.32498 loss_avg=1.98081 acc=0.78125 acc_top1_avg=0.81749 acc_top5_avg=0.93437 lr=0.00010 gn=30.61100 time=64.84it/s +epoch=98 global_step=38700 loss=1.89543 loss_avg=1.97794 acc=0.83594 acc_top1_avg=0.81796 acc_top5_avg=0.93460 lr=0.00010 gn=27.77241 time=60.79it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.67404 test_loss_avg=0.78249 acc=0.78906 test_acc_avg=0.77148 test_acc_top5_avg=0.98730 time=237.77it/s +epoch=98 global_step=38709 loss=0.20436 test_loss_avg=0.45253 acc=0.95312 test_acc_avg=0.86894 test_acc_top5_avg=0.99286 time=253.74it/s +epoch=98 global_step=38709 loss=0.17605 test_loss_avg=0.40216 acc=0.87500 test_acc_avg=0.88163 test_acc_top5_avg=0.99387 time=715.63it/s +curr_acc 0.8816 +BEST_ACC 0.8958 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=2.05075 loss_avg=1.99551 acc=0.82031 acc_top1_avg=0.81688 acc_top5_avg=0.93426 lr=0.00010 gn=23.00922 time=57.42it/s +epoch=99 global_step=38800 loss=1.63303 loss_avg=1.99063 acc=0.85156 acc_top1_avg=0.81748 acc_top5_avg=0.93389 lr=0.00010 gn=21.53269 time=61.27it/s +epoch=99 global_step=38850 loss=1.77468 loss_avg=1.98538 acc=0.85156 acc_top1_avg=0.81760 acc_top5_avg=0.93401 lr=0.00010 gn=24.35861 time=57.14it/s +epoch=99 global_step=38900 loss=2.35788 loss_avg=1.97487 acc=0.78125 acc_top1_avg=0.81847 acc_top5_avg=0.93525 lr=0.00010 gn=22.05216 time=56.42it/s +epoch=99 global_step=38950 loss=1.73641 loss_avg=1.98232 acc=0.85156 acc_top1_avg=0.81769 acc_top5_avg=0.93500 lr=0.00010 gn=21.75798 time=47.19it/s +epoch=99 global_step=39000 loss=2.24701 loss_avg=1.98624 acc=0.78906 acc_top1_avg=0.81701 acc_top5_avg=0.93465 lr=0.00010 gn=19.61203 time=54.61it/s +epoch=99 global_step=39050 loss=2.01302 loss_avg=1.98226 acc=0.81250 acc_top1_avg=0.81743 acc_top5_avg=0.93457 lr=0.00010 gn=21.84629 time=57.47it/s +epoch=99 global_step=39100 loss=1.93698 loss_avg=1.98113 acc=0.82500 acc_top1_avg=0.81761 acc_top5_avg=0.93439 lr=0.00010 gn=26.31873 time=81.03it/s +====================Eval==================== +epoch=99 global_step=39100 loss=0.50720 test_loss_avg=0.51439 acc=0.84375 test_acc_avg=0.85129 test_acc_top5_avg=0.99165 time=244.10it/s +epoch=99 global_step=39100 loss=0.13204 test_loss_avg=0.39592 acc=0.93750 test_acc_avg=0.88479 test_acc_top5_avg=0.99328 time=483.44it/s +epoch=99 global_step=39100 loss=0.13204 test_loss_avg=0.39592 acc=0.93750 test_acc_avg=0.88479 test_acc_top5_avg=0.99328 time=483.44it/s +curr_acc 0.8848 +BEST_ACC 0.8958 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=1.89448 loss_avg=1.97512 acc=0.82812 acc_top1_avg=0.81781 acc_top5_avg=0.93672 lr=0.00010 gn=24.26788 time=63.59it/s +epoch=100 global_step=39200 loss=2.06726 loss_avg=1.96570 acc=0.81250 acc_top1_avg=0.81867 acc_top5_avg=0.93461 lr=0.00010 gn=18.75479 time=64.57it/s +epoch=100 global_step=39250 loss=2.83288 loss_avg=1.98805 acc=0.71875 acc_top1_avg=0.81708 acc_top5_avg=0.93495 lr=0.00010 gn=29.21252 time=61.49it/s +epoch=100 global_step=39300 loss=1.61733 loss_avg=1.98456 acc=0.85156 acc_top1_avg=0.81723 acc_top5_avg=0.93410 lr=0.00010 gn=18.55377 time=54.61it/s +epoch=100 global_step=39350 loss=2.01777 loss_avg=1.97862 acc=0.80469 acc_top1_avg=0.81791 acc_top5_avg=0.93497 lr=0.00010 gn=18.69935 time=64.94it/s +epoch=100 global_step=39400 loss=2.14648 loss_avg=1.97949 acc=0.79688 acc_top1_avg=0.81792 acc_top5_avg=0.93471 lr=0.00010 gn=20.90725 time=60.69it/s +epoch=100 global_step=39450 loss=1.61489 loss_avg=1.98104 acc=0.85938 acc_top1_avg=0.81775 acc_top5_avg=0.93518 lr=0.00010 gn=27.11831 time=54.93it/s +====================Eval==================== +epoch=100 global_step=39491 loss=0.12239 test_loss_avg=0.48754 acc=0.96094 test_acc_avg=0.85719 test_acc_top5_avg=0.99203 time=257.07it/s +epoch=100 global_step=39491 loss=0.14864 test_loss_avg=0.39939 acc=0.93750 test_acc_avg=0.88301 test_acc_top5_avg=0.99367 time=553.05it/s +curr_acc 0.8830 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=2.21380 loss_avg=1.83361 acc=0.78906 acc_top1_avg=0.83247 acc_top5_avg=0.94358 lr=0.00010 gn=19.64288 time=63.63it/s +epoch=101 global_step=39550 loss=1.86717 loss_avg=1.97579 acc=0.82812 acc_top1_avg=0.81978 acc_top5_avg=0.93353 lr=0.00010 gn=13.43773 time=50.31it/s +epoch=101 global_step=39600 loss=1.73403 loss_avg=2.00372 acc=0.84375 acc_top1_avg=0.81673 acc_top5_avg=0.93334 lr=0.00010 gn=15.08116 time=64.36it/s +epoch=101 global_step=39650 loss=1.63163 loss_avg=1.95826 acc=0.85938 acc_top1_avg=0.82090 acc_top5_avg=0.93539 lr=0.00010 gn=29.59704 time=63.94it/s +epoch=101 global_step=39700 loss=2.77480 loss_avg=1.98052 acc=0.74219 acc_top1_avg=0.81897 acc_top5_avg=0.93402 lr=0.00010 gn=21.74050 time=61.40it/s +epoch=101 global_step=39750 loss=2.12813 loss_avg=1.97815 acc=0.79688 acc_top1_avg=0.81892 acc_top5_avg=0.93400 lr=0.00010 gn=23.59907 time=57.03it/s +epoch=101 global_step=39800 loss=1.89359 loss_avg=1.96845 acc=0.82812 acc_top1_avg=0.81981 acc_top5_avg=0.93439 lr=0.00010 gn=21.52291 time=53.20it/s +epoch=101 global_step=39850 loss=2.06474 loss_avg=1.97401 acc=0.81250 acc_top1_avg=0.81925 acc_top5_avg=0.93400 lr=0.00010 gn=25.54074 time=58.20it/s +====================Eval==================== +epoch=101 global_step=39882 loss=0.62558 test_loss_avg=0.50476 acc=0.81250 test_acc_avg=0.85193 test_acc_top5_avg=0.99405 time=247.66it/s +epoch=101 global_step=39882 loss=0.25862 test_loss_avg=0.39861 acc=0.90625 test_acc_avg=0.88413 test_acc_top5_avg=0.99373 time=260.45it/s +epoch=101 global_step=39882 loss=0.12236 test_loss_avg=0.39593 acc=0.93750 test_acc_avg=0.88350 test_acc_top5_avg=0.99367 time=867.49it/s +curr_acc 0.8835 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=1.91239 loss_avg=1.99800 acc=0.82812 acc_top1_avg=0.81727 acc_top5_avg=0.93446 lr=0.00010 gn=23.54989 time=54.91it/s +epoch=102 global_step=39950 loss=1.79221 loss_avg=1.94801 acc=0.83594 acc_top1_avg=0.82100 acc_top5_avg=0.93428 lr=0.00010 gn=27.93554 time=54.75it/s +epoch=102 global_step=40000 loss=1.86586 loss_avg=1.95586 acc=0.82812 acc_top1_avg=0.82018 acc_top5_avg=0.93532 lr=0.00010 gn=24.44029 time=57.26it/s +epoch=102 global_step=40050 loss=1.56084 loss_avg=1.96050 acc=0.85938 acc_top1_avg=0.81971 acc_top5_avg=0.93494 lr=0.00010 gn=19.93657 time=58.99it/s +epoch=102 global_step=40100 loss=1.55222 loss_avg=1.97167 acc=0.86719 acc_top1_avg=0.81866 acc_top5_avg=0.93524 lr=0.00010 gn=26.80878 time=64.33it/s +epoch=102 global_step=40150 loss=1.83384 loss_avg=1.96414 acc=0.83594 acc_top1_avg=0.81967 acc_top5_avg=0.93549 lr=0.00010 gn=21.50565 time=62.36it/s +epoch=102 global_step=40200 loss=1.96223 loss_avg=1.97184 acc=0.81250 acc_top1_avg=0.81899 acc_top5_avg=0.93531 lr=0.00010 gn=26.94758 time=60.55it/s +epoch=102 global_step=40250 loss=2.03937 loss_avg=1.98100 acc=0.79688 acc_top1_avg=0.81783 acc_top5_avg=0.93502 lr=0.00010 gn=23.13084 time=52.60it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.62357 test_loss_avg=0.48376 acc=0.84375 test_acc_avg=0.86012 test_acc_top5_avg=0.99126 time=248.14it/s +epoch=102 global_step=40273 loss=0.13768 test_loss_avg=0.40015 acc=0.87500 test_acc_avg=0.88321 test_acc_top5_avg=0.99298 time=859.66it/s +curr_acc 0.8832 +BEST_ACC 0.8958 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=2.35627 loss_avg=1.93948 acc=0.77344 acc_top1_avg=0.82292 acc_top5_avg=0.93287 lr=0.00010 gn=20.40678 time=53.32it/s +epoch=103 global_step=40350 loss=1.93549 loss_avg=1.94301 acc=0.82031 acc_top1_avg=0.82163 acc_top5_avg=0.93537 lr=0.00010 gn=23.07007 time=58.48it/s +epoch=103 global_step=40400 loss=2.25062 loss_avg=1.95941 acc=0.78906 acc_top1_avg=0.82000 acc_top5_avg=0.93418 lr=0.00010 gn=15.11788 time=53.45it/s +epoch=103 global_step=40450 loss=1.88315 loss_avg=1.93973 acc=0.82812 acc_top1_avg=0.82195 acc_top5_avg=0.93543 lr=0.00010 gn=25.94638 time=57.67it/s +epoch=103 global_step=40500 loss=2.24510 loss_avg=1.94926 acc=0.78125 acc_top1_avg=0.82097 acc_top5_avg=0.93550 lr=0.00010 gn=14.10317 time=55.53it/s +epoch=103 global_step=40550 loss=2.10329 loss_avg=1.96283 acc=0.81250 acc_top1_avg=0.81983 acc_top5_avg=0.93462 lr=0.00010 gn=23.80062 time=64.83it/s +epoch=103 global_step=40600 loss=1.96218 loss_avg=1.96338 acc=0.82812 acc_top1_avg=0.82005 acc_top5_avg=0.93513 lr=0.00010 gn=24.23592 time=58.61it/s +epoch=103 global_step=40650 loss=1.99018 loss_avg=1.96301 acc=0.81250 acc_top1_avg=0.82004 acc_top5_avg=0.93524 lr=0.00010 gn=19.07569 time=61.25it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.07112 test_loss_avg=0.47179 acc=0.99219 test_acc_avg=0.86358 test_acc_top5_avg=0.99339 time=247.47it/s +epoch=103 global_step=40664 loss=0.09523 test_loss_avg=0.42178 acc=0.96875 test_acc_avg=0.87736 test_acc_top5_avg=0.99293 time=246.71it/s +epoch=103 global_step=40664 loss=0.15197 test_loss_avg=0.39587 acc=0.93750 test_acc_avg=0.88439 test_acc_top5_avg=0.99347 time=549.57it/s +curr_acc 0.8844 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=2.04315 loss_avg=1.98237 acc=0.81250 acc_top1_avg=0.81901 acc_top5_avg=0.92665 lr=0.00010 gn=25.20052 time=64.70it/s +epoch=104 global_step=40750 loss=1.91412 loss_avg=2.00651 acc=0.82031 acc_top1_avg=0.81559 acc_top5_avg=0.92805 lr=0.00010 gn=20.85221 time=63.88it/s +epoch=104 global_step=40800 loss=2.21204 loss_avg=1.98598 acc=0.78906 acc_top1_avg=0.81819 acc_top5_avg=0.93210 lr=0.00010 gn=21.48008 time=62.02it/s +epoch=104 global_step=40850 loss=2.14290 loss_avg=1.97557 acc=0.81250 acc_top1_avg=0.81918 acc_top5_avg=0.93246 lr=0.00010 gn=24.09795 time=62.75it/s +epoch=104 global_step=40900 loss=1.78501 loss_avg=1.97036 acc=0.84375 acc_top1_avg=0.81952 acc_top5_avg=0.93356 lr=0.00010 gn=24.87152 time=58.65it/s +epoch=104 global_step=40950 loss=2.84884 loss_avg=1.96042 acc=0.73438 acc_top1_avg=0.82059 acc_top5_avg=0.93362 lr=0.00010 gn=29.51961 time=61.04it/s +epoch=104 global_step=41000 loss=2.42154 loss_avg=1.95623 acc=0.78125 acc_top1_avg=0.82089 acc_top5_avg=0.93385 lr=0.00010 gn=21.85068 time=56.51it/s +epoch=104 global_step=41050 loss=2.10357 loss_avg=1.95606 acc=0.80469 acc_top1_avg=0.82076 acc_top5_avg=0.93422 lr=0.00010 gn=22.20546 time=59.34it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.28499 test_loss_avg=0.51581 acc=0.89844 test_acc_avg=0.84490 test_acc_top5_avg=0.99104 time=245.50it/s +epoch=104 global_step=41055 loss=0.13405 test_loss_avg=0.39891 acc=0.87500 test_acc_avg=0.87994 test_acc_top5_avg=0.99328 time=873.81it/s +curr_acc 0.8799 +BEST_ACC 0.8958 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=1.88027 loss_avg=1.88718 acc=0.83594 acc_top1_avg=0.82743 acc_top5_avg=0.93403 lr=0.00010 gn=24.51403 time=57.47it/s +epoch=105 global_step=41150 loss=1.97011 loss_avg=1.95440 acc=0.82031 acc_top1_avg=0.82138 acc_top5_avg=0.93446 lr=0.00010 gn=23.13933 time=55.56it/s +epoch=105 global_step=41200 loss=1.96794 loss_avg=1.95207 acc=0.81250 acc_top1_avg=0.82101 acc_top5_avg=0.93518 lr=0.00010 gn=25.62948 time=61.84it/s +epoch=105 global_step=41250 loss=2.12302 loss_avg=1.95900 acc=0.80469 acc_top1_avg=0.82015 acc_top5_avg=0.93454 lr=0.00010 gn=22.45616 time=54.17it/s +epoch=105 global_step=41300 loss=1.78288 loss_avg=1.95087 acc=0.83594 acc_top1_avg=0.82108 acc_top5_avg=0.93520 lr=0.00010 gn=23.07253 time=63.16it/s +epoch=105 global_step=41350 loss=1.98742 loss_avg=1.95887 acc=0.82812 acc_top1_avg=0.82015 acc_top5_avg=0.93424 lr=0.00010 gn=24.43582 time=56.17it/s +epoch=105 global_step=41400 loss=2.40382 loss_avg=1.95791 acc=0.77344 acc_top1_avg=0.82043 acc_top5_avg=0.93442 lr=0.00010 gn=19.68457 time=49.66it/s +====================Eval==================== +epoch=105 global_step=41446 loss=0.49914 test_loss_avg=0.72190 acc=0.82812 test_acc_avg=0.78125 test_acc_top5_avg=0.98125 time=248.04it/s +epoch=105 global_step=41446 loss=0.23717 test_loss_avg=0.46000 acc=0.92188 test_acc_avg=0.86548 test_acc_top5_avg=0.99176 time=250.60it/s +epoch=105 global_step=41446 loss=0.14252 test_loss_avg=0.39490 acc=0.87500 test_acc_avg=0.88311 test_acc_top5_avg=0.99318 time=547.99it/s +curr_acc 0.8831 +BEST_ACC 0.8958 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=1.97172 loss_avg=1.82655 acc=0.81250 acc_top1_avg=0.83203 acc_top5_avg=0.93555 lr=0.00010 gn=13.27636 time=61.71it/s +epoch=106 global_step=41500 loss=1.43276 loss_avg=1.97035 acc=0.88281 acc_top1_avg=0.81959 acc_top5_avg=0.93692 lr=0.00010 gn=21.35629 time=64.82it/s +epoch=106 global_step=41550 loss=2.05705 loss_avg=1.95905 acc=0.80469 acc_top1_avg=0.82106 acc_top5_avg=0.93374 lr=0.00010 gn=24.55534 time=63.18it/s +epoch=106 global_step=41600 loss=1.34412 loss_avg=1.93755 acc=0.87500 acc_top1_avg=0.82366 acc_top5_avg=0.93501 lr=0.00010 gn=23.09920 time=62.57it/s +epoch=106 global_step=41650 loss=2.00572 loss_avg=1.96340 acc=0.82812 acc_top1_avg=0.82050 acc_top5_avg=0.93329 lr=0.00010 gn=25.59262 time=56.86it/s +epoch=106 global_step=41700 loss=1.85161 loss_avg=1.96278 acc=0.83594 acc_top1_avg=0.82056 acc_top5_avg=0.93350 lr=0.00010 gn=22.80384 time=58.55it/s +epoch=106 global_step=41750 loss=1.22856 loss_avg=1.95506 acc=0.89844 acc_top1_avg=0.82157 acc_top5_avg=0.93390 lr=0.00010 gn=32.97555 time=63.02it/s +epoch=106 global_step=41800 loss=2.25921 loss_avg=1.95190 acc=0.78125 acc_top1_avg=0.82175 acc_top5_avg=0.93472 lr=0.00010 gn=19.47418 time=53.52it/s +====================Eval==================== +epoch=106 global_step=41837 loss=0.75098 test_loss_avg=0.55631 acc=0.81250 test_acc_avg=0.83534 test_acc_top5_avg=0.99069 time=258.05it/s +epoch=106 global_step=41837 loss=0.39736 test_loss_avg=0.40649 acc=0.87500 test_acc_avg=0.88024 test_acc_top5_avg=0.99322 time=260.03it/s +epoch=106 global_step=41837 loss=0.12978 test_loss_avg=0.40358 acc=0.93750 test_acc_avg=0.88004 test_acc_top5_avg=0.99347 time=878.02it/s +curr_acc 0.8800 +BEST_ACC 0.8958 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=1.78611 loss_avg=2.04285 acc=0.82031 acc_top1_avg=0.81490 acc_top5_avg=0.93990 lr=0.00010 gn=23.40605 time=57.68it/s +epoch=107 global_step=41900 loss=1.76045 loss_avg=2.00965 acc=0.83594 acc_top1_avg=0.81572 acc_top5_avg=0.93304 lr=0.00010 gn=14.44763 time=63.34it/s +epoch=107 global_step=41950 loss=1.62832 loss_avg=2.02718 acc=0.85938 acc_top1_avg=0.81361 acc_top5_avg=0.93155 lr=0.00010 gn=21.37103 time=64.83it/s +epoch=107 global_step=42000 loss=1.98406 loss_avg=1.99366 acc=0.82031 acc_top1_avg=0.81696 acc_top5_avg=0.93304 lr=0.00010 gn=24.53564 time=54.34it/s +epoch=107 global_step=42050 loss=2.31976 loss_avg=1.96985 acc=0.78906 acc_top1_avg=0.81951 acc_top5_avg=0.93471 lr=0.00010 gn=25.97989 time=58.77it/s +epoch=107 global_step=42100 loss=1.89003 loss_avg=1.96821 acc=0.82812 acc_top1_avg=0.81933 acc_top5_avg=0.93429 lr=0.00010 gn=23.54655 time=58.15it/s +epoch=107 global_step=42150 loss=2.09519 loss_avg=1.95840 acc=0.80469 acc_top1_avg=0.82046 acc_top5_avg=0.93460 lr=0.00010 gn=31.75823 time=55.70it/s +epoch=107 global_step=42200 loss=1.48272 loss_avg=1.95352 acc=0.87500 acc_top1_avg=0.82096 acc_top5_avg=0.93500 lr=0.00010 gn=23.16022 time=56.74it/s +====================Eval==================== +epoch=107 global_step=42228 loss=0.40112 test_loss_avg=0.53042 acc=0.85938 test_acc_avg=0.84458 test_acc_top5_avg=0.99169 time=249.71it/s +epoch=107 global_step=42228 loss=0.16450 test_loss_avg=0.40934 acc=0.87500 test_acc_avg=0.87747 test_acc_top5_avg=0.99367 time=472.28it/s +curr_acc 0.8775 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=2.03152 loss_avg=1.89567 acc=0.80469 acc_top1_avg=0.82564 acc_top5_avg=0.94141 lr=0.00010 gn=26.83540 time=65.12it/s +epoch=108 global_step=42300 loss=1.64335 loss_avg=1.91718 acc=0.85938 acc_top1_avg=0.82389 acc_top5_avg=0.93761 lr=0.00010 gn=22.86538 time=61.98it/s +epoch=108 global_step=42350 loss=2.04170 loss_avg=1.94753 acc=0.82031 acc_top1_avg=0.82127 acc_top5_avg=0.93526 lr=0.00010 gn=23.34120 time=63.90it/s +epoch=108 global_step=42400 loss=1.65500 loss_avg=1.92023 acc=0.85156 acc_top1_avg=0.82395 acc_top5_avg=0.93636 lr=0.00010 gn=20.82145 time=63.45it/s +epoch=108 global_step=42450 loss=1.38484 loss_avg=1.93169 acc=0.87500 acc_top1_avg=0.82256 acc_top5_avg=0.93546 lr=0.00010 gn=13.83525 time=58.49it/s +epoch=108 global_step=42500 loss=1.42027 loss_avg=1.93339 acc=0.85938 acc_top1_avg=0.82250 acc_top5_avg=0.93532 lr=0.00010 gn=22.47104 time=57.06it/s +epoch=108 global_step=42550 loss=2.30253 loss_avg=1.94939 acc=0.77344 acc_top1_avg=0.82077 acc_top5_avg=0.93473 lr=0.00010 gn=26.54206 time=50.74it/s +epoch=108 global_step=42600 loss=2.08965 loss_avg=1.94699 acc=0.79688 acc_top1_avg=0.82098 acc_top5_avg=0.93452 lr=0.00010 gn=19.09133 time=60.10it/s +====================Eval==================== +epoch=108 global_step=42619 loss=0.44276 test_loss_avg=0.47101 acc=0.86719 test_acc_avg=0.86458 test_acc_top5_avg=0.99392 time=231.12it/s +epoch=108 global_step=42619 loss=0.24788 test_loss_avg=0.40729 acc=0.92969 test_acc_avg=0.88258 test_acc_top5_avg=0.99345 time=248.29it/s +epoch=108 global_step=42619 loss=0.13516 test_loss_avg=0.39400 acc=0.93750 test_acc_avg=0.88538 test_acc_top5_avg=0.99367 time=870.73it/s +curr_acc 0.8854 +BEST_ACC 0.8958 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=1.93855 loss_avg=1.94669 acc=0.82812 acc_top1_avg=0.82157 acc_top5_avg=0.93347 lr=0.00010 gn=22.91534 time=63.29it/s +epoch=109 global_step=42700 loss=1.81837 loss_avg=1.95010 acc=0.84375 acc_top1_avg=0.82022 acc_top5_avg=0.93297 lr=0.00010 gn=28.13560 time=58.02it/s +epoch=109 global_step=42750 loss=2.27759 loss_avg=1.97011 acc=0.78906 acc_top1_avg=0.81823 acc_top5_avg=0.93285 lr=0.00010 gn=30.11258 time=63.36it/s +epoch=109 global_step=42800 loss=1.73660 loss_avg=1.94163 acc=0.84375 acc_top1_avg=0.82131 acc_top5_avg=0.93448 lr=0.00010 gn=25.66793 time=61.00it/s +epoch=109 global_step=42850 loss=2.17918 loss_avg=1.94494 acc=0.80469 acc_top1_avg=0.82099 acc_top5_avg=0.93429 lr=0.00010 gn=24.34235 time=63.53it/s +epoch=109 global_step=42900 loss=2.09328 loss_avg=1.95631 acc=0.80469 acc_top1_avg=0.81981 acc_top5_avg=0.93397 lr=0.00010 gn=21.22320 time=60.92it/s +epoch=109 global_step=42950 loss=2.13691 loss_avg=1.95192 acc=0.78906 acc_top1_avg=0.82055 acc_top5_avg=0.93455 lr=0.00010 gn=20.96002 time=60.44it/s +epoch=109 global_step=43000 loss=1.37834 loss_avg=1.94338 acc=0.87500 acc_top1_avg=0.82165 acc_top5_avg=0.93512 lr=0.00010 gn=21.26811 time=63.78it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.13951 test_loss_avg=0.48020 acc=0.93750 test_acc_avg=0.86078 test_acc_top5_avg=0.99179 time=247.66it/s +epoch=109 global_step=43010 loss=0.19938 test_loss_avg=0.41753 acc=0.87500 test_acc_avg=0.87925 test_acc_top5_avg=0.99308 time=551.16it/s +curr_acc 0.8793 +BEST_ACC 0.8958 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=1.62312 loss_avg=1.99842 acc=0.85156 acc_top1_avg=0.81621 acc_top5_avg=0.93418 lr=0.00010 gn=15.68277 time=64.73it/s +epoch=110 global_step=43100 loss=2.10642 loss_avg=1.94614 acc=0.80469 acc_top1_avg=0.82153 acc_top5_avg=0.93533 lr=0.00010 gn=20.11145 time=53.40it/s +epoch=110 global_step=43150 loss=2.52598 loss_avg=1.93895 acc=0.77344 acc_top1_avg=0.82232 acc_top5_avg=0.93644 lr=0.00010 gn=28.30185 time=58.13it/s +epoch=110 global_step=43200 loss=1.52847 loss_avg=1.95725 acc=0.86719 acc_top1_avg=0.82039 acc_top5_avg=0.93561 lr=0.00010 gn=27.40379 time=59.53it/s +epoch=110 global_step=43250 loss=2.58285 loss_avg=1.95619 acc=0.76562 acc_top1_avg=0.82035 acc_top5_avg=0.93538 lr=0.00010 gn=28.16729 time=54.01it/s +epoch=110 global_step=43300 loss=2.19994 loss_avg=1.95412 acc=0.79688 acc_top1_avg=0.82074 acc_top5_avg=0.93537 lr=0.00010 gn=29.51389 time=58.35it/s +epoch=110 global_step=43350 loss=1.72930 loss_avg=1.94674 acc=0.83594 acc_top1_avg=0.82155 acc_top5_avg=0.93541 lr=0.00010 gn=18.20631 time=62.49it/s +epoch=110 global_step=43400 loss=1.66992 loss_avg=1.94853 acc=0.85156 acc_top1_avg=0.82149 acc_top5_avg=0.93564 lr=0.00010 gn=21.36157 time=57.40it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.15614 test_loss_avg=0.60875 acc=0.95312 test_acc_avg=0.82031 test_acc_top5_avg=0.98984 time=243.56it/s +epoch=110 global_step=43401 loss=0.11511 test_loss_avg=0.43993 acc=0.97656 test_acc_avg=0.87122 test_acc_top5_avg=0.99310 time=151.62it/s +epoch=110 global_step=43401 loss=0.09810 test_loss_avg=0.39454 acc=0.93750 test_acc_avg=0.88350 test_acc_top5_avg=0.99377 time=870.91it/s +curr_acc 0.8835 +BEST_ACC 0.8958 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=1.71773 loss_avg=1.91860 acc=0.85938 acc_top1_avg=0.82414 acc_top5_avg=0.93272 lr=0.00010 gn=28.51747 time=63.40it/s +epoch=111 global_step=43500 loss=1.61614 loss_avg=1.94982 acc=0.85938 acc_top1_avg=0.82102 acc_top5_avg=0.93411 lr=0.00010 gn=27.17046 time=52.98it/s +epoch=111 global_step=43550 loss=2.13314 loss_avg=1.96296 acc=0.81250 acc_top1_avg=0.82000 acc_top5_avg=0.93446 lr=0.00010 gn=29.26918 time=61.32it/s +epoch=111 global_step=43600 loss=2.51164 loss_avg=1.94740 acc=0.76562 acc_top1_avg=0.82173 acc_top5_avg=0.93503 lr=0.00010 gn=27.51304 time=57.51it/s +epoch=111 global_step=43650 loss=2.10639 loss_avg=1.94422 acc=0.80469 acc_top1_avg=0.82179 acc_top5_avg=0.93559 lr=0.00010 gn=19.50147 time=61.17it/s +epoch=111 global_step=43700 loss=2.58850 loss_avg=1.94052 acc=0.74219 acc_top1_avg=0.82243 acc_top5_avg=0.93549 lr=0.00010 gn=20.01824 time=63.47it/s +epoch=111 global_step=43750 loss=2.01119 loss_avg=1.94345 acc=0.81250 acc_top1_avg=0.82204 acc_top5_avg=0.93526 lr=0.00010 gn=19.37910 time=64.86it/s +====================Eval==================== +epoch=111 global_step=43792 loss=0.62871 test_loss_avg=0.51998 acc=0.82031 test_acc_avg=0.84526 test_acc_top5_avg=0.99042 time=245.53it/s +epoch=111 global_step=43792 loss=0.11898 test_loss_avg=0.39885 acc=0.93750 test_acc_avg=0.88034 test_acc_top5_avg=0.99308 time=868.75it/s +curr_acc 0.8803 +BEST_ACC 0.8958 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=1.96042 loss_avg=1.98979 acc=0.81250 acc_top1_avg=0.81445 acc_top5_avg=0.93066 lr=0.00010 gn=19.17914 time=61.23it/s +epoch=112 global_step=43850 loss=1.86383 loss_avg=1.93487 acc=0.82812 acc_top1_avg=0.82152 acc_top5_avg=0.93373 lr=0.00010 gn=24.87099 time=64.26it/s +epoch=112 global_step=43900 loss=2.40249 loss_avg=1.96420 acc=0.78125 acc_top1_avg=0.81908 acc_top5_avg=0.93258 lr=0.00010 gn=20.14144 time=56.91it/s +epoch=112 global_step=43950 loss=1.70165 loss_avg=1.96167 acc=0.85156 acc_top1_avg=0.81997 acc_top5_avg=0.93270 lr=0.00010 gn=19.70183 time=63.02it/s +epoch=112 global_step=44000 loss=1.68218 loss_avg=1.93724 acc=0.84375 acc_top1_avg=0.82264 acc_top5_avg=0.93487 lr=0.00010 gn=26.45873 time=64.61it/s +epoch=112 global_step=44050 loss=2.18838 loss_avg=1.93831 acc=0.79688 acc_top1_avg=0.82261 acc_top5_avg=0.93517 lr=0.00010 gn=31.29043 time=63.92it/s +epoch=112 global_step=44100 loss=2.09844 loss_avg=1.93517 acc=0.79688 acc_top1_avg=0.82303 acc_top5_avg=0.93529 lr=0.00010 gn=23.93555 time=58.93it/s +epoch=112 global_step=44150 loss=1.91258 loss_avg=1.93371 acc=0.82031 acc_top1_avg=0.82289 acc_top5_avg=0.93556 lr=0.00010 gn=18.18085 time=55.47it/s +====================Eval==================== +epoch=112 global_step=44183 loss=0.77582 test_loss_avg=0.78020 acc=0.77344 test_acc_avg=0.76953 test_acc_top5_avg=0.98047 time=236.18it/s +epoch=112 global_step=44183 loss=0.17176 test_loss_avg=0.46057 acc=0.94531 test_acc_avg=0.86463 test_acc_top5_avg=0.99294 time=241.77it/s +epoch=112 global_step=44183 loss=0.16552 test_loss_avg=0.39668 acc=0.93750 test_acc_avg=0.88291 test_acc_top5_avg=0.99387 time=541.69it/s +curr_acc 0.8829 +BEST_ACC 0.8958 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=1.81842 loss_avg=1.98421 acc=0.82812 acc_top1_avg=0.81801 acc_top5_avg=0.93290 lr=0.00010 gn=15.41438 time=54.36it/s +epoch=113 global_step=44250 loss=1.75137 loss_avg=1.94169 acc=0.83594 acc_top1_avg=0.82183 acc_top5_avg=0.93785 lr=0.00010 gn=22.32821 time=57.73it/s +epoch=113 global_step=44300 loss=1.90294 loss_avg=1.95129 acc=0.82812 acc_top1_avg=0.82125 acc_top5_avg=0.93643 lr=0.00010 gn=22.46091 time=54.53it/s +epoch=113 global_step=44350 loss=1.91042 loss_avg=1.96069 acc=0.82031 acc_top1_avg=0.81999 acc_top5_avg=0.93455 lr=0.00010 gn=24.39758 time=46.64it/s +epoch=113 global_step=44400 loss=2.19335 loss_avg=1.95113 acc=0.78906 acc_top1_avg=0.82089 acc_top5_avg=0.93527 lr=0.00010 gn=25.53448 time=58.36it/s +epoch=113 global_step=44450 loss=1.93036 loss_avg=1.94185 acc=0.82812 acc_top1_avg=0.82224 acc_top5_avg=0.93598 lr=0.00010 gn=24.28484 time=63.77it/s +epoch=113 global_step=44500 loss=0.93213 loss_avg=1.93287 acc=0.92188 acc_top1_avg=0.82327 acc_top5_avg=0.93614 lr=0.00010 gn=22.21383 time=60.93it/s +epoch=113 global_step=44550 loss=1.60955 loss_avg=1.93832 acc=0.86719 acc_top1_avg=0.82274 acc_top5_avg=0.93610 lr=0.00010 gn=22.98039 time=61.94it/s +====================Eval==================== +epoch=113 global_step=44574 loss=0.56164 test_loss_avg=0.50516 acc=0.80469 test_acc_avg=0.84545 test_acc_top5_avg=0.99219 time=242.12it/s +epoch=113 global_step=44574 loss=0.42522 test_loss_avg=0.40369 acc=0.89062 test_acc_avg=0.88121 test_acc_top5_avg=0.99251 time=263.00it/s +epoch=113 global_step=44574 loss=0.10939 test_loss_avg=0.39886 acc=0.93750 test_acc_avg=0.88212 test_acc_top5_avg=0.99298 time=878.20it/s +curr_acc 0.8821 +BEST_ACC 0.8958 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=2.20056 loss_avg=1.91703 acc=0.80469 acc_top1_avg=0.82452 acc_top5_avg=0.93690 lr=0.00010 gn=28.70356 time=53.04it/s +epoch=114 global_step=44650 loss=1.86189 loss_avg=1.93920 acc=0.82031 acc_top1_avg=0.82268 acc_top5_avg=0.93606 lr=0.00010 gn=19.88739 time=55.09it/s +epoch=114 global_step=44700 loss=2.03373 loss_avg=1.92492 acc=0.82031 acc_top1_avg=0.82391 acc_top5_avg=0.93750 lr=0.00010 gn=31.83490 time=53.21it/s +epoch=114 global_step=44750 loss=1.95160 loss_avg=1.92908 acc=0.80469 acc_top1_avg=0.82329 acc_top5_avg=0.93746 lr=0.00010 gn=29.01628 time=57.12it/s +epoch=114 global_step=44800 loss=1.89655 loss_avg=1.93956 acc=0.82031 acc_top1_avg=0.82197 acc_top5_avg=0.93660 lr=0.00010 gn=21.43774 time=51.34it/s +epoch=114 global_step=44850 loss=2.40309 loss_avg=1.93472 acc=0.76562 acc_top1_avg=0.82212 acc_top5_avg=0.93600 lr=0.00010 gn=17.76977 time=59.42it/s +epoch=114 global_step=44900 loss=2.25898 loss_avg=1.92386 acc=0.78906 acc_top1_avg=0.82348 acc_top5_avg=0.93609 lr=0.00010 gn=24.42126 time=64.26it/s +epoch=114 global_step=44950 loss=2.32654 loss_avg=1.92682 acc=0.78125 acc_top1_avg=0.82320 acc_top5_avg=0.93594 lr=0.00010 gn=26.31526 time=60.08it/s +====================Eval==================== +epoch=114 global_step=44965 loss=0.96325 test_loss_avg=0.51782 acc=0.77344 test_acc_avg=0.84908 test_acc_top5_avg=0.99094 time=239.84it/s +epoch=114 global_step=44965 loss=0.14281 test_loss_avg=0.41565 acc=0.93750 test_acc_avg=0.87747 test_acc_top5_avg=0.99268 time=542.81it/s +curr_acc 0.8775 +BEST_ACC 0.8958 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=1.86138 loss_avg=1.88879 acc=0.83594 acc_top1_avg=0.82656 acc_top5_avg=0.93080 lr=0.00010 gn=30.22027 time=63.83it/s +epoch=115 global_step=45050 loss=1.38795 loss_avg=1.93856 acc=0.88281 acc_top1_avg=0.82279 acc_top5_avg=0.93153 lr=0.00010 gn=22.17116 time=55.00it/s +epoch=115 global_step=45100 loss=2.12808 loss_avg=1.96100 acc=0.80469 acc_top1_avg=0.82031 acc_top5_avg=0.93131 lr=0.00010 gn=21.68458 time=56.02it/s +epoch=115 global_step=45150 loss=2.04903 loss_avg=1.93245 acc=0.81250 acc_top1_avg=0.82365 acc_top5_avg=0.93315 lr=0.00010 gn=23.58704 time=62.55it/s +epoch=115 global_step=45200 loss=1.75873 loss_avg=1.93302 acc=0.84375 acc_top1_avg=0.82347 acc_top5_avg=0.93351 lr=0.00010 gn=29.42536 time=58.45it/s +epoch=115 global_step=45250 loss=1.98754 loss_avg=1.93683 acc=0.82031 acc_top1_avg=0.82311 acc_top5_avg=0.93366 lr=0.00010 gn=28.19191 time=62.85it/s +epoch=115 global_step=45300 loss=1.16631 loss_avg=1.92993 acc=0.89844 acc_top1_avg=0.82369 acc_top5_avg=0.93449 lr=0.00010 gn=15.68032 time=51.07it/s +epoch=115 global_step=45350 loss=2.20987 loss_avg=1.92890 acc=0.78906 acc_top1_avg=0.82360 acc_top5_avg=0.93490 lr=0.00010 gn=29.16036 time=53.17it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.22748 test_loss_avg=0.51552 acc=0.91406 test_acc_avg=0.84740 test_acc_top5_avg=0.99323 time=246.97it/s +epoch=115 global_step=45356 loss=0.17085 test_loss_avg=0.44001 acc=0.92969 test_acc_avg=0.87007 test_acc_top5_avg=0.99207 time=251.26it/s +epoch=115 global_step=45356 loss=0.17050 test_loss_avg=0.41650 acc=0.93750 test_acc_avg=0.87678 test_acc_top5_avg=0.99248 time=889.00it/s +curr_acc 0.8768 +BEST_ACC 0.8958 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=2.02785 loss_avg=1.86038 acc=0.81250 acc_top1_avg=0.83043 acc_top5_avg=0.93519 lr=0.00010 gn=25.32654 time=57.21it/s +epoch=116 global_step=45450 loss=1.75972 loss_avg=1.89448 acc=0.83594 acc_top1_avg=0.82746 acc_top5_avg=0.93484 lr=0.00010 gn=30.07344 time=53.60it/s +epoch=116 global_step=45500 loss=1.75409 loss_avg=1.90384 acc=0.83594 acc_top1_avg=0.82644 acc_top5_avg=0.93500 lr=0.00010 gn=27.15096 time=64.88it/s +epoch=116 global_step=45550 loss=2.33014 loss_avg=1.90739 acc=0.78125 acc_top1_avg=0.82599 acc_top5_avg=0.93545 lr=0.00010 gn=27.63915 time=58.22it/s +epoch=116 global_step=45600 loss=1.93464 loss_avg=1.91267 acc=0.82031 acc_top1_avg=0.82569 acc_top5_avg=0.93625 lr=0.00010 gn=23.23314 time=63.55it/s +epoch=116 global_step=45650 loss=1.87192 loss_avg=1.91888 acc=0.82031 acc_top1_avg=0.82462 acc_top5_avg=0.93614 lr=0.00010 gn=15.94326 time=61.25it/s +epoch=116 global_step=45700 loss=1.70992 loss_avg=1.91935 acc=0.83594 acc_top1_avg=0.82445 acc_top5_avg=0.93639 lr=0.00010 gn=23.11205 time=58.51it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.30823 test_loss_avg=0.54254 acc=0.92188 test_acc_avg=0.84136 test_acc_top5_avg=0.99002 time=260.21it/s +epoch=116 global_step=45747 loss=0.14508 test_loss_avg=0.42759 acc=0.93750 test_acc_avg=0.87470 test_acc_top5_avg=0.99278 time=557.46it/s +curr_acc 0.8747 +BEST_ACC 0.8958 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=1.62041 loss_avg=1.74249 acc=0.85156 acc_top1_avg=0.83854 acc_top5_avg=0.94010 lr=0.00010 gn=30.97745 time=57.16it/s +epoch=117 global_step=45800 loss=2.23958 loss_avg=1.88031 acc=0.78906 acc_top1_avg=0.82798 acc_top5_avg=0.93691 lr=0.00010 gn=26.21055 time=56.11it/s +epoch=117 global_step=45850 loss=1.41102 loss_avg=1.90650 acc=0.88281 acc_top1_avg=0.82464 acc_top5_avg=0.93758 lr=0.00010 gn=28.59234 time=51.90it/s +epoch=117 global_step=45900 loss=1.47656 loss_avg=1.90460 acc=0.87500 acc_top1_avg=0.82511 acc_top5_avg=0.93699 lr=0.00010 gn=22.77888 time=62.04it/s +epoch=117 global_step=45950 loss=2.37737 loss_avg=1.92983 acc=0.77344 acc_top1_avg=0.82266 acc_top5_avg=0.93500 lr=0.00010 gn=15.78810 time=53.71it/s +epoch=117 global_step=46000 loss=1.87519 loss_avg=1.91517 acc=0.82812 acc_top1_avg=0.82414 acc_top5_avg=0.93552 lr=0.00010 gn=23.92499 time=54.67it/s +epoch=117 global_step=46050 loss=1.93500 loss_avg=1.91481 acc=0.82031 acc_top1_avg=0.82421 acc_top5_avg=0.93528 lr=0.00010 gn=18.82938 time=58.46it/s +epoch=117 global_step=46100 loss=1.31511 loss_avg=1.90715 acc=0.88281 acc_top1_avg=0.82512 acc_top5_avg=0.93540 lr=0.00010 gn=19.71643 time=61.76it/s +====================Eval==================== +epoch=117 global_step=46138 loss=0.63291 test_loss_avg=0.74812 acc=0.78125 test_acc_avg=0.78348 test_acc_top5_avg=0.98772 time=248.45it/s +epoch=117 global_step=46138 loss=0.26899 test_loss_avg=0.45923 acc=0.92969 test_acc_avg=0.86774 test_acc_top5_avg=0.99150 time=238.08it/s +epoch=117 global_step=46138 loss=0.14978 test_loss_avg=0.40900 acc=0.93750 test_acc_avg=0.88014 test_acc_top5_avg=0.99278 time=569.41it/s +curr_acc 0.8801 +BEST_ACC 0.8958 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=1.72063 loss_avg=1.92461 acc=0.83594 acc_top1_avg=0.82552 acc_top5_avg=0.93294 lr=0.00010 gn=24.87775 time=63.91it/s +epoch=118 global_step=46200 loss=1.74042 loss_avg=1.94749 acc=0.84375 acc_top1_avg=0.82308 acc_top5_avg=0.93322 lr=0.00010 gn=16.86007 time=53.81it/s +epoch=118 global_step=46250 loss=1.81253 loss_avg=1.94210 acc=0.83594 acc_top1_avg=0.82268 acc_top5_avg=0.93597 lr=0.00010 gn=25.31446 time=60.69it/s +epoch=118 global_step=46300 loss=1.51476 loss_avg=1.93138 acc=0.85938 acc_top1_avg=0.82374 acc_top5_avg=0.93470 lr=0.00010 gn=24.34481 time=54.74it/s +epoch=118 global_step=46350 loss=1.95179 loss_avg=1.93198 acc=0.83594 acc_top1_avg=0.82352 acc_top5_avg=0.93463 lr=0.00010 gn=28.33815 time=62.65it/s +epoch=118 global_step=46400 loss=1.27182 loss_avg=1.92255 acc=0.89844 acc_top1_avg=0.82398 acc_top5_avg=0.93401 lr=0.00010 gn=24.59835 time=61.72it/s +epoch=118 global_step=46450 loss=1.86714 loss_avg=1.92384 acc=0.83594 acc_top1_avg=0.82387 acc_top5_avg=0.93535 lr=0.00010 gn=24.74339 time=54.63it/s +epoch=118 global_step=46500 loss=1.95408 loss_avg=1.92141 acc=0.81250 acc_top1_avg=0.82420 acc_top5_avg=0.93580 lr=0.00010 gn=17.81195 time=65.11it/s +====================Eval==================== +epoch=118 global_step=46529 loss=0.38643 test_loss_avg=0.51973 acc=0.85938 test_acc_avg=0.84487 test_acc_top5_avg=0.98968 time=225.77it/s +epoch=118 global_step=46529 loss=0.39021 test_loss_avg=0.41507 acc=0.86719 test_acc_avg=0.87810 test_acc_top5_avg=0.99279 time=259.66it/s +epoch=118 global_step=46529 loss=0.13882 test_loss_avg=0.41157 acc=0.93750 test_acc_avg=0.87886 test_acc_top5_avg=0.99288 time=787.81it/s +curr_acc 0.8789 +BEST_ACC 0.8958 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9951 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=2.17289 loss_avg=1.84596 acc=0.81250 acc_top1_avg=0.83259 acc_top5_avg=0.93341 lr=0.00010 gn=24.56932 time=64.79it/s +epoch=119 global_step=46600 loss=2.01433 loss_avg=1.87995 acc=0.82031 acc_top1_avg=0.82956 acc_top5_avg=0.93508 lr=0.00010 gn=22.97647 time=64.18it/s +epoch=119 global_step=46650 loss=1.52399 loss_avg=1.87797 acc=0.85938 acc_top1_avg=0.82929 acc_top5_avg=0.93673 lr=0.00010 gn=23.85074 time=36.48it/s +epoch=119 global_step=46700 loss=1.55887 loss_avg=1.88481 acc=0.86719 acc_top1_avg=0.82817 acc_top5_avg=0.93599 lr=0.00010 gn=25.71212 time=65.18it/s +epoch=119 global_step=46750 loss=2.18105 loss_avg=1.91749 acc=0.78125 acc_top1_avg=0.82459 acc_top5_avg=0.93446 lr=0.00010 gn=26.16340 time=64.67it/s +epoch=119 global_step=46800 loss=1.93882 loss_avg=1.92040 acc=0.82031 acc_top1_avg=0.82412 acc_top5_avg=0.93433 lr=0.00010 gn=18.53374 time=62.88it/s +epoch=119 global_step=46850 loss=1.93897 loss_avg=1.91834 acc=0.82031 acc_top1_avg=0.82421 acc_top5_avg=0.93490 lr=0.00010 gn=15.44118 time=43.98it/s +epoch=119 global_step=46900 loss=1.57310 loss_avg=1.92348 acc=0.85938 acc_top1_avg=0.82374 acc_top5_avg=0.93449 lr=0.00010 gn=24.90584 time=54.49it/s +====================Eval==================== +epoch=119 global_step=46920 loss=0.12712 test_loss_avg=0.51781 acc=0.95312 test_acc_avg=0.84949 test_acc_top5_avg=0.99123 time=250.32it/s +epoch=119 global_step=46920 loss=0.21270 test_loss_avg=0.42809 acc=0.81250 test_acc_avg=0.87184 test_acc_top5_avg=0.99278 time=885.81it/s +curr_acc 0.8718 +BEST_ACC 0.8958 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9951 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_0_4.log b/other_methods/sceloss/sceloss_results/out_0_4.log new file mode 100644 index 0000000..1253698 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_0_4.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.0__noise_amount__0.4.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.19103 loss_avg=7.62847 acc=0.28125 acc_top1_avg=0.22281 acc_top5_avg=0.64969 lr=0.01000 gn=7.38238 time=62.68it/s +epoch=0 global_step=100 loss=6.47825 loss_avg=7.39039 acc=0.35156 acc_top1_avg=0.24922 acc_top5_avg=0.67242 lr=0.01000 gn=5.83524 time=49.04it/s +epoch=0 global_step=150 loss=6.93861 loss_avg=7.24718 acc=0.29688 acc_top1_avg=0.26401 acc_top5_avg=0.68646 lr=0.01000 gn=4.91090 time=58.42it/s +epoch=0 global_step=200 loss=6.26793 loss_avg=7.14783 acc=0.35156 acc_top1_avg=0.27469 acc_top5_avg=0.69766 lr=0.01000 gn=5.45683 time=56.23it/s +epoch=0 global_step=250 loss=6.77119 loss_avg=7.06856 acc=0.29688 acc_top1_avg=0.28309 acc_top5_avg=0.70453 lr=0.01000 gn=4.42326 time=66.59it/s +epoch=0 global_step=300 loss=6.67235 loss_avg=7.01236 acc=0.31250 acc_top1_avg=0.28911 acc_top5_avg=0.70951 lr=0.01000 gn=4.20304 time=59.10it/s +epoch=0 global_step=350 loss=5.95360 loss_avg=6.95461 acc=0.40625 acc_top1_avg=0.29482 acc_top5_avg=0.71422 lr=0.01000 gn=4.43775 time=65.11it/s +====================Eval==================== +epoch=0 global_step=391 loss=1.79036 test_loss_avg=2.73648 acc=0.53125 test_acc_avg=0.32297 test_acc_top5_avg=0.88266 time=255.69it/s +epoch=0 global_step=391 loss=0.64980 test_loss_avg=2.09249 acc=0.81250 test_acc_avg=0.47666 test_acc_top5_avg=0.90496 time=34.59it/s +curr_acc 0.4767 +BEST_ACC 0.0000 +curr_acc_top5 0.9050 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.46540 loss_avg=6.49912 acc=0.37500 acc_top1_avg=0.35330 acc_top5_avg=0.74740 lr=0.01000 gn=4.21797 time=51.85it/s +epoch=1 global_step=450 loss=6.64645 loss_avg=6.40365 acc=0.34375 acc_top1_avg=0.35593 acc_top5_avg=0.75132 lr=0.01000 gn=6.34805 time=58.60it/s +epoch=1 global_step=500 loss=6.75605 loss_avg=6.37059 acc=0.32812 acc_top1_avg=0.35981 acc_top5_avg=0.74950 lr=0.01000 gn=4.81661 time=61.66it/s +epoch=1 global_step=550 loss=5.90839 loss_avg=6.34311 acc=0.37500 acc_top1_avg=0.36144 acc_top5_avg=0.75172 lr=0.01000 gn=4.72012 time=53.59it/s +epoch=1 global_step=600 loss=6.07867 loss_avg=6.28717 acc=0.37500 acc_top1_avg=0.36745 acc_top5_avg=0.75378 lr=0.01000 gn=5.40023 time=56.86it/s +epoch=1 global_step=650 loss=5.93790 loss_avg=6.27051 acc=0.41406 acc_top1_avg=0.36882 acc_top5_avg=0.75573 lr=0.01000 gn=4.43524 time=61.71it/s +epoch=1 global_step=700 loss=5.53805 loss_avg=6.22972 acc=0.43750 acc_top1_avg=0.37338 acc_top5_avg=0.75870 lr=0.01000 gn=4.71369 time=57.18it/s +epoch=1 global_step=750 loss=6.21689 loss_avg=6.20160 acc=0.38281 acc_top1_avg=0.37678 acc_top5_avg=0.76025 lr=0.01000 gn=4.68444 time=46.40it/s +====================Eval==================== +epoch=1 global_step=782 loss=2.45611 test_loss_avg=2.06082 acc=0.36719 test_acc_avg=0.51265 test_acc_top5_avg=0.93638 time=246.26it/s +epoch=1 global_step=782 loss=0.41308 test_loss_avg=1.90035 acc=0.85156 test_acc_avg=0.52597 test_acc_top5_avg=0.91461 time=245.05it/s +epoch=1 global_step=782 loss=0.16689 test_loss_avg=1.75769 acc=0.87500 test_acc_avg=0.56013 test_acc_top5_avg=0.92158 time=467.64it/s +curr_acc 0.5601 +BEST_ACC 0.4767 +curr_acc_top5 0.9216 +BEST_ACC_top5 0.9050 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=6.13419 loss_avg=5.80037 acc=0.36719 acc_top1_avg=0.42144 acc_top5_avg=0.77995 lr=0.01000 gn=4.11997 time=32.34it/s +epoch=2 global_step=850 loss=5.72188 loss_avg=5.84910 acc=0.43750 acc_top1_avg=0.41636 acc_top5_avg=0.77642 lr=0.01000 gn=4.30768 time=52.52it/s +epoch=2 global_step=900 loss=5.66096 loss_avg=5.87377 acc=0.45312 acc_top1_avg=0.41234 acc_top5_avg=0.77794 lr=0.01000 gn=4.78645 time=56.47it/s +epoch=2 global_step=950 loss=5.80480 loss_avg=5.87876 acc=0.45312 acc_top1_avg=0.41248 acc_top5_avg=0.77558 lr=0.01000 gn=4.80437 time=56.98it/s +epoch=2 global_step=1000 loss=5.26767 loss_avg=5.86944 acc=0.46094 acc_top1_avg=0.41327 acc_top5_avg=0.77677 lr=0.01000 gn=6.26979 time=57.98it/s +epoch=2 global_step=1050 loss=5.86792 loss_avg=5.84508 acc=0.41406 acc_top1_avg=0.41581 acc_top5_avg=0.77991 lr=0.01000 gn=5.85198 time=55.79it/s +epoch=2 global_step=1100 loss=5.71860 loss_avg=5.83887 acc=0.44531 acc_top1_avg=0.41645 acc_top5_avg=0.77970 lr=0.01000 gn=5.15819 time=62.39it/s +epoch=2 global_step=1150 loss=6.39823 loss_avg=5.82452 acc=0.36719 acc_top1_avg=0.41807 acc_top5_avg=0.78150 lr=0.01000 gn=5.75286 time=59.66it/s +====================Eval==================== +epoch=2 global_step=1173 loss=3.79203 test_loss_avg=2.37050 acc=0.29688 test_acc_avg=0.40848 test_acc_top5_avg=0.91146 time=164.84it/s +epoch=2 global_step=1173 loss=0.38981 test_loss_avg=1.94114 acc=0.93750 test_acc_avg=0.52710 test_acc_top5_avg=0.90111 time=831.21it/s +curr_acc 0.5271 +BEST_ACC 0.5601 +curr_acc_top5 0.9011 +BEST_ACC_top5 0.9216 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=5.70628 loss_avg=5.59710 acc=0.44531 acc_top1_avg=0.43981 acc_top5_avg=0.78328 lr=0.01000 gn=4.40336 time=57.39it/s +epoch=3 global_step=1250 loss=5.57242 loss_avg=5.68228 acc=0.45312 acc_top1_avg=0.43192 acc_top5_avg=0.78744 lr=0.01000 gn=5.64855 time=60.21it/s +epoch=3 global_step=1300 loss=5.98238 loss_avg=5.68939 acc=0.39844 acc_top1_avg=0.43258 acc_top5_avg=0.78900 lr=0.01000 gn=5.26674 time=52.74it/s +epoch=3 global_step=1350 loss=5.97773 loss_avg=5.65347 acc=0.41406 acc_top1_avg=0.43684 acc_top5_avg=0.79334 lr=0.01000 gn=4.55741 time=54.44it/s +epoch=3 global_step=1400 loss=6.29119 loss_avg=5.66170 acc=0.37500 acc_top1_avg=0.43616 acc_top5_avg=0.79226 lr=0.01000 gn=4.79944 time=55.06it/s +epoch=3 global_step=1450 loss=5.05101 loss_avg=5.66195 acc=0.50781 acc_top1_avg=0.43620 acc_top5_avg=0.79273 lr=0.01000 gn=5.75092 time=54.76it/s +epoch=3 global_step=1500 loss=5.73635 loss_avg=5.65962 acc=0.41406 acc_top1_avg=0.43681 acc_top5_avg=0.79217 lr=0.01000 gn=4.37463 time=59.20it/s +epoch=3 global_step=1550 loss=5.48406 loss_avg=5.64716 acc=0.46094 acc_top1_avg=0.43802 acc_top5_avg=0.79205 lr=0.01000 gn=4.39223 time=60.13it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.96485 test_loss_avg=1.27872 acc=0.76562 test_acc_avg=0.65565 test_acc_top5_avg=0.96274 time=239.17it/s +epoch=3 global_step=1564 loss=0.78949 test_loss_avg=1.27276 acc=0.75000 test_acc_avg=0.62785 test_acc_top5_avg=0.95672 time=238.56it/s +epoch=3 global_step=1564 loss=0.12978 test_loss_avg=1.12689 acc=0.93750 test_acc_avg=0.67000 test_acc_top5_avg=0.96094 time=489.19it/s +curr_acc 0.6700 +BEST_ACC 0.5601 +curr_acc_top5 0.9609 +BEST_ACC_top5 0.9216 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=5.83765 loss_avg=5.52077 acc=0.40625 acc_top1_avg=0.45182 acc_top5_avg=0.79991 lr=0.01000 gn=5.08915 time=54.88it/s +epoch=4 global_step=1650 loss=5.42308 loss_avg=5.58809 acc=0.46875 acc_top1_avg=0.44604 acc_top5_avg=0.79933 lr=0.01000 gn=4.50341 time=59.31it/s +epoch=4 global_step=1700 loss=5.53895 loss_avg=5.55812 acc=0.43750 acc_top1_avg=0.44853 acc_top5_avg=0.79992 lr=0.01000 gn=4.48795 time=63.10it/s +epoch=4 global_step=1750 loss=6.03097 loss_avg=5.54593 acc=0.39844 acc_top1_avg=0.44943 acc_top5_avg=0.80028 lr=0.01000 gn=3.36252 time=55.22it/s +epoch=4 global_step=1800 loss=5.81502 loss_avg=5.53578 acc=0.40625 acc_top1_avg=0.45001 acc_top5_avg=0.79906 lr=0.01000 gn=5.68139 time=56.73it/s +epoch=4 global_step=1850 loss=5.35117 loss_avg=5.53439 acc=0.49219 acc_top1_avg=0.44979 acc_top5_avg=0.79840 lr=0.01000 gn=4.77589 time=58.18it/s +epoch=4 global_step=1900 loss=4.56945 loss_avg=5.51623 acc=0.54688 acc_top1_avg=0.45166 acc_top5_avg=0.79978 lr=0.01000 gn=5.80121 time=52.44it/s +epoch=4 global_step=1950 loss=4.87044 loss_avg=5.51365 acc=0.53125 acc_top1_avg=0.45203 acc_top5_avg=0.79924 lr=0.01000 gn=4.32241 time=60.29it/s +====================Eval==================== +epoch=4 global_step=1955 loss=0.83287 test_loss_avg=1.70232 acc=0.75000 test_acc_avg=0.54573 test_acc_top5_avg=0.93382 time=238.83it/s +epoch=4 global_step=1955 loss=0.23251 test_loss_avg=1.27359 acc=0.93750 test_acc_avg=0.66238 test_acc_top5_avg=0.94324 time=855.46it/s +curr_acc 0.6624 +BEST_ACC 0.6700 +curr_acc_top5 0.9432 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=5.31248 loss_avg=5.46260 acc=0.46094 acc_top1_avg=0.45781 acc_top5_avg=0.80139 lr=0.01000 gn=4.60273 time=57.23it/s +epoch=5 global_step=2050 loss=4.85749 loss_avg=5.41549 acc=0.51562 acc_top1_avg=0.46250 acc_top5_avg=0.80025 lr=0.01000 gn=4.00717 time=59.77it/s +epoch=5 global_step=2100 loss=5.64247 loss_avg=5.43479 acc=0.42969 acc_top1_avg=0.46051 acc_top5_avg=0.79941 lr=0.01000 gn=5.21709 time=51.98it/s +epoch=5 global_step=2150 loss=5.69947 loss_avg=5.44405 acc=0.43750 acc_top1_avg=0.45946 acc_top5_avg=0.79968 lr=0.01000 gn=5.04921 time=58.67it/s +epoch=5 global_step=2200 loss=5.83315 loss_avg=5.45115 acc=0.39844 acc_top1_avg=0.45871 acc_top5_avg=0.79974 lr=0.01000 gn=4.21041 time=55.72it/s +epoch=5 global_step=2250 loss=5.27850 loss_avg=5.43406 acc=0.47656 acc_top1_avg=0.46012 acc_top5_avg=0.80034 lr=0.01000 gn=4.11126 time=52.91it/s +epoch=5 global_step=2300 loss=5.77178 loss_avg=5.42471 acc=0.43750 acc_top1_avg=0.46155 acc_top5_avg=0.80014 lr=0.01000 gn=5.00102 time=55.19it/s +====================Eval==================== +epoch=5 global_step=2346 loss=1.01357 test_loss_avg=0.98838 acc=0.71094 test_acc_avg=0.72031 test_acc_top5_avg=0.98750 time=241.11it/s +epoch=5 global_step=2346 loss=1.26205 test_loss_avg=1.30028 acc=0.64844 test_acc_avg=0.64702 test_acc_top5_avg=0.93111 time=257.90it/s +epoch=5 global_step=2346 loss=0.50847 test_loss_avg=1.13515 acc=0.87500 test_acc_avg=0.69007 test_acc_top5_avg=0.94462 time=704.81it/s +curr_acc 0.6901 +BEST_ACC 0.6700 +curr_acc_top5 0.9446 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=5.56111 loss_avg=5.49202 acc=0.44531 acc_top1_avg=0.45703 acc_top5_avg=0.79492 lr=0.01000 gn=4.87107 time=61.05it/s +epoch=6 global_step=2400 loss=5.22454 loss_avg=5.44805 acc=0.47656 acc_top1_avg=0.46108 acc_top5_avg=0.80252 lr=0.01000 gn=5.40145 time=57.55it/s +epoch=6 global_step=2450 loss=6.07252 loss_avg=5.42496 acc=0.39844 acc_top1_avg=0.46221 acc_top5_avg=0.80431 lr=0.01000 gn=5.66667 time=57.45it/s +epoch=6 global_step=2500 loss=4.97337 loss_avg=5.38876 acc=0.52344 acc_top1_avg=0.46611 acc_top5_avg=0.80454 lr=0.01000 gn=5.19688 time=57.67it/s +epoch=6 global_step=2550 loss=5.01874 loss_avg=5.38894 acc=0.50781 acc_top1_avg=0.46645 acc_top5_avg=0.80465 lr=0.01000 gn=4.03092 time=61.09it/s +epoch=6 global_step=2600 loss=5.33009 loss_avg=5.38103 acc=0.46094 acc_top1_avg=0.46703 acc_top5_avg=0.80552 lr=0.01000 gn=4.99612 time=54.60it/s +epoch=6 global_step=2650 loss=5.35528 loss_avg=5.38341 acc=0.48438 acc_top1_avg=0.46685 acc_top5_avg=0.80502 lr=0.01000 gn=4.84986 time=54.74it/s +epoch=6 global_step=2700 loss=5.57334 loss_avg=5.37155 acc=0.44531 acc_top1_avg=0.46796 acc_top5_avg=0.80495 lr=0.01000 gn=5.91840 time=61.15it/s +====================Eval==================== +epoch=6 global_step=2737 loss=3.52054 test_loss_avg=1.14508 acc=0.24219 test_acc_avg=0.70403 test_acc_top5_avg=0.96034 time=229.79it/s +epoch=6 global_step=2737 loss=0.69941 test_loss_avg=1.43292 acc=0.83594 test_acc_avg=0.65101 test_acc_top5_avg=0.92835 time=258.11it/s +epoch=6 global_step=2737 loss=0.39227 test_loss_avg=1.39920 acc=0.87500 test_acc_avg=0.65902 test_acc_top5_avg=0.93097 time=690.76it/s +curr_acc 0.6590 +BEST_ACC 0.6901 +curr_acc_top5 0.9310 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=5.25126 loss_avg=5.18912 acc=0.49219 acc_top1_avg=0.48618 acc_top5_avg=0.81911 lr=0.01000 gn=5.95706 time=64.33it/s +epoch=7 global_step=2800 loss=4.85453 loss_avg=5.33648 acc=0.54688 acc_top1_avg=0.47532 acc_top5_avg=0.80605 lr=0.01000 gn=4.60617 time=55.92it/s +epoch=7 global_step=2850 loss=4.97816 loss_avg=5.31549 acc=0.50000 acc_top1_avg=0.47642 acc_top5_avg=0.80303 lr=0.01000 gn=5.46143 time=50.78it/s +epoch=7 global_step=2900 loss=5.08422 loss_avg=5.29557 acc=0.50000 acc_top1_avg=0.47786 acc_top5_avg=0.80426 lr=0.01000 gn=5.63834 time=56.48it/s +epoch=7 global_step=2950 loss=5.79375 loss_avg=5.31287 acc=0.40625 acc_top1_avg=0.47447 acc_top5_avg=0.80439 lr=0.01000 gn=5.58821 time=59.67it/s +epoch=7 global_step=3000 loss=5.61658 loss_avg=5.30625 acc=0.44531 acc_top1_avg=0.47478 acc_top5_avg=0.80700 lr=0.01000 gn=5.53248 time=57.93it/s +epoch=7 global_step=3050 loss=5.51866 loss_avg=5.31197 acc=0.45312 acc_top1_avg=0.47399 acc_top5_avg=0.80658 lr=0.01000 gn=5.45331 time=51.57it/s +epoch=7 global_step=3100 loss=5.58446 loss_avg=5.32063 acc=0.42969 acc_top1_avg=0.47299 acc_top5_avg=0.80682 lr=0.01000 gn=6.40425 time=55.55it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.37983 test_loss_avg=1.54810 acc=0.59375 test_acc_avg=0.59641 test_acc_top5_avg=0.93783 time=70.48it/s +epoch=7 global_step=3128 loss=0.13335 test_loss_avg=1.15047 acc=0.93750 test_acc_avg=0.69353 test_acc_top5_avg=0.95550 time=745.65it/s +curr_acc 0.6935 +BEST_ACC 0.6901 +curr_acc_top5 0.9555 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=5.62987 loss_avg=5.23291 acc=0.44531 acc_top1_avg=0.48295 acc_top5_avg=0.81428 lr=0.01000 gn=5.39662 time=62.88it/s +epoch=8 global_step=3200 loss=5.07419 loss_avg=5.18983 acc=0.51562 acc_top1_avg=0.48774 acc_top5_avg=0.81022 lr=0.01000 gn=6.15488 time=51.38it/s +epoch=8 global_step=3250 loss=5.17361 loss_avg=5.25539 acc=0.48438 acc_top1_avg=0.48066 acc_top5_avg=0.80795 lr=0.01000 gn=5.44541 time=56.91it/s +epoch=8 global_step=3300 loss=5.24485 loss_avg=5.27928 acc=0.46094 acc_top1_avg=0.47738 acc_top5_avg=0.80782 lr=0.01000 gn=5.79291 time=55.72it/s +epoch=8 global_step=3350 loss=5.83603 loss_avg=5.29721 acc=0.41406 acc_top1_avg=0.47508 acc_top5_avg=0.80743 lr=0.01000 gn=6.45636 time=54.62it/s +epoch=8 global_step=3400 loss=5.86576 loss_avg=5.29061 acc=0.39844 acc_top1_avg=0.47527 acc_top5_avg=0.80920 lr=0.01000 gn=6.06171 time=59.07it/s +epoch=8 global_step=3450 loss=5.04760 loss_avg=5.28635 acc=0.49219 acc_top1_avg=0.47579 acc_top5_avg=0.80988 lr=0.01000 gn=5.75264 time=55.02it/s +epoch=8 global_step=3500 loss=5.10726 loss_avg=5.28473 acc=0.50000 acc_top1_avg=0.47585 acc_top5_avg=0.81015 lr=0.01000 gn=7.15163 time=52.85it/s +====================Eval==================== +epoch=8 global_step=3519 loss=1.87610 test_loss_avg=1.32600 acc=0.50781 test_acc_avg=0.64887 test_acc_top5_avg=0.97179 time=240.97it/s +epoch=8 global_step=3519 loss=0.15641 test_loss_avg=1.15233 acc=0.96094 test_acc_avg=0.68222 test_acc_top5_avg=0.96186 time=234.23it/s +epoch=8 global_step=3519 loss=0.28089 test_loss_avg=1.06959 acc=0.93750 test_acc_avg=0.70560 test_acc_top5_avg=0.96568 time=220.81it/s +curr_acc 0.7056 +BEST_ACC 0.6935 +curr_acc_top5 0.9657 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=5.09744 loss_avg=5.25403 acc=0.50781 acc_top1_avg=0.48160 acc_top5_avg=0.80595 lr=0.01000 gn=5.17506 time=52.48it/s +epoch=9 global_step=3600 loss=5.24394 loss_avg=5.17749 acc=0.47656 acc_top1_avg=0.48987 acc_top5_avg=0.82022 lr=0.01000 gn=4.92664 time=57.20it/s +epoch=9 global_step=3650 loss=5.03150 loss_avg=5.18755 acc=0.50781 acc_top1_avg=0.48909 acc_top5_avg=0.81954 lr=0.01000 gn=5.75369 time=54.94it/s +epoch=9 global_step=3700 loss=4.89653 loss_avg=5.20076 acc=0.51562 acc_top1_avg=0.48610 acc_top5_avg=0.81790 lr=0.01000 gn=5.52448 time=52.49it/s +epoch=9 global_step=3750 loss=5.13548 loss_avg=5.22398 acc=0.47656 acc_top1_avg=0.48295 acc_top5_avg=0.81669 lr=0.01000 gn=7.35637 time=55.66it/s +epoch=9 global_step=3800 loss=5.03940 loss_avg=5.23338 acc=0.50000 acc_top1_avg=0.48159 acc_top5_avg=0.81481 lr=0.01000 gn=8.27491 time=55.84it/s +epoch=9 global_step=3850 loss=5.01554 loss_avg=5.23443 acc=0.50781 acc_top1_avg=0.48138 acc_top5_avg=0.81467 lr=0.01000 gn=6.69043 time=56.35it/s +epoch=9 global_step=3900 loss=5.17245 loss_avg=5.25717 acc=0.50781 acc_top1_avg=0.47890 acc_top5_avg=0.81314 lr=0.01000 gn=6.35788 time=60.91it/s +====================Eval==================== +epoch=9 global_step=3910 loss=2.05431 test_loss_avg=2.17377 acc=0.55469 test_acc_avg=0.52163 test_acc_top5_avg=0.94131 time=238.58it/s +epoch=9 global_step=3910 loss=1.23007 test_loss_avg=1.88770 acc=0.56250 test_acc_avg=0.59405 test_acc_top5_avg=0.94007 time=861.43it/s +curr_acc 0.5940 +BEST_ACC 0.7056 +curr_acc_top5 0.9401 +BEST_ACC_top5 0.9657 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=5.27175 loss_avg=5.37934 acc=0.47656 acc_top1_avg=0.46445 acc_top5_avg=0.80781 lr=0.01000 gn=4.96138 time=61.93it/s +epoch=10 global_step=4000 loss=4.81754 loss_avg=5.28093 acc=0.53125 acc_top1_avg=0.47804 acc_top5_avg=0.81172 lr=0.01000 gn=5.99567 time=52.77it/s +epoch=10 global_step=4050 loss=5.13812 loss_avg=5.23827 acc=0.50000 acc_top1_avg=0.48276 acc_top5_avg=0.81295 lr=0.01000 gn=5.52535 time=53.57it/s +epoch=10 global_step=4100 loss=4.69595 loss_avg=5.23849 acc=0.52344 acc_top1_avg=0.48150 acc_top5_avg=0.81345 lr=0.01000 gn=5.22300 time=54.19it/s +epoch=10 global_step=4150 loss=5.16065 loss_avg=5.24264 acc=0.47656 acc_top1_avg=0.48060 acc_top5_avg=0.81370 lr=0.01000 gn=6.22626 time=54.29it/s +epoch=10 global_step=4200 loss=5.53099 loss_avg=5.23363 acc=0.43750 acc_top1_avg=0.48192 acc_top5_avg=0.81398 lr=0.01000 gn=6.14938 time=60.79it/s +epoch=10 global_step=4250 loss=4.96067 loss_avg=5.23408 acc=0.50781 acc_top1_avg=0.48219 acc_top5_avg=0.81436 lr=0.01000 gn=5.62023 time=51.52it/s +epoch=10 global_step=4300 loss=4.73731 loss_avg=5.23559 acc=0.53125 acc_top1_avg=0.48175 acc_top5_avg=0.81456 lr=0.01000 gn=5.49509 time=64.91it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.34234 test_loss_avg=0.75494 acc=0.89062 test_acc_avg=0.76719 test_acc_top5_avg=0.99531 time=245.25it/s +epoch=10 global_step=4301 loss=0.77850 test_loss_avg=1.38694 acc=0.78906 test_acc_avg=0.62669 test_acc_top5_avg=0.93776 time=243.04it/s +epoch=10 global_step=4301 loss=0.15145 test_loss_avg=1.16186 acc=0.87500 test_acc_avg=0.68661 test_acc_top5_avg=0.94838 time=860.72it/s +curr_acc 0.6866 +BEST_ACC 0.7056 +curr_acc_top5 0.9484 +BEST_ACC_top5 0.9657 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=5.16910 loss_avg=5.12524 acc=0.49219 acc_top1_avg=0.49474 acc_top5_avg=0.81760 lr=0.01000 gn=6.78765 time=54.66it/s +epoch=11 global_step=4400 loss=5.31725 loss_avg=5.17533 acc=0.48438 acc_top1_avg=0.48935 acc_top5_avg=0.81597 lr=0.01000 gn=6.34438 time=55.16it/s +epoch=11 global_step=4450 loss=4.73923 loss_avg=5.16902 acc=0.53125 acc_top1_avg=0.48957 acc_top5_avg=0.81822 lr=0.01000 gn=6.06222 time=63.58it/s +epoch=11 global_step=4500 loss=5.70962 loss_avg=5.16133 acc=0.42188 acc_top1_avg=0.49007 acc_top5_avg=0.81737 lr=0.01000 gn=4.69784 time=51.38it/s +epoch=11 global_step=4550 loss=5.06336 loss_avg=5.17386 acc=0.50781 acc_top1_avg=0.48870 acc_top5_avg=0.81743 lr=0.01000 gn=6.17622 time=56.18it/s +epoch=11 global_step=4600 loss=5.34713 loss_avg=5.17866 acc=0.46094 acc_top1_avg=0.48816 acc_top5_avg=0.81658 lr=0.01000 gn=6.10933 time=58.45it/s +epoch=11 global_step=4650 loss=4.97705 loss_avg=5.18582 acc=0.49219 acc_top1_avg=0.48749 acc_top5_avg=0.81563 lr=0.01000 gn=6.50755 time=54.89it/s +====================Eval==================== +epoch=11 global_step=4692 loss=1.63043 test_loss_avg=1.14863 acc=0.57031 test_acc_avg=0.67969 test_acc_top5_avg=0.96673 time=252.96it/s +epoch=11 global_step=4692 loss=0.17308 test_loss_avg=0.92594 acc=0.93750 test_acc_avg=0.73981 test_acc_top5_avg=0.96915 time=719.19it/s +curr_acc 0.7398 +BEST_ACC 0.7056 +curr_acc_top5 0.9691 +BEST_ACC_top5 0.9657 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=4.95878 loss_avg=5.31200 acc=0.53125 acc_top1_avg=0.47461 acc_top5_avg=0.82227 lr=0.01000 gn=7.24033 time=62.49it/s +epoch=12 global_step=4750 loss=5.47464 loss_avg=5.10795 acc=0.45312 acc_top1_avg=0.49380 acc_top5_avg=0.82072 lr=0.01000 gn=6.55058 time=54.75it/s +epoch=12 global_step=4800 loss=5.30419 loss_avg=5.13571 acc=0.46094 acc_top1_avg=0.49161 acc_top5_avg=0.82024 lr=0.01000 gn=5.89438 time=63.52it/s +epoch=12 global_step=4850 loss=5.16851 loss_avg=5.15343 acc=0.48438 acc_top1_avg=0.49061 acc_top5_avg=0.81883 lr=0.01000 gn=6.98307 time=57.14it/s +epoch=12 global_step=4900 loss=5.02990 loss_avg=5.16161 acc=0.49219 acc_top1_avg=0.48892 acc_top5_avg=0.81941 lr=0.01000 gn=5.30978 time=53.13it/s +epoch=12 global_step=4950 loss=5.04046 loss_avg=5.17891 acc=0.51562 acc_top1_avg=0.48731 acc_top5_avg=0.81883 lr=0.01000 gn=7.03153 time=55.06it/s +epoch=12 global_step=5000 loss=5.98699 loss_avg=5.16908 acc=0.38281 acc_top1_avg=0.48864 acc_top5_avg=0.81684 lr=0.01000 gn=5.47513 time=56.34it/s +epoch=12 global_step=5050 loss=5.57613 loss_avg=5.18388 acc=0.46875 acc_top1_avg=0.48736 acc_top5_avg=0.81730 lr=0.01000 gn=7.33079 time=63.73it/s +====================Eval==================== +epoch=12 global_step=5083 loss=0.94246 test_loss_avg=1.00611 acc=0.75000 test_acc_avg=0.73438 test_acc_top5_avg=0.97656 time=236.09it/s +epoch=12 global_step=5083 loss=0.67945 test_loss_avg=1.01977 acc=0.82812 test_acc_avg=0.71184 test_acc_top5_avg=0.96875 time=241.30it/s +epoch=12 global_step=5083 loss=0.33635 test_loss_avg=0.92561 acc=0.81250 test_acc_avg=0.73912 test_acc_top5_avg=0.96776 time=754.51it/s +curr_acc 0.7391 +BEST_ACC 0.7398 +curr_acc_top5 0.9678 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=5.18412 loss_avg=5.22150 acc=0.47656 acc_top1_avg=0.48116 acc_top5_avg=0.80836 lr=0.01000 gn=6.15036 time=56.97it/s +epoch=13 global_step=5150 loss=4.97713 loss_avg=5.20641 acc=0.50781 acc_top1_avg=0.48519 acc_top5_avg=0.81810 lr=0.01000 gn=5.26657 time=32.72it/s +epoch=13 global_step=5200 loss=4.55636 loss_avg=5.17177 acc=0.55469 acc_top1_avg=0.48892 acc_top5_avg=0.81824 lr=0.01000 gn=5.75330 time=59.15it/s +epoch=13 global_step=5250 loss=4.92709 loss_avg=5.17168 acc=0.51562 acc_top1_avg=0.48966 acc_top5_avg=0.81961 lr=0.01000 gn=6.34871 time=54.39it/s +epoch=13 global_step=5300 loss=4.97214 loss_avg=5.16463 acc=0.53125 acc_top1_avg=0.49060 acc_top5_avg=0.81887 lr=0.01000 gn=6.62337 time=55.83it/s +epoch=13 global_step=5350 loss=5.24478 loss_avg=5.17610 acc=0.46875 acc_top1_avg=0.48914 acc_top5_avg=0.81876 lr=0.01000 gn=6.52816 time=52.67it/s +epoch=13 global_step=5400 loss=5.39664 loss_avg=5.17564 acc=0.46094 acc_top1_avg=0.48935 acc_top5_avg=0.81856 lr=0.01000 gn=6.63783 time=50.49it/s +epoch=13 global_step=5450 loss=5.46819 loss_avg=5.17540 acc=0.47656 acc_top1_avg=0.48929 acc_top5_avg=0.81912 lr=0.01000 gn=6.40870 time=40.31it/s +====================Eval==================== +epoch=13 global_step=5474 loss=3.23326 test_loss_avg=1.87138 acc=0.31250 test_acc_avg=0.56963 test_acc_top5_avg=0.94701 time=254.31it/s +epoch=13 global_step=5474 loss=0.46982 test_loss_avg=1.58743 acc=0.89844 test_acc_avg=0.63345 test_acc_top5_avg=0.94253 time=258.21it/s +epoch=13 global_step=5474 loss=0.02285 test_loss_avg=1.48051 acc=1.00000 test_acc_avg=0.65763 test_acc_top5_avg=0.94680 time=843.58it/s +curr_acc 0.6576 +BEST_ACC 0.7398 +curr_acc_top5 0.9468 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=5.40542 loss_avg=5.24608 acc=0.46094 acc_top1_avg=0.48197 acc_top5_avg=0.81310 lr=0.01000 gn=6.79842 time=57.29it/s +epoch=14 global_step=5550 loss=5.64463 loss_avg=5.15514 acc=0.44531 acc_top1_avg=0.49085 acc_top5_avg=0.81517 lr=0.01000 gn=5.23474 time=42.49it/s +epoch=14 global_step=5600 loss=5.01017 loss_avg=5.17208 acc=0.50000 acc_top1_avg=0.48896 acc_top5_avg=0.81523 lr=0.01000 gn=5.57683 time=62.05it/s +epoch=14 global_step=5650 loss=5.67278 loss_avg=5.16575 acc=0.42969 acc_top1_avg=0.49063 acc_top5_avg=0.81556 lr=0.01000 gn=5.63180 time=59.79it/s +epoch=14 global_step=5700 loss=4.48103 loss_avg=5.15644 acc=0.56250 acc_top1_avg=0.49084 acc_top5_avg=0.81810 lr=0.01000 gn=5.33832 time=55.81it/s +epoch=14 global_step=5750 loss=6.07745 loss_avg=5.17860 acc=0.39844 acc_top1_avg=0.48842 acc_top5_avg=0.81709 lr=0.01000 gn=6.85961 time=55.68it/s +epoch=14 global_step=5800 loss=5.26315 loss_avg=5.16068 acc=0.49219 acc_top1_avg=0.49080 acc_top5_avg=0.81739 lr=0.01000 gn=6.37395 time=57.20it/s +epoch=14 global_step=5850 loss=5.55772 loss_avg=5.16862 acc=0.44531 acc_top1_avg=0.48969 acc_top5_avg=0.81697 lr=0.01000 gn=7.07644 time=63.81it/s +====================Eval==================== +epoch=14 global_step=5865 loss=0.87095 test_loss_avg=1.24211 acc=0.76562 test_acc_avg=0.66211 test_acc_top5_avg=0.95224 time=253.17it/s +epoch=14 global_step=5865 loss=0.22561 test_loss_avg=0.94001 acc=0.87500 test_acc_avg=0.73616 test_acc_top5_avg=0.96361 time=870.19it/s +curr_acc 0.7362 +BEST_ACC 0.7398 +curr_acc_top5 0.9636 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=4.65975 loss_avg=5.12642 acc=0.53125 acc_top1_avg=0.49375 acc_top5_avg=0.81853 lr=0.01000 gn=5.53623 time=50.00it/s +epoch=15 global_step=5950 loss=5.53316 loss_avg=5.15896 acc=0.45312 acc_top1_avg=0.48961 acc_top5_avg=0.81820 lr=0.01000 gn=7.01965 time=55.10it/s +epoch=15 global_step=6000 loss=4.63758 loss_avg=5.14586 acc=0.57812 acc_top1_avg=0.49184 acc_top5_avg=0.82002 lr=0.01000 gn=6.84819 time=51.79it/s +epoch=15 global_step=6050 loss=5.26758 loss_avg=5.15454 acc=0.50000 acc_top1_avg=0.49046 acc_top5_avg=0.82078 lr=0.01000 gn=7.76129 time=56.86it/s +epoch=15 global_step=6100 loss=5.93907 loss_avg=5.14253 acc=0.39844 acc_top1_avg=0.49176 acc_top5_avg=0.82128 lr=0.01000 gn=6.78524 time=55.11it/s +epoch=15 global_step=6150 loss=5.45229 loss_avg=5.14847 acc=0.48438 acc_top1_avg=0.49095 acc_top5_avg=0.82185 lr=0.01000 gn=7.38506 time=54.15it/s +epoch=15 global_step=6200 loss=5.09837 loss_avg=5.14846 acc=0.50000 acc_top1_avg=0.49102 acc_top5_avg=0.82134 lr=0.01000 gn=6.51640 time=63.81it/s +epoch=15 global_step=6250 loss=5.10144 loss_avg=5.16194 acc=0.50000 acc_top1_avg=0.48983 acc_top5_avg=0.82045 lr=0.01000 gn=6.33695 time=63.80it/s +====================Eval==================== +epoch=15 global_step=6256 loss=0.26003 test_loss_avg=0.55847 acc=0.91406 test_acc_avg=0.83958 test_acc_top5_avg=0.99792 time=245.40it/s +epoch=15 global_step=6256 loss=0.21455 test_loss_avg=1.03574 acc=0.91406 test_acc_avg=0.70974 test_acc_top5_avg=0.95745 time=92.14it/s +epoch=15 global_step=6256 loss=0.71814 test_loss_avg=0.99118 acc=0.75000 test_acc_avg=0.72557 test_acc_top5_avg=0.95995 time=454.32it/s +curr_acc 0.7256 +BEST_ACC 0.7398 +curr_acc_top5 0.9599 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=5.17917 loss_avg=5.08855 acc=0.50781 acc_top1_avg=0.49556 acc_top5_avg=0.82972 lr=0.01000 gn=5.96608 time=55.49it/s +epoch=16 global_step=6350 loss=5.52840 loss_avg=5.12364 acc=0.45312 acc_top1_avg=0.49335 acc_top5_avg=0.82522 lr=0.01000 gn=5.25106 time=53.81it/s +epoch=16 global_step=6400 loss=5.38155 loss_avg=5.14719 acc=0.45312 acc_top1_avg=0.49061 acc_top5_avg=0.82503 lr=0.01000 gn=7.23088 time=60.26it/s +epoch=16 global_step=6450 loss=5.25977 loss_avg=5.12423 acc=0.46875 acc_top1_avg=0.49336 acc_top5_avg=0.82446 lr=0.01000 gn=6.97057 time=55.22it/s +epoch=16 global_step=6500 loss=4.76303 loss_avg=5.13540 acc=0.53125 acc_top1_avg=0.49267 acc_top5_avg=0.82236 lr=0.01000 gn=7.78298 time=58.48it/s +epoch=16 global_step=6550 loss=4.90783 loss_avg=5.14135 acc=0.52344 acc_top1_avg=0.49229 acc_top5_avg=0.82281 lr=0.01000 gn=5.81367 time=55.01it/s +epoch=16 global_step=6600 loss=5.15462 loss_avg=5.14047 acc=0.52344 acc_top1_avg=0.49237 acc_top5_avg=0.82199 lr=0.01000 gn=8.14842 time=55.98it/s +====================Eval==================== +epoch=16 global_step=6647 loss=0.73618 test_loss_avg=1.21682 acc=0.78125 test_acc_avg=0.67687 test_acc_top5_avg=0.96137 time=243.84it/s +epoch=16 global_step=6647 loss=0.59617 test_loss_avg=1.07242 acc=0.87500 test_acc_avg=0.70896 test_acc_top5_avg=0.96381 time=842.23it/s +curr_acc 0.7090 +BEST_ACC 0.7398 +curr_acc_top5 0.9638 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=5.68463 loss_avg=5.69366 acc=0.44531 acc_top1_avg=0.43490 acc_top5_avg=0.77083 lr=0.01000 gn=7.51663 time=54.80it/s +epoch=17 global_step=6700 loss=5.31146 loss_avg=5.19020 acc=0.46875 acc_top1_avg=0.48614 acc_top5_avg=0.81781 lr=0.01000 gn=8.04601 time=58.51it/s +epoch=17 global_step=6750 loss=5.15793 loss_avg=5.10671 acc=0.48438 acc_top1_avg=0.49575 acc_top5_avg=0.81690 lr=0.01000 gn=7.52380 time=55.90it/s +epoch=17 global_step=6800 loss=4.89199 loss_avg=5.08905 acc=0.50000 acc_top1_avg=0.49821 acc_top5_avg=0.82006 lr=0.01000 gn=6.12050 time=55.76it/s +epoch=17 global_step=6850 loss=6.01667 loss_avg=5.11671 acc=0.35938 acc_top1_avg=0.49484 acc_top5_avg=0.81947 lr=0.01000 gn=5.56314 time=57.97it/s +epoch=17 global_step=6900 loss=4.76191 loss_avg=5.12320 acc=0.54688 acc_top1_avg=0.49407 acc_top5_avg=0.82068 lr=0.01000 gn=8.12767 time=55.35it/s +epoch=17 global_step=6950 loss=4.53515 loss_avg=5.13157 acc=0.56250 acc_top1_avg=0.49296 acc_top5_avg=0.82155 lr=0.01000 gn=6.38620 time=51.56it/s +epoch=17 global_step=7000 loss=5.57192 loss_avg=5.12955 acc=0.44531 acc_top1_avg=0.49323 acc_top5_avg=0.82224 lr=0.01000 gn=6.12983 time=55.48it/s +====================Eval==================== +epoch=17 global_step=7038 loss=2.29064 test_loss_avg=2.20346 acc=0.42969 test_acc_avg=0.45424 test_acc_top5_avg=0.92076 time=227.49it/s +epoch=17 global_step=7038 loss=0.36509 test_loss_avg=1.63323 acc=0.92188 test_acc_avg=0.58786 test_acc_top5_avg=0.94010 time=250.60it/s +epoch=17 global_step=7038 loss=0.00092 test_loss_avg=1.32869 acc=1.00000 test_acc_avg=0.66228 test_acc_top5_avg=0.95194 time=767.48it/s +curr_acc 0.6623 +BEST_ACC 0.7398 +curr_acc_top5 0.9519 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=4.62729 loss_avg=4.96644 acc=0.53906 acc_top1_avg=0.52214 acc_top5_avg=0.83333 lr=0.01000 gn=6.38983 time=56.96it/s +epoch=18 global_step=7100 loss=4.22822 loss_avg=5.06789 acc=0.59375 acc_top1_avg=0.50164 acc_top5_avg=0.82409 lr=0.01000 gn=6.27322 time=54.15it/s +epoch=18 global_step=7150 loss=4.98769 loss_avg=5.11292 acc=0.50781 acc_top1_avg=0.49735 acc_top5_avg=0.82234 lr=0.01000 gn=7.25999 time=53.47it/s +epoch=18 global_step=7200 loss=5.12331 loss_avg=5.09769 acc=0.48438 acc_top1_avg=0.49817 acc_top5_avg=0.82215 lr=0.01000 gn=6.62691 time=53.98it/s +epoch=18 global_step=7250 loss=5.17203 loss_avg=5.10536 acc=0.49219 acc_top1_avg=0.49757 acc_top5_avg=0.82090 lr=0.01000 gn=6.16154 time=55.81it/s +epoch=18 global_step=7300 loss=5.10300 loss_avg=5.11136 acc=0.49219 acc_top1_avg=0.49690 acc_top5_avg=0.81936 lr=0.01000 gn=6.73185 time=63.02it/s +epoch=18 global_step=7350 loss=5.16611 loss_avg=5.11406 acc=0.47656 acc_top1_avg=0.49647 acc_top5_avg=0.81894 lr=0.01000 gn=6.05776 time=54.24it/s +epoch=18 global_step=7400 loss=5.47542 loss_avg=5.12953 acc=0.45312 acc_top1_avg=0.49426 acc_top5_avg=0.81999 lr=0.01000 gn=5.89766 time=62.86it/s +====================Eval==================== +epoch=18 global_step=7429 loss=2.23570 test_loss_avg=1.06351 acc=0.38281 test_acc_avg=0.69141 test_acc_top5_avg=0.96484 time=245.38it/s +epoch=18 global_step=7429 loss=0.67657 test_loss_avg=0.95888 acc=0.78906 test_acc_avg=0.71845 test_acc_top5_avg=0.96645 time=243.08it/s +epoch=18 global_step=7429 loss=0.86954 test_loss_avg=0.95775 acc=0.81250 test_acc_avg=0.71964 test_acc_top5_avg=0.96608 time=464.02it/s +curr_acc 0.7196 +BEST_ACC 0.7398 +curr_acc_top5 0.9661 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=5.85566 loss_avg=5.20101 acc=0.42969 acc_top1_avg=0.48289 acc_top5_avg=0.82626 lr=0.01000 gn=7.41992 time=58.43it/s +epoch=19 global_step=7500 loss=5.63059 loss_avg=5.14595 acc=0.42969 acc_top1_avg=0.48900 acc_top5_avg=0.82174 lr=0.01000 gn=6.66184 time=51.42it/s +epoch=19 global_step=7550 loss=5.33425 loss_avg=5.12323 acc=0.46094 acc_top1_avg=0.49206 acc_top5_avg=0.82109 lr=0.01000 gn=6.55107 time=55.51it/s +epoch=19 global_step=7600 loss=5.42562 loss_avg=5.11948 acc=0.47656 acc_top1_avg=0.49388 acc_top5_avg=0.82237 lr=0.01000 gn=7.20638 time=51.69it/s +epoch=19 global_step=7650 loss=5.12982 loss_avg=5.12539 acc=0.49219 acc_top1_avg=0.49353 acc_top5_avg=0.82212 lr=0.01000 gn=5.58277 time=54.55it/s +epoch=19 global_step=7700 loss=5.23951 loss_avg=5.11450 acc=0.47656 acc_top1_avg=0.49507 acc_top5_avg=0.81991 lr=0.01000 gn=8.50830 time=59.61it/s +epoch=19 global_step=7750 loss=5.73152 loss_avg=5.10987 acc=0.42188 acc_top1_avg=0.49591 acc_top5_avg=0.82126 lr=0.01000 gn=5.56910 time=63.67it/s +epoch=19 global_step=7800 loss=5.62326 loss_avg=5.11657 acc=0.42969 acc_top1_avg=0.49511 acc_top5_avg=0.82115 lr=0.01000 gn=6.54321 time=49.67it/s +====================Eval==================== +epoch=19 global_step=7820 loss=1.37819 test_loss_avg=1.14190 acc=0.67188 test_acc_avg=0.69659 test_acc_top5_avg=0.96253 time=239.25it/s +epoch=19 global_step=7820 loss=0.41562 test_loss_avg=1.11643 acc=0.87500 test_acc_avg=0.70184 test_acc_top5_avg=0.96054 time=876.37it/s +curr_acc 0.7018 +BEST_ACC 0.7398 +curr_acc_top5 0.9605 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=5.23191 loss_avg=5.21311 acc=0.47656 acc_top1_avg=0.48255 acc_top5_avg=0.81276 lr=0.01000 gn=5.82945 time=56.42it/s +epoch=20 global_step=7900 loss=4.74124 loss_avg=5.19848 acc=0.53906 acc_top1_avg=0.48535 acc_top5_avg=0.81777 lr=0.01000 gn=8.02406 time=54.78it/s +epoch=20 global_step=7950 loss=4.96684 loss_avg=5.17778 acc=0.49219 acc_top1_avg=0.48798 acc_top5_avg=0.81544 lr=0.01000 gn=7.49603 time=59.79it/s +epoch=20 global_step=8000 loss=5.34870 loss_avg=5.17400 acc=0.47656 acc_top1_avg=0.48859 acc_top5_avg=0.81589 lr=0.01000 gn=5.77900 time=58.93it/s +epoch=20 global_step=8050 loss=5.16161 loss_avg=5.14704 acc=0.49219 acc_top1_avg=0.49141 acc_top5_avg=0.81664 lr=0.01000 gn=5.93938 time=56.15it/s +epoch=20 global_step=8100 loss=5.33304 loss_avg=5.13766 acc=0.46875 acc_top1_avg=0.49261 acc_top5_avg=0.81855 lr=0.01000 gn=8.02968 time=56.56it/s +epoch=20 global_step=8150 loss=4.62973 loss_avg=5.12541 acc=0.55469 acc_top1_avg=0.49382 acc_top5_avg=0.81970 lr=0.01000 gn=6.36561 time=63.48it/s +epoch=20 global_step=8200 loss=4.88822 loss_avg=5.10816 acc=0.51562 acc_top1_avg=0.49579 acc_top5_avg=0.82079 lr=0.01000 gn=6.87831 time=60.23it/s +====================Eval==================== +epoch=20 global_step=8211 loss=2.21551 test_loss_avg=1.23455 acc=0.43750 test_acc_avg=0.68945 test_acc_top5_avg=0.95195 time=174.48it/s +epoch=20 global_step=8211 loss=0.54433 test_loss_avg=1.35971 acc=0.84375 test_acc_avg=0.65100 test_acc_top5_avg=0.94308 time=250.02it/s +epoch=20 global_step=8211 loss=0.08974 test_loss_avg=1.22415 acc=0.93750 test_acc_avg=0.68493 test_acc_top5_avg=0.94937 time=849.39it/s +curr_acc 0.6849 +BEST_ACC 0.7398 +curr_acc_top5 0.9494 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=4.85197 loss_avg=5.05994 acc=0.51562 acc_top1_avg=0.50160 acc_top5_avg=0.81911 lr=0.01000 gn=6.46827 time=56.84it/s +epoch=21 global_step=8300 loss=5.14745 loss_avg=5.00438 acc=0.48438 acc_top1_avg=0.50878 acc_top5_avg=0.82532 lr=0.01000 gn=6.75518 time=55.45it/s +epoch=21 global_step=8350 loss=5.50726 loss_avg=5.04703 acc=0.44531 acc_top1_avg=0.50450 acc_top5_avg=0.82475 lr=0.01000 gn=7.58722 time=51.09it/s +epoch=21 global_step=8400 loss=5.32413 loss_avg=5.07254 acc=0.46875 acc_top1_avg=0.50095 acc_top5_avg=0.82271 lr=0.01000 gn=6.99959 time=57.07it/s +epoch=21 global_step=8450 loss=5.29118 loss_avg=5.07723 acc=0.50000 acc_top1_avg=0.50065 acc_top5_avg=0.82303 lr=0.01000 gn=6.15444 time=56.80it/s +epoch=21 global_step=8500 loss=5.58790 loss_avg=5.09333 acc=0.43750 acc_top1_avg=0.49838 acc_top5_avg=0.82253 lr=0.01000 gn=5.93833 time=50.72it/s +epoch=21 global_step=8550 loss=5.36172 loss_avg=5.09606 acc=0.46875 acc_top1_avg=0.49770 acc_top5_avg=0.82248 lr=0.01000 gn=6.75506 time=59.77it/s +epoch=21 global_step=8600 loss=5.65341 loss_avg=5.10597 acc=0.45312 acc_top1_avg=0.49683 acc_top5_avg=0.82222 lr=0.01000 gn=6.65042 time=53.51it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.56919 test_loss_avg=1.49734 acc=0.64062 test_acc_avg=0.62271 test_acc_top5_avg=0.94093 time=236.15it/s +epoch=21 global_step=8602 loss=0.00382 test_loss_avg=1.29963 acc=1.00000 test_acc_avg=0.67108 test_acc_top5_avg=0.94314 time=648.07it/s +curr_acc 0.6711 +BEST_ACC 0.7398 +curr_acc_top5 0.9431 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=5.24225 loss_avg=5.00947 acc=0.48438 acc_top1_avg=0.50537 acc_top5_avg=0.82438 lr=0.01000 gn=6.17112 time=57.93it/s +epoch=22 global_step=8700 loss=4.80732 loss_avg=5.06940 acc=0.53125 acc_top1_avg=0.49880 acc_top5_avg=0.81991 lr=0.01000 gn=7.43803 time=56.05it/s +epoch=22 global_step=8750 loss=4.93351 loss_avg=5.07946 acc=0.50000 acc_top1_avg=0.49821 acc_top5_avg=0.82232 lr=0.01000 gn=8.63025 time=61.91it/s +epoch=22 global_step=8800 loss=5.15627 loss_avg=5.08041 acc=0.50781 acc_top1_avg=0.49791 acc_top5_avg=0.82264 lr=0.01000 gn=7.34567 time=55.26it/s +epoch=22 global_step=8850 loss=5.66912 loss_avg=5.10726 acc=0.42969 acc_top1_avg=0.49477 acc_top5_avg=0.82167 lr=0.01000 gn=7.00789 time=62.88it/s +epoch=22 global_step=8900 loss=4.82626 loss_avg=5.10895 acc=0.51562 acc_top1_avg=0.49486 acc_top5_avg=0.82123 lr=0.01000 gn=6.89523 time=58.63it/s +epoch=22 global_step=8950 loss=5.37658 loss_avg=5.09373 acc=0.43750 acc_top1_avg=0.49679 acc_top5_avg=0.82209 lr=0.01000 gn=6.81645 time=54.29it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.41243 test_loss_avg=1.45721 acc=0.85938 test_acc_avg=0.60091 test_acc_top5_avg=0.93555 time=238.16it/s +epoch=22 global_step=8993 loss=1.40830 test_loss_avg=1.32671 acc=0.60156 test_acc_avg=0.64894 test_acc_top5_avg=0.94871 time=140.97it/s +epoch=22 global_step=8993 loss=0.35534 test_loss_avg=1.13712 acc=0.87500 test_acc_avg=0.69571 test_acc_top5_avg=0.95678 time=504.18it/s +curr_acc 0.6957 +BEST_ACC 0.7398 +curr_acc_top5 0.9568 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=4.93943 loss_avg=5.00263 acc=0.50781 acc_top1_avg=0.51116 acc_top5_avg=0.81585 lr=0.01000 gn=7.61821 time=48.31it/s +epoch=23 global_step=9050 loss=4.54341 loss_avg=5.13989 acc=0.55469 acc_top1_avg=0.49219 acc_top5_avg=0.81593 lr=0.01000 gn=6.38705 time=61.19it/s +epoch=23 global_step=9100 loss=4.74543 loss_avg=5.11500 acc=0.53125 acc_top1_avg=0.49547 acc_top5_avg=0.82141 lr=0.01000 gn=6.19478 time=53.79it/s +epoch=23 global_step=9150 loss=5.40605 loss_avg=5.12275 acc=0.45312 acc_top1_avg=0.49468 acc_top5_avg=0.82041 lr=0.01000 gn=9.15952 time=49.90it/s +epoch=23 global_step=9200 loss=4.63052 loss_avg=5.09521 acc=0.57812 acc_top1_avg=0.49706 acc_top5_avg=0.82326 lr=0.01000 gn=7.58501 time=53.20it/s +epoch=23 global_step=9250 loss=5.02845 loss_avg=5.08098 acc=0.49219 acc_top1_avg=0.49869 acc_top5_avg=0.82317 lr=0.01000 gn=7.56140 time=60.72it/s +epoch=23 global_step=9300 loss=4.36477 loss_avg=5.07777 acc=0.60938 acc_top1_avg=0.49924 acc_top5_avg=0.82375 lr=0.01000 gn=8.04367 time=56.19it/s +epoch=23 global_step=9350 loss=5.09350 loss_avg=5.09112 acc=0.50781 acc_top1_avg=0.49801 acc_top5_avg=0.82215 lr=0.01000 gn=7.75166 time=61.06it/s +====================Eval==================== +epoch=23 global_step=9384 loss=1.14906 test_loss_avg=1.90967 acc=0.70312 test_acc_avg=0.54451 test_acc_top5_avg=0.94815 time=239.17it/s +epoch=23 global_step=9384 loss=0.01075 test_loss_avg=1.35276 acc=1.00000 test_acc_avg=0.66357 test_acc_top5_avg=0.94907 time=864.45it/s +curr_acc 0.6636 +BEST_ACC 0.7398 +curr_acc_top5 0.9491 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=4.42467 loss_avg=5.01764 acc=0.57812 acc_top1_avg=0.50635 acc_top5_avg=0.82568 lr=0.01000 gn=6.05699 time=54.34it/s +epoch=24 global_step=9450 loss=5.00420 loss_avg=5.01572 acc=0.47656 acc_top1_avg=0.50414 acc_top5_avg=0.82528 lr=0.01000 gn=7.51960 time=37.40it/s +epoch=24 global_step=9500 loss=5.12815 loss_avg=5.01790 acc=0.50000 acc_top1_avg=0.50485 acc_top5_avg=0.82617 lr=0.01000 gn=7.40858 time=55.88it/s +epoch=24 global_step=9550 loss=4.66256 loss_avg=5.05021 acc=0.53906 acc_top1_avg=0.50108 acc_top5_avg=0.82568 lr=0.01000 gn=7.55350 time=54.53it/s +epoch=24 global_step=9600 loss=5.52726 loss_avg=5.07034 acc=0.45312 acc_top1_avg=0.49866 acc_top5_avg=0.82512 lr=0.01000 gn=6.08773 time=57.09it/s +epoch=24 global_step=9650 loss=5.10445 loss_avg=5.08306 acc=0.49219 acc_top1_avg=0.49715 acc_top5_avg=0.82431 lr=0.01000 gn=5.95097 time=47.22it/s +epoch=24 global_step=9700 loss=4.65948 loss_avg=5.08478 acc=0.55469 acc_top1_avg=0.49740 acc_top5_avg=0.82496 lr=0.01000 gn=6.64240 time=62.04it/s +epoch=24 global_step=9750 loss=5.08299 loss_avg=5.09633 acc=0.51562 acc_top1_avg=0.49609 acc_top5_avg=0.82386 lr=0.01000 gn=7.65094 time=53.76it/s +====================Eval==================== +epoch=24 global_step=9775 loss=1.37107 test_loss_avg=1.29082 acc=0.63281 test_acc_avg=0.64258 test_acc_top5_avg=0.97070 time=252.53it/s +epoch=24 global_step=9775 loss=1.11909 test_loss_avg=1.03324 acc=0.69531 test_acc_avg=0.69980 test_acc_top5_avg=0.95978 time=245.55it/s +epoch=24 global_step=9775 loss=0.35353 test_loss_avg=0.87902 acc=0.87500 test_acc_avg=0.74604 test_acc_top5_avg=0.96232 time=521.61it/s +curr_acc 0.7460 +BEST_ACC 0.7398 +curr_acc_top5 0.9623 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=4.90462 loss_avg=5.07063 acc=0.52344 acc_top1_avg=0.50156 acc_top5_avg=0.82375 lr=0.01000 gn=6.06033 time=55.75it/s +epoch=25 global_step=9850 loss=4.58436 loss_avg=5.09843 acc=0.53125 acc_top1_avg=0.49646 acc_top5_avg=0.82156 lr=0.01000 gn=7.71329 time=57.24it/s +epoch=25 global_step=9900 loss=4.92843 loss_avg=5.11595 acc=0.53125 acc_top1_avg=0.49338 acc_top5_avg=0.82150 lr=0.01000 gn=7.37247 time=57.36it/s +epoch=25 global_step=9950 loss=5.23494 loss_avg=5.10620 acc=0.46094 acc_top1_avg=0.49366 acc_top5_avg=0.82201 lr=0.01000 gn=7.61201 time=61.28it/s +epoch=25 global_step=10000 loss=5.49412 loss_avg=5.09074 acc=0.44531 acc_top1_avg=0.49569 acc_top5_avg=0.82476 lr=0.01000 gn=5.97134 time=62.39it/s +epoch=25 global_step=10050 loss=4.20918 loss_avg=5.09094 acc=0.60156 acc_top1_avg=0.49676 acc_top5_avg=0.82392 lr=0.01000 gn=6.48230 time=42.00it/s +epoch=25 global_step=10100 loss=4.47080 loss_avg=5.09946 acc=0.54688 acc_top1_avg=0.49608 acc_top5_avg=0.82397 lr=0.01000 gn=6.73982 time=62.10it/s +epoch=25 global_step=10150 loss=5.15992 loss_avg=5.09306 acc=0.50000 acc_top1_avg=0.49652 acc_top5_avg=0.82444 lr=0.01000 gn=6.66616 time=54.42it/s +====================Eval==================== +epoch=25 global_step=10166 loss=2.45397 test_loss_avg=1.04580 acc=0.39844 test_acc_avg=0.71719 test_acc_top5_avg=0.97469 time=227.20it/s +epoch=25 global_step=10166 loss=0.17548 test_loss_avg=1.02375 acc=0.94531 test_acc_avg=0.73083 test_acc_top5_avg=0.96198 time=246.40it/s +epoch=25 global_step=10166 loss=0.16312 test_loss_avg=0.98637 acc=0.93750 test_acc_avg=0.74031 test_acc_top5_avg=0.96381 time=521.36it/s +curr_acc 0.7403 +BEST_ACC 0.7460 +curr_acc_top5 0.9638 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=5.18638 loss_avg=4.91370 acc=0.46875 acc_top1_avg=0.51631 acc_top5_avg=0.83272 lr=0.01000 gn=6.97672 time=57.75it/s +epoch=26 global_step=10250 loss=5.85187 loss_avg=5.01337 acc=0.42969 acc_top1_avg=0.50716 acc_top5_avg=0.82487 lr=0.01000 gn=6.12797 time=55.18it/s +epoch=26 global_step=10300 loss=4.42068 loss_avg=5.00732 acc=0.57031 acc_top1_avg=0.50805 acc_top5_avg=0.82667 lr=0.01000 gn=6.57759 time=52.50it/s +epoch=26 global_step=10350 loss=4.70270 loss_avg=5.06111 acc=0.53125 acc_top1_avg=0.50161 acc_top5_avg=0.82371 lr=0.01000 gn=6.00883 time=55.21it/s +epoch=26 global_step=10400 loss=5.22190 loss_avg=5.06849 acc=0.48438 acc_top1_avg=0.50057 acc_top5_avg=0.82345 lr=0.01000 gn=6.89804 time=61.60it/s +epoch=26 global_step=10450 loss=4.48527 loss_avg=5.06800 acc=0.56250 acc_top1_avg=0.50080 acc_top5_avg=0.82392 lr=0.01000 gn=8.25423 time=53.35it/s +epoch=26 global_step=10500 loss=5.54023 loss_avg=5.08525 acc=0.43750 acc_top1_avg=0.49890 acc_top5_avg=0.82490 lr=0.01000 gn=6.50001 time=60.79it/s +epoch=26 global_step=10550 loss=4.53440 loss_avg=5.07939 acc=0.56250 acc_top1_avg=0.49917 acc_top5_avg=0.82452 lr=0.01000 gn=6.22762 time=54.89it/s +====================Eval==================== +epoch=26 global_step=10557 loss=1.19155 test_loss_avg=1.49099 acc=0.67188 test_acc_avg=0.64606 test_acc_top5_avg=0.93359 time=243.74it/s +epoch=26 global_step=10557 loss=0.21185 test_loss_avg=1.46058 acc=0.93750 test_acc_avg=0.65012 test_acc_top5_avg=0.93008 time=557.31it/s +curr_acc 0.6501 +BEST_ACC 0.7460 +curr_acc_top5 0.9301 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=5.48372 loss_avg=4.98368 acc=0.43750 acc_top1_avg=0.50545 acc_top5_avg=0.83467 lr=0.01000 gn=5.65692 time=52.11it/s +epoch=27 global_step=10650 loss=5.01194 loss_avg=5.00440 acc=0.50781 acc_top1_avg=0.50487 acc_top5_avg=0.83249 lr=0.01000 gn=7.40788 time=55.86it/s +epoch=27 global_step=10700 loss=5.07245 loss_avg=5.04078 acc=0.50781 acc_top1_avg=0.50148 acc_top5_avg=0.82960 lr=0.01000 gn=7.43305 time=58.48it/s +epoch=27 global_step=10750 loss=5.27983 loss_avg=5.07485 acc=0.47656 acc_top1_avg=0.49794 acc_top5_avg=0.82910 lr=0.01000 gn=7.40532 time=57.26it/s +epoch=27 global_step=10800 loss=4.76049 loss_avg=5.06088 acc=0.53125 acc_top1_avg=0.49929 acc_top5_avg=0.82893 lr=0.01000 gn=8.47101 time=56.47it/s +epoch=27 global_step=10850 loss=5.29684 loss_avg=5.07702 acc=0.46094 acc_top1_avg=0.49784 acc_top5_avg=0.82807 lr=0.01000 gn=5.63658 time=46.78it/s +epoch=27 global_step=10900 loss=5.64494 loss_avg=5.07775 acc=0.46094 acc_top1_avg=0.49811 acc_top5_avg=0.82769 lr=0.01000 gn=7.01828 time=56.55it/s +====================Eval==================== +epoch=27 global_step=10948 loss=2.24284 test_loss_avg=1.18454 acc=0.44531 test_acc_avg=0.70680 test_acc_top5_avg=0.98208 time=228.30it/s +epoch=27 global_step=10948 loss=0.21051 test_loss_avg=1.55518 acc=0.93750 test_acc_avg=0.61742 test_acc_top5_avg=0.92607 time=233.30it/s +epoch=27 global_step=10948 loss=0.00340 test_loss_avg=1.34042 acc=1.00000 test_acc_avg=0.66831 test_acc_top5_avg=0.93730 time=511.94it/s +curr_acc 0.6683 +BEST_ACC 0.7460 +curr_acc_top5 0.9373 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=4.60757 loss_avg=4.95447 acc=0.52344 acc_top1_avg=0.50000 acc_top5_avg=0.81641 lr=0.01000 gn=6.77057 time=48.74it/s +epoch=28 global_step=11000 loss=5.42333 loss_avg=5.08061 acc=0.44531 acc_top1_avg=0.49624 acc_top5_avg=0.82557 lr=0.01000 gn=9.04741 time=58.26it/s +epoch=28 global_step=11050 loss=5.56553 loss_avg=5.05470 acc=0.46094 acc_top1_avg=0.50031 acc_top5_avg=0.82667 lr=0.01000 gn=7.24814 time=53.39it/s +epoch=28 global_step=11100 loss=5.58263 loss_avg=5.04843 acc=0.44531 acc_top1_avg=0.50154 acc_top5_avg=0.82550 lr=0.01000 gn=5.48853 time=46.62it/s +epoch=28 global_step=11150 loss=4.83600 loss_avg=5.04294 acc=0.54688 acc_top1_avg=0.50217 acc_top5_avg=0.82720 lr=0.01000 gn=7.96319 time=54.52it/s +epoch=28 global_step=11200 loss=4.75978 loss_avg=5.04964 acc=0.55469 acc_top1_avg=0.50133 acc_top5_avg=0.82558 lr=0.01000 gn=6.43997 time=59.29it/s +epoch=28 global_step=11250 loss=5.52270 loss_avg=5.06155 acc=0.46094 acc_top1_avg=0.50005 acc_top5_avg=0.82458 lr=0.01000 gn=5.82500 time=59.96it/s +epoch=28 global_step=11300 loss=5.09692 loss_avg=5.06139 acc=0.50000 acc_top1_avg=0.50038 acc_top5_avg=0.82486 lr=0.01000 gn=6.98373 time=51.97it/s +====================Eval==================== +epoch=28 global_step=11339 loss=1.06269 test_loss_avg=1.28301 acc=0.72656 test_acc_avg=0.63775 test_acc_top5_avg=0.94942 time=239.70it/s +epoch=28 global_step=11339 loss=0.44455 test_loss_avg=0.97939 acc=0.87500 test_acc_avg=0.71974 test_acc_top5_avg=0.95609 time=469.90it/s +curr_acc 0.7197 +BEST_ACC 0.7460 +curr_acc_top5 0.9561 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=4.68198 loss_avg=4.85081 acc=0.55469 acc_top1_avg=0.52770 acc_top5_avg=0.83239 lr=0.01000 gn=7.16859 time=63.91it/s +epoch=29 global_step=11400 loss=5.00111 loss_avg=4.97503 acc=0.50000 acc_top1_avg=0.51345 acc_top5_avg=0.82428 lr=0.01000 gn=7.53693 time=57.56it/s +epoch=29 global_step=11450 loss=5.26628 loss_avg=4.97959 acc=0.50781 acc_top1_avg=0.51112 acc_top5_avg=0.82651 lr=0.01000 gn=7.08103 time=54.47it/s +epoch=29 global_step=11500 loss=4.30942 loss_avg=5.02083 acc=0.58594 acc_top1_avg=0.50631 acc_top5_avg=0.82463 lr=0.01000 gn=7.02342 time=57.69it/s +epoch=29 global_step=11550 loss=5.47959 loss_avg=5.05295 acc=0.46094 acc_top1_avg=0.50293 acc_top5_avg=0.82376 lr=0.01000 gn=6.69486 time=57.73it/s +epoch=29 global_step=11600 loss=4.62333 loss_avg=5.04900 acc=0.53906 acc_top1_avg=0.50260 acc_top5_avg=0.82534 lr=0.01000 gn=6.89574 time=42.15it/s +epoch=29 global_step=11650 loss=4.98421 loss_avg=5.06018 acc=0.49219 acc_top1_avg=0.50153 acc_top5_avg=0.82529 lr=0.01000 gn=6.49282 time=49.60it/s +epoch=29 global_step=11700 loss=4.79089 loss_avg=5.06436 acc=0.53125 acc_top1_avg=0.50113 acc_top5_avg=0.82505 lr=0.01000 gn=6.19440 time=63.18it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.42936 test_loss_avg=1.96469 acc=0.89062 test_acc_avg=0.53906 test_acc_top5_avg=0.95399 time=247.36it/s +epoch=29 global_step=11730 loss=2.38980 test_loss_avg=1.57522 acc=0.54688 test_acc_avg=0.61653 test_acc_top5_avg=0.95299 time=240.21it/s +epoch=29 global_step=11730 loss=0.23754 test_loss_avg=1.37544 acc=0.87500 test_acc_avg=0.66218 test_acc_top5_avg=0.95580 time=162.82it/s +curr_acc 0.6622 +BEST_ACC 0.7460 +curr_acc_top5 0.9558 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=4.75376 loss_avg=4.88059 acc=0.51562 acc_top1_avg=0.51641 acc_top5_avg=0.82617 lr=0.01000 gn=5.64339 time=54.43it/s +epoch=30 global_step=11800 loss=4.67288 loss_avg=4.99855 acc=0.54688 acc_top1_avg=0.50926 acc_top5_avg=0.82388 lr=0.01000 gn=6.22203 time=58.30it/s +epoch=30 global_step=11850 loss=4.43731 loss_avg=5.04341 acc=0.55469 acc_top1_avg=0.50384 acc_top5_avg=0.82539 lr=0.01000 gn=6.75318 time=56.83it/s +epoch=30 global_step=11900 loss=5.10475 loss_avg=5.05564 acc=0.48438 acc_top1_avg=0.50221 acc_top5_avg=0.82711 lr=0.01000 gn=9.52296 time=61.67it/s +epoch=30 global_step=11950 loss=5.23740 loss_avg=5.05570 acc=0.50000 acc_top1_avg=0.50146 acc_top5_avg=0.82670 lr=0.01000 gn=6.50706 time=58.35it/s +epoch=30 global_step=12000 loss=5.03421 loss_avg=5.06373 acc=0.50781 acc_top1_avg=0.50041 acc_top5_avg=0.82613 lr=0.01000 gn=5.20116 time=59.08it/s +epoch=30 global_step=12050 loss=4.85991 loss_avg=5.07930 acc=0.53906 acc_top1_avg=0.49878 acc_top5_avg=0.82632 lr=0.01000 gn=7.92908 time=57.76it/s +epoch=30 global_step=12100 loss=5.02216 loss_avg=5.06911 acc=0.49219 acc_top1_avg=0.49922 acc_top5_avg=0.82728 lr=0.01000 gn=6.00232 time=59.95it/s +====================Eval==================== +epoch=30 global_step=12121 loss=1.60418 test_loss_avg=2.49203 acc=0.46875 test_acc_avg=0.41745 test_acc_top5_avg=0.94271 time=242.01it/s +epoch=30 global_step=12121 loss=0.00153 test_loss_avg=1.63295 acc=1.00000 test_acc_avg=0.59701 test_acc_top5_avg=0.93958 time=799.83it/s +curr_acc 0.5970 +BEST_ACC 0.7460 +curr_acc_top5 0.9396 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=4.86647 loss_avg=5.11771 acc=0.53125 acc_top1_avg=0.49488 acc_top5_avg=0.82247 lr=0.01000 gn=6.81898 time=50.12it/s +epoch=31 global_step=12200 loss=5.09340 loss_avg=5.02771 acc=0.47656 acc_top1_avg=0.50307 acc_top5_avg=0.82911 lr=0.01000 gn=8.17894 time=48.77it/s +epoch=31 global_step=12250 loss=5.16764 loss_avg=5.04682 acc=0.48438 acc_top1_avg=0.50224 acc_top5_avg=0.82631 lr=0.01000 gn=9.40930 time=55.31it/s +epoch=31 global_step=12300 loss=4.24140 loss_avg=5.04036 acc=0.58594 acc_top1_avg=0.50236 acc_top5_avg=0.82887 lr=0.01000 gn=6.79611 time=56.53it/s +epoch=31 global_step=12350 loss=5.04298 loss_avg=5.04387 acc=0.50000 acc_top1_avg=0.50252 acc_top5_avg=0.82932 lr=0.01000 gn=8.18176 time=62.79it/s +epoch=31 global_step=12400 loss=4.92274 loss_avg=5.04985 acc=0.53125 acc_top1_avg=0.50204 acc_top5_avg=0.82897 lr=0.01000 gn=7.14237 time=62.78it/s +epoch=31 global_step=12450 loss=5.73660 loss_avg=5.06000 acc=0.39844 acc_top1_avg=0.50081 acc_top5_avg=0.82846 lr=0.01000 gn=7.20366 time=52.32it/s +epoch=31 global_step=12500 loss=5.09858 loss_avg=5.07469 acc=0.50000 acc_top1_avg=0.49924 acc_top5_avg=0.82691 lr=0.01000 gn=7.02664 time=59.17it/s +====================Eval==================== +epoch=31 global_step=12512 loss=0.63123 test_loss_avg=0.63123 acc=0.83594 test_acc_avg=0.83594 test_acc_top5_avg=0.98438 time=209.03it/s +epoch=31 global_step=12512 loss=1.22356 test_loss_avg=1.15269 acc=0.66406 test_acc_avg=0.68505 test_acc_top5_avg=0.96109 time=238.07it/s +epoch=31 global_step=12512 loss=0.00426 test_loss_avg=0.93017 acc=1.00000 test_acc_avg=0.74426 test_acc_top5_avg=0.96489 time=475.87it/s +curr_acc 0.7443 +BEST_ACC 0.7460 +curr_acc_top5 0.9649 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=5.47107 loss_avg=5.00529 acc=0.43750 acc_top1_avg=0.50905 acc_top5_avg=0.82072 lr=0.01000 gn=7.42843 time=49.24it/s +epoch=32 global_step=12600 loss=5.19804 loss_avg=5.04760 acc=0.46875 acc_top1_avg=0.50240 acc_top5_avg=0.82076 lr=0.01000 gn=6.38354 time=39.61it/s +epoch=32 global_step=12650 loss=4.37379 loss_avg=5.07119 acc=0.57812 acc_top1_avg=0.49802 acc_top5_avg=0.82139 lr=0.01000 gn=8.06256 time=60.54it/s +epoch=32 global_step=12700 loss=5.02558 loss_avg=5.08297 acc=0.50781 acc_top1_avg=0.49680 acc_top5_avg=0.82526 lr=0.01000 gn=7.87937 time=51.59it/s +epoch=32 global_step=12750 loss=4.60831 loss_avg=5.05753 acc=0.55469 acc_top1_avg=0.50023 acc_top5_avg=0.82639 lr=0.01000 gn=6.74777 time=53.74it/s +epoch=32 global_step=12800 loss=5.40278 loss_avg=5.06213 acc=0.46875 acc_top1_avg=0.50019 acc_top5_avg=0.82693 lr=0.01000 gn=7.29920 time=54.46it/s +epoch=32 global_step=12850 loss=5.13974 loss_avg=5.05576 acc=0.48438 acc_top1_avg=0.50099 acc_top5_avg=0.82833 lr=0.01000 gn=5.88839 time=57.09it/s +epoch=32 global_step=12900 loss=5.72109 loss_avg=5.06330 acc=0.41406 acc_top1_avg=0.50020 acc_top5_avg=0.82710 lr=0.01000 gn=7.18645 time=60.19it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.15157 test_loss_avg=1.21544 acc=0.71875 test_acc_avg=0.68572 test_acc_top5_avg=0.98295 time=243.30it/s +epoch=32 global_step=12903 loss=0.80376 test_loss_avg=1.80594 acc=0.78906 test_acc_avg=0.59581 test_acc_top5_avg=0.94607 time=236.25it/s +epoch=32 global_step=12903 loss=0.53481 test_loss_avg=1.70333 acc=0.87500 test_acc_avg=0.61679 test_acc_top5_avg=0.94937 time=473.13it/s +curr_acc 0.6168 +BEST_ACC 0.7460 +curr_acc_top5 0.9494 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=5.38531 loss_avg=5.05730 acc=0.46094 acc_top1_avg=0.50332 acc_top5_avg=0.82613 lr=0.01000 gn=7.37981 time=50.76it/s +epoch=33 global_step=13000 loss=4.38072 loss_avg=4.99604 acc=0.57812 acc_top1_avg=0.50910 acc_top5_avg=0.83183 lr=0.01000 gn=7.26899 time=25.32it/s +epoch=33 global_step=13050 loss=4.95216 loss_avg=5.01195 acc=0.50781 acc_top1_avg=0.50744 acc_top5_avg=0.82961 lr=0.01000 gn=7.93636 time=55.80it/s +epoch=33 global_step=13100 loss=5.90554 loss_avg=5.03263 acc=0.39844 acc_top1_avg=0.50512 acc_top5_avg=0.82816 lr=0.01000 gn=7.06619 time=54.68it/s +epoch=33 global_step=13150 loss=4.97115 loss_avg=5.04252 acc=0.49219 acc_top1_avg=0.50316 acc_top5_avg=0.82790 lr=0.01000 gn=8.17222 time=54.53it/s +epoch=33 global_step=13200 loss=4.85897 loss_avg=5.04173 acc=0.53125 acc_top1_avg=0.50310 acc_top5_avg=0.82812 lr=0.01000 gn=6.04165 time=50.72it/s +epoch=33 global_step=13250 loss=5.25159 loss_avg=5.04103 acc=0.48438 acc_top1_avg=0.50331 acc_top5_avg=0.82817 lr=0.01000 gn=6.04760 time=54.70it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.55960 test_loss_avg=1.07106 acc=0.82812 test_acc_avg=0.69677 test_acc_top5_avg=0.96875 time=236.07it/s +epoch=33 global_step=13294 loss=0.75318 test_loss_avg=0.88380 acc=0.87500 test_acc_avg=0.74575 test_acc_top5_avg=0.97102 time=485.56it/s +curr_acc 0.7457 +BEST_ACC 0.7460 +curr_acc_top5 0.9710 +BEST_ACC_top5 0.9691 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=5.66347 loss_avg=5.27561 acc=0.45312 acc_top1_avg=0.47526 acc_top5_avg=0.82161 lr=0.01000 gn=7.32593 time=59.97it/s +epoch=34 global_step=13350 loss=4.93784 loss_avg=5.06873 acc=0.52344 acc_top1_avg=0.50042 acc_top5_avg=0.82896 lr=0.01000 gn=7.92978 time=39.02it/s +epoch=34 global_step=13400 loss=5.52729 loss_avg=5.03541 acc=0.46875 acc_top1_avg=0.50501 acc_top5_avg=0.82849 lr=0.01000 gn=8.25564 time=54.83it/s +epoch=34 global_step=13450 loss=4.54046 loss_avg=5.03208 acc=0.57812 acc_top1_avg=0.50456 acc_top5_avg=0.83228 lr=0.01000 gn=8.07591 time=56.18it/s +epoch=34 global_step=13500 loss=5.09523 loss_avg=5.04541 acc=0.46875 acc_top1_avg=0.50235 acc_top5_avg=0.83048 lr=0.01000 gn=6.36174 time=51.09it/s +epoch=34 global_step=13550 loss=4.29665 loss_avg=5.04833 acc=0.57812 acc_top1_avg=0.50192 acc_top5_avg=0.82974 lr=0.01000 gn=6.08374 time=53.52it/s +epoch=34 global_step=13600 loss=4.63486 loss_avg=5.06127 acc=0.55469 acc_top1_avg=0.50033 acc_top5_avg=0.82869 lr=0.01000 gn=6.05578 time=51.91it/s +epoch=34 global_step=13650 loss=5.40172 loss_avg=5.07560 acc=0.46094 acc_top1_avg=0.49886 acc_top5_avg=0.82867 lr=0.01000 gn=7.17890 time=54.69it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.91459 test_loss_avg=0.68742 acc=0.77344 test_acc_avg=0.80804 test_acc_top5_avg=0.99330 time=230.96it/s +epoch=34 global_step=13685 loss=0.34812 test_loss_avg=1.23424 acc=0.89062 test_acc_avg=0.66296 test_acc_top5_avg=0.91602 time=242.38it/s +epoch=34 global_step=13685 loss=0.44640 test_loss_avg=1.10792 acc=0.87500 test_acc_avg=0.69571 test_acc_top5_avg=0.92919 time=717.34it/s +curr_acc 0.6957 +BEST_ACC 0.7460 +curr_acc_top5 0.9292 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=5.10980 loss_avg=5.06918 acc=0.48438 acc_top1_avg=0.49792 acc_top5_avg=0.82031 lr=0.01000 gn=6.90778 time=60.54it/s +epoch=35 global_step=13750 loss=4.45268 loss_avg=4.97705 acc=0.55469 acc_top1_avg=0.51214 acc_top5_avg=0.83185 lr=0.01000 gn=6.74595 time=55.19it/s +epoch=35 global_step=13800 loss=5.34012 loss_avg=4.99217 acc=0.46875 acc_top1_avg=0.50992 acc_top5_avg=0.82894 lr=0.01000 gn=6.05960 time=55.61it/s +epoch=35 global_step=13850 loss=4.86255 loss_avg=5.01221 acc=0.50000 acc_top1_avg=0.50696 acc_top5_avg=0.82666 lr=0.01000 gn=6.38996 time=55.18it/s +epoch=35 global_step=13900 loss=4.60224 loss_avg=5.00587 acc=0.53906 acc_top1_avg=0.50767 acc_top5_avg=0.82533 lr=0.01000 gn=8.91774 time=63.98it/s +epoch=35 global_step=13950 loss=5.25257 loss_avg=5.02992 acc=0.46094 acc_top1_avg=0.50422 acc_top5_avg=0.82485 lr=0.01000 gn=7.25005 time=53.83it/s +epoch=35 global_step=14000 loss=5.44690 loss_avg=5.03257 acc=0.45312 acc_top1_avg=0.50427 acc_top5_avg=0.82649 lr=0.01000 gn=6.28752 time=58.63it/s +epoch=35 global_step=14050 loss=5.51467 loss_avg=5.04973 acc=0.42969 acc_top1_avg=0.50227 acc_top5_avg=0.82564 lr=0.01000 gn=7.25777 time=62.53it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.92985 test_loss_avg=1.16546 acc=0.72656 test_acc_avg=0.68214 test_acc_top5_avg=0.95692 time=243.06it/s +epoch=35 global_step=14076 loss=0.01034 test_loss_avg=0.98215 acc=1.00000 test_acc_avg=0.73299 test_acc_top5_avg=0.96183 time=481.00it/s +curr_acc 0.7330 +BEST_ACC 0.7460 +curr_acc_top5 0.9618 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=5.00901 loss_avg=4.91687 acc=0.52344 acc_top1_avg=0.51888 acc_top5_avg=0.83496 lr=0.01000 gn=7.84591 time=54.11it/s +epoch=36 global_step=14150 loss=5.08451 loss_avg=5.02844 acc=0.49219 acc_top1_avg=0.50549 acc_top5_avg=0.82812 lr=0.01000 gn=7.24676 time=56.23it/s +epoch=36 global_step=14200 loss=4.84325 loss_avg=5.03422 acc=0.53125 acc_top1_avg=0.50447 acc_top5_avg=0.82800 lr=0.01000 gn=6.34717 time=53.34it/s +epoch=36 global_step=14250 loss=4.32163 loss_avg=5.03557 acc=0.57031 acc_top1_avg=0.50418 acc_top5_avg=0.82539 lr=0.01000 gn=6.95821 time=57.00it/s +epoch=36 global_step=14300 loss=5.26955 loss_avg=5.04858 acc=0.46875 acc_top1_avg=0.50265 acc_top5_avg=0.82561 lr=0.01000 gn=7.73984 time=60.50it/s +epoch=36 global_step=14350 loss=5.05817 loss_avg=5.04830 acc=0.50781 acc_top1_avg=0.50279 acc_top5_avg=0.82781 lr=0.01000 gn=7.75145 time=54.20it/s +epoch=36 global_step=14400 loss=4.66973 loss_avg=5.05664 acc=0.57031 acc_top1_avg=0.50152 acc_top5_avg=0.82694 lr=0.01000 gn=6.53568 time=56.03it/s +epoch=36 global_step=14450 loss=5.37954 loss_avg=5.04194 acc=0.43750 acc_top1_avg=0.50343 acc_top5_avg=0.82702 lr=0.01000 gn=7.02567 time=60.91it/s +====================Eval==================== +epoch=36 global_step=14467 loss=1.18274 test_loss_avg=1.13817 acc=0.72656 test_acc_avg=0.71875 test_acc_top5_avg=1.00000 time=238.20it/s +epoch=36 global_step=14467 loss=1.06975 test_loss_avg=1.63347 acc=0.71875 test_acc_avg=0.59208 test_acc_top5_avg=0.93192 time=186.80it/s +epoch=36 global_step=14467 loss=0.43842 test_loss_avg=1.33202 acc=0.87500 test_acc_avg=0.66416 test_acc_top5_avg=0.94660 time=286.99it/s +curr_acc 0.6642 +BEST_ACC 0.7460 +curr_acc_top5 0.9466 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=5.09057 loss_avg=5.06978 acc=0.53906 acc_top1_avg=0.50047 acc_top5_avg=0.82907 lr=0.01000 gn=7.71601 time=56.87it/s +epoch=37 global_step=14550 loss=5.03726 loss_avg=5.04554 acc=0.48438 acc_top1_avg=0.50169 acc_top5_avg=0.83020 lr=0.01000 gn=6.76500 time=63.17it/s +epoch=37 global_step=14600 loss=4.79700 loss_avg=5.02684 acc=0.53125 acc_top1_avg=0.50394 acc_top5_avg=0.83106 lr=0.01000 gn=7.12300 time=53.00it/s +epoch=37 global_step=14650 loss=5.78826 loss_avg=5.02718 acc=0.42188 acc_top1_avg=0.50363 acc_top5_avg=0.83090 lr=0.01000 gn=6.66669 time=59.56it/s +epoch=37 global_step=14700 loss=4.86155 loss_avg=4.99945 acc=0.53906 acc_top1_avg=0.50657 acc_top5_avg=0.83024 lr=0.01000 gn=7.90415 time=49.67it/s +epoch=37 global_step=14750 loss=4.76366 loss_avg=5.02811 acc=0.54688 acc_top1_avg=0.50417 acc_top5_avg=0.82939 lr=0.01000 gn=7.38612 time=58.34it/s +epoch=37 global_step=14800 loss=4.79974 loss_avg=5.02460 acc=0.53125 acc_top1_avg=0.50457 acc_top5_avg=0.82881 lr=0.01000 gn=7.30387 time=55.20it/s +epoch=37 global_step=14850 loss=4.74323 loss_avg=5.04128 acc=0.53125 acc_top1_avg=0.50235 acc_top5_avg=0.82829 lr=0.01000 gn=6.75698 time=61.07it/s +====================Eval==================== +epoch=37 global_step=14858 loss=2.38204 test_loss_avg=0.88595 acc=0.35938 test_acc_avg=0.74711 test_acc_top5_avg=0.96701 time=249.62it/s +epoch=37 global_step=14858 loss=1.00140 test_loss_avg=0.90808 acc=0.74219 test_acc_avg=0.75477 test_acc_top5_avg=0.95586 time=245.40it/s +epoch=37 global_step=14858 loss=0.52790 test_loss_avg=0.90162 acc=0.87500 test_acc_avg=0.75653 test_acc_top5_avg=0.95669 time=672.16it/s +curr_acc 0.7565 +BEST_ACC 0.7460 +curr_acc_top5 0.9567 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=6.16811 loss_avg=4.96107 acc=0.38281 acc_top1_avg=0.51283 acc_top5_avg=0.82571 lr=0.01000 gn=7.44918 time=27.82it/s +epoch=38 global_step=14950 loss=4.89167 loss_avg=4.99736 acc=0.51562 acc_top1_avg=0.50968 acc_top5_avg=0.82583 lr=0.01000 gn=7.46167 time=59.50it/s +epoch=38 global_step=15000 loss=5.84119 loss_avg=5.05636 acc=0.40625 acc_top1_avg=0.50358 acc_top5_avg=0.82422 lr=0.01000 gn=6.06370 time=55.67it/s +epoch=38 global_step=15050 loss=5.69192 loss_avg=5.04475 acc=0.42188 acc_top1_avg=0.50419 acc_top5_avg=0.82324 lr=0.01000 gn=8.22557 time=55.06it/s +epoch=38 global_step=15100 loss=4.81382 loss_avg=5.04210 acc=0.54688 acc_top1_avg=0.50410 acc_top5_avg=0.82432 lr=0.01000 gn=7.88681 time=59.93it/s +epoch=38 global_step=15150 loss=5.10932 loss_avg=5.05267 acc=0.50000 acc_top1_avg=0.50249 acc_top5_avg=0.82400 lr=0.01000 gn=5.93304 time=56.76it/s +epoch=38 global_step=15200 loss=5.03782 loss_avg=5.05118 acc=0.50781 acc_top1_avg=0.50233 acc_top5_avg=0.82529 lr=0.01000 gn=9.06967 time=53.85it/s +====================Eval==================== +epoch=38 global_step=15249 loss=2.29500 test_loss_avg=1.51390 acc=0.47656 test_acc_avg=0.63493 test_acc_top5_avg=0.94808 time=229.06it/s +epoch=38 global_step=15249 loss=0.00369 test_loss_avg=1.19161 acc=1.00000 test_acc_avg=0.70708 test_acc_top5_avg=0.95481 time=464.49it/s +curr_acc 0.7071 +BEST_ACC 0.7565 +curr_acc_top5 0.9548 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=4.63312 loss_avg=4.63312 acc=0.54688 acc_top1_avg=0.54688 acc_top5_avg=0.86719 lr=0.01000 gn=7.63019 time=48.70it/s +epoch=39 global_step=15300 loss=5.12777 loss_avg=5.01843 acc=0.49219 acc_top1_avg=0.50398 acc_top5_avg=0.82491 lr=0.01000 gn=6.50345 time=42.56it/s +epoch=39 global_step=15350 loss=5.63769 loss_avg=5.03634 acc=0.42969 acc_top1_avg=0.50294 acc_top5_avg=0.82433 lr=0.01000 gn=6.82762 time=59.80it/s +epoch=39 global_step=15400 loss=4.61451 loss_avg=5.04259 acc=0.57031 acc_top1_avg=0.50238 acc_top5_avg=0.82533 lr=0.01000 gn=7.80346 time=59.23it/s +epoch=39 global_step=15450 loss=4.95297 loss_avg=5.04268 acc=0.51562 acc_top1_avg=0.50190 acc_top5_avg=0.82649 lr=0.01000 gn=6.84126 time=46.35it/s +epoch=39 global_step=15500 loss=4.35988 loss_avg=5.03780 acc=0.57812 acc_top1_avg=0.50246 acc_top5_avg=0.82654 lr=0.01000 gn=6.75655 time=62.60it/s +epoch=39 global_step=15550 loss=5.37297 loss_avg=5.02495 acc=0.46875 acc_top1_avg=0.50420 acc_top5_avg=0.82735 lr=0.01000 gn=7.18768 time=59.97it/s +epoch=39 global_step=15600 loss=5.87699 loss_avg=5.02932 acc=0.42188 acc_top1_avg=0.50425 acc_top5_avg=0.82708 lr=0.01000 gn=7.22203 time=55.28it/s +====================Eval==================== +epoch=39 global_step=15640 loss=1.39954 test_loss_avg=0.54820 acc=0.67188 test_acc_avg=0.84334 test_acc_top5_avg=0.98890 time=239.78it/s +epoch=39 global_step=15640 loss=0.41845 test_loss_avg=1.14376 acc=0.87500 test_acc_avg=0.70675 test_acc_top5_avg=0.95358 time=251.88it/s +epoch=39 global_step=15640 loss=2.24912 test_loss_avg=1.19097 acc=0.43750 test_acc_avg=0.69828 test_acc_top5_avg=0.95055 time=505.46it/s +curr_acc 0.6983 +BEST_ACC 0.7565 +curr_acc_top5 0.9506 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=4.90157 loss_avg=5.02729 acc=0.50781 acc_top1_avg=0.50000 acc_top5_avg=0.80937 lr=0.00100 gn=5.22699 time=53.58it/s +epoch=40 global_step=15700 loss=4.70604 loss_avg=4.90363 acc=0.54688 acc_top1_avg=0.51654 acc_top5_avg=0.82578 lr=0.00100 gn=5.81126 time=54.87it/s +epoch=40 global_step=15750 loss=4.96326 loss_avg=4.86796 acc=0.50000 acc_top1_avg=0.51953 acc_top5_avg=0.83097 lr=0.00100 gn=5.03246 time=56.87it/s +epoch=40 global_step=15800 loss=5.26730 loss_avg=4.82704 acc=0.47656 acc_top1_avg=0.52432 acc_top5_avg=0.83218 lr=0.00100 gn=7.13005 time=53.75it/s +epoch=40 global_step=15850 loss=3.98134 loss_avg=4.80539 acc=0.60938 acc_top1_avg=0.52693 acc_top5_avg=0.83460 lr=0.00100 gn=5.95833 time=56.27it/s +epoch=40 global_step=15900 loss=4.81541 loss_avg=4.78390 acc=0.53906 acc_top1_avg=0.52918 acc_top5_avg=0.83696 lr=0.00100 gn=7.21308 time=55.24it/s +epoch=40 global_step=15950 loss=4.57287 loss_avg=4.76014 acc=0.53125 acc_top1_avg=0.53120 acc_top5_avg=0.83929 lr=0.00100 gn=4.96642 time=60.22it/s +epoch=40 global_step=16000 loss=4.48702 loss_avg=4.74677 acc=0.54688 acc_top1_avg=0.53264 acc_top5_avg=0.84049 lr=0.00100 gn=6.99604 time=53.08it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.74612 test_loss_avg=0.62525 acc=0.81250 test_acc_avg=0.82285 test_acc_top5_avg=0.97930 time=238.58it/s +epoch=40 global_step=16031 loss=0.05861 test_loss_avg=0.52017 acc=0.93750 test_acc_avg=0.84909 test_acc_top5_avg=0.98269 time=486.63it/s +curr_acc 0.8491 +BEST_ACC 0.7565 +curr_acc_top5 0.9827 +BEST_ACC_top5 0.9710 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=4.87421 loss_avg=4.47656 acc=0.51562 acc_top1_avg=0.56086 acc_top5_avg=0.84457 lr=0.00100 gn=5.39371 time=64.02it/s +epoch=41 global_step=16100 loss=5.15058 loss_avg=4.59898 acc=0.49219 acc_top1_avg=0.54744 acc_top5_avg=0.84318 lr=0.00100 gn=7.18330 time=55.47it/s +epoch=41 global_step=16150 loss=4.96686 loss_avg=4.60842 acc=0.50781 acc_top1_avg=0.54661 acc_top5_avg=0.84217 lr=0.00100 gn=5.61723 time=55.47it/s +epoch=41 global_step=16200 loss=4.91955 loss_avg=4.61641 acc=0.50781 acc_top1_avg=0.54590 acc_top5_avg=0.84148 lr=0.00100 gn=7.59694 time=60.50it/s +epoch=41 global_step=16250 loss=4.80339 loss_avg=4.62208 acc=0.53906 acc_top1_avg=0.54484 acc_top5_avg=0.84364 lr=0.00100 gn=6.70597 time=51.41it/s +epoch=41 global_step=16300 loss=4.20474 loss_avg=4.60976 acc=0.57812 acc_top1_avg=0.54624 acc_top5_avg=0.84491 lr=0.00100 gn=4.21563 time=55.45it/s +epoch=41 global_step=16350 loss=4.46192 loss_avg=4.62195 acc=0.55469 acc_top1_avg=0.54521 acc_top5_avg=0.84468 lr=0.00100 gn=6.03941 time=52.61it/s +epoch=41 global_step=16400 loss=4.37826 loss_avg=4.61443 acc=0.58594 acc_top1_avg=0.54594 acc_top5_avg=0.84527 lr=0.00100 gn=8.01785 time=53.77it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.26687 test_loss_avg=0.60230 acc=0.92969 test_acc_avg=0.83878 test_acc_top5_avg=0.99432 time=243.97it/s +epoch=41 global_step=16422 loss=0.15802 test_loss_avg=0.65855 acc=0.94531 test_acc_avg=0.81634 test_acc_top5_avg=0.97682 time=240.14it/s +epoch=41 global_step=16422 loss=0.03205 test_loss_avg=0.54143 acc=1.00000 test_acc_avg=0.84889 test_acc_top5_avg=0.98101 time=707.66it/s +curr_acc 0.8489 +BEST_ACC 0.8491 +curr_acc_top5 0.9810 +BEST_ACC_top5 0.9827 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=3.85504 loss_avg=4.46109 acc=0.63281 acc_top1_avg=0.56334 acc_top5_avg=0.84570 lr=0.00100 gn=5.89589 time=56.34it/s +epoch=42 global_step=16500 loss=4.66411 loss_avg=4.48973 acc=0.53906 acc_top1_avg=0.55779 acc_top5_avg=0.85036 lr=0.00100 gn=7.61115 time=55.97it/s +epoch=42 global_step=16550 loss=4.42803 loss_avg=4.51385 acc=0.56250 acc_top1_avg=0.55536 acc_top5_avg=0.85034 lr=0.00100 gn=6.10003 time=54.19it/s +epoch=42 global_step=16600 loss=4.29971 loss_avg=4.50388 acc=0.60156 acc_top1_avg=0.55675 acc_top5_avg=0.85126 lr=0.00100 gn=8.45012 time=54.09it/s +epoch=42 global_step=16650 loss=4.16464 loss_avg=4.52388 acc=0.60156 acc_top1_avg=0.55513 acc_top5_avg=0.85095 lr=0.00100 gn=7.77015 time=54.76it/s +epoch=42 global_step=16700 loss=4.48540 loss_avg=4.54080 acc=0.57812 acc_top1_avg=0.55331 acc_top5_avg=0.85066 lr=0.00100 gn=7.76805 time=52.26it/s +epoch=42 global_step=16750 loss=4.54184 loss_avg=4.55398 acc=0.55469 acc_top1_avg=0.55169 acc_top5_avg=0.84918 lr=0.00100 gn=5.88950 time=56.03it/s +epoch=42 global_step=16800 loss=4.29277 loss_avg=4.57186 acc=0.58594 acc_top1_avg=0.54985 acc_top5_avg=0.84826 lr=0.00100 gn=6.60134 time=62.38it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.50640 test_loss_avg=0.67316 acc=0.85156 test_acc_avg=0.81030 test_acc_top5_avg=0.97803 time=242.46it/s +epoch=42 global_step=16813 loss=0.17158 test_loss_avg=0.51074 acc=0.93750 test_acc_avg=0.85354 test_acc_top5_avg=0.98319 time=458.49it/s +curr_acc 0.8535 +BEST_ACC 0.8491 +curr_acc_top5 0.9832 +BEST_ACC_top5 0.9827 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=4.76613 loss_avg=4.65965 acc=0.53906 acc_top1_avg=0.54751 acc_top5_avg=0.84481 lr=0.00100 gn=6.46559 time=46.77it/s +epoch=43 global_step=16900 loss=4.29427 loss_avg=4.59508 acc=0.57812 acc_top1_avg=0.54975 acc_top5_avg=0.84824 lr=0.00100 gn=8.93420 time=60.30it/s +epoch=43 global_step=16950 loss=4.57562 loss_avg=4.55385 acc=0.54688 acc_top1_avg=0.55372 acc_top5_avg=0.85025 lr=0.00100 gn=7.31302 time=55.49it/s +epoch=43 global_step=17000 loss=4.17254 loss_avg=4.54549 acc=0.59375 acc_top1_avg=0.55460 acc_top5_avg=0.85077 lr=0.00100 gn=7.64097 time=62.52it/s +epoch=43 global_step=17050 loss=5.18401 loss_avg=4.53291 acc=0.48438 acc_top1_avg=0.55515 acc_top5_avg=0.85044 lr=0.00100 gn=5.99059 time=59.39it/s +epoch=43 global_step=17100 loss=4.83235 loss_avg=4.51807 acc=0.52344 acc_top1_avg=0.55676 acc_top5_avg=0.85115 lr=0.00100 gn=5.94904 time=55.71it/s +epoch=43 global_step=17150 loss=4.10238 loss_avg=4.51809 acc=0.60938 acc_top1_avg=0.55668 acc_top5_avg=0.85047 lr=0.00100 gn=7.93621 time=55.14it/s +epoch=43 global_step=17200 loss=4.52030 loss_avg=4.52488 acc=0.54688 acc_top1_avg=0.55584 acc_top5_avg=0.85092 lr=0.00100 gn=7.06433 time=60.26it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.55346 test_loss_avg=0.57196 acc=0.85156 test_acc_avg=0.84115 test_acc_top5_avg=0.98698 time=221.48it/s +epoch=43 global_step=17204 loss=0.51864 test_loss_avg=0.61101 acc=0.84375 test_acc_avg=0.82709 test_acc_top5_avg=0.97730 time=231.59it/s +epoch=43 global_step=17204 loss=0.03313 test_loss_avg=0.48861 acc=1.00000 test_acc_avg=0.86056 test_acc_top5_avg=0.98230 time=493.16it/s +curr_acc 0.8606 +BEST_ACC 0.8535 +curr_acc_top5 0.9823 +BEST_ACC_top5 0.9832 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=4.23143 loss_avg=4.51233 acc=0.60938 acc_top1_avg=0.55622 acc_top5_avg=0.85411 lr=0.00100 gn=8.51722 time=54.42it/s +epoch=44 global_step=17300 loss=3.95906 loss_avg=4.50127 acc=0.62500 acc_top1_avg=0.55802 acc_top5_avg=0.85156 lr=0.00100 gn=7.37706 time=55.23it/s +epoch=44 global_step=17350 loss=4.43989 loss_avg=4.49955 acc=0.57812 acc_top1_avg=0.55891 acc_top5_avg=0.85113 lr=0.00100 gn=8.32959 time=55.05it/s +epoch=44 global_step=17400 loss=4.22985 loss_avg=4.49907 acc=0.58594 acc_top1_avg=0.55839 acc_top5_avg=0.85144 lr=0.00100 gn=7.09642 time=55.26it/s +epoch=44 global_step=17450 loss=4.18343 loss_avg=4.48478 acc=0.57812 acc_top1_avg=0.56015 acc_top5_avg=0.85210 lr=0.00100 gn=8.84920 time=50.30it/s +epoch=44 global_step=17500 loss=4.36928 loss_avg=4.49002 acc=0.55469 acc_top1_avg=0.55976 acc_top5_avg=0.85122 lr=0.00100 gn=7.94893 time=57.23it/s +epoch=44 global_step=17550 loss=5.16158 loss_avg=4.50161 acc=0.48438 acc_top1_avg=0.55857 acc_top5_avg=0.85059 lr=0.00100 gn=9.16994 time=54.11it/s +====================Eval==================== +epoch=44 global_step=17595 loss=1.15195 test_loss_avg=0.50948 acc=0.70312 test_acc_avg=0.85059 test_acc_top5_avg=0.98796 time=254.91it/s +epoch=44 global_step=17595 loss=0.15881 test_loss_avg=0.50753 acc=0.96875 test_acc_avg=0.85579 test_acc_top5_avg=0.98152 time=234.95it/s +epoch=44 global_step=17595 loss=0.04278 test_loss_avg=0.48146 acc=1.00000 test_acc_avg=0.86303 test_acc_top5_avg=0.98259 time=506.62it/s +curr_acc 0.8630 +BEST_ACC 0.8606 +curr_acc_top5 0.9826 +BEST_ACC_top5 0.9832 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=5.33299 loss_avg=4.31795 acc=0.46875 acc_top1_avg=0.58125 acc_top5_avg=0.86250 lr=0.00100 gn=9.09721 time=57.61it/s +epoch=45 global_step=17650 loss=4.19839 loss_avg=4.45827 acc=0.58594 acc_top1_avg=0.56250 acc_top5_avg=0.84815 lr=0.00100 gn=8.15292 time=50.57it/s +epoch=45 global_step=17700 loss=4.87985 loss_avg=4.46508 acc=0.53125 acc_top1_avg=0.56220 acc_top5_avg=0.84710 lr=0.00100 gn=8.11665 time=55.84it/s +epoch=45 global_step=17750 loss=5.03883 loss_avg=4.48492 acc=0.50000 acc_top1_avg=0.56008 acc_top5_avg=0.84894 lr=0.00100 gn=6.83563 time=58.41it/s +epoch=45 global_step=17800 loss=5.02624 loss_avg=4.47976 acc=0.50781 acc_top1_avg=0.56101 acc_top5_avg=0.85004 lr=0.00100 gn=7.53568 time=52.75it/s +epoch=45 global_step=17850 loss=4.76022 loss_avg=4.47413 acc=0.54688 acc_top1_avg=0.56210 acc_top5_avg=0.84942 lr=0.00100 gn=6.32921 time=25.63it/s +epoch=45 global_step=17900 loss=4.94062 loss_avg=4.45554 acc=0.51562 acc_top1_avg=0.56396 acc_top5_avg=0.85118 lr=0.00100 gn=9.19557 time=55.74it/s +epoch=45 global_step=17950 loss=4.69616 loss_avg=4.45609 acc=0.52344 acc_top1_avg=0.56382 acc_top5_avg=0.85143 lr=0.00100 gn=7.71892 time=60.72it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.62888 test_loss_avg=0.58570 acc=0.79688 test_acc_avg=0.83316 test_acc_top5_avg=0.98142 time=238.20it/s +epoch=45 global_step=17986 loss=0.05969 test_loss_avg=0.47139 acc=0.93750 test_acc_avg=0.86363 test_acc_top5_avg=0.98457 time=489.70it/s +curr_acc 0.8636 +BEST_ACC 0.8630 +curr_acc_top5 0.9846 +BEST_ACC_top5 0.9832 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=4.29481 loss_avg=4.33788 acc=0.57031 acc_top1_avg=0.57143 acc_top5_avg=0.85100 lr=0.00100 gn=6.78987 time=54.94it/s +epoch=46 global_step=18050 loss=4.88095 loss_avg=4.45084 acc=0.53125 acc_top1_avg=0.56189 acc_top5_avg=0.85706 lr=0.00100 gn=12.21894 time=57.08it/s +epoch=46 global_step=18100 loss=4.71458 loss_avg=4.45570 acc=0.53906 acc_top1_avg=0.56305 acc_top5_avg=0.85430 lr=0.00100 gn=8.36394 time=60.13it/s +epoch=46 global_step=18150 loss=4.57857 loss_avg=4.43641 acc=0.55469 acc_top1_avg=0.56541 acc_top5_avg=0.85528 lr=0.00100 gn=8.91045 time=56.42it/s +epoch=46 global_step=18200 loss=4.89213 loss_avg=4.43903 acc=0.51562 acc_top1_avg=0.56553 acc_top5_avg=0.85401 lr=0.00100 gn=8.43129 time=51.16it/s +epoch=46 global_step=18250 loss=4.81415 loss_avg=4.43983 acc=0.53125 acc_top1_avg=0.56528 acc_top5_avg=0.85337 lr=0.00100 gn=9.87573 time=53.69it/s +epoch=46 global_step=18300 loss=3.49840 loss_avg=4.43575 acc=0.68750 acc_top1_avg=0.56556 acc_top5_avg=0.85233 lr=0.00100 gn=8.82545 time=54.76it/s +epoch=46 global_step=18350 loss=4.46710 loss_avg=4.42457 acc=0.56250 acc_top1_avg=0.56666 acc_top5_avg=0.85229 lr=0.00100 gn=7.91514 time=55.51it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.58243 test_loss_avg=0.42518 acc=0.85938 test_acc_avg=0.88135 test_acc_top5_avg=0.99463 time=251.77it/s +epoch=46 global_step=18377 loss=0.12190 test_loss_avg=0.54414 acc=0.96875 test_acc_avg=0.84399 test_acc_top5_avg=0.98047 time=235.53it/s +epoch=46 global_step=18377 loss=0.03277 test_loss_avg=0.47418 acc=1.00000 test_acc_avg=0.86353 test_acc_top5_avg=0.98319 time=620.73it/s +curr_acc 0.8635 +BEST_ACC 0.8636 +curr_acc_top5 0.9832 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=4.37174 loss_avg=4.49254 acc=0.57031 acc_top1_avg=0.55673 acc_top5_avg=0.84205 lr=0.00100 gn=8.56693 time=57.55it/s +epoch=47 global_step=18450 loss=4.48510 loss_avg=4.42064 acc=0.55469 acc_top1_avg=0.56550 acc_top5_avg=0.85220 lr=0.00100 gn=11.83710 time=56.81it/s +epoch=47 global_step=18500 loss=4.17946 loss_avg=4.40441 acc=0.58594 acc_top1_avg=0.56847 acc_top5_avg=0.85448 lr=0.00100 gn=8.62786 time=53.03it/s +epoch=47 global_step=18550 loss=3.79449 loss_avg=4.40592 acc=0.64844 acc_top1_avg=0.56814 acc_top5_avg=0.85278 lr=0.00100 gn=11.49762 time=48.02it/s +epoch=47 global_step=18600 loss=3.82671 loss_avg=4.38908 acc=0.63281 acc_top1_avg=0.56996 acc_top5_avg=0.85293 lr=0.00100 gn=8.28126 time=60.39it/s +epoch=47 global_step=18650 loss=4.52918 loss_avg=4.38508 acc=0.56250 acc_top1_avg=0.57028 acc_top5_avg=0.85331 lr=0.00100 gn=6.70289 time=58.13it/s +epoch=47 global_step=18700 loss=4.22310 loss_avg=4.41185 acc=0.57031 acc_top1_avg=0.56707 acc_top5_avg=0.85214 lr=0.00100 gn=8.89911 time=60.20it/s +epoch=47 global_step=18750 loss=4.70999 loss_avg=4.40878 acc=0.53125 acc_top1_avg=0.56769 acc_top5_avg=0.85225 lr=0.00100 gn=8.31983 time=58.20it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.22825 test_loss_avg=0.62652 acc=0.93750 test_acc_avg=0.82285 test_acc_top5_avg=0.98226 time=240.78it/s +epoch=47 global_step=18768 loss=0.10509 test_loss_avg=0.49160 acc=0.93750 test_acc_avg=0.85987 test_acc_top5_avg=0.98408 time=518.33it/s +curr_acc 0.8599 +BEST_ACC 0.8636 +curr_acc_top5 0.9841 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=3.61503 loss_avg=4.32660 acc=0.64844 acc_top1_avg=0.57397 acc_top5_avg=0.84448 lr=0.00100 gn=9.65713 time=50.26it/s +epoch=48 global_step=18850 loss=4.40458 loss_avg=4.35860 acc=0.55469 acc_top1_avg=0.57117 acc_top5_avg=0.85004 lr=0.00100 gn=8.48843 time=55.13it/s +epoch=48 global_step=18900 loss=4.47059 loss_avg=4.34837 acc=0.54688 acc_top1_avg=0.57363 acc_top5_avg=0.85085 lr=0.00100 gn=6.62826 time=55.30it/s +epoch=48 global_step=18950 loss=4.63157 loss_avg=4.35700 acc=0.54688 acc_top1_avg=0.57259 acc_top5_avg=0.85345 lr=0.00100 gn=10.65991 time=58.51it/s +epoch=48 global_step=19000 loss=4.11632 loss_avg=4.36287 acc=0.60938 acc_top1_avg=0.57193 acc_top5_avg=0.85436 lr=0.00100 gn=10.92605 time=59.66it/s +epoch=48 global_step=19050 loss=4.15366 loss_avg=4.36447 acc=0.59375 acc_top1_avg=0.57175 acc_top5_avg=0.85550 lr=0.00100 gn=9.91582 time=55.20it/s +epoch=48 global_step=19100 loss=4.42350 loss_avg=4.38387 acc=0.56250 acc_top1_avg=0.57001 acc_top5_avg=0.85420 lr=0.00100 gn=9.32018 time=62.67it/s +epoch=48 global_step=19150 loss=4.37998 loss_avg=4.39341 acc=0.55469 acc_top1_avg=0.56884 acc_top5_avg=0.85418 lr=0.00100 gn=8.00315 time=55.58it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.27580 test_loss_avg=0.46109 acc=0.88281 test_acc_avg=0.85742 test_acc_top5_avg=0.99414 time=242.80it/s +epoch=48 global_step=19159 loss=0.36837 test_loss_avg=0.62627 acc=0.92969 test_acc_avg=0.82220 test_acc_top5_avg=0.97252 time=239.88it/s +epoch=48 global_step=19159 loss=0.22100 test_loss_avg=0.50878 acc=0.87500 test_acc_avg=0.85473 test_acc_top5_avg=0.97844 time=818.56it/s +curr_acc 0.8547 +BEST_ACC 0.8636 +curr_acc_top5 0.9784 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=4.45075 loss_avg=4.41533 acc=0.54688 acc_top1_avg=0.56593 acc_top5_avg=0.85804 lr=0.00100 gn=7.73356 time=55.64it/s +epoch=49 global_step=19250 loss=4.39968 loss_avg=4.39117 acc=0.57812 acc_top1_avg=0.56928 acc_top5_avg=0.85268 lr=0.00100 gn=11.56624 time=55.95it/s +epoch=49 global_step=19300 loss=4.19792 loss_avg=4.37629 acc=0.58594 acc_top1_avg=0.57048 acc_top5_avg=0.85577 lr=0.00100 gn=8.77786 time=54.89it/s +epoch=49 global_step=19350 loss=3.90661 loss_avg=4.37571 acc=0.62500 acc_top1_avg=0.57105 acc_top5_avg=0.85565 lr=0.00100 gn=12.15040 time=53.86it/s +epoch=49 global_step=19400 loss=3.85538 loss_avg=4.37466 acc=0.63281 acc_top1_avg=0.57135 acc_top5_avg=0.85510 lr=0.00100 gn=10.24405 time=63.67it/s +epoch=49 global_step=19450 loss=4.90180 loss_avg=4.35649 acc=0.51562 acc_top1_avg=0.57348 acc_top5_avg=0.85556 lr=0.00100 gn=10.28951 time=62.87it/s +epoch=49 global_step=19500 loss=3.90713 loss_avg=4.35935 acc=0.61719 acc_top1_avg=0.57329 acc_top5_avg=0.85557 lr=0.00100 gn=10.59479 time=63.75it/s +epoch=49 global_step=19550 loss=4.55073 loss_avg=4.35805 acc=0.55000 acc_top1_avg=0.57356 acc_top5_avg=0.85435 lr=0.00100 gn=12.11170 time=85.51it/s +====================Eval==================== +epoch=49 global_step=19550 loss=1.00383 test_loss_avg=0.59271 acc=0.71094 test_acc_avg=0.83082 test_acc_top5_avg=0.98033 time=87.57it/s +epoch=49 global_step=19550 loss=0.14118 test_loss_avg=0.48109 acc=0.87500 test_acc_avg=0.85829 test_acc_top5_avg=0.98408 time=485.34it/s +epoch=49 global_step=19550 loss=0.14118 test_loss_avg=0.48109 acc=0.87500 test_acc_avg=0.85829 test_acc_top5_avg=0.98408 time=485.34it/s +curr_acc 0.8583 +BEST_ACC 0.8636 +curr_acc_top5 0.9841 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=4.65793 loss_avg=4.38165 acc=0.54688 acc_top1_avg=0.56797 acc_top5_avg=0.85156 lr=0.00100 gn=10.16491 time=49.66it/s +epoch=50 global_step=19650 loss=3.69037 loss_avg=4.36819 acc=0.65625 acc_top1_avg=0.57273 acc_top5_avg=0.85000 lr=0.00100 gn=12.07659 time=56.95it/s +epoch=50 global_step=19700 loss=4.31190 loss_avg=4.38224 acc=0.59375 acc_top1_avg=0.57099 acc_top5_avg=0.85010 lr=0.00100 gn=10.05226 time=55.87it/s +epoch=50 global_step=19750 loss=4.55180 loss_avg=4.35903 acc=0.55469 acc_top1_avg=0.57371 acc_top5_avg=0.85363 lr=0.00100 gn=9.93072 time=53.48it/s +epoch=50 global_step=19800 loss=4.20386 loss_avg=4.35504 acc=0.58594 acc_top1_avg=0.57391 acc_top5_avg=0.85372 lr=0.00100 gn=10.50855 time=56.19it/s +epoch=50 global_step=19850 loss=4.56997 loss_avg=4.34794 acc=0.55469 acc_top1_avg=0.57451 acc_top5_avg=0.85411 lr=0.00100 gn=12.15865 time=51.72it/s +epoch=50 global_step=19900 loss=4.25844 loss_avg=4.33952 acc=0.57031 acc_top1_avg=0.57556 acc_top5_avg=0.85435 lr=0.00100 gn=10.81989 time=52.52it/s +====================Eval==================== +epoch=50 global_step=19941 loss=0.81771 test_loss_avg=0.59699 acc=0.79688 test_acc_avg=0.82859 test_acc_top5_avg=0.97828 time=67.75it/s +epoch=50 global_step=19941 loss=0.24455 test_loss_avg=0.48852 acc=0.87500 test_acc_avg=0.85799 test_acc_top5_avg=0.98259 time=817.60it/s +curr_acc 0.8580 +BEST_ACC 0.8636 +curr_acc_top5 0.9826 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=4.68580 loss_avg=4.44412 acc=0.53906 acc_top1_avg=0.56337 acc_top5_avg=0.85069 lr=0.00100 gn=10.05639 time=52.73it/s +epoch=51 global_step=20000 loss=4.06648 loss_avg=4.33917 acc=0.60938 acc_top1_avg=0.57548 acc_top5_avg=0.85739 lr=0.00100 gn=12.76212 time=57.07it/s +epoch=51 global_step=20050 loss=4.08433 loss_avg=4.32486 acc=0.58594 acc_top1_avg=0.57597 acc_top5_avg=0.85565 lr=0.00100 gn=8.57759 time=58.31it/s +epoch=51 global_step=20100 loss=4.74133 loss_avg=4.33639 acc=0.56250 acc_top1_avg=0.57528 acc_top5_avg=0.85323 lr=0.00100 gn=13.79130 time=59.35it/s +epoch=51 global_step=20150 loss=4.05223 loss_avg=4.32519 acc=0.61719 acc_top1_avg=0.57689 acc_top5_avg=0.85425 lr=0.00100 gn=9.12326 time=54.50it/s +epoch=51 global_step=20200 loss=4.71015 loss_avg=4.32178 acc=0.53125 acc_top1_avg=0.57737 acc_top5_avg=0.85431 lr=0.00100 gn=10.76767 time=52.83it/s +epoch=51 global_step=20250 loss=3.91502 loss_avg=4.31761 acc=0.63281 acc_top1_avg=0.57770 acc_top5_avg=0.85493 lr=0.00100 gn=11.29309 time=59.01it/s +epoch=51 global_step=20300 loss=3.52898 loss_avg=4.32780 acc=0.65625 acc_top1_avg=0.57667 acc_top5_avg=0.85461 lr=0.00100 gn=9.85423 time=50.11it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.40971 test_loss_avg=0.50739 acc=0.86719 test_acc_avg=0.85528 test_acc_top5_avg=0.99070 time=230.98it/s +epoch=51 global_step=20332 loss=0.08093 test_loss_avg=0.53157 acc=0.96875 test_acc_avg=0.84815 test_acc_top5_avg=0.97997 time=146.63it/s +epoch=51 global_step=20332 loss=0.02649 test_loss_avg=0.49177 acc=1.00000 test_acc_avg=0.85938 test_acc_top5_avg=0.98170 time=514.70it/s +curr_acc 0.8594 +BEST_ACC 0.8636 +curr_acc_top5 0.9817 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=3.78337 loss_avg=4.18950 acc=0.64062 acc_top1_avg=0.59245 acc_top5_avg=0.86024 lr=0.00100 gn=10.09384 time=63.21it/s +epoch=52 global_step=20400 loss=4.66952 loss_avg=4.24999 acc=0.53906 acc_top1_avg=0.58617 acc_top5_avg=0.85432 lr=0.00100 gn=13.77644 time=64.16it/s +epoch=52 global_step=20450 loss=4.34853 loss_avg=4.28338 acc=0.56250 acc_top1_avg=0.58283 acc_top5_avg=0.85388 lr=0.00100 gn=6.62392 time=59.14it/s +epoch=52 global_step=20500 loss=3.38053 loss_avg=4.27847 acc=0.68750 acc_top1_avg=0.58305 acc_top5_avg=0.85338 lr=0.00100 gn=11.03423 time=63.43it/s +epoch=52 global_step=20550 loss=3.91465 loss_avg=4.29560 acc=0.62500 acc_top1_avg=0.58071 acc_top5_avg=0.85393 lr=0.00100 gn=9.31139 time=52.80it/s +epoch=52 global_step=20600 loss=4.01468 loss_avg=4.30705 acc=0.61719 acc_top1_avg=0.57932 acc_top5_avg=0.85474 lr=0.00100 gn=11.54943 time=54.97it/s +epoch=52 global_step=20650 loss=3.86187 loss_avg=4.30120 acc=0.62500 acc_top1_avg=0.58002 acc_top5_avg=0.85466 lr=0.00100 gn=11.98534 time=59.84it/s +epoch=52 global_step=20700 loss=4.54934 loss_avg=4.30659 acc=0.56250 acc_top1_avg=0.57900 acc_top5_avg=0.85521 lr=0.00100 gn=12.05050 time=51.61it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.67287 test_loss_avg=0.58616 acc=0.82031 test_acc_avg=0.83017 test_acc_top5_avg=0.97805 time=237.36it/s +epoch=52 global_step=20723 loss=0.06404 test_loss_avg=0.49110 acc=0.93750 test_acc_avg=0.85354 test_acc_top5_avg=0.98250 time=796.79it/s +curr_acc 0.8535 +BEST_ACC 0.8636 +curr_acc_top5 0.9825 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=4.40209 loss_avg=4.25286 acc=0.56250 acc_top1_avg=0.58160 acc_top5_avg=0.85822 lr=0.00100 gn=11.66688 time=56.21it/s +epoch=53 global_step=20800 loss=5.07014 loss_avg=4.22829 acc=0.48438 acc_top1_avg=0.58644 acc_top5_avg=0.85907 lr=0.00100 gn=10.51487 time=57.56it/s +epoch=53 global_step=20850 loss=4.65672 loss_avg=4.26847 acc=0.55469 acc_top1_avg=0.58317 acc_top5_avg=0.85556 lr=0.00100 gn=11.73349 time=59.67it/s +epoch=53 global_step=20900 loss=4.43943 loss_avg=4.26220 acc=0.57812 acc_top1_avg=0.58377 acc_top5_avg=0.85567 lr=0.00100 gn=12.02393 time=55.59it/s +epoch=53 global_step=20950 loss=4.64488 loss_avg=4.25708 acc=0.53906 acc_top1_avg=0.58429 acc_top5_avg=0.85566 lr=0.00100 gn=13.86844 time=58.50it/s +epoch=53 global_step=21000 loss=4.89084 loss_avg=4.27664 acc=0.50781 acc_top1_avg=0.58255 acc_top5_avg=0.85514 lr=0.00100 gn=12.55977 time=59.68it/s +epoch=53 global_step=21050 loss=4.12003 loss_avg=4.27363 acc=0.60938 acc_top1_avg=0.58305 acc_top5_avg=0.85522 lr=0.00100 gn=11.01126 time=57.52it/s +epoch=53 global_step=21100 loss=4.17118 loss_avg=4.27749 acc=0.60156 acc_top1_avg=0.58281 acc_top5_avg=0.85577 lr=0.00100 gn=12.75151 time=54.28it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.10075 test_loss_avg=0.52613 acc=0.96094 test_acc_avg=0.84135 test_acc_top5_avg=0.99279 time=235.91it/s +epoch=53 global_step=21114 loss=0.17246 test_loss_avg=0.64330 acc=0.96094 test_acc_avg=0.82006 test_acc_top5_avg=0.97594 time=241.77it/s +epoch=53 global_step=21114 loss=0.02735 test_loss_avg=0.54261 acc=1.00000 test_acc_avg=0.84810 test_acc_top5_avg=0.98012 time=485.90it/s +curr_acc 0.8481 +BEST_ACC 0.8636 +curr_acc_top5 0.9801 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=4.45507 loss_avg=4.17792 acc=0.57812 acc_top1_avg=0.59353 acc_top5_avg=0.85981 lr=0.00100 gn=12.39108 time=59.70it/s +epoch=54 global_step=21200 loss=3.92936 loss_avg=4.25051 acc=0.60156 acc_top1_avg=0.58712 acc_top5_avg=0.85947 lr=0.00100 gn=10.40997 time=55.06it/s +epoch=54 global_step=21250 loss=4.06064 loss_avg=4.23712 acc=0.60156 acc_top1_avg=0.58801 acc_top5_avg=0.85794 lr=0.00100 gn=12.43999 time=55.80it/s +epoch=54 global_step=21300 loss=4.45104 loss_avg=4.26059 acc=0.56250 acc_top1_avg=0.58409 acc_top5_avg=0.85442 lr=0.00100 gn=11.96259 time=50.76it/s +epoch=54 global_step=21350 loss=4.47181 loss_avg=4.25749 acc=0.57031 acc_top1_avg=0.58461 acc_top5_avg=0.85481 lr=0.00100 gn=11.73423 time=54.75it/s +epoch=54 global_step=21400 loss=4.34473 loss_avg=4.25000 acc=0.57812 acc_top1_avg=0.58536 acc_top5_avg=0.85637 lr=0.00100 gn=13.92548 time=59.01it/s +epoch=54 global_step=21450 loss=4.63168 loss_avg=4.26564 acc=0.55469 acc_top1_avg=0.58357 acc_top5_avg=0.85628 lr=0.00100 gn=14.08632 time=55.51it/s +epoch=54 global_step=21500 loss=4.33484 loss_avg=4.26490 acc=0.58594 acc_top1_avg=0.58393 acc_top5_avg=0.85612 lr=0.00100 gn=15.03267 time=63.97it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.25766 test_loss_avg=0.64072 acc=0.90625 test_acc_avg=0.81273 test_acc_top5_avg=0.97955 time=237.73it/s +epoch=54 global_step=21505 loss=0.07107 test_loss_avg=0.49691 acc=0.93750 test_acc_avg=0.85453 test_acc_top5_avg=0.98161 time=527.78it/s +curr_acc 0.8545 +BEST_ACC 0.8636 +curr_acc_top5 0.9816 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=3.70194 loss_avg=4.19169 acc=0.64062 acc_top1_avg=0.59253 acc_top5_avg=0.85469 lr=0.00100 gn=14.02822 time=52.60it/s +epoch=55 global_step=21600 loss=4.19493 loss_avg=4.23795 acc=0.60156 acc_top1_avg=0.58750 acc_top5_avg=0.85559 lr=0.00100 gn=13.52945 time=54.09it/s +epoch=55 global_step=21650 loss=4.22541 loss_avg=4.22144 acc=0.58594 acc_top1_avg=0.58842 acc_top5_avg=0.85528 lr=0.00100 gn=13.18242 time=55.38it/s +epoch=55 global_step=21700 loss=3.90934 loss_avg=4.20112 acc=0.62500 acc_top1_avg=0.59030 acc_top5_avg=0.85765 lr=0.00100 gn=11.99934 time=63.81it/s +epoch=55 global_step=21750 loss=3.95154 loss_avg=4.22167 acc=0.60156 acc_top1_avg=0.58788 acc_top5_avg=0.85740 lr=0.00100 gn=10.12999 time=50.70it/s +epoch=55 global_step=21800 loss=4.42373 loss_avg=4.22179 acc=0.55469 acc_top1_avg=0.58782 acc_top5_avg=0.85625 lr=0.00100 gn=11.39073 time=61.27it/s +epoch=55 global_step=21850 loss=4.19362 loss_avg=4.22976 acc=0.57812 acc_top1_avg=0.58693 acc_top5_avg=0.85693 lr=0.00100 gn=14.89421 time=59.79it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.28686 test_loss_avg=0.47507 acc=0.89844 test_acc_avg=0.85156 test_acc_top5_avg=0.99375 time=244.65it/s +epoch=55 global_step=21896 loss=0.79404 test_loss_avg=0.68642 acc=0.75000 test_acc_avg=0.80312 test_acc_top5_avg=0.97571 time=236.82it/s +epoch=55 global_step=21896 loss=0.02254 test_loss_avg=0.52728 acc=1.00000 test_acc_avg=0.84929 test_acc_top5_avg=0.98170 time=819.36it/s +curr_acc 0.8493 +BEST_ACC 0.8636 +curr_acc_top5 0.9817 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=3.99745 loss_avg=4.08568 acc=0.60938 acc_top1_avg=0.59766 acc_top5_avg=0.85742 lr=0.00100 gn=14.18459 time=54.36it/s +epoch=56 global_step=21950 loss=4.92656 loss_avg=4.21695 acc=0.53125 acc_top1_avg=0.58898 acc_top5_avg=0.85503 lr=0.00100 gn=16.21574 time=52.80it/s +epoch=56 global_step=22000 loss=4.16782 loss_avg=4.18111 acc=0.59375 acc_top1_avg=0.59255 acc_top5_avg=0.85870 lr=0.00100 gn=12.13128 time=63.48it/s +epoch=56 global_step=22050 loss=4.42001 loss_avg=4.20651 acc=0.57031 acc_top1_avg=0.58979 acc_top5_avg=0.85897 lr=0.00100 gn=14.85748 time=46.66it/s +epoch=56 global_step=22100 loss=4.23581 loss_avg=4.19210 acc=0.59375 acc_top1_avg=0.59134 acc_top5_avg=0.85773 lr=0.00100 gn=11.75039 time=50.73it/s +epoch=56 global_step=22150 loss=4.10731 loss_avg=4.21015 acc=0.60156 acc_top1_avg=0.58981 acc_top5_avg=0.85845 lr=0.00100 gn=11.31980 time=31.50it/s +epoch=56 global_step=22200 loss=3.92012 loss_avg=4.22070 acc=0.64062 acc_top1_avg=0.58887 acc_top5_avg=0.85768 lr=0.00100 gn=14.10210 time=54.22it/s +epoch=56 global_step=22250 loss=4.39497 loss_avg=4.23371 acc=0.57812 acc_top1_avg=0.58790 acc_top5_avg=0.85719 lr=0.00100 gn=10.56903 time=63.61it/s +====================Eval==================== +epoch=56 global_step=22287 loss=1.05796 test_loss_avg=0.54916 acc=0.71094 test_acc_avg=0.83864 test_acc_top5_avg=0.98377 time=214.04it/s +epoch=56 global_step=22287 loss=0.18135 test_loss_avg=0.51027 acc=0.94531 test_acc_avg=0.85074 test_acc_top5_avg=0.98037 time=243.13it/s +epoch=56 global_step=22287 loss=0.06191 test_loss_avg=0.49550 acc=0.93750 test_acc_avg=0.85423 test_acc_top5_avg=0.98111 time=485.96it/s +curr_acc 0.8542 +BEST_ACC 0.8636 +curr_acc_top5 0.9811 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=3.27044 loss_avg=4.30031 acc=0.71094 acc_top1_avg=0.57993 acc_top5_avg=0.85577 lr=0.00100 gn=20.00301 time=56.95it/s +epoch=57 global_step=22350 loss=3.52951 loss_avg=4.12338 acc=0.66406 acc_top1_avg=0.60069 acc_top5_avg=0.86508 lr=0.00100 gn=15.51180 time=54.33it/s +epoch=57 global_step=22400 loss=4.35038 loss_avg=4.17351 acc=0.57812 acc_top1_avg=0.59340 acc_top5_avg=0.86235 lr=0.00100 gn=17.25018 time=53.66it/s +epoch=57 global_step=22450 loss=3.71715 loss_avg=4.17020 acc=0.63281 acc_top1_avg=0.59442 acc_top5_avg=0.85990 lr=0.00100 gn=9.64896 time=55.69it/s +epoch=57 global_step=22500 loss=4.24015 loss_avg=4.18731 acc=0.58594 acc_top1_avg=0.59236 acc_top5_avg=0.85827 lr=0.00100 gn=11.57553 time=45.44it/s +epoch=57 global_step=22550 loss=5.11694 loss_avg=4.21560 acc=0.50781 acc_top1_avg=0.58918 acc_top5_avg=0.85700 lr=0.00100 gn=15.43187 time=58.69it/s +epoch=57 global_step=22600 loss=5.01741 loss_avg=4.22047 acc=0.50781 acc_top1_avg=0.58853 acc_top5_avg=0.85735 lr=0.00100 gn=12.74513 time=56.34it/s +epoch=57 global_step=22650 loss=3.69486 loss_avg=4.21564 acc=0.64062 acc_top1_avg=0.58904 acc_top5_avg=0.85795 lr=0.00100 gn=17.54885 time=60.65it/s +====================Eval==================== +epoch=57 global_step=22678 loss=0.62238 test_loss_avg=0.63541 acc=0.78906 test_acc_avg=0.81732 test_acc_top5_avg=0.97357 time=233.91it/s +epoch=57 global_step=22678 loss=0.03004 test_loss_avg=0.51589 acc=1.00000 test_acc_avg=0.85047 test_acc_top5_avg=0.97992 time=758.88it/s +curr_acc 0.8505 +BEST_ACC 0.8636 +curr_acc_top5 0.9799 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=3.91472 loss_avg=4.23704 acc=0.64062 acc_top1_avg=0.58913 acc_top5_avg=0.85369 lr=0.00100 gn=15.67197 time=62.88it/s +epoch=58 global_step=22750 loss=4.43735 loss_avg=4.19479 acc=0.55469 acc_top1_avg=0.59321 acc_top5_avg=0.85992 lr=0.00100 gn=14.91101 time=53.44it/s +epoch=58 global_step=22800 loss=4.41922 loss_avg=4.20826 acc=0.56250 acc_top1_avg=0.59106 acc_top5_avg=0.85893 lr=0.00100 gn=13.29715 time=38.41it/s +epoch=58 global_step=22850 loss=3.78136 loss_avg=4.20157 acc=0.64062 acc_top1_avg=0.59248 acc_top5_avg=0.85906 lr=0.00100 gn=16.54846 time=61.95it/s +epoch=58 global_step=22900 loss=4.35508 loss_avg=4.19271 acc=0.59375 acc_top1_avg=0.59340 acc_top5_avg=0.85959 lr=0.00100 gn=16.71538 time=55.52it/s +epoch=58 global_step=22950 loss=4.38133 loss_avg=4.20171 acc=0.58594 acc_top1_avg=0.59254 acc_top5_avg=0.85892 lr=0.00100 gn=14.42126 time=58.07it/s +epoch=58 global_step=23000 loss=4.25931 loss_avg=4.19948 acc=0.57812 acc_top1_avg=0.59244 acc_top5_avg=0.85765 lr=0.00100 gn=13.03623 time=58.26it/s +epoch=58 global_step=23050 loss=3.88596 loss_avg=4.19049 acc=0.62500 acc_top1_avg=0.59287 acc_top5_avg=0.85790 lr=0.00100 gn=13.83861 time=55.04it/s +====================Eval==================== +epoch=58 global_step=23069 loss=0.61928 test_loss_avg=0.48165 acc=0.84375 test_acc_avg=0.86632 test_acc_top5_avg=0.98872 time=238.08it/s +epoch=58 global_step=23069 loss=0.14194 test_loss_avg=0.56702 acc=0.94531 test_acc_avg=0.83525 test_acc_top5_avg=0.97737 time=254.66it/s +epoch=58 global_step=23069 loss=0.05744 test_loss_avg=0.50555 acc=0.93750 test_acc_avg=0.85235 test_acc_top5_avg=0.98012 time=846.48it/s +curr_acc 0.8524 +BEST_ACC 0.8636 +curr_acc_top5 0.9801 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=3.48593 loss_avg=4.14773 acc=0.67969 acc_top1_avg=0.59703 acc_top5_avg=0.85660 lr=0.00100 gn=12.40993 time=58.72it/s +epoch=59 global_step=23150 loss=3.55620 loss_avg=4.16126 acc=0.67188 acc_top1_avg=0.59664 acc_top5_avg=0.85860 lr=0.00100 gn=17.76532 time=52.81it/s +epoch=59 global_step=23200 loss=4.38174 loss_avg=4.09885 acc=0.58594 acc_top1_avg=0.60317 acc_top5_avg=0.86331 lr=0.00100 gn=15.40369 time=52.84it/s +epoch=59 global_step=23250 loss=3.63207 loss_avg=4.13157 acc=0.65625 acc_top1_avg=0.59945 acc_top5_avg=0.86093 lr=0.00100 gn=15.08784 time=62.48it/s +epoch=59 global_step=23300 loss=3.57315 loss_avg=4.12692 acc=0.64844 acc_top1_avg=0.60014 acc_top5_avg=0.86063 lr=0.00100 gn=16.80536 time=64.01it/s +epoch=59 global_step=23350 loss=4.28199 loss_avg=4.13707 acc=0.57812 acc_top1_avg=0.59925 acc_top5_avg=0.86074 lr=0.00100 gn=11.89358 time=62.94it/s +epoch=59 global_step=23400 loss=4.07162 loss_avg=4.16602 acc=0.60156 acc_top1_avg=0.59599 acc_top5_avg=0.85848 lr=0.00100 gn=13.79662 time=55.71it/s +epoch=59 global_step=23450 loss=3.86183 loss_avg=4.16737 acc=0.62500 acc_top1_avg=0.59605 acc_top5_avg=0.85868 lr=0.00100 gn=17.01681 time=54.71it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.22125 test_loss_avg=0.64083 acc=0.93750 test_acc_avg=0.81731 test_acc_top5_avg=0.97877 time=236.70it/s +epoch=59 global_step=23460 loss=0.03464 test_loss_avg=0.51726 acc=1.00000 test_acc_avg=0.85196 test_acc_top5_avg=0.98141 time=496.84it/s +curr_acc 0.8520 +BEST_ACC 0.8636 +curr_acc_top5 0.9814 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=4.39245 loss_avg=4.13979 acc=0.56250 acc_top1_avg=0.59922 acc_top5_avg=0.85313 lr=0.00100 gn=16.30781 time=54.77it/s +epoch=60 global_step=23550 loss=4.47017 loss_avg=4.18878 acc=0.57031 acc_top1_avg=0.59453 acc_top5_avg=0.85156 lr=0.00100 gn=18.73891 time=50.46it/s +epoch=60 global_step=23600 loss=4.21663 loss_avg=4.17094 acc=0.58594 acc_top1_avg=0.59542 acc_top5_avg=0.85257 lr=0.00100 gn=17.23034 time=54.87it/s +epoch=60 global_step=23650 loss=4.02433 loss_avg=4.15215 acc=0.62500 acc_top1_avg=0.59749 acc_top5_avg=0.85535 lr=0.00100 gn=17.58716 time=58.13it/s +epoch=60 global_step=23700 loss=3.73334 loss_avg=4.14322 acc=0.65625 acc_top1_avg=0.59795 acc_top5_avg=0.85615 lr=0.00100 gn=16.30186 time=62.73it/s +epoch=60 global_step=23750 loss=3.57400 loss_avg=4.14711 acc=0.67188 acc_top1_avg=0.59774 acc_top5_avg=0.85603 lr=0.00100 gn=18.06305 time=62.71it/s +epoch=60 global_step=23800 loss=3.95201 loss_avg=4.15319 acc=0.60156 acc_top1_avg=0.59665 acc_top5_avg=0.85655 lr=0.00100 gn=11.17318 time=54.49it/s +epoch=60 global_step=23850 loss=5.02676 loss_avg=4.15225 acc=0.50000 acc_top1_avg=0.59683 acc_top5_avg=0.85719 lr=0.00100 gn=14.38062 time=60.36it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.16560 test_loss_avg=0.51715 acc=0.95312 test_acc_avg=0.84844 test_acc_top5_avg=0.99219 time=79.87it/s +epoch=60 global_step=23851 loss=0.36954 test_loss_avg=0.63362 acc=0.89844 test_acc_avg=0.81602 test_acc_top5_avg=0.97409 time=234.19it/s +epoch=60 global_step=23851 loss=0.01842 test_loss_avg=0.52526 acc=1.00000 test_acc_avg=0.84771 test_acc_top5_avg=0.97923 time=515.21it/s +curr_acc 0.8477 +BEST_ACC 0.8636 +curr_acc_top5 0.9792 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=3.98110 loss_avg=4.03148 acc=0.61719 acc_top1_avg=0.61145 acc_top5_avg=0.85890 lr=0.00100 gn=17.09218 time=58.96it/s +epoch=61 global_step=23950 loss=4.05681 loss_avg=4.12457 acc=0.60938 acc_top1_avg=0.60164 acc_top5_avg=0.85756 lr=0.00100 gn=13.61903 time=51.62it/s +epoch=61 global_step=24000 loss=4.10446 loss_avg=4.15882 acc=0.58594 acc_top1_avg=0.59753 acc_top5_avg=0.85481 lr=0.00100 gn=13.79533 time=58.73it/s +epoch=61 global_step=24050 loss=3.72554 loss_avg=4.13864 acc=0.63281 acc_top1_avg=0.59925 acc_top5_avg=0.85541 lr=0.00100 gn=17.91450 time=46.45it/s +epoch=61 global_step=24100 loss=4.14792 loss_avg=4.13481 acc=0.59375 acc_top1_avg=0.59974 acc_top5_avg=0.85705 lr=0.00100 gn=11.13989 time=57.78it/s +epoch=61 global_step=24150 loss=4.63210 loss_avg=4.12193 acc=0.53125 acc_top1_avg=0.60086 acc_top5_avg=0.85708 lr=0.00100 gn=15.88256 time=62.25it/s +epoch=61 global_step=24200 loss=4.80477 loss_avg=4.12771 acc=0.53125 acc_top1_avg=0.60022 acc_top5_avg=0.85718 lr=0.00100 gn=13.15251 time=55.22it/s +====================Eval==================== +epoch=61 global_step=24242 loss=1.57631 test_loss_avg=0.87205 acc=0.56250 test_acc_avg=0.75630 test_acc_top5_avg=0.96421 time=236.38it/s +epoch=61 global_step=24242 loss=0.10114 test_loss_avg=0.58871 acc=0.93750 test_acc_avg=0.83228 test_acc_top5_avg=0.97735 time=823.54it/s +curr_acc 0.8323 +BEST_ACC 0.8636 +curr_acc_top5 0.9774 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=3.55991 loss_avg=3.85423 acc=0.64844 acc_top1_avg=0.62695 acc_top5_avg=0.87695 lr=0.00100 gn=13.95064 time=54.97it/s +epoch=62 global_step=24300 loss=3.81111 loss_avg=4.04907 acc=0.62500 acc_top1_avg=0.61086 acc_top5_avg=0.86328 lr=0.00100 gn=15.31460 time=50.75it/s +epoch=62 global_step=24350 loss=4.44456 loss_avg=4.07790 acc=0.55469 acc_top1_avg=0.60670 acc_top5_avg=0.85858 lr=0.00100 gn=16.18199 time=56.75it/s +epoch=62 global_step=24400 loss=3.87012 loss_avg=4.11728 acc=0.64062 acc_top1_avg=0.60151 acc_top5_avg=0.85631 lr=0.00100 gn=16.46152 time=62.99it/s +epoch=62 global_step=24450 loss=3.70645 loss_avg=4.10798 acc=0.63281 acc_top1_avg=0.60258 acc_top5_avg=0.85874 lr=0.00100 gn=15.69380 time=53.80it/s +epoch=62 global_step=24500 loss=4.15785 loss_avg=4.11814 acc=0.60938 acc_top1_avg=0.60144 acc_top5_avg=0.85877 lr=0.00100 gn=18.88020 time=53.25it/s +epoch=62 global_step=24550 loss=4.08828 loss_avg=4.14220 acc=0.61719 acc_top1_avg=0.59910 acc_top5_avg=0.85869 lr=0.00100 gn=19.73208 time=63.85it/s +epoch=62 global_step=24600 loss=4.43424 loss_avg=4.12639 acc=0.58594 acc_top1_avg=0.60106 acc_top5_avg=0.85920 lr=0.00100 gn=22.21758 time=38.57it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.81881 test_loss_avg=0.84438 acc=0.77344 test_acc_avg=0.75781 test_acc_top5_avg=0.98047 time=215.75it/s +epoch=62 global_step=24633 loss=1.28972 test_loss_avg=0.75664 acc=0.56250 test_acc_avg=0.77494 test_acc_top5_avg=0.97236 time=235.24it/s +epoch=62 global_step=24633 loss=0.01830 test_loss_avg=0.59360 acc=1.00000 test_acc_avg=0.82130 test_acc_top5_avg=0.97864 time=816.49it/s +curr_acc 0.8213 +BEST_ACC 0.8636 +curr_acc_top5 0.9786 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=3.57850 loss_avg=4.10638 acc=0.65625 acc_top1_avg=0.60064 acc_top5_avg=0.85202 lr=0.00100 gn=21.24286 time=52.73it/s +epoch=63 global_step=24700 loss=3.81986 loss_avg=4.08972 acc=0.64062 acc_top1_avg=0.60506 acc_top5_avg=0.85949 lr=0.00100 gn=18.90226 time=52.01it/s +epoch=63 global_step=24750 loss=4.34200 loss_avg=4.12813 acc=0.57031 acc_top1_avg=0.59936 acc_top5_avg=0.85724 lr=0.00100 gn=14.33456 time=57.98it/s +epoch=63 global_step=24800 loss=3.33640 loss_avg=4.13085 acc=0.66406 acc_top1_avg=0.59922 acc_top5_avg=0.85704 lr=0.00100 gn=18.08445 time=51.35it/s +epoch=63 global_step=24850 loss=3.35895 loss_avg=4.10822 acc=0.68750 acc_top1_avg=0.60271 acc_top5_avg=0.85883 lr=0.00100 gn=19.40787 time=57.31it/s +epoch=63 global_step=24900 loss=4.52491 loss_avg=4.09476 acc=0.55469 acc_top1_avg=0.60387 acc_top5_avg=0.85882 lr=0.00100 gn=21.47958 time=50.59it/s +epoch=63 global_step=24950 loss=3.73612 loss_avg=4.11015 acc=0.64844 acc_top1_avg=0.60245 acc_top5_avg=0.85824 lr=0.00100 gn=17.90304 time=55.61it/s +epoch=63 global_step=25000 loss=4.36912 loss_avg=4.10798 acc=0.57031 acc_top1_avg=0.60307 acc_top5_avg=0.85825 lr=0.00100 gn=18.13343 time=55.44it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.90439 test_loss_avg=0.64699 acc=0.75781 test_acc_avg=0.81624 test_acc_top5_avg=0.98302 time=236.47it/s +epoch=63 global_step=25024 loss=0.15023 test_loss_avg=0.63077 acc=0.93750 test_acc_avg=0.81732 test_acc_top5_avg=0.97795 time=244.21it/s +epoch=63 global_step=25024 loss=0.04397 test_loss_avg=0.59280 acc=1.00000 test_acc_avg=0.82812 test_acc_top5_avg=0.97943 time=518.33it/s +curr_acc 0.8281 +BEST_ACC 0.8636 +curr_acc_top5 0.9794 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=4.08247 loss_avg=4.13940 acc=0.61719 acc_top1_avg=0.59976 acc_top5_avg=0.84916 lr=0.00100 gn=18.27457 time=60.05it/s +epoch=64 global_step=25100 loss=4.55713 loss_avg=4.01091 acc=0.54688 acc_top1_avg=0.61390 acc_top5_avg=0.86184 lr=0.00100 gn=17.81799 time=63.58it/s +epoch=64 global_step=25150 loss=3.98108 loss_avg=4.05707 acc=0.63281 acc_top1_avg=0.60888 acc_top5_avg=0.85944 lr=0.00100 gn=22.46134 time=54.71it/s +epoch=64 global_step=25200 loss=3.88054 loss_avg=4.07395 acc=0.60156 acc_top1_avg=0.60689 acc_top5_avg=0.85853 lr=0.00100 gn=21.95201 time=52.61it/s +epoch=64 global_step=25250 loss=3.47506 loss_avg=4.08219 acc=0.66406 acc_top1_avg=0.60547 acc_top5_avg=0.85872 lr=0.00100 gn=20.07932 time=51.58it/s +epoch=64 global_step=25300 loss=3.88482 loss_avg=4.07257 acc=0.62500 acc_top1_avg=0.60660 acc_top5_avg=0.85918 lr=0.00100 gn=19.81463 time=56.09it/s +epoch=64 global_step=25350 loss=4.92014 loss_avg=4.08080 acc=0.50000 acc_top1_avg=0.60568 acc_top5_avg=0.86000 lr=0.00100 gn=18.95803 time=58.79it/s +epoch=64 global_step=25400 loss=4.73961 loss_avg=4.09883 acc=0.54688 acc_top1_avg=0.60339 acc_top5_avg=0.85850 lr=0.00100 gn=19.78788 time=54.69it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.55007 test_loss_avg=0.65580 acc=0.86719 test_acc_avg=0.81250 test_acc_top5_avg=0.97887 time=241.59it/s +epoch=64 global_step=25415 loss=0.09931 test_loss_avg=0.56247 acc=0.93750 test_acc_avg=0.83564 test_acc_top5_avg=0.98042 time=505.28it/s +curr_acc 0.8356 +BEST_ACC 0.8636 +curr_acc_top5 0.9804 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=3.62820 loss_avg=4.15761 acc=0.67969 acc_top1_avg=0.59643 acc_top5_avg=0.85938 lr=0.00100 gn=21.53192 time=59.07it/s +epoch=65 global_step=25500 loss=4.60202 loss_avg=4.06136 acc=0.55469 acc_top1_avg=0.60643 acc_top5_avg=0.86241 lr=0.00100 gn=16.57657 time=58.37it/s +epoch=65 global_step=25550 loss=3.75804 loss_avg=4.07060 acc=0.62500 acc_top1_avg=0.60666 acc_top5_avg=0.86273 lr=0.00100 gn=16.56573 time=54.87it/s +epoch=65 global_step=25600 loss=4.01659 loss_avg=4.06162 acc=0.61719 acc_top1_avg=0.60739 acc_top5_avg=0.86195 lr=0.00100 gn=20.04442 time=53.75it/s +epoch=65 global_step=25650 loss=4.11209 loss_avg=4.06306 acc=0.60938 acc_top1_avg=0.60711 acc_top5_avg=0.86011 lr=0.00100 gn=19.73826 time=57.65it/s +epoch=65 global_step=25700 loss=3.67767 loss_avg=4.06056 acc=0.67188 acc_top1_avg=0.60825 acc_top5_avg=0.86077 lr=0.00100 gn=19.59549 time=54.28it/s +epoch=65 global_step=25750 loss=4.00253 loss_avg=4.06069 acc=0.62500 acc_top1_avg=0.60884 acc_top5_avg=0.86105 lr=0.00100 gn=24.60488 time=57.46it/s +epoch=65 global_step=25800 loss=4.32628 loss_avg=4.06858 acc=0.57812 acc_top1_avg=0.60777 acc_top5_avg=0.86116 lr=0.00100 gn=17.55564 time=61.82it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.28706 test_loss_avg=0.47395 acc=0.92969 test_acc_avg=0.86302 test_acc_top5_avg=0.99583 time=235.57it/s +epoch=65 global_step=25806 loss=0.14589 test_loss_avg=0.72562 acc=0.94531 test_acc_avg=0.79171 test_acc_top5_avg=0.97127 time=230.87it/s +epoch=65 global_step=25806 loss=0.02361 test_loss_avg=0.62392 acc=1.00000 test_acc_avg=0.82021 test_acc_top5_avg=0.97587 time=809.71it/s +curr_acc 0.8202 +BEST_ACC 0.8636 +curr_acc_top5 0.9759 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=3.44154 loss_avg=3.97032 acc=0.67969 acc_top1_avg=0.62163 acc_top5_avg=0.86790 lr=0.00100 gn=25.69544 time=63.48it/s +epoch=66 global_step=25900 loss=4.06097 loss_avg=3.98147 acc=0.60938 acc_top1_avg=0.61960 acc_top5_avg=0.86702 lr=0.00100 gn=18.64959 time=54.40it/s +epoch=66 global_step=25950 loss=4.34478 loss_avg=4.00369 acc=0.57031 acc_top1_avg=0.61740 acc_top5_avg=0.86377 lr=0.00100 gn=24.21140 time=63.49it/s +epoch=66 global_step=26000 loss=4.06692 loss_avg=4.01290 acc=0.63281 acc_top1_avg=0.61614 acc_top5_avg=0.86264 lr=0.00100 gn=24.36816 time=46.17it/s +epoch=66 global_step=26050 loss=4.06065 loss_avg=4.02658 acc=0.61719 acc_top1_avg=0.61440 acc_top5_avg=0.86120 lr=0.00100 gn=23.86355 time=63.63it/s +epoch=66 global_step=26100 loss=4.04197 loss_avg=4.02906 acc=0.60938 acc_top1_avg=0.61408 acc_top5_avg=0.86108 lr=0.00100 gn=17.48287 time=60.05it/s +epoch=66 global_step=26150 loss=4.35442 loss_avg=4.03933 acc=0.57031 acc_top1_avg=0.61274 acc_top5_avg=0.86085 lr=0.00100 gn=23.09886 time=55.63it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.36820 test_loss_avg=0.74133 acc=0.88281 test_acc_avg=0.78385 test_acc_top5_avg=0.97418 time=237.85it/s +epoch=66 global_step=26197 loss=0.10088 test_loss_avg=0.66281 acc=0.93750 test_acc_avg=0.80400 test_acc_top5_avg=0.97429 time=450.56it/s +curr_acc 0.8040 +BEST_ACC 0.8636 +curr_acc_top5 0.9743 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=4.46416 loss_avg=4.18881 acc=0.57031 acc_top1_avg=0.60156 acc_top5_avg=0.82552 lr=0.00100 gn=20.66052 time=47.98it/s +epoch=67 global_step=26250 loss=4.55555 loss_avg=4.05709 acc=0.54688 acc_top1_avg=0.60820 acc_top5_avg=0.85775 lr=0.00100 gn=20.51462 time=61.77it/s +epoch=67 global_step=26300 loss=3.79541 loss_avg=4.05541 acc=0.62500 acc_top1_avg=0.60915 acc_top5_avg=0.85975 lr=0.00100 gn=19.27101 time=49.84it/s +epoch=67 global_step=26350 loss=4.03211 loss_avg=4.02498 acc=0.62500 acc_top1_avg=0.61218 acc_top5_avg=0.86009 lr=0.00100 gn=19.47622 time=50.80it/s +epoch=67 global_step=26400 loss=4.30852 loss_avg=4.04498 acc=0.57812 acc_top1_avg=0.61053 acc_top5_avg=0.85984 lr=0.00100 gn=17.54730 time=58.93it/s +epoch=67 global_step=26450 loss=5.13788 loss_avg=4.03553 acc=0.49219 acc_top1_avg=0.61129 acc_top5_avg=0.86101 lr=0.00100 gn=18.89837 time=57.74it/s +epoch=67 global_step=26500 loss=3.59950 loss_avg=4.03284 acc=0.64844 acc_top1_avg=0.61188 acc_top5_avg=0.86090 lr=0.00100 gn=18.40661 time=52.22it/s +epoch=67 global_step=26550 loss=3.86573 loss_avg=4.03354 acc=0.64062 acc_top1_avg=0.61205 acc_top5_avg=0.86099 lr=0.00100 gn=19.64366 time=55.20it/s +====================Eval==================== +epoch=67 global_step=26588 loss=0.74431 test_loss_avg=0.79360 acc=0.78906 test_acc_avg=0.77009 test_acc_top5_avg=0.98884 time=236.71it/s +epoch=67 global_step=26588 loss=0.38026 test_loss_avg=0.70445 acc=0.91406 test_acc_avg=0.79427 test_acc_top5_avg=0.97327 time=239.52it/s +epoch=67 global_step=26588 loss=0.10592 test_loss_avg=0.56717 acc=0.93750 test_acc_avg=0.83455 test_acc_top5_avg=0.97913 time=766.50it/s +curr_acc 0.8346 +BEST_ACC 0.8636 +curr_acc_top5 0.9791 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=4.07240 loss_avg=3.96737 acc=0.60938 acc_top1_avg=0.61784 acc_top5_avg=0.87891 lr=0.00100 gn=22.68673 time=61.96it/s +epoch=68 global_step=26650 loss=3.91169 loss_avg=4.03674 acc=0.61719 acc_top1_avg=0.61190 acc_top5_avg=0.86051 lr=0.00100 gn=20.18268 time=59.92it/s +epoch=68 global_step=26700 loss=3.43222 loss_avg=4.00863 acc=0.65625 acc_top1_avg=0.61489 acc_top5_avg=0.85972 lr=0.00100 gn=19.14980 time=56.00it/s +epoch=68 global_step=26750 loss=3.67502 loss_avg=4.02666 acc=0.67969 acc_top1_avg=0.61270 acc_top5_avg=0.85966 lr=0.00100 gn=24.47031 time=54.63it/s +epoch=68 global_step=26800 loss=4.43532 loss_avg=4.02165 acc=0.55469 acc_top1_avg=0.61317 acc_top5_avg=0.85971 lr=0.00100 gn=19.60399 time=54.89it/s +epoch=68 global_step=26850 loss=3.76930 loss_avg=4.02186 acc=0.66406 acc_top1_avg=0.61349 acc_top5_avg=0.86015 lr=0.00100 gn=23.16718 time=63.56it/s +epoch=68 global_step=26900 loss=4.80022 loss_avg=4.01854 acc=0.53125 acc_top1_avg=0.61393 acc_top5_avg=0.86138 lr=0.00100 gn=18.66002 time=52.71it/s +epoch=68 global_step=26950 loss=4.62998 loss_avg=4.01712 acc=0.53125 acc_top1_avg=0.61432 acc_top5_avg=0.86106 lr=0.00100 gn=21.44995 time=53.87it/s +====================Eval==================== +epoch=68 global_step=26979 loss=1.52765 test_loss_avg=0.76395 acc=0.54688 test_acc_avg=0.78488 test_acc_top5_avg=0.97517 time=226.57it/s +epoch=68 global_step=26979 loss=0.15868 test_loss_avg=0.60745 acc=0.93750 test_acc_avg=0.82222 test_acc_top5_avg=0.97837 time=244.88it/s +epoch=68 global_step=26979 loss=0.02754 test_loss_avg=0.60011 acc=1.00000 test_acc_avg=0.82447 test_acc_top5_avg=0.97864 time=483.55it/s +curr_acc 0.8245 +BEST_ACC 0.8636 +curr_acc_top5 0.9786 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=2.96870 loss_avg=3.93182 acc=0.74219 acc_top1_avg=0.62277 acc_top5_avg=0.85640 lr=0.00100 gn=24.28589 time=55.78it/s +epoch=69 global_step=27050 loss=3.46191 loss_avg=3.95140 acc=0.67969 acc_top1_avg=0.62093 acc_top5_avg=0.85904 lr=0.00100 gn=27.64712 time=52.61it/s +epoch=69 global_step=27100 loss=3.60545 loss_avg=3.98567 acc=0.66406 acc_top1_avg=0.61738 acc_top5_avg=0.85815 lr=0.00100 gn=20.48115 time=58.06it/s +epoch=69 global_step=27150 loss=4.05037 loss_avg=4.00081 acc=0.60938 acc_top1_avg=0.61577 acc_top5_avg=0.85755 lr=0.00100 gn=19.66017 time=63.36it/s +epoch=69 global_step=27200 loss=4.09547 loss_avg=3.99783 acc=0.60938 acc_top1_avg=0.61669 acc_top5_avg=0.85923 lr=0.00100 gn=27.81608 time=53.30it/s +epoch=69 global_step=27250 loss=3.28794 loss_avg=3.99045 acc=0.70312 acc_top1_avg=0.61768 acc_top5_avg=0.85958 lr=0.00100 gn=26.54150 time=54.10it/s +epoch=69 global_step=27300 loss=3.52933 loss_avg=3.99503 acc=0.67969 acc_top1_avg=0.61675 acc_top5_avg=0.85984 lr=0.00100 gn=21.67799 time=55.68it/s +epoch=69 global_step=27350 loss=3.52660 loss_avg=4.00517 acc=0.65625 acc_top1_avg=0.61561 acc_top5_avg=0.85986 lr=0.00100 gn=20.12141 time=51.78it/s +====================Eval==================== +epoch=69 global_step=27370 loss=1.47998 test_loss_avg=0.78087 acc=0.55469 test_acc_avg=0.77599 test_acc_top5_avg=0.96620 time=114.11it/s +epoch=69 global_step=27370 loss=0.18516 test_loss_avg=0.66730 acc=0.93750 test_acc_avg=0.80746 test_acc_top5_avg=0.97360 time=805.67it/s +curr_acc 0.8075 +BEST_ACC 0.8636 +curr_acc_top5 0.9736 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=4.57934 loss_avg=3.95720 acc=0.55469 acc_top1_avg=0.61927 acc_top5_avg=0.86172 lr=0.00100 gn=29.18402 time=51.68it/s +epoch=70 global_step=27450 loss=4.00789 loss_avg=3.96773 acc=0.62500 acc_top1_avg=0.61914 acc_top5_avg=0.86035 lr=0.00100 gn=22.39643 time=51.98it/s +epoch=70 global_step=27500 loss=4.36734 loss_avg=3.96280 acc=0.57812 acc_top1_avg=0.62055 acc_top5_avg=0.85956 lr=0.00100 gn=18.25747 time=52.48it/s +epoch=70 global_step=27550 loss=3.84400 loss_avg=4.00112 acc=0.61719 acc_top1_avg=0.61628 acc_top5_avg=0.85907 lr=0.00100 gn=20.21126 time=59.92it/s +epoch=70 global_step=27600 loss=4.27703 loss_avg=3.98576 acc=0.57031 acc_top1_avg=0.61793 acc_top5_avg=0.86016 lr=0.00100 gn=21.63210 time=55.45it/s +epoch=70 global_step=27650 loss=3.86421 loss_avg=3.98268 acc=0.63281 acc_top1_avg=0.61819 acc_top5_avg=0.86127 lr=0.00100 gn=21.62394 time=54.87it/s +epoch=70 global_step=27700 loss=4.22747 loss_avg=3.97991 acc=0.57031 acc_top1_avg=0.61830 acc_top5_avg=0.86136 lr=0.00100 gn=21.99402 time=61.52it/s +epoch=70 global_step=27750 loss=4.77042 loss_avg=3.99776 acc=0.53125 acc_top1_avg=0.61639 acc_top5_avg=0.86063 lr=0.00100 gn=21.92653 time=54.81it/s +====================Eval==================== +epoch=70 global_step=27761 loss=0.56891 test_loss_avg=0.64891 acc=0.83594 test_acc_avg=0.81289 test_acc_top5_avg=0.98867 time=173.10it/s +epoch=70 global_step=27761 loss=0.09263 test_loss_avg=0.75232 acc=0.96094 test_acc_avg=0.78203 test_acc_top5_avg=0.96987 time=239.59it/s +epoch=70 global_step=27761 loss=0.24667 test_loss_avg=0.68807 acc=0.93750 test_acc_avg=0.80024 test_acc_top5_avg=0.97300 time=508.52it/s +curr_acc 0.8002 +BEST_ACC 0.8636 +curr_acc_top5 0.9730 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=4.12646 loss_avg=3.91494 acc=0.61719 acc_top1_avg=0.62480 acc_top5_avg=0.86579 lr=0.00100 gn=25.30540 time=61.48it/s +epoch=71 global_step=27850 loss=4.24531 loss_avg=3.91479 acc=0.59375 acc_top1_avg=0.62579 acc_top5_avg=0.86517 lr=0.00100 gn=21.97737 time=62.63it/s +epoch=71 global_step=27900 loss=4.07390 loss_avg=3.92818 acc=0.59375 acc_top1_avg=0.62522 acc_top5_avg=0.86252 lr=0.00100 gn=24.60396 time=58.61it/s +epoch=71 global_step=27950 loss=3.00195 loss_avg=3.92887 acc=0.72656 acc_top1_avg=0.62463 acc_top5_avg=0.86322 lr=0.00100 gn=21.68462 time=60.92it/s +epoch=71 global_step=28000 loss=3.97513 loss_avg=3.93485 acc=0.61719 acc_top1_avg=0.62444 acc_top5_avg=0.86274 lr=0.00100 gn=20.75876 time=60.56it/s +epoch=71 global_step=28050 loss=3.76502 loss_avg=3.94410 acc=0.62500 acc_top1_avg=0.62335 acc_top5_avg=0.86343 lr=0.00100 gn=27.14444 time=59.97it/s +epoch=71 global_step=28100 loss=3.57977 loss_avg=3.95693 acc=0.67188 acc_top1_avg=0.62228 acc_top5_avg=0.86209 lr=0.00100 gn=27.93557 time=55.59it/s +epoch=71 global_step=28150 loss=3.54742 loss_avg=3.97167 acc=0.64844 acc_top1_avg=0.62068 acc_top5_avg=0.86168 lr=0.00100 gn=20.47868 time=55.50it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.48119 test_loss_avg=0.86823 acc=0.83594 test_acc_avg=0.75152 test_acc_top5_avg=0.96208 time=237.41it/s +epoch=71 global_step=28152 loss=0.14193 test_loss_avg=0.72718 acc=0.93750 test_acc_avg=0.79094 test_acc_top5_avg=0.96648 time=847.16it/s +curr_acc 0.7909 +BEST_ACC 0.8636 +curr_acc_top5 0.9665 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=4.30670 loss_avg=3.96398 acc=0.57031 acc_top1_avg=0.61833 acc_top5_avg=0.86312 lr=0.00100 gn=20.17835 time=56.67it/s +epoch=72 global_step=28250 loss=3.88036 loss_avg=4.00570 acc=0.62500 acc_top1_avg=0.61623 acc_top5_avg=0.85858 lr=0.00100 gn=24.88911 time=56.04it/s +epoch=72 global_step=28300 loss=4.30418 loss_avg=3.99153 acc=0.57031 acc_top1_avg=0.61793 acc_top5_avg=0.86017 lr=0.00100 gn=28.54612 time=59.53it/s +epoch=72 global_step=28350 loss=4.21676 loss_avg=3.96225 acc=0.59375 acc_top1_avg=0.62086 acc_top5_avg=0.86226 lr=0.00100 gn=25.56777 time=64.16it/s +epoch=72 global_step=28400 loss=4.31066 loss_avg=3.96412 acc=0.58594 acc_top1_avg=0.62113 acc_top5_avg=0.86221 lr=0.00100 gn=23.95127 time=31.77it/s +epoch=72 global_step=28450 loss=4.51739 loss_avg=3.97299 acc=0.57031 acc_top1_avg=0.62005 acc_top5_avg=0.86166 lr=0.00100 gn=24.12650 time=55.19it/s +epoch=72 global_step=28500 loss=3.75765 loss_avg=3.96345 acc=0.65625 acc_top1_avg=0.62096 acc_top5_avg=0.86184 lr=0.00100 gn=25.19112 time=61.38it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.15646 test_loss_avg=0.77665 acc=0.95312 test_acc_avg=0.76758 test_acc_top5_avg=0.98828 time=243.44it/s +epoch=72 global_step=28543 loss=0.31878 test_loss_avg=0.76161 acc=0.91406 test_acc_avg=0.77659 test_acc_top5_avg=0.96951 time=235.26it/s +epoch=72 global_step=28543 loss=0.18101 test_loss_avg=0.63940 acc=0.93750 test_acc_avg=0.81191 test_acc_top5_avg=0.97498 time=841.72it/s +curr_acc 0.8119 +BEST_ACC 0.8636 +curr_acc_top5 0.9750 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=4.11636 loss_avg=3.88159 acc=0.61719 acc_top1_avg=0.62835 acc_top5_avg=0.85938 lr=0.00100 gn=25.56696 time=52.92it/s +epoch=73 global_step=28600 loss=3.85511 loss_avg=3.88633 acc=0.61719 acc_top1_avg=0.62952 acc_top5_avg=0.86472 lr=0.00100 gn=32.24813 time=53.65it/s +epoch=73 global_step=28650 loss=3.43139 loss_avg=3.87375 acc=0.67969 acc_top1_avg=0.63011 acc_top5_avg=0.86711 lr=0.00100 gn=27.73435 time=63.47it/s +epoch=73 global_step=28700 loss=3.99274 loss_avg=3.88763 acc=0.62500 acc_top1_avg=0.62928 acc_top5_avg=0.86614 lr=0.00100 gn=27.72113 time=58.36it/s +epoch=73 global_step=28750 loss=3.91875 loss_avg=3.92022 acc=0.62500 acc_top1_avg=0.62564 acc_top5_avg=0.86334 lr=0.00100 gn=23.16907 time=58.11it/s +epoch=73 global_step=28800 loss=3.53043 loss_avg=3.93093 acc=0.68750 acc_top1_avg=0.62482 acc_top5_avg=0.86275 lr=0.00100 gn=28.60689 time=50.85it/s +epoch=73 global_step=28850 loss=4.22199 loss_avg=3.94354 acc=0.59375 acc_top1_avg=0.62337 acc_top5_avg=0.86380 lr=0.00100 gn=23.34314 time=63.04it/s +epoch=73 global_step=28900 loss=3.83771 loss_avg=3.95032 acc=0.64062 acc_top1_avg=0.62253 acc_top5_avg=0.86397 lr=0.00100 gn=24.11694 time=47.56it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.42696 test_loss_avg=0.73791 acc=0.89062 test_acc_avg=0.78196 test_acc_top5_avg=0.97633 time=114.51it/s +epoch=73 global_step=28934 loss=0.02878 test_loss_avg=0.60613 acc=1.00000 test_acc_avg=0.81982 test_acc_top5_avg=0.97834 time=542.60it/s +curr_acc 0.8198 +BEST_ACC 0.8636 +curr_acc_top5 0.9783 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=4.26854 loss_avg=3.98361 acc=0.57031 acc_top1_avg=0.61572 acc_top5_avg=0.85303 lr=0.00100 gn=17.83542 time=55.98it/s +epoch=74 global_step=29000 loss=4.65770 loss_avg=3.94630 acc=0.55469 acc_top1_avg=0.62382 acc_top5_avg=0.86328 lr=0.00100 gn=28.06654 time=56.40it/s +epoch=74 global_step=29050 loss=3.31810 loss_avg=3.91827 acc=0.68750 acc_top1_avg=0.62702 acc_top5_avg=0.86268 lr=0.00100 gn=27.20315 time=53.12it/s +epoch=74 global_step=29100 loss=3.56126 loss_avg=3.92082 acc=0.67969 acc_top1_avg=0.62627 acc_top5_avg=0.86328 lr=0.00100 gn=25.55177 time=55.98it/s +epoch=74 global_step=29150 loss=3.54110 loss_avg=3.93470 acc=0.66406 acc_top1_avg=0.62486 acc_top5_avg=0.86086 lr=0.00100 gn=27.74324 time=55.72it/s +epoch=74 global_step=29200 loss=3.53265 loss_avg=3.92415 acc=0.67969 acc_top1_avg=0.62644 acc_top5_avg=0.86181 lr=0.00100 gn=26.26488 time=61.54it/s +epoch=74 global_step=29250 loss=3.55753 loss_avg=3.92821 acc=0.67188 acc_top1_avg=0.62631 acc_top5_avg=0.86229 lr=0.00100 gn=26.33235 time=57.04it/s +epoch=74 global_step=29300 loss=3.94468 loss_avg=3.92928 acc=0.60156 acc_top1_avg=0.62632 acc_top5_avg=0.86253 lr=0.00100 gn=22.43874 time=53.11it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.94116 test_loss_avg=1.07697 acc=0.72656 test_acc_avg=0.70312 test_acc_top5_avg=0.96289 time=236.59it/s +epoch=74 global_step=29325 loss=1.28491 test_loss_avg=0.81562 acc=0.60156 test_acc_avg=0.76056 test_acc_top5_avg=0.96976 time=177.24it/s +epoch=74 global_step=29325 loss=0.04160 test_loss_avg=0.62961 acc=1.00000 test_acc_avg=0.81349 test_acc_top5_avg=0.97706 time=819.52it/s +curr_acc 0.8135 +BEST_ACC 0.8636 +curr_acc_top5 0.9771 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=3.69239 loss_avg=3.86436 acc=0.64844 acc_top1_avg=0.63031 acc_top5_avg=0.86094 lr=0.00100 gn=24.43517 time=53.34it/s +epoch=75 global_step=29400 loss=3.81095 loss_avg=3.88785 acc=0.63281 acc_top1_avg=0.62844 acc_top5_avg=0.86125 lr=0.00100 gn=26.66655 time=54.97it/s +epoch=75 global_step=29450 loss=4.12478 loss_avg=3.89176 acc=0.59375 acc_top1_avg=0.62900 acc_top5_avg=0.86256 lr=0.00100 gn=22.27179 time=61.33it/s +epoch=75 global_step=29500 loss=4.20168 loss_avg=3.92826 acc=0.58594 acc_top1_avg=0.62518 acc_top5_avg=0.86076 lr=0.00100 gn=18.44659 time=58.62it/s +epoch=75 global_step=29550 loss=3.69329 loss_avg=3.91026 acc=0.66406 acc_top1_avg=0.62750 acc_top5_avg=0.86177 lr=0.00100 gn=23.93642 time=58.56it/s +epoch=75 global_step=29600 loss=4.03485 loss_avg=3.91167 acc=0.61719 acc_top1_avg=0.62713 acc_top5_avg=0.86128 lr=0.00100 gn=26.63362 time=58.81it/s +epoch=75 global_step=29650 loss=4.12888 loss_avg=3.92108 acc=0.59375 acc_top1_avg=0.62647 acc_top5_avg=0.86031 lr=0.00100 gn=25.78273 time=60.44it/s +epoch=75 global_step=29700 loss=4.03388 loss_avg=3.93532 acc=0.60938 acc_top1_avg=0.62523 acc_top5_avg=0.85979 lr=0.00100 gn=23.00652 time=54.02it/s +====================Eval==================== +epoch=75 global_step=29716 loss=1.42031 test_loss_avg=0.74337 acc=0.61719 test_acc_avg=0.78469 test_acc_top5_avg=0.97250 time=222.27it/s +epoch=75 global_step=29716 loss=0.14567 test_loss_avg=0.69529 acc=0.96094 test_acc_avg=0.79854 test_acc_top5_avg=0.96781 time=244.04it/s +epoch=75 global_step=29716 loss=0.09950 test_loss_avg=0.66681 acc=0.93750 test_acc_avg=0.80597 test_acc_top5_avg=0.96944 time=495.78it/s +curr_acc 0.8060 +BEST_ACC 0.8636 +curr_acc_top5 0.9694 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=4.16693 loss_avg=3.91379 acc=0.60156 acc_top1_avg=0.62753 acc_top5_avg=0.85823 lr=0.00100 gn=32.62614 time=58.76it/s +epoch=76 global_step=29800 loss=3.46441 loss_avg=3.85238 acc=0.67969 acc_top1_avg=0.63393 acc_top5_avg=0.85882 lr=0.00100 gn=26.82054 time=58.72it/s +epoch=76 global_step=29850 loss=4.02375 loss_avg=3.88201 acc=0.62500 acc_top1_avg=0.63054 acc_top5_avg=0.85908 lr=0.00100 gn=24.39656 time=55.80it/s +epoch=76 global_step=29900 loss=3.38862 loss_avg=3.89695 acc=0.67188 acc_top1_avg=0.62886 acc_top5_avg=0.85938 lr=0.00100 gn=27.11804 time=58.93it/s +epoch=76 global_step=29950 loss=4.01028 loss_avg=3.89875 acc=0.60938 acc_top1_avg=0.62837 acc_top5_avg=0.86108 lr=0.00100 gn=26.53061 time=61.75it/s +epoch=76 global_step=30000 loss=3.67477 loss_avg=3.89564 acc=0.64844 acc_top1_avg=0.62904 acc_top5_avg=0.86259 lr=0.00100 gn=23.98614 time=52.21it/s +epoch=76 global_step=30050 loss=4.23756 loss_avg=3.90329 acc=0.58594 acc_top1_avg=0.62825 acc_top5_avg=0.86352 lr=0.00100 gn=28.99977 time=55.78it/s +epoch=76 global_step=30100 loss=3.75611 loss_avg=3.91269 acc=0.64844 acc_top1_avg=0.62699 acc_top5_avg=0.86351 lr=0.00100 gn=28.30518 time=55.61it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.86626 test_loss_avg=0.81208 acc=0.75000 test_acc_avg=0.75832 test_acc_top5_avg=0.96179 time=198.41it/s +epoch=76 global_step=30107 loss=0.08016 test_loss_avg=0.64906 acc=0.93750 test_acc_avg=0.80439 test_acc_top5_avg=0.97280 time=512.50it/s +curr_acc 0.8044 +BEST_ACC 0.8636 +curr_acc_top5 0.9728 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=3.58482 loss_avg=3.88361 acc=0.64844 acc_top1_avg=0.62918 acc_top5_avg=0.86701 lr=0.00100 gn=17.31623 time=56.00it/s +epoch=77 global_step=30200 loss=3.74936 loss_avg=3.87698 acc=0.63281 acc_top1_avg=0.63071 acc_top5_avg=0.86685 lr=0.00100 gn=21.96262 time=60.65it/s +epoch=77 global_step=30250 loss=3.64355 loss_avg=3.88878 acc=0.66406 acc_top1_avg=0.63019 acc_top5_avg=0.86222 lr=0.00100 gn=27.76740 time=55.41it/s +epoch=77 global_step=30300 loss=3.95169 loss_avg=3.89796 acc=0.62500 acc_top1_avg=0.63014 acc_top5_avg=0.86241 lr=0.00100 gn=23.91107 time=51.73it/s +epoch=77 global_step=30350 loss=3.55984 loss_avg=3.88642 acc=0.64062 acc_top1_avg=0.63072 acc_top5_avg=0.86259 lr=0.00100 gn=29.06981 time=63.93it/s +epoch=77 global_step=30400 loss=4.11801 loss_avg=3.90092 acc=0.60938 acc_top1_avg=0.62964 acc_top5_avg=0.86260 lr=0.00100 gn=27.19179 time=57.59it/s +epoch=77 global_step=30450 loss=3.62427 loss_avg=3.89730 acc=0.64844 acc_top1_avg=0.63022 acc_top5_avg=0.86388 lr=0.00100 gn=25.33328 time=57.95it/s +====================Eval==================== +epoch=77 global_step=30498 loss=1.19318 test_loss_avg=0.51701 acc=0.71875 test_acc_avg=0.85616 test_acc_top5_avg=0.98805 time=225.40it/s +epoch=77 global_step=30498 loss=0.23465 test_loss_avg=0.75812 acc=0.92188 test_acc_avg=0.77472 test_acc_top5_avg=0.97062 time=240.13it/s +epoch=77 global_step=30498 loss=0.03818 test_loss_avg=0.66887 acc=1.00000 test_acc_avg=0.80133 test_acc_top5_avg=0.97419 time=635.50it/s +curr_acc 0.8013 +BEST_ACC 0.8636 +curr_acc_top5 0.9742 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=4.44423 loss_avg=4.06719 acc=0.56250 acc_top1_avg=0.59766 acc_top5_avg=0.86328 lr=0.00100 gn=24.39378 time=49.02it/s +epoch=78 global_step=30550 loss=4.49139 loss_avg=3.85446 acc=0.57812 acc_top1_avg=0.63657 acc_top5_avg=0.86373 lr=0.00100 gn=30.54629 time=55.85it/s +epoch=78 global_step=30600 loss=3.71275 loss_avg=3.88873 acc=0.66406 acc_top1_avg=0.63197 acc_top5_avg=0.86098 lr=0.00100 gn=30.24018 time=59.74it/s +epoch=78 global_step=30650 loss=3.27541 loss_avg=3.86553 acc=0.67969 acc_top1_avg=0.63425 acc_top5_avg=0.86503 lr=0.00100 gn=25.71459 time=53.21it/s +epoch=78 global_step=30700 loss=4.88146 loss_avg=3.87337 acc=0.52344 acc_top1_avg=0.63390 acc_top5_avg=0.86545 lr=0.00100 gn=27.36733 time=53.92it/s +epoch=78 global_step=30750 loss=3.64353 loss_avg=3.87497 acc=0.65625 acc_top1_avg=0.63340 acc_top5_avg=0.86471 lr=0.00100 gn=29.23072 time=48.97it/s +epoch=78 global_step=30800 loss=4.20157 loss_avg=3.88101 acc=0.58594 acc_top1_avg=0.63242 acc_top5_avg=0.86398 lr=0.00100 gn=30.06378 time=55.43it/s +epoch=78 global_step=30850 loss=3.40887 loss_avg=3.89145 acc=0.69531 acc_top1_avg=0.63108 acc_top5_avg=0.86406 lr=0.00100 gn=27.58502 time=56.97it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.39088 test_loss_avg=0.87821 acc=0.89844 test_acc_avg=0.74363 test_acc_top5_avg=0.96238 time=222.51it/s +epoch=78 global_step=30889 loss=0.09252 test_loss_avg=0.66200 acc=0.93750 test_acc_avg=0.80409 test_acc_top5_avg=0.97221 time=812.69it/s +curr_acc 0.8041 +BEST_ACC 0.8636 +curr_acc_top5 0.9722 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=3.76522 loss_avg=3.98885 acc=0.64062 acc_top1_avg=0.61790 acc_top5_avg=0.84304 lr=0.00100 gn=25.04372 time=60.08it/s +epoch=79 global_step=30950 loss=3.77017 loss_avg=3.89629 acc=0.64062 acc_top1_avg=0.62782 acc_top5_avg=0.86053 lr=0.00100 gn=29.05727 time=56.33it/s +epoch=79 global_step=31000 loss=4.03311 loss_avg=3.86684 acc=0.62500 acc_top1_avg=0.63260 acc_top5_avg=0.86578 lr=0.00100 gn=27.23842 time=60.25it/s +epoch=79 global_step=31050 loss=3.73979 loss_avg=3.86540 acc=0.64062 acc_top1_avg=0.63267 acc_top5_avg=0.86447 lr=0.00100 gn=27.60243 time=54.29it/s +epoch=79 global_step=31100 loss=3.17155 loss_avg=3.87829 acc=0.70312 acc_top1_avg=0.63203 acc_top5_avg=0.86274 lr=0.00100 gn=27.88032 time=21.62it/s +epoch=79 global_step=31150 loss=4.18331 loss_avg=3.86729 acc=0.58594 acc_top1_avg=0.63371 acc_top5_avg=0.86303 lr=0.00100 gn=28.89186 time=51.15it/s +epoch=79 global_step=31200 loss=3.98190 loss_avg=3.87091 acc=0.61719 acc_top1_avg=0.63379 acc_top5_avg=0.86319 lr=0.00100 gn=28.93215 time=48.10it/s +epoch=79 global_step=31250 loss=3.55128 loss_avg=3.88082 acc=0.67969 acc_top1_avg=0.63251 acc_top5_avg=0.86342 lr=0.00100 gn=33.45170 time=45.49it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.19230 test_loss_avg=0.83511 acc=0.94531 test_acc_avg=0.75087 test_acc_top5_avg=0.98698 time=229.81it/s +epoch=79 global_step=31280 loss=0.32561 test_loss_avg=0.82881 acc=0.90625 test_acc_avg=0.75225 test_acc_top5_avg=0.96610 time=234.08it/s +epoch=79 global_step=31280 loss=0.23151 test_loss_avg=0.66747 acc=0.93750 test_acc_avg=0.79984 test_acc_top5_avg=0.97369 time=537.52it/s +curr_acc 0.7998 +BEST_ACC 0.8636 +curr_acc_top5 0.9737 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=3.94153 loss_avg=3.89930 acc=0.62500 acc_top1_avg=0.62852 acc_top5_avg=0.87539 lr=0.00010 gn=28.74890 time=47.96it/s +epoch=80 global_step=31350 loss=3.74895 loss_avg=3.84398 acc=0.64844 acc_top1_avg=0.63426 acc_top5_avg=0.86618 lr=0.00010 gn=28.68902 time=52.46it/s +epoch=80 global_step=31400 loss=3.62667 loss_avg=3.77848 acc=0.66406 acc_top1_avg=0.64212 acc_top5_avg=0.86706 lr=0.00010 gn=31.39411 time=54.18it/s +epoch=80 global_step=31450 loss=4.16086 loss_avg=3.76435 acc=0.60156 acc_top1_avg=0.64380 acc_top5_avg=0.86677 lr=0.00010 gn=24.72194 time=54.19it/s +epoch=80 global_step=31500 loss=3.78423 loss_avg=3.77266 acc=0.62500 acc_top1_avg=0.64265 acc_top5_avg=0.86612 lr=0.00010 gn=26.21049 time=54.31it/s +epoch=80 global_step=31550 loss=3.84047 loss_avg=3.75357 acc=0.64062 acc_top1_avg=0.64473 acc_top5_avg=0.86713 lr=0.00010 gn=21.52716 time=53.30it/s +epoch=80 global_step=31600 loss=4.53771 loss_avg=3.75512 acc=0.55469 acc_top1_avg=0.64424 acc_top5_avg=0.86663 lr=0.00010 gn=31.43757 time=64.26it/s +epoch=80 global_step=31650 loss=4.29499 loss_avg=3.75450 acc=0.57812 acc_top1_avg=0.64396 acc_top5_avg=0.86615 lr=0.00010 gn=27.25677 time=54.23it/s +====================Eval==================== +epoch=80 global_step=31671 loss=1.51331 test_loss_avg=0.90982 acc=0.59375 test_acc_avg=0.73724 test_acc_top5_avg=0.96276 time=237.84it/s +epoch=80 global_step=31671 loss=0.13425 test_loss_avg=0.67069 acc=0.93750 test_acc_avg=0.79935 test_acc_top5_avg=0.97389 time=753.42it/s +curr_acc 0.7993 +BEST_ACC 0.8636 +curr_acc_top5 0.9739 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=3.83725 loss_avg=3.80548 acc=0.62500 acc_top1_avg=0.63605 acc_top5_avg=0.86638 lr=0.00010 gn=36.28930 time=56.30it/s +epoch=81 global_step=31750 loss=3.53991 loss_avg=3.75055 acc=0.66406 acc_top1_avg=0.64488 acc_top5_avg=0.86383 lr=0.00010 gn=29.04592 time=55.77it/s +epoch=81 global_step=31800 loss=3.45576 loss_avg=3.72157 acc=0.68750 acc_top1_avg=0.64771 acc_top5_avg=0.86628 lr=0.00010 gn=27.11383 time=57.45it/s +epoch=81 global_step=31850 loss=3.69394 loss_avg=3.70739 acc=0.65625 acc_top1_avg=0.64935 acc_top5_avg=0.86754 lr=0.00010 gn=27.27861 time=49.84it/s +epoch=81 global_step=31900 loss=3.99172 loss_avg=3.70903 acc=0.61719 acc_top1_avg=0.64919 acc_top5_avg=0.86575 lr=0.00010 gn=23.69837 time=57.18it/s +epoch=81 global_step=31950 loss=3.86671 loss_avg=3.70876 acc=0.63281 acc_top1_avg=0.64964 acc_top5_avg=0.86607 lr=0.00010 gn=27.15241 time=54.82it/s +epoch=81 global_step=32000 loss=3.84378 loss_avg=3.69891 acc=0.62500 acc_top1_avg=0.65065 acc_top5_avg=0.86709 lr=0.00010 gn=24.37759 time=60.04it/s +epoch=81 global_step=32050 loss=3.67465 loss_avg=3.70245 acc=0.62500 acc_top1_avg=0.65019 acc_top5_avg=0.86626 lr=0.00010 gn=27.08455 time=55.79it/s +====================Eval==================== +epoch=81 global_step=32062 loss=1.27966 test_loss_avg=1.27966 acc=0.67188 test_acc_avg=0.67188 test_acc_top5_avg=0.96094 time=210.37it/s +epoch=81 global_step=32062 loss=1.33009 test_loss_avg=0.85527 acc=0.55469 test_acc_avg=0.74617 test_acc_top5_avg=0.96615 time=241.15it/s +epoch=81 global_step=32062 loss=0.08647 test_loss_avg=0.68172 acc=0.93750 test_acc_avg=0.79470 test_acc_top5_avg=0.97389 time=653.22it/s +curr_acc 0.7947 +BEST_ACC 0.8636 +curr_acc_top5 0.9739 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=4.00222 loss_avg=3.62863 acc=0.62500 acc_top1_avg=0.65440 acc_top5_avg=0.86616 lr=0.00010 gn=26.93449 time=54.44it/s +epoch=82 global_step=32150 loss=4.46957 loss_avg=3.70399 acc=0.57031 acc_top1_avg=0.64711 acc_top5_avg=0.86435 lr=0.00010 gn=28.98723 time=54.70it/s +epoch=82 global_step=32200 loss=3.61275 loss_avg=3.68480 acc=0.67188 acc_top1_avg=0.64991 acc_top5_avg=0.86673 lr=0.00010 gn=29.82066 time=59.75it/s +epoch=82 global_step=32250 loss=2.95269 loss_avg=3.66723 acc=0.72656 acc_top1_avg=0.65205 acc_top5_avg=0.86719 lr=0.00010 gn=23.24112 time=57.02it/s +epoch=82 global_step=32300 loss=3.79587 loss_avg=3.68014 acc=0.64062 acc_top1_avg=0.65119 acc_top5_avg=0.86653 lr=0.00010 gn=27.49973 time=51.36it/s +epoch=82 global_step=32350 loss=4.00792 loss_avg=3.67421 acc=0.60156 acc_top1_avg=0.65202 acc_top5_avg=0.86732 lr=0.00010 gn=25.01564 time=47.76it/s +epoch=82 global_step=32400 loss=4.61846 loss_avg=3.68227 acc=0.56250 acc_top1_avg=0.65105 acc_top5_avg=0.86693 lr=0.00010 gn=27.32500 time=60.06it/s +epoch=82 global_step=32450 loss=3.83886 loss_avg=3.68691 acc=0.64844 acc_top1_avg=0.65085 acc_top5_avg=0.86709 lr=0.00010 gn=35.05589 time=63.43it/s +====================Eval==================== +epoch=82 global_step=32453 loss=0.97891 test_loss_avg=0.72014 acc=0.69531 test_acc_avg=0.78516 test_acc_top5_avg=0.97905 time=232.42it/s +epoch=82 global_step=32453 loss=0.17891 test_loss_avg=0.74653 acc=0.94531 test_acc_avg=0.77756 test_acc_top5_avg=0.97027 time=242.53it/s +epoch=82 global_step=32453 loss=0.11330 test_loss_avg=0.69232 acc=0.93750 test_acc_avg=0.79312 test_acc_top5_avg=0.97261 time=502.07it/s +curr_acc 0.7931 +BEST_ACC 0.8636 +curr_acc_top5 0.9726 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=3.56002 loss_avg=3.64855 acc=0.67188 acc_top1_avg=0.65525 acc_top5_avg=0.87450 lr=0.00010 gn=23.02224 time=59.92it/s +epoch=83 global_step=32550 loss=4.08037 loss_avg=3.61307 acc=0.60156 acc_top1_avg=0.65891 acc_top5_avg=0.87226 lr=0.00010 gn=27.02885 time=52.30it/s +epoch=83 global_step=32600 loss=3.23692 loss_avg=3.63182 acc=0.68750 acc_top1_avg=0.65742 acc_top5_avg=0.87107 lr=0.00010 gn=19.96912 time=56.20it/s +epoch=83 global_step=32650 loss=3.35874 loss_avg=3.66998 acc=0.69531 acc_top1_avg=0.65367 acc_top5_avg=0.86897 lr=0.00010 gn=30.28038 time=56.97it/s +epoch=83 global_step=32700 loss=4.00335 loss_avg=3.68118 acc=0.61719 acc_top1_avg=0.65198 acc_top5_avg=0.86791 lr=0.00010 gn=25.88018 time=62.76it/s +epoch=83 global_step=32750 loss=3.91674 loss_avg=3.67024 acc=0.64844 acc_top1_avg=0.65299 acc_top5_avg=0.86924 lr=0.00010 gn=34.15172 time=55.23it/s +epoch=83 global_step=32800 loss=3.44499 loss_avg=3.65634 acc=0.67969 acc_top1_avg=0.65458 acc_top5_avg=0.86921 lr=0.00010 gn=30.05010 time=54.99it/s +====================Eval==================== +epoch=83 global_step=32844 loss=0.82744 test_loss_avg=0.78993 acc=0.74219 test_acc_avg=0.76726 test_acc_top5_avg=0.96748 time=236.89it/s +epoch=83 global_step=32844 loss=0.12296 test_loss_avg=0.68823 acc=0.93750 test_acc_avg=0.79529 test_acc_top5_avg=0.97201 time=804.28it/s +curr_acc 0.7953 +BEST_ACC 0.8636 +curr_acc_top5 0.9720 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=3.21953 loss_avg=3.45843 acc=0.71094 acc_top1_avg=0.67969 acc_top5_avg=0.88411 lr=0.00010 gn=29.52287 time=53.88it/s +epoch=84 global_step=32900 loss=3.19076 loss_avg=3.63754 acc=0.71875 acc_top1_avg=0.65765 acc_top5_avg=0.86342 lr=0.00010 gn=32.05440 time=54.77it/s +epoch=84 global_step=32950 loss=4.02913 loss_avg=3.69438 acc=0.62500 acc_top1_avg=0.65153 acc_top5_avg=0.86210 lr=0.00010 gn=36.94234 time=56.33it/s +epoch=84 global_step=33000 loss=3.70042 loss_avg=3.67743 acc=0.64062 acc_top1_avg=0.65294 acc_top5_avg=0.86313 lr=0.00010 gn=27.40578 time=55.95it/s +epoch=84 global_step=33050 loss=3.53276 loss_avg=3.67144 acc=0.65625 acc_top1_avg=0.65356 acc_top5_avg=0.86457 lr=0.00010 gn=32.11219 time=54.73it/s +epoch=84 global_step=33100 loss=3.44600 loss_avg=3.66848 acc=0.67969 acc_top1_avg=0.65427 acc_top5_avg=0.86514 lr=0.00010 gn=23.14301 time=56.09it/s +epoch=84 global_step=33150 loss=3.03179 loss_avg=3.66013 acc=0.72656 acc_top1_avg=0.65490 acc_top5_avg=0.86691 lr=0.00010 gn=27.31854 time=60.11it/s +epoch=84 global_step=33200 loss=3.07414 loss_avg=3.65791 acc=0.71875 acc_top1_avg=0.65515 acc_top5_avg=0.86714 lr=0.00010 gn=28.13895 time=57.57it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.27548 test_loss_avg=0.62540 acc=0.91406 test_acc_avg=0.81250 test_acc_top5_avg=0.98549 time=241.04it/s +epoch=84 global_step=33235 loss=0.27973 test_loss_avg=0.80109 acc=0.91406 test_acc_avg=0.75806 test_acc_top5_avg=0.96790 time=155.93it/s +epoch=84 global_step=33235 loss=0.09211 test_loss_avg=0.67699 acc=0.93750 test_acc_avg=0.79559 test_acc_top5_avg=0.97340 time=849.74it/s +curr_acc 0.7956 +BEST_ACC 0.8636 +curr_acc_top5 0.9734 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=3.92520 loss_avg=3.50881 acc=0.63281 acc_top1_avg=0.66927 acc_top5_avg=0.87760 lr=0.00010 gn=29.46617 time=54.82it/s +epoch=85 global_step=33300 loss=3.43977 loss_avg=3.62458 acc=0.68750 acc_top1_avg=0.65661 acc_top5_avg=0.86671 lr=0.00010 gn=33.57211 time=54.00it/s +epoch=85 global_step=33350 loss=4.04338 loss_avg=3.63955 acc=0.60156 acc_top1_avg=0.65482 acc_top5_avg=0.86671 lr=0.00010 gn=25.93548 time=51.90it/s +epoch=85 global_step=33400 loss=3.71660 loss_avg=3.61591 acc=0.64844 acc_top1_avg=0.65748 acc_top5_avg=0.86932 lr=0.00010 gn=30.87617 time=59.21it/s +epoch=85 global_step=33450 loss=3.54665 loss_avg=3.60013 acc=0.66406 acc_top1_avg=0.65890 acc_top5_avg=0.86900 lr=0.00010 gn=29.22247 time=54.90it/s +epoch=85 global_step=33500 loss=3.43787 loss_avg=3.61212 acc=0.69531 acc_top1_avg=0.65793 acc_top5_avg=0.86846 lr=0.00010 gn=29.88367 time=56.35it/s +epoch=85 global_step=33550 loss=3.65719 loss_avg=3.63083 acc=0.66406 acc_top1_avg=0.65642 acc_top5_avg=0.86806 lr=0.00010 gn=32.89413 time=54.82it/s +epoch=85 global_step=33600 loss=3.98704 loss_avg=3.62646 acc=0.61719 acc_top1_avg=0.65679 acc_top5_avg=0.86811 lr=0.00010 gn=24.79333 time=51.54it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.24321 test_loss_avg=0.89342 acc=0.90625 test_acc_avg=0.73839 test_acc_top5_avg=0.96228 time=231.30it/s +epoch=85 global_step=33626 loss=0.10811 test_loss_avg=0.70503 acc=0.93750 test_acc_avg=0.78985 test_acc_top5_avg=0.97073 time=825.33it/s +curr_acc 0.7899 +BEST_ACC 0.8636 +curr_acc_top5 0.9707 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=3.58201 loss_avg=3.58264 acc=0.66406 acc_top1_avg=0.66048 acc_top5_avg=0.86393 lr=0.00010 gn=27.50302 time=56.53it/s +epoch=86 global_step=33700 loss=3.90577 loss_avg=3.59558 acc=0.63281 acc_top1_avg=0.66026 acc_top5_avg=0.86592 lr=0.00010 gn=27.35541 time=55.90it/s +epoch=86 global_step=33750 loss=3.93758 loss_avg=3.59537 acc=0.61719 acc_top1_avg=0.66116 acc_top5_avg=0.86883 lr=0.00010 gn=25.41029 time=56.20it/s +epoch=86 global_step=33800 loss=3.33375 loss_avg=3.62174 acc=0.70312 acc_top1_avg=0.65841 acc_top5_avg=0.86620 lr=0.00010 gn=39.30360 time=62.17it/s +epoch=86 global_step=33850 loss=3.27240 loss_avg=3.60783 acc=0.71094 acc_top1_avg=0.66009 acc_top5_avg=0.86649 lr=0.00010 gn=29.20352 time=61.19it/s +epoch=86 global_step=33900 loss=3.49476 loss_avg=3.62465 acc=0.68750 acc_top1_avg=0.65839 acc_top5_avg=0.86568 lr=0.00010 gn=35.30293 time=47.48it/s +epoch=86 global_step=33950 loss=4.01355 loss_avg=3.63405 acc=0.60938 acc_top1_avg=0.65746 acc_top5_avg=0.86574 lr=0.00010 gn=29.75833 time=57.42it/s +epoch=86 global_step=34000 loss=3.32708 loss_avg=3.63715 acc=0.67969 acc_top1_avg=0.65727 acc_top5_avg=0.86633 lr=0.00010 gn=34.78676 time=55.45it/s +====================Eval==================== +epoch=86 global_step=34017 loss=1.26053 test_loss_avg=1.10435 acc=0.64844 test_acc_avg=0.67708 test_acc_top5_avg=0.97526 time=246.03it/s +epoch=86 global_step=34017 loss=0.41345 test_loss_avg=0.94187 acc=0.89062 test_acc_avg=0.72070 test_acc_top5_avg=0.95871 time=227.12it/s +epoch=86 global_step=34017 loss=0.11619 test_loss_avg=0.72403 acc=0.93750 test_acc_avg=0.78501 test_acc_top5_avg=0.96954 time=501.05it/s +curr_acc 0.7850 +BEST_ACC 0.8636 +curr_acc_top5 0.9695 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=3.55205 loss_avg=3.59425 acc=0.65625 acc_top1_avg=0.66051 acc_top5_avg=0.86790 lr=0.00010 gn=26.03846 time=53.55it/s +epoch=87 global_step=34100 loss=3.63717 loss_avg=3.59717 acc=0.64062 acc_top1_avg=0.66002 acc_top5_avg=0.86380 lr=0.00010 gn=30.92325 time=54.00it/s +epoch=87 global_step=34150 loss=3.33047 loss_avg=3.58964 acc=0.68750 acc_top1_avg=0.66207 acc_top5_avg=0.86531 lr=0.00010 gn=24.45876 time=55.55it/s +epoch=87 global_step=34200 loss=3.23676 loss_avg=3.63542 acc=0.69531 acc_top1_avg=0.65702 acc_top5_avg=0.86488 lr=0.00010 gn=21.95501 time=59.78it/s +epoch=87 global_step=34250 loss=3.94706 loss_avg=3.63919 acc=0.62500 acc_top1_avg=0.65635 acc_top5_avg=0.86568 lr=0.00010 gn=29.67735 time=49.04it/s +epoch=87 global_step=34300 loss=4.19031 loss_avg=3.64540 acc=0.58594 acc_top1_avg=0.65570 acc_top5_avg=0.86595 lr=0.00010 gn=25.88607 time=54.18it/s +epoch=87 global_step=34350 loss=4.22115 loss_avg=3.63347 acc=0.58594 acc_top1_avg=0.65698 acc_top5_avg=0.86684 lr=0.00010 gn=26.76430 time=56.31it/s +epoch=87 global_step=34400 loss=3.31103 loss_avg=3.62666 acc=0.71094 acc_top1_avg=0.65800 acc_top5_avg=0.86782 lr=0.00010 gn=33.21133 time=60.63it/s +====================Eval==================== +epoch=87 global_step=34408 loss=1.69180 test_loss_avg=0.85640 acc=0.54688 test_acc_avg=0.74971 test_acc_top5_avg=0.96759 time=236.94it/s +epoch=87 global_step=34408 loss=0.19290 test_loss_avg=0.72662 acc=0.94531 test_acc_avg=0.78490 test_acc_top5_avg=0.96966 time=244.55it/s +epoch=87 global_step=34408 loss=0.19457 test_loss_avg=0.71221 acc=0.93750 test_acc_avg=0.78906 test_acc_top5_avg=0.97043 time=687.37it/s +curr_acc 0.7891 +BEST_ACC 0.8636 +curr_acc_top5 0.9704 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=3.42791 loss_avg=3.53369 acc=0.65625 acc_top1_avg=0.66685 acc_top5_avg=0.87574 lr=0.00010 gn=24.44306 time=59.78it/s +epoch=88 global_step=34500 loss=3.52992 loss_avg=3.58901 acc=0.66406 acc_top1_avg=0.66202 acc_top5_avg=0.87254 lr=0.00010 gn=24.61890 time=54.53it/s +epoch=88 global_step=34550 loss=3.86128 loss_avg=3.58269 acc=0.62500 acc_top1_avg=0.66329 acc_top5_avg=0.87384 lr=0.00010 gn=27.12729 time=54.76it/s +epoch=88 global_step=34600 loss=4.05439 loss_avg=3.60091 acc=0.62500 acc_top1_avg=0.66109 acc_top5_avg=0.87162 lr=0.00010 gn=27.25819 time=57.07it/s +epoch=88 global_step=34650 loss=4.23935 loss_avg=3.62209 acc=0.59375 acc_top1_avg=0.65883 acc_top5_avg=0.87106 lr=0.00010 gn=35.62562 time=53.14it/s +epoch=88 global_step=34700 loss=3.80302 loss_avg=3.61161 acc=0.64844 acc_top1_avg=0.65984 acc_top5_avg=0.87053 lr=0.00010 gn=29.48998 time=41.87it/s +epoch=88 global_step=34750 loss=3.98811 loss_avg=3.61699 acc=0.63281 acc_top1_avg=0.65911 acc_top5_avg=0.86959 lr=0.00010 gn=32.38277 time=50.29it/s +====================Eval==================== +epoch=88 global_step=34799 loss=1.70782 test_loss_avg=0.83550 acc=0.46875 test_acc_avg=0.75130 test_acc_top5_avg=0.96501 time=245.83it/s +epoch=88 global_step=34799 loss=0.15118 test_loss_avg=0.71260 acc=0.93750 test_acc_avg=0.78807 test_acc_top5_avg=0.96964 time=495.37it/s +curr_acc 0.7881 +BEST_ACC 0.8636 +curr_acc_top5 0.9696 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=2.98252 loss_avg=2.98252 acc=0.72656 acc_top1_avg=0.72656 acc_top5_avg=0.90625 lr=0.00010 gn=34.40080 time=42.57it/s +epoch=89 global_step=34850 loss=3.50908 loss_avg=3.53826 acc=0.66406 acc_top1_avg=0.66820 acc_top5_avg=0.87393 lr=0.00010 gn=24.28550 time=55.08it/s +epoch=89 global_step=34900 loss=3.82823 loss_avg=3.58269 acc=0.64062 acc_top1_avg=0.66337 acc_top5_avg=0.87075 lr=0.00010 gn=33.31040 time=56.96it/s +epoch=89 global_step=34950 loss=2.64974 loss_avg=3.57189 acc=0.76562 acc_top1_avg=0.66437 acc_top5_avg=0.87081 lr=0.00010 gn=25.85703 time=50.08it/s +epoch=89 global_step=35000 loss=3.83115 loss_avg=3.61194 acc=0.62500 acc_top1_avg=0.65971 acc_top5_avg=0.86948 lr=0.00010 gn=30.32208 time=57.91it/s +epoch=89 global_step=35050 loss=3.76858 loss_avg=3.58940 acc=0.64062 acc_top1_avg=0.66216 acc_top5_avg=0.87008 lr=0.00010 gn=26.96502 time=56.16it/s +epoch=89 global_step=35100 loss=3.82522 loss_avg=3.59679 acc=0.64844 acc_top1_avg=0.66139 acc_top5_avg=0.86882 lr=0.00010 gn=33.59564 time=52.98it/s +epoch=89 global_step=35150 loss=3.89389 loss_avg=3.59933 acc=0.63281 acc_top1_avg=0.66101 acc_top5_avg=0.86832 lr=0.00010 gn=30.84683 time=52.97it/s +====================Eval==================== +epoch=89 global_step=35190 loss=0.82808 test_loss_avg=0.72971 acc=0.77344 test_acc_avg=0.78824 test_acc_top5_avg=0.97862 time=237.10it/s +epoch=89 global_step=35190 loss=0.18670 test_loss_avg=0.79319 acc=0.95312 test_acc_avg=0.76540 test_acc_top5_avg=0.96581 time=244.20it/s +epoch=89 global_step=35190 loss=0.16103 test_loss_avg=0.71112 acc=0.93750 test_acc_avg=0.78906 test_acc_top5_avg=0.97004 time=503.64it/s +curr_acc 0.7891 +BEST_ACC 0.8636 +curr_acc_top5 0.9700 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=3.35352 loss_avg=3.73441 acc=0.67188 acc_top1_avg=0.64062 acc_top5_avg=0.85391 lr=0.00010 gn=27.65233 time=58.20it/s +epoch=90 global_step=35250 loss=3.31224 loss_avg=3.65582 acc=0.69531 acc_top1_avg=0.65430 acc_top5_avg=0.86576 lr=0.00010 gn=25.60419 time=53.65it/s +epoch=90 global_step=35300 loss=3.54066 loss_avg=3.59682 acc=0.69531 acc_top1_avg=0.66087 acc_top5_avg=0.86825 lr=0.00010 gn=32.13875 time=55.00it/s +epoch=90 global_step=35350 loss=2.68916 loss_avg=3.59601 acc=0.75000 acc_top1_avg=0.66123 acc_top5_avg=0.86934 lr=0.00010 gn=33.65620 time=55.77it/s +epoch=90 global_step=35400 loss=3.05513 loss_avg=3.59625 acc=0.70312 acc_top1_avg=0.66120 acc_top5_avg=0.86968 lr=0.00010 gn=31.70805 time=57.08it/s +epoch=90 global_step=35450 loss=3.80473 loss_avg=3.60012 acc=0.64062 acc_top1_avg=0.66124 acc_top5_avg=0.86938 lr=0.00010 gn=31.62915 time=54.54it/s +epoch=90 global_step=35500 loss=3.58970 loss_avg=3.59689 acc=0.65625 acc_top1_avg=0.66149 acc_top5_avg=0.86930 lr=0.00010 gn=35.58453 time=56.97it/s +epoch=90 global_step=35550 loss=3.95987 loss_avg=3.58893 acc=0.61719 acc_top1_avg=0.66211 acc_top5_avg=0.86914 lr=0.00010 gn=26.31342 time=53.46it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.79771 test_loss_avg=0.84330 acc=0.76562 test_acc_avg=0.75527 test_acc_top5_avg=0.96582 time=237.31it/s +epoch=90 global_step=35581 loss=0.14474 test_loss_avg=0.72291 acc=0.93750 test_acc_avg=0.78699 test_acc_top5_avg=0.96974 time=853.72it/s +curr_acc 0.7870 +BEST_ACC 0.8636 +curr_acc_top5 0.9697 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=3.55906 loss_avg=3.72869 acc=0.65625 acc_top1_avg=0.64638 acc_top5_avg=0.86472 lr=0.00010 gn=34.93790 time=55.98it/s +epoch=91 global_step=35650 loss=3.10833 loss_avg=3.61458 acc=0.70312 acc_top1_avg=0.65897 acc_top5_avg=0.86945 lr=0.00010 gn=31.65944 time=53.95it/s +epoch=91 global_step=35700 loss=3.76210 loss_avg=3.62025 acc=0.65625 acc_top1_avg=0.65815 acc_top5_avg=0.86594 lr=0.00010 gn=33.96536 time=54.23it/s +epoch=91 global_step=35750 loss=3.32744 loss_avg=3.59664 acc=0.69531 acc_top1_avg=0.66073 acc_top5_avg=0.86894 lr=0.00010 gn=35.46347 time=59.75it/s +epoch=91 global_step=35800 loss=3.63731 loss_avg=3.57391 acc=0.65625 acc_top1_avg=0.66310 acc_top5_avg=0.87004 lr=0.00010 gn=24.43561 time=63.18it/s +epoch=91 global_step=35850 loss=3.71761 loss_avg=3.58097 acc=0.63281 acc_top1_avg=0.66229 acc_top5_avg=0.86931 lr=0.00010 gn=26.74218 time=61.51it/s +epoch=91 global_step=35900 loss=3.67420 loss_avg=3.58647 acc=0.66406 acc_top1_avg=0.66232 acc_top5_avg=0.86866 lr=0.00010 gn=34.38908 time=61.95it/s +epoch=91 global_step=35950 loss=2.64748 loss_avg=3.58934 acc=0.75000 acc_top1_avg=0.66156 acc_top5_avg=0.86806 lr=0.00010 gn=33.34236 time=51.38it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.27215 test_loss_avg=0.87573 acc=0.92969 test_acc_avg=0.74290 test_acc_top5_avg=0.97798 time=222.23it/s +epoch=91 global_step=35972 loss=0.17323 test_loss_avg=0.88542 acc=0.94531 test_acc_avg=0.73783 test_acc_top5_avg=0.96004 time=238.07it/s +epoch=91 global_step=35972 loss=0.15188 test_loss_avg=0.72194 acc=0.93750 test_acc_avg=0.78570 test_acc_top5_avg=0.96826 time=531.19it/s +curr_acc 0.7857 +BEST_ACC 0.8636 +curr_acc_top5 0.9683 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=3.56366 loss_avg=3.70129 acc=0.66406 acc_top1_avg=0.65206 acc_top5_avg=0.87026 lr=0.00010 gn=36.79374 time=59.81it/s +epoch=92 global_step=36050 loss=3.86487 loss_avg=3.65071 acc=0.65625 acc_top1_avg=0.65695 acc_top5_avg=0.86819 lr=0.00010 gn=39.16993 time=54.09it/s +epoch=92 global_step=36100 loss=3.81989 loss_avg=3.61138 acc=0.62500 acc_top1_avg=0.66052 acc_top5_avg=0.86920 lr=0.00010 gn=29.88499 time=58.04it/s +epoch=92 global_step=36150 loss=3.34209 loss_avg=3.60663 acc=0.69531 acc_top1_avg=0.66055 acc_top5_avg=0.86947 lr=0.00010 gn=36.40741 time=56.37it/s +epoch=92 global_step=36200 loss=3.24177 loss_avg=3.60013 acc=0.69531 acc_top1_avg=0.66142 acc_top5_avg=0.86722 lr=0.00010 gn=26.54506 time=44.78it/s +epoch=92 global_step=36250 loss=3.96483 loss_avg=3.60126 acc=0.61719 acc_top1_avg=0.66103 acc_top5_avg=0.86783 lr=0.00010 gn=27.55556 time=62.85it/s +epoch=92 global_step=36300 loss=3.78879 loss_avg=3.59274 acc=0.63281 acc_top1_avg=0.66190 acc_top5_avg=0.86869 lr=0.00010 gn=29.78184 time=58.09it/s +epoch=92 global_step=36350 loss=3.58171 loss_avg=3.58610 acc=0.66406 acc_top1_avg=0.66222 acc_top5_avg=0.86936 lr=0.00010 gn=26.53519 time=58.88it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.51130 test_loss_avg=0.96922 acc=0.80469 test_acc_avg=0.71411 test_acc_top5_avg=0.95898 time=236.85it/s +epoch=92 global_step=36363 loss=0.15118 test_loss_avg=0.72146 acc=0.93750 test_acc_avg=0.78610 test_acc_top5_avg=0.97033 time=845.46it/s +curr_acc 0.7861 +BEST_ACC 0.8636 +curr_acc_top5 0.9703 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=2.80712 loss_avg=3.55398 acc=0.75000 acc_top1_avg=0.66765 acc_top5_avg=0.86318 lr=0.00010 gn=26.58712 time=58.37it/s +epoch=93 global_step=36450 loss=4.01563 loss_avg=3.56327 acc=0.59375 acc_top1_avg=0.66658 acc_top5_avg=0.86548 lr=0.00010 gn=24.02023 time=63.08it/s +epoch=93 global_step=36500 loss=3.21714 loss_avg=3.53678 acc=0.70312 acc_top1_avg=0.66902 acc_top5_avg=0.87027 lr=0.00010 gn=24.79872 time=51.06it/s +epoch=93 global_step=36550 loss=3.64236 loss_avg=3.56577 acc=0.65625 acc_top1_avg=0.66569 acc_top5_avg=0.86852 lr=0.00010 gn=27.63058 time=62.39it/s +epoch=93 global_step=36600 loss=3.46264 loss_avg=3.57443 acc=0.67969 acc_top1_avg=0.66508 acc_top5_avg=0.86663 lr=0.00010 gn=29.30701 time=53.14it/s +epoch=93 global_step=36650 loss=2.68828 loss_avg=3.56384 acc=0.76562 acc_top1_avg=0.66616 acc_top5_avg=0.86814 lr=0.00010 gn=34.92557 time=60.27it/s +epoch=93 global_step=36700 loss=4.02144 loss_avg=3.57425 acc=0.62500 acc_top1_avg=0.66504 acc_top5_avg=0.86856 lr=0.00010 gn=38.32785 time=55.38it/s +epoch=93 global_step=36750 loss=3.64363 loss_avg=3.57614 acc=0.65625 acc_top1_avg=0.66459 acc_top5_avg=0.86904 lr=0.00010 gn=30.30607 time=62.96it/s +====================Eval==================== +epoch=93 global_step=36754 loss=1.13471 test_loss_avg=1.19462 acc=0.68750 test_acc_avg=0.66927 test_acc_top5_avg=0.96615 time=237.19it/s +epoch=93 global_step=36754 loss=1.67960 test_loss_avg=0.92919 acc=0.49219 test_acc_avg=0.72818 test_acc_top5_avg=0.95799 time=239.66it/s +epoch=93 global_step=36754 loss=0.14716 test_loss_avg=0.72531 acc=0.93750 test_acc_avg=0.78689 test_acc_top5_avg=0.96855 time=822.09it/s +curr_acc 0.7869 +BEST_ACC 0.8636 +curr_acc_top5 0.9686 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=3.64265 loss_avg=3.68288 acc=0.64062 acc_top1_avg=0.65285 acc_top5_avg=0.86583 lr=0.00010 gn=26.88399 time=60.06it/s +epoch=94 global_step=36850 loss=3.39125 loss_avg=3.61095 acc=0.67969 acc_top1_avg=0.66016 acc_top5_avg=0.86776 lr=0.00010 gn=22.85632 time=52.36it/s +epoch=94 global_step=36900 loss=3.37143 loss_avg=3.59817 acc=0.68750 acc_top1_avg=0.66080 acc_top5_avg=0.86992 lr=0.00010 gn=30.82390 time=52.86it/s +epoch=94 global_step=36950 loss=2.96121 loss_avg=3.59013 acc=0.72656 acc_top1_avg=0.66227 acc_top5_avg=0.86814 lr=0.00010 gn=26.43424 time=54.04it/s +epoch=94 global_step=37000 loss=3.47400 loss_avg=3.58669 acc=0.67969 acc_top1_avg=0.66247 acc_top5_avg=0.86868 lr=0.00010 gn=29.76795 time=52.70it/s +epoch=94 global_step=37050 loss=3.92665 loss_avg=3.57602 acc=0.64062 acc_top1_avg=0.66375 acc_top5_avg=0.86859 lr=0.00010 gn=35.83192 time=54.89it/s +epoch=94 global_step=37100 loss=3.88135 loss_avg=3.56748 acc=0.62500 acc_top1_avg=0.66485 acc_top5_avg=0.86911 lr=0.00010 gn=34.62074 time=47.32it/s +====================Eval==================== +epoch=94 global_step=37145 loss=1.52418 test_loss_avg=0.77966 acc=0.57812 test_acc_avg=0.77702 test_acc_top5_avg=0.97233 time=235.89it/s +epoch=94 global_step=37145 loss=0.16379 test_loss_avg=0.75653 acc=0.95312 test_acc_avg=0.77829 test_acc_top5_avg=0.96727 time=241.79it/s +epoch=94 global_step=37145 loss=0.11802 test_loss_avg=0.71699 acc=0.93750 test_acc_avg=0.78946 test_acc_top5_avg=0.96934 time=782.96it/s +curr_acc 0.7895 +BEST_ACC 0.8636 +curr_acc_top5 0.9693 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=3.29236 loss_avg=3.58620 acc=0.71094 acc_top1_avg=0.65938 acc_top5_avg=0.85156 lr=0.00010 gn=38.08711 time=51.01it/s +epoch=95 global_step=37200 loss=3.80912 loss_avg=3.54807 acc=0.64844 acc_top1_avg=0.66676 acc_top5_avg=0.86690 lr=0.00010 gn=34.68763 time=57.68it/s +epoch=95 global_step=37250 loss=3.23159 loss_avg=3.58525 acc=0.71094 acc_top1_avg=0.66302 acc_top5_avg=0.86592 lr=0.00010 gn=36.46453 time=57.86it/s +epoch=95 global_step=37300 loss=3.55795 loss_avg=3.58677 acc=0.66406 acc_top1_avg=0.66114 acc_top5_avg=0.86648 lr=0.00010 gn=25.81269 time=58.83it/s +epoch=95 global_step=37350 loss=3.92150 loss_avg=3.60453 acc=0.64062 acc_top1_avg=0.66029 acc_top5_avg=0.86547 lr=0.00010 gn=33.85844 time=50.02it/s +epoch=95 global_step=37400 loss=3.57242 loss_avg=3.59492 acc=0.67188 acc_top1_avg=0.66170 acc_top5_avg=0.86602 lr=0.00010 gn=31.39452 time=30.86it/s +epoch=95 global_step=37450 loss=3.39259 loss_avg=3.59343 acc=0.66406 acc_top1_avg=0.66214 acc_top5_avg=0.86603 lr=0.00010 gn=30.31961 time=30.92it/s +epoch=95 global_step=37500 loss=2.67459 loss_avg=3.57301 acc=0.77344 acc_top1_avg=0.66452 acc_top5_avg=0.86794 lr=0.00010 gn=29.61894 time=49.92it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.62276 test_loss_avg=0.87285 acc=0.82031 test_acc_avg=0.74427 test_acc_top5_avg=0.96215 time=239.29it/s +epoch=95 global_step=37536 loss=0.14365 test_loss_avg=0.76706 acc=0.93750 test_acc_avg=0.77561 test_acc_top5_avg=0.96628 time=505.89it/s +curr_acc 0.7756 +BEST_ACC 0.8636 +curr_acc_top5 0.9663 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=3.38282 loss_avg=3.43294 acc=0.67969 acc_top1_avg=0.67690 acc_top5_avg=0.86663 lr=0.00010 gn=38.43290 time=56.58it/s +epoch=96 global_step=37600 loss=3.94865 loss_avg=3.51028 acc=0.62500 acc_top1_avg=0.66980 acc_top5_avg=0.87146 lr=0.00010 gn=28.53250 time=53.80it/s +epoch=96 global_step=37650 loss=3.73067 loss_avg=3.54881 acc=0.64062 acc_top1_avg=0.66715 acc_top5_avg=0.87137 lr=0.00010 gn=31.95927 time=54.88it/s +epoch=96 global_step=37700 loss=4.21673 loss_avg=3.55485 acc=0.59375 acc_top1_avg=0.66716 acc_top5_avg=0.86971 lr=0.00010 gn=29.97071 time=62.63it/s +epoch=96 global_step=37750 loss=3.95446 loss_avg=3.55516 acc=0.62500 acc_top1_avg=0.66713 acc_top5_avg=0.86861 lr=0.00010 gn=30.77791 time=64.04it/s +epoch=96 global_step=37800 loss=3.45594 loss_avg=3.57396 acc=0.67188 acc_top1_avg=0.66483 acc_top5_avg=0.86796 lr=0.00010 gn=27.85348 time=63.05it/s +epoch=96 global_step=37850 loss=3.40332 loss_avg=3.57723 acc=0.67188 acc_top1_avg=0.66454 acc_top5_avg=0.86751 lr=0.00010 gn=37.10732 time=55.18it/s +epoch=96 global_step=37900 loss=3.58713 loss_avg=3.56414 acc=0.67188 acc_top1_avg=0.66576 acc_top5_avg=0.86850 lr=0.00010 gn=36.27945 time=63.19it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.65672 test_loss_avg=0.70239 acc=0.81250 test_acc_avg=0.79785 test_acc_top5_avg=0.98242 time=238.42it/s +epoch=96 global_step=37927 loss=0.16645 test_loss_avg=0.83839 acc=0.96094 test_acc_avg=0.75391 test_acc_top5_avg=0.96342 time=230.79it/s +epoch=96 global_step=37927 loss=0.13700 test_loss_avg=0.72632 acc=0.93750 test_acc_avg=0.78639 test_acc_top5_avg=0.96895 time=779.61it/s +curr_acc 0.7864 +BEST_ACC 0.8636 +curr_acc_top5 0.9689 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=3.11519 loss_avg=3.44118 acc=0.74219 acc_top1_avg=0.68207 acc_top5_avg=0.87058 lr=0.00010 gn=36.87285 time=63.28it/s +epoch=97 global_step=38000 loss=3.05334 loss_avg=3.50821 acc=0.72656 acc_top1_avg=0.67198 acc_top5_avg=0.86687 lr=0.00010 gn=38.00050 time=61.06it/s +epoch=97 global_step=38050 loss=3.15369 loss_avg=3.56689 acc=0.69531 acc_top1_avg=0.66495 acc_top5_avg=0.86554 lr=0.00010 gn=34.61324 time=57.03it/s +epoch=97 global_step=38100 loss=3.91905 loss_avg=3.57216 acc=0.63281 acc_top1_avg=0.66451 acc_top5_avg=0.86619 lr=0.00010 gn=35.56013 time=61.15it/s +epoch=97 global_step=38150 loss=3.62233 loss_avg=3.56944 acc=0.67969 acc_top1_avg=0.66494 acc_top5_avg=0.86596 lr=0.00010 gn=34.82812 time=63.77it/s +epoch=97 global_step=38200 loss=2.68567 loss_avg=3.56012 acc=0.75000 acc_top1_avg=0.66615 acc_top5_avg=0.86590 lr=0.00010 gn=29.10557 time=59.90it/s +epoch=97 global_step=38250 loss=3.57542 loss_avg=3.54855 acc=0.67188 acc_top1_avg=0.66706 acc_top5_avg=0.86733 lr=0.00010 gn=30.07063 time=56.52it/s +epoch=97 global_step=38300 loss=3.03765 loss_avg=3.54983 acc=0.72656 acc_top1_avg=0.66681 acc_top5_avg=0.86761 lr=0.00010 gn=33.31199 time=52.48it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.28474 test_loss_avg=0.90641 acc=0.91406 test_acc_avg=0.73670 test_acc_top5_avg=0.95756 time=231.21it/s +epoch=97 global_step=38318 loss=0.18052 test_loss_avg=0.74188 acc=0.93750 test_acc_avg=0.78165 test_acc_top5_avg=0.96657 time=506.74it/s +curr_acc 0.7816 +BEST_ACC 0.8636 +curr_acc_top5 0.9666 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=3.21596 loss_avg=3.69320 acc=0.70312 acc_top1_avg=0.64795 acc_top5_avg=0.87354 lr=0.00010 gn=30.63432 time=60.03it/s +epoch=98 global_step=38400 loss=3.70949 loss_avg=3.60878 acc=0.67969 acc_top1_avg=0.65978 acc_top5_avg=0.86976 lr=0.00010 gn=37.96258 time=60.59it/s +epoch=98 global_step=38450 loss=3.20420 loss_avg=3.55079 acc=0.69531 acc_top1_avg=0.66613 acc_top5_avg=0.87050 lr=0.00010 gn=35.33264 time=55.71it/s +epoch=98 global_step=38500 loss=3.54680 loss_avg=3.55074 acc=0.65625 acc_top1_avg=0.66574 acc_top5_avg=0.87032 lr=0.00010 gn=33.14500 time=59.20it/s +epoch=98 global_step=38550 loss=3.99183 loss_avg=3.55190 acc=0.61719 acc_top1_avg=0.66524 acc_top5_avg=0.86870 lr=0.00010 gn=27.08165 time=55.28it/s +epoch=98 global_step=38600 loss=3.45087 loss_avg=3.56939 acc=0.66406 acc_top1_avg=0.66359 acc_top5_avg=0.86818 lr=0.00010 gn=31.60773 time=52.12it/s +epoch=98 global_step=38650 loss=3.25967 loss_avg=3.55824 acc=0.71094 acc_top1_avg=0.66498 acc_top5_avg=0.86822 lr=0.00010 gn=33.46618 time=60.63it/s +epoch=98 global_step=38700 loss=3.68991 loss_avg=3.55529 acc=0.65625 acc_top1_avg=0.66560 acc_top5_avg=0.86778 lr=0.00010 gn=36.66031 time=55.64it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.77038 test_loss_avg=1.03178 acc=0.75781 test_acc_avg=0.69922 test_acc_top5_avg=0.97656 time=241.15it/s +epoch=98 global_step=38709 loss=0.36932 test_loss_avg=0.92578 acc=0.87500 test_acc_avg=0.73235 test_acc_top5_avg=0.95730 time=238.27it/s +epoch=98 global_step=38709 loss=0.16850 test_loss_avg=0.72886 acc=0.93750 test_acc_avg=0.78877 test_acc_top5_avg=0.96766 time=764.41it/s +curr_acc 0.7888 +BEST_ACC 0.8636 +curr_acc_top5 0.9677 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=3.09076 loss_avg=3.51753 acc=0.71875 acc_top1_avg=0.67054 acc_top5_avg=0.87138 lr=0.00010 gn=30.05900 time=56.27it/s +epoch=99 global_step=38800 loss=2.97943 loss_avg=3.51159 acc=0.74219 acc_top1_avg=0.67153 acc_top5_avg=0.86745 lr=0.00010 gn=32.59088 time=48.46it/s +epoch=99 global_step=38850 loss=3.69548 loss_avg=3.52666 acc=0.66406 acc_top1_avg=0.67038 acc_top5_avg=0.86802 lr=0.00010 gn=34.76803 time=51.61it/s +epoch=99 global_step=38900 loss=3.34652 loss_avg=3.53256 acc=0.69531 acc_top1_avg=0.66979 acc_top5_avg=0.86870 lr=0.00010 gn=30.11352 time=54.68it/s +epoch=99 global_step=38950 loss=3.81428 loss_avg=3.52850 acc=0.62500 acc_top1_avg=0.66983 acc_top5_avg=0.86884 lr=0.00010 gn=34.27120 time=52.79it/s +epoch=99 global_step=39000 loss=3.67813 loss_avg=3.52757 acc=0.65625 acc_top1_avg=0.66994 acc_top5_avg=0.86958 lr=0.00010 gn=33.45655 time=58.66it/s +epoch=99 global_step=39050 loss=3.63112 loss_avg=3.53938 acc=0.70312 acc_top1_avg=0.66864 acc_top5_avg=0.86886 lr=0.00010 gn=39.16128 time=47.50it/s +epoch=99 global_step=39100 loss=3.85092 loss_avg=3.54485 acc=0.63750 acc_top1_avg=0.66777 acc_top5_avg=0.86840 lr=0.00010 gn=45.89874 time=80.33it/s +====================Eval==================== +epoch=99 global_step=39100 loss=1.67862 test_loss_avg=0.97931 acc=0.53906 test_acc_avg=0.72037 test_acc_top5_avg=0.95555 time=241.65it/s +epoch=99 global_step=39100 loss=0.12389 test_loss_avg=0.76488 acc=0.93750 test_acc_avg=0.77631 test_acc_top5_avg=0.96539 time=544.01it/s +epoch=99 global_step=39100 loss=0.12389 test_loss_avg=0.76488 acc=0.93750 test_acc_avg=0.77631 test_acc_top5_avg=0.96539 time=544.01it/s +curr_acc 0.7763 +BEST_ACC 0.8636 +curr_acc_top5 0.9654 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=3.60414 loss_avg=3.58294 acc=0.64844 acc_top1_avg=0.66156 acc_top5_avg=0.85906 lr=0.00010 gn=25.98973 time=55.92it/s +epoch=100 global_step=39200 loss=3.48659 loss_avg=3.54330 acc=0.66406 acc_top1_avg=0.66711 acc_top5_avg=0.86586 lr=0.00010 gn=24.48216 time=56.16it/s +epoch=100 global_step=39250 loss=3.71725 loss_avg=3.53787 acc=0.64844 acc_top1_avg=0.66755 acc_top5_avg=0.86479 lr=0.00010 gn=30.98398 time=56.92it/s +epoch=100 global_step=39300 loss=3.56573 loss_avg=3.54668 acc=0.65625 acc_top1_avg=0.66668 acc_top5_avg=0.86637 lr=0.00010 gn=32.88804 time=55.04it/s +epoch=100 global_step=39350 loss=3.16699 loss_avg=3.53601 acc=0.72656 acc_top1_avg=0.66828 acc_top5_avg=0.86706 lr=0.00010 gn=32.78162 time=58.93it/s +epoch=100 global_step=39400 loss=3.48438 loss_avg=3.53174 acc=0.67969 acc_top1_avg=0.66883 acc_top5_avg=0.86786 lr=0.00010 gn=34.33478 time=56.41it/s +epoch=100 global_step=39450 loss=3.58941 loss_avg=3.52794 acc=0.67188 acc_top1_avg=0.66946 acc_top5_avg=0.86893 lr=0.00010 gn=31.66387 time=54.12it/s +====================Eval==================== +epoch=100 global_step=39491 loss=1.74443 test_loss_avg=0.90592 acc=0.50000 test_acc_avg=0.73813 test_acc_top5_avg=0.95859 time=158.31it/s +epoch=100 global_step=39491 loss=0.17700 test_loss_avg=0.73833 acc=0.93750 test_acc_avg=0.78441 test_acc_top5_avg=0.96628 time=833.69it/s +curr_acc 0.7844 +BEST_ACC 0.8636 +curr_acc_top5 0.9663 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=4.25990 loss_avg=3.44003 acc=0.57031 acc_top1_avg=0.67708 acc_top5_avg=0.86372 lr=0.00010 gn=35.65988 time=55.88it/s +epoch=101 global_step=39550 loss=2.84582 loss_avg=3.58119 acc=0.75000 acc_top1_avg=0.66102 acc_top5_avg=0.87010 lr=0.00010 gn=35.60865 time=55.75it/s +epoch=101 global_step=39600 loss=3.65034 loss_avg=3.54378 acc=0.64844 acc_top1_avg=0.66628 acc_top5_avg=0.87027 lr=0.00010 gn=25.19988 time=57.53it/s +epoch=101 global_step=39650 loss=3.77598 loss_avg=3.51961 acc=0.64844 acc_top1_avg=0.66922 acc_top5_avg=0.87009 lr=0.00010 gn=34.19525 time=61.76it/s +epoch=101 global_step=39700 loss=3.96971 loss_avg=3.52496 acc=0.61719 acc_top1_avg=0.66859 acc_top5_avg=0.86891 lr=0.00010 gn=32.32980 time=57.94it/s +epoch=101 global_step=39750 loss=3.50026 loss_avg=3.53337 acc=0.65625 acc_top1_avg=0.66816 acc_top5_avg=0.86948 lr=0.00010 gn=28.07776 time=57.44it/s +epoch=101 global_step=39800 loss=3.44870 loss_avg=3.53414 acc=0.66406 acc_top1_avg=0.66826 acc_top5_avg=0.86934 lr=0.00010 gn=31.61032 time=52.81it/s +epoch=101 global_step=39850 loss=3.08098 loss_avg=3.53420 acc=0.71875 acc_top1_avg=0.66831 acc_top5_avg=0.86960 lr=0.00010 gn=23.01315 time=49.98it/s +====================Eval==================== +epoch=101 global_step=39882 loss=0.93501 test_loss_avg=0.75015 acc=0.69531 test_acc_avg=0.78274 test_acc_top5_avg=0.97582 time=109.52it/s +epoch=101 global_step=39882 loss=0.11592 test_loss_avg=0.81936 acc=0.96094 test_acc_avg=0.76254 test_acc_top5_avg=0.96116 time=219.04it/s +epoch=101 global_step=39882 loss=0.20696 test_loss_avg=0.75284 acc=0.93750 test_acc_avg=0.78194 test_acc_top5_avg=0.96479 time=501.95it/s +curr_acc 0.7819 +BEST_ACC 0.8636 +curr_acc_top5 0.9648 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=4.20375 loss_avg=3.56024 acc=0.59375 acc_top1_avg=0.66406 acc_top5_avg=0.87109 lr=0.00010 gn=35.75769 time=52.94it/s +epoch=102 global_step=39950 loss=2.67837 loss_avg=3.46882 acc=0.75781 acc_top1_avg=0.67590 acc_top5_avg=0.87006 lr=0.00010 gn=36.24083 time=51.41it/s +epoch=102 global_step=40000 loss=3.69731 loss_avg=3.50265 acc=0.65625 acc_top1_avg=0.67135 acc_top5_avg=0.86752 lr=0.00010 gn=37.36857 time=58.04it/s +epoch=102 global_step=40050 loss=3.46012 loss_avg=3.49448 acc=0.66406 acc_top1_avg=0.67313 acc_top5_avg=0.87137 lr=0.00010 gn=30.79583 time=58.12it/s +epoch=102 global_step=40100 loss=3.58778 loss_avg=3.48073 acc=0.65625 acc_top1_avg=0.67503 acc_top5_avg=0.87059 lr=0.00010 gn=32.10088 time=53.70it/s +epoch=102 global_step=40150 loss=3.30233 loss_avg=3.50457 acc=0.68750 acc_top1_avg=0.67167 acc_top5_avg=0.86891 lr=0.00010 gn=32.02493 time=58.52it/s +epoch=102 global_step=40200 loss=3.22288 loss_avg=3.50890 acc=0.68750 acc_top1_avg=0.67124 acc_top5_avg=0.86952 lr=0.00010 gn=28.92225 time=49.33it/s +epoch=102 global_step=40250 loss=3.48116 loss_avg=3.51687 acc=0.67188 acc_top1_avg=0.67075 acc_top5_avg=0.86882 lr=0.00010 gn=31.84026 time=55.16it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.77380 test_loss_avg=0.89555 acc=0.77344 test_acc_avg=0.74349 test_acc_top5_avg=0.95926 time=223.08it/s +epoch=102 global_step=40273 loss=0.17514 test_loss_avg=0.77418 acc=0.93750 test_acc_avg=0.77720 test_acc_top5_avg=0.96361 time=805.98it/s +curr_acc 0.7772 +BEST_ACC 0.8636 +curr_acc_top5 0.9636 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=3.94456 loss_avg=3.34026 acc=0.64062 acc_top1_avg=0.69271 acc_top5_avg=0.88021 lr=0.00010 gn=37.92301 time=54.46it/s +epoch=103 global_step=40350 loss=3.29304 loss_avg=3.46637 acc=0.70312 acc_top1_avg=0.67482 acc_top5_avg=0.87013 lr=0.00010 gn=35.21970 time=56.92it/s +epoch=103 global_step=40400 loss=3.52742 loss_avg=3.52956 acc=0.67188 acc_top1_avg=0.66905 acc_top5_avg=0.86934 lr=0.00010 gn=37.95280 time=56.35it/s +epoch=103 global_step=40450 loss=3.95156 loss_avg=3.54495 acc=0.60156 acc_top1_avg=0.66728 acc_top5_avg=0.86864 lr=0.00010 gn=28.94875 time=62.02it/s +epoch=103 global_step=40500 loss=4.07662 loss_avg=3.54268 acc=0.61719 acc_top1_avg=0.66830 acc_top5_avg=0.86843 lr=0.00010 gn=31.87556 time=42.81it/s +epoch=103 global_step=40550 loss=3.47170 loss_avg=3.53122 acc=0.68750 acc_top1_avg=0.66922 acc_top5_avg=0.86905 lr=0.00010 gn=39.32309 time=53.57it/s +epoch=103 global_step=40600 loss=2.83811 loss_avg=3.51276 acc=0.74219 acc_top1_avg=0.67149 acc_top5_avg=0.86931 lr=0.00010 gn=33.37504 time=54.23it/s +epoch=103 global_step=40650 loss=3.70970 loss_avg=3.52205 acc=0.64062 acc_top1_avg=0.67049 acc_top5_avg=0.86965 lr=0.00010 gn=33.69876 time=57.71it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.20285 test_loss_avg=0.82513 acc=0.93750 test_acc_avg=0.76442 test_acc_top5_avg=0.97837 time=239.18it/s +epoch=103 global_step=40664 loss=0.14442 test_loss_avg=0.89404 acc=0.92969 test_acc_avg=0.73859 test_acc_top5_avg=0.95883 time=208.64it/s +epoch=103 global_step=40664 loss=0.18892 test_loss_avg=0.74609 acc=0.93750 test_acc_avg=0.78234 test_acc_top5_avg=0.96657 time=535.47it/s +curr_acc 0.7823 +BEST_ACC 0.8636 +curr_acc_top5 0.9666 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=3.61729 loss_avg=3.45555 acc=0.66406 acc_top1_avg=0.67730 acc_top5_avg=0.87174 lr=0.00010 gn=33.84797 time=55.08it/s +epoch=104 global_step=40750 loss=3.96090 loss_avg=3.53914 acc=0.60938 acc_top1_avg=0.66742 acc_top5_avg=0.86982 lr=0.00010 gn=31.28448 time=59.39it/s +epoch=104 global_step=40800 loss=3.08237 loss_avg=3.54526 acc=0.71875 acc_top1_avg=0.66751 acc_top5_avg=0.86966 lr=0.00010 gn=38.72753 time=60.01it/s +epoch=104 global_step=40850 loss=3.32077 loss_avg=3.54803 acc=0.70312 acc_top1_avg=0.66805 acc_top5_avg=0.86874 lr=0.00010 gn=36.46918 time=58.63it/s +epoch=104 global_step=40900 loss=3.25898 loss_avg=3.53629 acc=0.71875 acc_top1_avg=0.66913 acc_top5_avg=0.86841 lr=0.00010 gn=38.16866 time=60.74it/s +epoch=104 global_step=40950 loss=3.68601 loss_avg=3.53809 acc=0.66406 acc_top1_avg=0.66849 acc_top5_avg=0.86842 lr=0.00010 gn=41.05649 time=52.23it/s +epoch=104 global_step=41000 loss=3.26481 loss_avg=3.52388 acc=0.70312 acc_top1_avg=0.67008 acc_top5_avg=0.86900 lr=0.00010 gn=34.58751 time=56.96it/s +epoch=104 global_step=41050 loss=3.20772 loss_avg=3.51827 acc=0.70312 acc_top1_avg=0.67068 acc_top5_avg=0.86911 lr=0.00010 gn=35.38481 time=63.23it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.39727 test_loss_avg=0.99936 acc=0.87500 test_acc_avg=0.71369 test_acc_top5_avg=0.95427 time=223.64it/s +epoch=104 global_step=41055 loss=0.14303 test_loss_avg=0.77714 acc=0.93750 test_acc_avg=0.77453 test_acc_top5_avg=0.96440 time=837.52it/s +curr_acc 0.7745 +BEST_ACC 0.8636 +curr_acc_top5 0.9644 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=3.70899 loss_avg=3.60062 acc=0.64844 acc_top1_avg=0.66285 acc_top5_avg=0.86649 lr=0.00010 gn=26.37231 time=55.11it/s +epoch=105 global_step=41150 loss=4.25021 loss_avg=3.54685 acc=0.58594 acc_top1_avg=0.66850 acc_top5_avg=0.87237 lr=0.00010 gn=30.24337 time=59.96it/s +epoch=105 global_step=41200 loss=3.26365 loss_avg=3.53763 acc=0.70312 acc_top1_avg=0.66902 acc_top5_avg=0.87247 lr=0.00010 gn=39.25138 time=64.14it/s +epoch=105 global_step=41250 loss=3.33924 loss_avg=3.52336 acc=0.71094 acc_top1_avg=0.67095 acc_top5_avg=0.87312 lr=0.00010 gn=41.82192 time=50.78it/s +epoch=105 global_step=41300 loss=3.57923 loss_avg=3.50958 acc=0.67188 acc_top1_avg=0.67245 acc_top5_avg=0.87395 lr=0.00010 gn=38.52766 time=58.67it/s +epoch=105 global_step=41350 loss=3.44254 loss_avg=3.51166 acc=0.66406 acc_top1_avg=0.67217 acc_top5_avg=0.87288 lr=0.00010 gn=38.20453 time=54.57it/s +epoch=105 global_step=41400 loss=3.96388 loss_avg=3.51002 acc=0.63281 acc_top1_avg=0.67215 acc_top5_avg=0.87271 lr=0.00010 gn=32.26539 time=60.32it/s +====================Eval==================== +epoch=105 global_step=41446 loss=0.85690 test_loss_avg=1.15033 acc=0.74219 test_acc_avg=0.67812 test_acc_top5_avg=0.97031 time=238.45it/s +epoch=105 global_step=41446 loss=1.30589 test_loss_avg=1.01691 acc=0.62500 test_acc_avg=0.70696 test_acc_top5_avg=0.94886 time=234.50it/s +epoch=105 global_step=41446 loss=0.12799 test_loss_avg=0.76778 acc=0.93750 test_acc_avg=0.77838 test_acc_top5_avg=0.96311 time=500.04it/s +curr_acc 0.7784 +BEST_ACC 0.8636 +curr_acc_top5 0.9631 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=3.60462 loss_avg=3.46381 acc=0.65625 acc_top1_avg=0.67383 acc_top5_avg=0.86719 lr=0.00010 gn=32.33038 time=52.39it/s +epoch=106 global_step=41500 loss=3.23719 loss_avg=3.44672 acc=0.70312 acc_top1_avg=0.68027 acc_top5_avg=0.87297 lr=0.00010 gn=35.45186 time=54.47it/s +epoch=106 global_step=41550 loss=3.25488 loss_avg=3.44151 acc=0.69531 acc_top1_avg=0.68014 acc_top5_avg=0.87252 lr=0.00010 gn=28.55500 time=62.21it/s +epoch=106 global_step=41600 loss=3.63504 loss_avg=3.46313 acc=0.67969 acc_top1_avg=0.67766 acc_top5_avg=0.87241 lr=0.00010 gn=40.63398 time=63.97it/s +epoch=106 global_step=41650 loss=3.54015 loss_avg=3.47508 acc=0.66406 acc_top1_avg=0.67678 acc_top5_avg=0.87113 lr=0.00010 gn=29.73454 time=52.99it/s +epoch=106 global_step=41700 loss=3.79597 loss_avg=3.49844 acc=0.62500 acc_top1_avg=0.67412 acc_top5_avg=0.86959 lr=0.00010 gn=40.75800 time=54.02it/s +epoch=106 global_step=41750 loss=3.58852 loss_avg=3.49611 acc=0.65625 acc_top1_avg=0.67409 acc_top5_avg=0.87048 lr=0.00010 gn=39.23838 time=60.88it/s +epoch=106 global_step=41800 loss=3.04517 loss_avg=3.50488 acc=0.72656 acc_top1_avg=0.67302 acc_top5_avg=0.87006 lr=0.00010 gn=30.43335 time=51.62it/s +====================Eval==================== +epoch=106 global_step=41837 loss=1.94751 test_loss_avg=0.91495 acc=0.39062 test_acc_avg=0.73498 test_acc_top5_avg=0.96184 time=71.19it/s +epoch=106 global_step=41837 loss=0.16612 test_loss_avg=0.77232 acc=0.96094 test_acc_avg=0.77292 test_acc_top5_avg=0.96556 time=242.95it/s +epoch=106 global_step=41837 loss=0.22362 test_loss_avg=0.74916 acc=0.93750 test_acc_avg=0.77977 test_acc_top5_avg=0.96687 time=535.12it/s +curr_acc 0.7798 +BEST_ACC 0.8636 +curr_acc_top5 0.9669 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=3.66128 loss_avg=3.45933 acc=0.64844 acc_top1_avg=0.67488 acc_top5_avg=0.86358 lr=0.00010 gn=30.50004 time=52.47it/s +epoch=107 global_step=41900 loss=3.38516 loss_avg=3.43578 acc=0.67969 acc_top1_avg=0.67721 acc_top5_avg=0.87438 lr=0.00010 gn=30.06863 time=53.81it/s +epoch=107 global_step=41950 loss=4.14565 loss_avg=3.45492 acc=0.60156 acc_top1_avg=0.67595 acc_top5_avg=0.87334 lr=0.00010 gn=39.39448 time=52.24it/s +epoch=107 global_step=42000 loss=4.10638 loss_avg=3.46156 acc=0.59375 acc_top1_avg=0.67609 acc_top5_avg=0.87256 lr=0.00010 gn=26.67338 time=51.27it/s +epoch=107 global_step=42050 loss=3.62336 loss_avg=3.47840 acc=0.65625 acc_top1_avg=0.67455 acc_top5_avg=0.87126 lr=0.00010 gn=33.22845 time=49.15it/s +epoch=107 global_step=42100 loss=3.92474 loss_avg=3.48283 acc=0.60938 acc_top1_avg=0.67452 acc_top5_avg=0.87099 lr=0.00010 gn=35.92334 time=50.73it/s +epoch=107 global_step=42150 loss=4.01768 loss_avg=3.49676 acc=0.62500 acc_top1_avg=0.67280 acc_top5_avg=0.86981 lr=0.00010 gn=32.30456 time=60.27it/s +epoch=107 global_step=42200 loss=3.19269 loss_avg=3.49331 acc=0.71094 acc_top1_avg=0.67325 acc_top5_avg=0.86990 lr=0.00010 gn=30.75269 time=53.23it/s +====================Eval==================== +epoch=107 global_step=42228 loss=0.86858 test_loss_avg=0.88916 acc=0.75000 test_acc_avg=0.74252 test_acc_top5_avg=0.95928 time=236.14it/s +epoch=107 global_step=42228 loss=0.23058 test_loss_avg=0.76801 acc=0.93750 test_acc_avg=0.77591 test_acc_top5_avg=0.96588 time=512.81it/s +curr_acc 0.7759 +BEST_ACC 0.8636 +curr_acc_top5 0.9659 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=3.92328 loss_avg=3.57938 acc=0.62500 acc_top1_avg=0.66726 acc_top5_avg=0.86612 lr=0.00010 gn=36.90598 time=56.59it/s +epoch=108 global_step=42300 loss=3.43821 loss_avg=3.50208 acc=0.68750 acc_top1_avg=0.67263 acc_top5_avg=0.87012 lr=0.00010 gn=31.85184 time=55.73it/s +epoch=108 global_step=42350 loss=3.50161 loss_avg=3.46551 acc=0.67969 acc_top1_avg=0.67636 acc_top5_avg=0.86968 lr=0.00010 gn=34.16122 time=49.55it/s +epoch=108 global_step=42400 loss=3.92816 loss_avg=3.48903 acc=0.63281 acc_top1_avg=0.67415 acc_top5_avg=0.87073 lr=0.00010 gn=37.82829 time=59.66it/s +epoch=108 global_step=42450 loss=4.05901 loss_avg=3.49755 acc=0.59375 acc_top1_avg=0.67325 acc_top5_avg=0.87007 lr=0.00010 gn=33.91604 time=52.03it/s +epoch=108 global_step=42500 loss=3.73816 loss_avg=3.49031 acc=0.65625 acc_top1_avg=0.67380 acc_top5_avg=0.87058 lr=0.00010 gn=32.70776 time=55.32it/s +epoch=108 global_step=42550 loss=4.17652 loss_avg=3.48677 acc=0.57812 acc_top1_avg=0.67425 acc_top5_avg=0.86937 lr=0.00010 gn=38.67514 time=55.69it/s +epoch=108 global_step=42600 loss=4.44453 loss_avg=3.48246 acc=0.57812 acc_top1_avg=0.67490 acc_top5_avg=0.86899 lr=0.00010 gn=34.75246 time=54.53it/s +====================Eval==================== +epoch=108 global_step=42619 loss=1.03785 test_loss_avg=0.75450 acc=0.70312 test_acc_avg=0.78776 test_acc_top5_avg=0.97700 time=243.95it/s +epoch=108 global_step=42619 loss=0.09852 test_loss_avg=0.86368 acc=0.97656 test_acc_avg=0.74851 test_acc_top5_avg=0.95990 time=230.82it/s +epoch=108 global_step=42619 loss=0.18549 test_loss_avg=0.76895 acc=0.93750 test_acc_avg=0.77601 test_acc_top5_avg=0.96499 time=496.66it/s +curr_acc 0.7760 +BEST_ACC 0.8636 +curr_acc_top5 0.9650 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=3.64722 loss_avg=3.48970 acc=0.66406 acc_top1_avg=0.67566 acc_top5_avg=0.87349 lr=0.00010 gn=36.45264 time=53.71it/s +epoch=109 global_step=42700 loss=3.70317 loss_avg=3.49850 acc=0.64844 acc_top1_avg=0.67342 acc_top5_avg=0.86989 lr=0.00010 gn=29.87578 time=53.18it/s +epoch=109 global_step=42750 loss=3.47093 loss_avg=3.50571 acc=0.67188 acc_top1_avg=0.67199 acc_top5_avg=0.86951 lr=0.00010 gn=26.64168 time=50.22it/s +epoch=109 global_step=42800 loss=3.33053 loss_avg=3.51722 acc=0.67188 acc_top1_avg=0.67023 acc_top5_avg=0.86835 lr=0.00010 gn=26.24812 time=63.99it/s +epoch=109 global_step=42850 loss=3.60106 loss_avg=3.51094 acc=0.65625 acc_top1_avg=0.67150 acc_top5_avg=0.86739 lr=0.00010 gn=34.85808 time=55.49it/s +epoch=109 global_step=42900 loss=3.62706 loss_avg=3.51423 acc=0.67188 acc_top1_avg=0.67149 acc_top5_avg=0.86747 lr=0.00010 gn=40.78396 time=60.32it/s +epoch=109 global_step=42950 loss=3.51396 loss_avg=3.49814 acc=0.66406 acc_top1_avg=0.67336 acc_top5_avg=0.86936 lr=0.00010 gn=36.45500 time=59.42it/s +epoch=109 global_step=43000 loss=3.38129 loss_avg=3.49085 acc=0.68750 acc_top1_avg=0.67372 acc_top5_avg=0.86998 lr=0.00010 gn=29.31807 time=56.53it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.25693 test_loss_avg=0.92624 acc=0.93750 test_acc_avg=0.73057 test_acc_top5_avg=0.95693 time=239.95it/s +epoch=109 global_step=43010 loss=0.18391 test_loss_avg=0.77749 acc=0.93750 test_acc_avg=0.77097 test_acc_top5_avg=0.96400 time=779.61it/s +curr_acc 0.7710 +BEST_ACC 0.8636 +curr_acc_top5 0.9640 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=3.25482 loss_avg=3.42040 acc=0.70312 acc_top1_avg=0.68281 acc_top5_avg=0.87051 lr=0.00010 gn=34.37670 time=55.07it/s +epoch=110 global_step=43100 loss=3.58787 loss_avg=3.45496 acc=0.66406 acc_top1_avg=0.67830 acc_top5_avg=0.87057 lr=0.00010 gn=36.07853 time=60.43it/s +epoch=110 global_step=43150 loss=3.49858 loss_avg=3.47052 acc=0.67188 acc_top1_avg=0.67645 acc_top5_avg=0.86942 lr=0.00010 gn=41.41761 time=62.53it/s +epoch=110 global_step=43200 loss=3.19266 loss_avg=3.47525 acc=0.69531 acc_top1_avg=0.67648 acc_top5_avg=0.86990 lr=0.00010 gn=33.81076 time=62.47it/s +epoch=110 global_step=43250 loss=3.61365 loss_avg=3.48940 acc=0.65625 acc_top1_avg=0.67487 acc_top5_avg=0.86872 lr=0.00010 gn=33.30719 time=57.86it/s +epoch=110 global_step=43300 loss=3.32577 loss_avg=3.48468 acc=0.69531 acc_top1_avg=0.67535 acc_top5_avg=0.86967 lr=0.00010 gn=33.62951 time=57.77it/s +epoch=110 global_step=43350 loss=3.01564 loss_avg=3.47687 acc=0.72656 acc_top1_avg=0.67626 acc_top5_avg=0.86994 lr=0.00010 gn=35.13850 time=58.94it/s +epoch=110 global_step=43400 loss=3.31838 loss_avg=3.48250 acc=0.69531 acc_top1_avg=0.67548 acc_top5_avg=0.86961 lr=0.00010 gn=39.17042 time=53.90it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.17095 test_loss_avg=1.01270 acc=0.93750 test_acc_avg=0.71328 test_acc_top5_avg=0.96797 time=241.72it/s +epoch=110 global_step=43401 loss=0.31833 test_loss_avg=0.95547 acc=0.93750 test_acc_avg=0.72174 test_acc_top5_avg=0.95469 time=223.66it/s +epoch=110 global_step=43401 loss=0.18856 test_loss_avg=0.76944 acc=0.93750 test_acc_avg=0.77611 test_acc_top5_avg=0.96479 time=690.99it/s +curr_acc 0.7761 +BEST_ACC 0.8636 +curr_acc_top5 0.9648 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=3.54127 loss_avg=3.43218 acc=0.66406 acc_top1_avg=0.68001 acc_top5_avg=0.86894 lr=0.00010 gn=33.20441 time=56.95it/s +epoch=111 global_step=43500 loss=3.54347 loss_avg=3.47914 acc=0.66406 acc_top1_avg=0.67558 acc_top5_avg=0.86640 lr=0.00010 gn=37.26106 time=57.88it/s +epoch=111 global_step=43550 loss=3.62998 loss_avg=3.45334 acc=0.65625 acc_top1_avg=0.67838 acc_top5_avg=0.87086 lr=0.00010 gn=35.66031 time=53.81it/s +epoch=111 global_step=43600 loss=3.16819 loss_avg=3.45741 acc=0.70312 acc_top1_avg=0.67820 acc_top5_avg=0.87092 lr=0.00010 gn=29.75210 time=61.53it/s +epoch=111 global_step=43650 loss=3.46838 loss_avg=3.46315 acc=0.67188 acc_top1_avg=0.67755 acc_top5_avg=0.87014 lr=0.00010 gn=35.95467 time=59.28it/s +epoch=111 global_step=43700 loss=3.75844 loss_avg=3.46755 acc=0.64062 acc_top1_avg=0.67676 acc_top5_avg=0.87098 lr=0.00010 gn=33.38600 time=61.91it/s +epoch=111 global_step=43750 loss=3.84095 loss_avg=3.46453 acc=0.64062 acc_top1_avg=0.67716 acc_top5_avg=0.87126 lr=0.00010 gn=35.09736 time=62.18it/s +====================Eval==================== +epoch=111 global_step=43792 loss=1.58361 test_loss_avg=1.02380 acc=0.55469 test_acc_avg=0.70514 test_acc_top5_avg=0.95060 time=251.82it/s +epoch=111 global_step=43792 loss=0.16869 test_loss_avg=0.80764 acc=0.93750 test_acc_avg=0.76879 test_acc_top5_avg=0.96034 time=498.02it/s +curr_acc 0.7688 +BEST_ACC 0.8636 +curr_acc_top5 0.9603 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=3.81003 loss_avg=3.55269 acc=0.64844 acc_top1_avg=0.66797 acc_top5_avg=0.88086 lr=0.00010 gn=41.02001 time=57.17it/s +epoch=112 global_step=43850 loss=3.49730 loss_avg=3.43242 acc=0.67969 acc_top1_avg=0.68036 acc_top5_avg=0.87271 lr=0.00010 gn=36.83694 time=56.26it/s +epoch=112 global_step=43900 loss=3.32143 loss_avg=3.41558 acc=0.69531 acc_top1_avg=0.68128 acc_top5_avg=0.87247 lr=0.00010 gn=35.28661 time=55.55it/s +epoch=112 global_step=43950 loss=3.57895 loss_avg=3.45090 acc=0.66406 acc_top1_avg=0.67806 acc_top5_avg=0.87085 lr=0.00010 gn=35.80272 time=53.65it/s +epoch=112 global_step=44000 loss=3.43870 loss_avg=3.44589 acc=0.68750 acc_top1_avg=0.67871 acc_top5_avg=0.87053 lr=0.00010 gn=41.21872 time=49.83it/s +epoch=112 global_step=44050 loss=2.86240 loss_avg=3.46710 acc=0.75781 acc_top1_avg=0.67739 acc_top5_avg=0.87006 lr=0.00010 gn=40.61832 time=56.02it/s +epoch=112 global_step=44100 loss=3.40747 loss_avg=3.46834 acc=0.67188 acc_top1_avg=0.67733 acc_top5_avg=0.87043 lr=0.00010 gn=39.15979 time=59.59it/s +epoch=112 global_step=44150 loss=3.46285 loss_avg=3.46685 acc=0.68750 acc_top1_avg=0.67785 acc_top5_avg=0.87068 lr=0.00010 gn=37.96319 time=55.61it/s +====================Eval==================== +epoch=112 global_step=44183 loss=1.22819 test_loss_avg=1.28887 acc=0.64062 test_acc_avg=0.63281 test_acc_top5_avg=0.96094 time=236.02it/s +epoch=112 global_step=44183 loss=2.02206 test_loss_avg=0.99705 acc=0.45312 test_acc_avg=0.71229 test_acc_top5_avg=0.95027 time=227.88it/s +epoch=112 global_step=44183 loss=0.13410 test_loss_avg=0.78728 acc=0.93750 test_acc_avg=0.77235 test_acc_top5_avg=0.96282 time=488.22it/s +curr_acc 0.7723 +BEST_ACC 0.8636 +curr_acc_top5 0.9628 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=4.78409 loss_avg=3.71049 acc=0.54688 acc_top1_avg=0.65441 acc_top5_avg=0.86351 lr=0.00010 gn=36.75296 time=55.26it/s +epoch=113 global_step=44250 loss=3.35964 loss_avg=3.57387 acc=0.68750 acc_top1_avg=0.66569 acc_top5_avg=0.86241 lr=0.00010 gn=32.48735 time=54.19it/s +epoch=113 global_step=44300 loss=3.77462 loss_avg=3.48848 acc=0.65625 acc_top1_avg=0.67481 acc_top5_avg=0.86986 lr=0.00010 gn=38.47874 time=56.50it/s +epoch=113 global_step=44350 loss=3.96929 loss_avg=3.48176 acc=0.64062 acc_top1_avg=0.67580 acc_top5_avg=0.86934 lr=0.00010 gn=40.87575 time=56.38it/s +epoch=113 global_step=44400 loss=3.54734 loss_avg=3.47743 acc=0.67188 acc_top1_avg=0.67630 acc_top5_avg=0.86949 lr=0.00010 gn=39.95995 time=43.39it/s +epoch=113 global_step=44450 loss=3.56759 loss_avg=3.48538 acc=0.67969 acc_top1_avg=0.67506 acc_top5_avg=0.86903 lr=0.00010 gn=33.86473 time=58.43it/s +epoch=113 global_step=44500 loss=3.77905 loss_avg=3.47635 acc=0.63281 acc_top1_avg=0.67582 acc_top5_avg=0.87000 lr=0.00010 gn=24.95608 time=60.27it/s +epoch=113 global_step=44550 loss=3.55347 loss_avg=3.47694 acc=0.67188 acc_top1_avg=0.67590 acc_top5_avg=0.87013 lr=0.00010 gn=42.77244 time=54.23it/s +====================Eval==================== +epoch=113 global_step=44574 loss=0.96204 test_loss_avg=0.84598 acc=0.71875 test_acc_avg=0.75917 test_acc_top5_avg=0.97181 time=228.19it/s +epoch=113 global_step=44574 loss=0.19221 test_loss_avg=0.82860 acc=0.93750 test_acc_avg=0.75760 test_acc_top5_avg=0.96211 time=242.53it/s +epoch=113 global_step=44574 loss=0.15110 test_loss_avg=0.77631 acc=0.93750 test_acc_avg=0.77265 test_acc_top5_avg=0.96489 time=833.86it/s +curr_acc 0.7726 +BEST_ACC 0.8636 +curr_acc_top5 0.9649 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=2.82863 loss_avg=3.42681 acc=0.75781 acc_top1_avg=0.68630 acc_top5_avg=0.86899 lr=0.00010 gn=33.13347 time=48.66it/s +epoch=114 global_step=44650 loss=3.60047 loss_avg=3.47070 acc=0.65625 acc_top1_avg=0.67630 acc_top5_avg=0.87048 lr=0.00010 gn=38.95214 time=55.91it/s +epoch=114 global_step=44700 loss=3.13714 loss_avg=3.44204 acc=0.73438 acc_top1_avg=0.67863 acc_top5_avg=0.87264 lr=0.00010 gn=42.42185 time=52.20it/s +epoch=114 global_step=44750 loss=3.73624 loss_avg=3.45551 acc=0.64062 acc_top1_avg=0.67747 acc_top5_avg=0.87127 lr=0.00010 gn=35.67499 time=58.13it/s +epoch=114 global_step=44800 loss=2.62968 loss_avg=3.44827 acc=0.78125 acc_top1_avg=0.67869 acc_top5_avg=0.87137 lr=0.00010 gn=40.34449 time=52.51it/s +epoch=114 global_step=44850 loss=2.92591 loss_avg=3.45866 acc=0.72656 acc_top1_avg=0.67776 acc_top5_avg=0.87177 lr=0.00010 gn=33.72995 time=53.46it/s +epoch=114 global_step=44900 loss=3.34811 loss_avg=3.45818 acc=0.71094 acc_top1_avg=0.67784 acc_top5_avg=0.87186 lr=0.00010 gn=37.93554 time=53.50it/s +epoch=114 global_step=44950 loss=4.02079 loss_avg=3.46025 acc=0.61719 acc_top1_avg=0.67757 acc_top5_avg=0.87188 lr=0.00010 gn=33.94958 time=51.75it/s +====================Eval==================== +epoch=114 global_step=44965 loss=0.80198 test_loss_avg=0.92254 acc=0.76562 test_acc_avg=0.74006 test_acc_top5_avg=0.95472 time=240.86it/s +epoch=114 global_step=44965 loss=0.14351 test_loss_avg=0.78978 acc=0.93750 test_acc_avg=0.77373 test_acc_top5_avg=0.96114 time=543.44it/s +curr_acc 0.7737 +BEST_ACC 0.8636 +curr_acc_top5 0.9611 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=3.31631 loss_avg=3.44206 acc=0.69531 acc_top1_avg=0.67656 acc_top5_avg=0.86853 lr=0.00010 gn=34.79752 time=60.13it/s +epoch=115 global_step=45050 loss=3.40577 loss_avg=3.45803 acc=0.69531 acc_top1_avg=0.67702 acc_top5_avg=0.87151 lr=0.00010 gn=36.72702 time=52.76it/s +epoch=115 global_step=45100 loss=3.59144 loss_avg=3.48193 acc=0.66406 acc_top1_avg=0.67517 acc_top5_avg=0.87002 lr=0.00010 gn=27.45185 time=45.55it/s +epoch=115 global_step=45150 loss=3.54598 loss_avg=3.45706 acc=0.67188 acc_top1_avg=0.67825 acc_top5_avg=0.87213 lr=0.00010 gn=34.83906 time=58.88it/s +epoch=115 global_step=45200 loss=2.96118 loss_avg=3.46862 acc=0.73438 acc_top1_avg=0.67680 acc_top5_avg=0.87154 lr=0.00010 gn=33.27765 time=51.29it/s +epoch=115 global_step=45250 loss=2.92724 loss_avg=3.46248 acc=0.74219 acc_top1_avg=0.67741 acc_top5_avg=0.87135 lr=0.00010 gn=34.13851 time=52.79it/s +epoch=115 global_step=45300 loss=3.57690 loss_avg=3.44981 acc=0.68750 acc_top1_avg=0.67878 acc_top5_avg=0.87213 lr=0.00010 gn=37.88924 time=54.26it/s +epoch=115 global_step=45350 loss=3.27970 loss_avg=3.45959 acc=0.70312 acc_top1_avg=0.67762 acc_top5_avg=0.87092 lr=0.00010 gn=36.81149 time=58.05it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.26466 test_loss_avg=0.73579 acc=0.93750 test_acc_avg=0.78750 test_acc_top5_avg=0.97917 time=227.40it/s +epoch=115 global_step=45356 loss=0.14797 test_loss_avg=0.89978 acc=0.96094 test_acc_avg=0.73762 test_acc_top5_avg=0.95661 time=239.06it/s +epoch=115 global_step=45356 loss=0.12535 test_loss_avg=0.76820 acc=0.93750 test_acc_avg=0.77591 test_acc_top5_avg=0.96381 time=509.39it/s +curr_acc 0.7759 +BEST_ACC 0.8636 +curr_acc_top5 0.9638 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=3.48933 loss_avg=3.46621 acc=0.67969 acc_top1_avg=0.67702 acc_top5_avg=0.86577 lr=0.00010 gn=37.09544 time=51.22it/s +epoch=116 global_step=45450 loss=2.95584 loss_avg=3.45494 acc=0.71875 acc_top1_avg=0.67836 acc_top5_avg=0.86694 lr=0.00010 gn=26.88875 time=54.79it/s +epoch=116 global_step=45500 loss=3.48286 loss_avg=3.43692 acc=0.67969 acc_top1_avg=0.68094 acc_top5_avg=0.86925 lr=0.00010 gn=41.59529 time=55.31it/s +epoch=116 global_step=45550 loss=3.39835 loss_avg=3.41639 acc=0.70312 acc_top1_avg=0.68335 acc_top5_avg=0.86981 lr=0.00010 gn=41.40962 time=52.89it/s +epoch=116 global_step=45600 loss=3.86511 loss_avg=3.42189 acc=0.64062 acc_top1_avg=0.68267 acc_top5_avg=0.87049 lr=0.00010 gn=35.66611 time=63.59it/s +epoch=116 global_step=45650 loss=3.93891 loss_avg=3.43456 acc=0.61719 acc_top1_avg=0.68118 acc_top5_avg=0.87038 lr=0.00010 gn=40.13876 time=57.84it/s +epoch=116 global_step=45700 loss=3.57258 loss_avg=3.43860 acc=0.65625 acc_top1_avg=0.68051 acc_top5_avg=0.87030 lr=0.00010 gn=26.96302 time=60.10it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.43865 test_loss_avg=0.98685 acc=0.86719 test_acc_avg=0.71810 test_acc_top5_avg=0.95312 time=241.33it/s +epoch=116 global_step=45747 loss=0.14088 test_loss_avg=0.79439 acc=0.93750 test_acc_avg=0.77126 test_acc_top5_avg=0.96094 time=800.75it/s +curr_acc 0.7713 +BEST_ACC 0.8636 +curr_acc_top5 0.9609 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=2.86047 loss_avg=3.25865 acc=0.73438 acc_top1_avg=0.69271 acc_top5_avg=0.89583 lr=0.00010 gn=30.96381 time=37.01it/s +epoch=117 global_step=45800 loss=3.82968 loss_avg=3.42720 acc=0.63281 acc_top1_avg=0.68087 acc_top5_avg=0.87028 lr=0.00010 gn=38.41898 time=52.04it/s +epoch=117 global_step=45850 loss=3.58545 loss_avg=3.41467 acc=0.66406 acc_top1_avg=0.68265 acc_top5_avg=0.87151 lr=0.00010 gn=35.15866 time=53.99it/s +epoch=117 global_step=45900 loss=3.12736 loss_avg=3.43310 acc=0.71875 acc_top1_avg=0.68061 acc_top5_avg=0.87153 lr=0.00010 gn=37.30764 time=61.11it/s +epoch=117 global_step=45950 loss=3.63745 loss_avg=3.43009 acc=0.67188 acc_top1_avg=0.68069 acc_top5_avg=0.87184 lr=0.00010 gn=37.27839 time=57.71it/s +epoch=117 global_step=46000 loss=3.49649 loss_avg=3.44052 acc=0.67969 acc_top1_avg=0.67919 acc_top5_avg=0.87271 lr=0.00010 gn=37.51780 time=63.58it/s +epoch=117 global_step=46050 loss=3.11651 loss_avg=3.43735 acc=0.71094 acc_top1_avg=0.67953 acc_top5_avg=0.87258 lr=0.00010 gn=30.72536 time=59.42it/s +epoch=117 global_step=46100 loss=3.30880 loss_avg=3.45572 acc=0.70312 acc_top1_avg=0.67798 acc_top5_avg=0.87161 lr=0.00010 gn=41.71659 time=61.46it/s +====================Eval==================== +epoch=117 global_step=46138 loss=1.54241 test_loss_avg=1.40717 acc=0.57031 test_acc_avg=0.59598 test_acc_top5_avg=0.93973 time=144.57it/s +epoch=117 global_step=46138 loss=0.36112 test_loss_avg=1.05172 acc=0.88281 test_acc_avg=0.69668 test_acc_top5_avg=0.94668 time=233.32it/s +epoch=117 global_step=46138 loss=0.12204 test_loss_avg=0.81751 acc=0.93750 test_acc_avg=0.76394 test_acc_top5_avg=0.96044 time=850.43it/s +curr_acc 0.7639 +BEST_ACC 0.8636 +curr_acc_top5 0.9604 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=3.88561 loss_avg=3.62295 acc=0.64844 acc_top1_avg=0.66146 acc_top5_avg=0.85547 lr=0.00010 gn=42.50521 time=60.47it/s +epoch=118 global_step=46200 loss=3.47998 loss_avg=3.47446 acc=0.67188 acc_top1_avg=0.67654 acc_top5_avg=0.86933 lr=0.00010 gn=32.72326 time=56.61it/s +epoch=118 global_step=46250 loss=3.23828 loss_avg=3.47052 acc=0.70312 acc_top1_avg=0.67676 acc_top5_avg=0.87221 lr=0.00010 gn=38.47304 time=59.70it/s +epoch=118 global_step=46300 loss=2.77817 loss_avg=3.45916 acc=0.74219 acc_top1_avg=0.67824 acc_top5_avg=0.87119 lr=0.00010 gn=34.42484 time=61.72it/s +epoch=118 global_step=46350 loss=3.02864 loss_avg=3.45882 acc=0.72656 acc_top1_avg=0.67829 acc_top5_avg=0.87198 lr=0.00010 gn=34.14369 time=50.76it/s +epoch=118 global_step=46400 loss=3.76273 loss_avg=3.46128 acc=0.63281 acc_top1_avg=0.67787 acc_top5_avg=0.86996 lr=0.00010 gn=23.10682 time=55.11it/s +epoch=118 global_step=46450 loss=3.37687 loss_avg=3.44916 acc=0.68750 acc_top1_avg=0.67916 acc_top5_avg=0.87112 lr=0.00010 gn=37.40936 time=49.76it/s +epoch=118 global_step=46500 loss=3.65019 loss_avg=3.45290 acc=0.67188 acc_top1_avg=0.67913 acc_top5_avg=0.87109 lr=0.00010 gn=45.24629 time=52.35it/s +====================Eval==================== +epoch=118 global_step=46529 loss=1.81728 test_loss_avg=1.02189 acc=0.50781 test_acc_avg=0.71429 test_acc_top5_avg=0.95089 time=243.09it/s +epoch=118 global_step=46529 loss=0.10044 test_loss_avg=0.81288 acc=0.97656 test_acc_avg=0.76843 test_acc_top5_avg=0.95823 time=246.03it/s +epoch=118 global_step=46529 loss=0.14738 test_loss_avg=0.80446 acc=0.93750 test_acc_avg=0.77057 test_acc_top5_avg=0.95876 time=825.65it/s +curr_acc 0.7706 +BEST_ACC 0.8636 +curr_acc_top5 0.9588 +BEST_ACC_top5 0.9846 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=3.10431 loss_avg=3.29669 acc=0.71094 acc_top1_avg=0.69643 acc_top5_avg=0.88356 lr=0.00010 gn=31.43923 time=50.81it/s +epoch=119 global_step=46600 loss=3.45176 loss_avg=3.35822 acc=0.66406 acc_top1_avg=0.68904 acc_top5_avg=0.87775 lr=0.00010 gn=39.60205 time=61.02it/s +epoch=119 global_step=46650 loss=3.96835 loss_avg=3.37929 acc=0.62500 acc_top1_avg=0.68634 acc_top5_avg=0.87448 lr=0.00010 gn=32.42696 time=55.10it/s +epoch=119 global_step=46700 loss=2.66212 loss_avg=3.37376 acc=0.76562 acc_top1_avg=0.68755 acc_top5_avg=0.87632 lr=0.00010 gn=34.98332 time=56.08it/s +epoch=119 global_step=46750 loss=4.01774 loss_avg=3.40492 acc=0.60938 acc_top1_avg=0.68347 acc_top5_avg=0.87405 lr=0.00010 gn=37.53465 time=56.78it/s +epoch=119 global_step=46800 loss=3.38540 loss_avg=3.40808 acc=0.68750 acc_top1_avg=0.68341 acc_top5_avg=0.87408 lr=0.00010 gn=36.36833 time=61.63it/s +epoch=119 global_step=46850 loss=2.96921 loss_avg=3.42418 acc=0.75000 acc_top1_avg=0.68178 acc_top5_avg=0.87232 lr=0.00010 gn=40.90182 time=49.68it/s +epoch=119 global_step=46900 loss=2.90164 loss_avg=3.43712 acc=0.72656 acc_top1_avg=0.68036 acc_top5_avg=0.87119 lr=0.00010 gn=41.41741 time=64.30it/s +====================Eval==================== +epoch=119 global_step=46920 loss=1.96000 test_loss_avg=0.96646 acc=0.42969 test_acc_avg=0.72066 test_acc_top5_avg=0.95281 time=249.97it/s +epoch=119 global_step=46920 loss=0.14474 test_loss_avg=0.82145 acc=0.93750 test_acc_avg=0.76325 test_acc_top5_avg=0.95936 time=557.16it/s +curr_acc 0.7633 +BEST_ACC 0.8636 +curr_acc_top5 0.9594 +BEST_ACC_top5 0.9846 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_0_6.log b/other_methods/sceloss/sceloss_results/out_0_6.log new file mode 100644 index 0000000..82062a5 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_0_6.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.0__noise_amount__0.6.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.40195 loss_avg=7.92849 acc=0.26562 acc_top1_avg=0.19109 acc_top5_avg=0.57891 lr=0.01000 gn=5.49110 time=62.95it/s +epoch=0 global_step=100 loss=7.17559 loss_avg=7.71583 acc=0.26562 acc_top1_avg=0.21375 acc_top5_avg=0.61531 lr=0.01000 gn=3.43508 time=65.94it/s +epoch=0 global_step=150 loss=8.04950 loss_avg=7.65055 acc=0.17188 acc_top1_avg=0.22146 acc_top5_avg=0.62292 lr=0.01000 gn=3.00095 time=59.36it/s +epoch=0 global_step=200 loss=7.85511 loss_avg=7.63787 acc=0.21094 acc_top1_avg=0.22164 acc_top5_avg=0.62969 lr=0.01000 gn=2.88263 time=63.11it/s +epoch=0 global_step=250 loss=7.92053 loss_avg=7.60973 acc=0.17969 acc_top1_avg=0.22447 acc_top5_avg=0.63547 lr=0.01000 gn=2.49037 time=59.32it/s +epoch=0 global_step=300 loss=6.63910 loss_avg=7.57177 acc=0.32812 acc_top1_avg=0.22833 acc_top5_avg=0.64112 lr=0.01000 gn=2.83353 time=62.15it/s +epoch=0 global_step=350 loss=7.62995 loss_avg=7.55565 acc=0.21094 acc_top1_avg=0.22987 acc_top5_avg=0.64272 lr=0.01000 gn=2.29683 time=65.19it/s +====================Eval==================== +epoch=0 global_step=391 loss=4.70195 test_loss_avg=4.85623 acc=0.00000 test_acc_avg=0.02141 test_acc_top5_avg=0.52297 time=241.97it/s +epoch=0 global_step=391 loss=5.57848 test_loss_avg=4.12940 acc=0.00000 test_acc_avg=0.17227 test_acc_top5_avg=0.61383 time=32.65it/s +curr_acc 0.1723 +BEST_ACC 0.0000 +curr_acc_top5 0.6138 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=7.67438 loss_avg=7.35534 acc=0.20312 acc_top1_avg=0.25434 acc_top5_avg=0.67274 lr=0.01000 gn=2.18131 time=60.33it/s +epoch=1 global_step=450 loss=7.27577 loss_avg=7.40040 acc=0.26562 acc_top1_avg=0.24523 acc_top5_avg=0.66618 lr=0.01000 gn=2.65639 time=64.33it/s +epoch=1 global_step=500 loss=7.06510 loss_avg=7.38899 acc=0.27344 acc_top1_avg=0.24677 acc_top5_avg=0.66750 lr=0.01000 gn=2.20351 time=57.63it/s +epoch=1 global_step=550 loss=7.49457 loss_avg=7.39600 acc=0.24219 acc_top1_avg=0.24597 acc_top5_avg=0.66883 lr=0.01000 gn=3.22966 time=61.23it/s +epoch=1 global_step=600 loss=7.41907 loss_avg=7.38442 acc=0.24219 acc_top1_avg=0.24735 acc_top5_avg=0.67184 lr=0.01000 gn=2.99490 time=61.69it/s +epoch=1 global_step=650 loss=7.43262 loss_avg=7.37167 acc=0.23438 acc_top1_avg=0.24819 acc_top5_avg=0.67359 lr=0.01000 gn=2.28044 time=57.35it/s +epoch=1 global_step=700 loss=7.48705 loss_avg=7.36992 acc=0.24219 acc_top1_avg=0.24869 acc_top5_avg=0.67440 lr=0.01000 gn=2.88583 time=55.55it/s +epoch=1 global_step=750 loss=7.09787 loss_avg=7.37052 acc=0.28125 acc_top1_avg=0.24852 acc_top5_avg=0.67407 lr=0.01000 gn=3.06890 time=58.32it/s +====================Eval==================== +epoch=1 global_step=782 loss=4.34778 test_loss_avg=3.79140 acc=0.00000 test_acc_avg=0.07329 test_acc_top5_avg=0.83036 time=243.78it/s +epoch=1 global_step=782 loss=3.32158 test_loss_avg=3.33391 acc=0.29688 test_acc_avg=0.21897 test_acc_top5_avg=0.68145 time=249.20it/s +epoch=1 global_step=782 loss=4.89038 test_loss_avg=3.46922 acc=0.00000 test_acc_avg=0.19680 test_acc_top5_avg=0.66831 time=840.54it/s +curr_acc 0.1968 +BEST_ACC 0.1723 +curr_acc_top5 0.6683 +BEST_ACC_top5 0.6138 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=7.28976 loss_avg=7.26352 acc=0.25781 acc_top1_avg=0.26345 acc_top5_avg=0.69141 lr=0.01000 gn=2.27251 time=46.39it/s +epoch=2 global_step=850 loss=7.63050 loss_avg=7.31344 acc=0.21094 acc_top1_avg=0.25666 acc_top5_avg=0.67911 lr=0.01000 gn=3.25484 time=63.90it/s +epoch=2 global_step=900 loss=7.28447 loss_avg=7.30194 acc=0.26562 acc_top1_avg=0.25775 acc_top5_avg=0.67942 lr=0.01000 gn=2.74611 time=58.04it/s +epoch=2 global_step=950 loss=7.04222 loss_avg=7.29886 acc=0.29688 acc_top1_avg=0.25739 acc_top5_avg=0.67560 lr=0.01000 gn=2.75469 time=57.17it/s +epoch=2 global_step=1000 loss=7.58835 loss_avg=7.29192 acc=0.23438 acc_top1_avg=0.25706 acc_top5_avg=0.67872 lr=0.01000 gn=2.70138 time=62.77it/s +epoch=2 global_step=1050 loss=7.85269 loss_avg=7.29107 acc=0.18750 acc_top1_avg=0.25618 acc_top5_avg=0.67995 lr=0.01000 gn=2.76188 time=60.39it/s +epoch=2 global_step=1100 loss=7.36046 loss_avg=7.29168 acc=0.25000 acc_top1_avg=0.25639 acc_top5_avg=0.68146 lr=0.01000 gn=2.84406 time=61.68it/s +epoch=2 global_step=1150 loss=7.39531 loss_avg=7.29491 acc=0.24219 acc_top1_avg=0.25567 acc_top5_avg=0.68306 lr=0.01000 gn=2.16102 time=60.12it/s +====================Eval==================== +epoch=2 global_step=1173 loss=2.63067 test_loss_avg=3.09082 acc=0.13281 test_acc_avg=0.16555 test_acc_top5_avg=0.57013 time=234.10it/s +epoch=2 global_step=1173 loss=4.28579 test_loss_avg=2.81505 acc=0.00000 test_acc_avg=0.25682 test_acc_top5_avg=0.68394 time=486.41it/s +curr_acc 0.2568 +BEST_ACC 0.1968 +curr_acc_top5 0.6839 +BEST_ACC_top5 0.6683 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=7.25443 loss_avg=7.18666 acc=0.26562 acc_top1_avg=0.26794 acc_top5_avg=0.69676 lr=0.01000 gn=2.70842 time=55.08it/s +epoch=3 global_step=1250 loss=6.43043 loss_avg=7.20718 acc=0.33594 acc_top1_avg=0.26573 acc_top5_avg=0.69724 lr=0.01000 gn=2.69130 time=61.47it/s +epoch=3 global_step=1300 loss=6.90923 loss_avg=7.23426 acc=0.30469 acc_top1_avg=0.26390 acc_top5_avg=0.69648 lr=0.01000 gn=2.57206 time=57.04it/s +epoch=3 global_step=1350 loss=7.35732 loss_avg=7.23167 acc=0.25781 acc_top1_avg=0.26457 acc_top5_avg=0.69500 lr=0.01000 gn=2.56329 time=59.44it/s +epoch=3 global_step=1400 loss=7.70547 loss_avg=7.22336 acc=0.20312 acc_top1_avg=0.26532 acc_top5_avg=0.69617 lr=0.01000 gn=3.30366 time=52.69it/s +epoch=3 global_step=1450 loss=7.45708 loss_avg=7.21581 acc=0.25000 acc_top1_avg=0.26610 acc_top5_avg=0.69607 lr=0.01000 gn=2.94126 time=60.30it/s +epoch=3 global_step=1500 loss=7.05985 loss_avg=7.22253 acc=0.28125 acc_top1_avg=0.26553 acc_top5_avg=0.69687 lr=0.01000 gn=3.05808 time=57.62it/s +epoch=3 global_step=1550 loss=6.90913 loss_avg=7.21827 acc=0.27344 acc_top1_avg=0.26538 acc_top5_avg=0.69575 lr=0.01000 gn=2.94556 time=61.42it/s +====================Eval==================== +epoch=3 global_step=1564 loss=1.46456 test_loss_avg=3.11187 acc=0.56250 test_acc_avg=0.23017 test_acc_top5_avg=0.96394 time=244.87it/s +epoch=3 global_step=1564 loss=0.39241 test_loss_avg=3.70571 acc=0.89062 test_acc_avg=0.18378 test_acc_top5_avg=0.64819 time=236.29it/s +epoch=3 global_step=1564 loss=5.25837 test_loss_avg=3.52393 acc=0.00000 test_acc_avg=0.23477 test_acc_top5_avg=0.69670 time=492.29it/s +curr_acc 0.2348 +BEST_ACC 0.2568 +curr_acc_top5 0.6967 +BEST_ACC_top5 0.6839 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=6.59713 loss_avg=7.16358 acc=0.32031 acc_top1_avg=0.27105 acc_top5_avg=0.69596 lr=0.01000 gn=3.20355 time=55.27it/s +epoch=4 global_step=1650 loss=7.41578 loss_avg=7.16623 acc=0.24219 acc_top1_avg=0.26989 acc_top5_avg=0.69468 lr=0.01000 gn=2.29382 time=63.44it/s +epoch=4 global_step=1700 loss=6.79011 loss_avg=7.20082 acc=0.29688 acc_top1_avg=0.26729 acc_top5_avg=0.69497 lr=0.01000 gn=2.79211 time=61.59it/s +epoch=4 global_step=1750 loss=7.12546 loss_avg=7.15899 acc=0.27344 acc_top1_avg=0.27180 acc_top5_avg=0.69943 lr=0.01000 gn=3.09070 time=56.00it/s +epoch=4 global_step=1800 loss=7.28933 loss_avg=7.16818 acc=0.26562 acc_top1_avg=0.27033 acc_top5_avg=0.69786 lr=0.01000 gn=3.13801 time=62.46it/s +epoch=4 global_step=1850 loss=7.90424 loss_avg=7.17555 acc=0.17188 acc_top1_avg=0.26893 acc_top5_avg=0.69884 lr=0.01000 gn=2.70508 time=53.98it/s +epoch=4 global_step=1900 loss=6.87538 loss_avg=7.16325 acc=0.31250 acc_top1_avg=0.27035 acc_top5_avg=0.70031 lr=0.01000 gn=4.62583 time=55.34it/s +epoch=4 global_step=1950 loss=6.93978 loss_avg=7.16711 acc=0.29688 acc_top1_avg=0.27016 acc_top5_avg=0.70064 lr=0.01000 gn=3.08386 time=61.48it/s +====================Eval==================== +epoch=4 global_step=1955 loss=5.12753 test_loss_avg=4.02692 acc=0.00000 test_acc_avg=0.15901 test_acc_top5_avg=0.58709 time=190.02it/s +epoch=4 global_step=1955 loss=5.62403 test_loss_avg=3.53388 acc=0.00000 test_acc_avg=0.24792 test_acc_top5_avg=0.67277 time=498.85it/s +curr_acc 0.2479 +BEST_ACC 0.2568 +curr_acc_top5 0.6728 +BEST_ACC_top5 0.6967 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=7.22956 loss_avg=7.14344 acc=0.25781 acc_top1_avg=0.27361 acc_top5_avg=0.69965 lr=0.01000 gn=3.39157 time=53.15it/s +epoch=5 global_step=2050 loss=6.81322 loss_avg=7.16223 acc=0.32031 acc_top1_avg=0.27204 acc_top5_avg=0.70197 lr=0.01000 gn=3.11484 time=49.33it/s +epoch=5 global_step=2100 loss=6.70613 loss_avg=7.13312 acc=0.32031 acc_top1_avg=0.27559 acc_top5_avg=0.69892 lr=0.01000 gn=3.60075 time=58.41it/s +epoch=5 global_step=2150 loss=7.09241 loss_avg=7.13263 acc=0.29688 acc_top1_avg=0.27560 acc_top5_avg=0.69940 lr=0.01000 gn=3.43225 time=57.37it/s +epoch=5 global_step=2200 loss=7.55267 loss_avg=7.13844 acc=0.21875 acc_top1_avg=0.27478 acc_top5_avg=0.70032 lr=0.01000 gn=3.62262 time=64.27it/s +epoch=5 global_step=2250 loss=7.07247 loss_avg=7.13390 acc=0.25781 acc_top1_avg=0.27439 acc_top5_avg=0.70273 lr=0.01000 gn=2.00257 time=63.94it/s +epoch=5 global_step=2300 loss=7.24053 loss_avg=7.13391 acc=0.25000 acc_top1_avg=0.27464 acc_top5_avg=0.70274 lr=0.01000 gn=3.03673 time=51.88it/s +====================Eval==================== +epoch=5 global_step=2346 loss=3.49879 test_loss_avg=3.35058 acc=0.01562 test_acc_avg=0.05312 test_acc_top5_avg=0.98750 time=234.54it/s +epoch=5 global_step=2346 loss=3.27791 test_loss_avg=3.52959 acc=0.22656 test_acc_avg=0.14119 test_acc_top5_avg=0.62884 time=241.79it/s +epoch=5 global_step=2346 loss=4.34194 test_loss_avg=3.04010 acc=0.00000 test_acc_avg=0.26592 test_acc_top5_avg=0.70856 time=485.28it/s +curr_acc 0.2659 +BEST_ACC 0.2568 +curr_acc_top5 0.7086 +BEST_ACC_top5 0.6967 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=6.73569 loss_avg=6.96287 acc=0.31250 acc_top1_avg=0.29688 acc_top5_avg=0.71484 lr=0.01000 gn=3.62647 time=31.97it/s +epoch=6 global_step=2400 loss=7.48553 loss_avg=7.11296 acc=0.23438 acc_top1_avg=0.27836 acc_top5_avg=0.70761 lr=0.01000 gn=3.05285 time=54.67it/s +epoch=6 global_step=2450 loss=6.72851 loss_avg=7.12325 acc=0.32812 acc_top1_avg=0.27607 acc_top5_avg=0.70478 lr=0.01000 gn=3.84699 time=61.31it/s +epoch=6 global_step=2500 loss=6.41663 loss_avg=7.12199 acc=0.36719 acc_top1_avg=0.27668 acc_top5_avg=0.70399 lr=0.01000 gn=3.47476 time=63.40it/s +epoch=6 global_step=2550 loss=7.48117 loss_avg=7.12245 acc=0.21094 acc_top1_avg=0.27608 acc_top5_avg=0.70312 lr=0.01000 gn=2.81673 time=56.22it/s +epoch=6 global_step=2600 loss=7.30216 loss_avg=7.11769 acc=0.22656 acc_top1_avg=0.27633 acc_top5_avg=0.70512 lr=0.01000 gn=3.47198 time=54.46it/s +epoch=6 global_step=2650 loss=7.04538 loss_avg=7.11370 acc=0.28125 acc_top1_avg=0.27670 acc_top5_avg=0.70521 lr=0.01000 gn=2.80297 time=57.10it/s +epoch=6 global_step=2700 loss=7.30264 loss_avg=7.11340 acc=0.24219 acc_top1_avg=0.27664 acc_top5_avg=0.70520 lr=0.01000 gn=3.71258 time=42.05it/s +====================Eval==================== +epoch=6 global_step=2737 loss=4.73588 test_loss_avg=3.14858 acc=0.00000 test_acc_avg=0.22596 test_acc_top5_avg=0.68209 time=236.81it/s +epoch=6 global_step=2737 loss=6.34088 test_loss_avg=3.16129 acc=0.00000 test_acc_avg=0.28032 test_acc_top5_avg=0.70384 time=253.83it/s +epoch=6 global_step=2737 loss=6.84601 test_loss_avg=3.28309 acc=0.00000 test_acc_avg=0.26968 test_acc_top5_avg=0.71193 time=523.24it/s +curr_acc 0.2697 +BEST_ACC 0.2659 +curr_acc_top5 0.7119 +BEST_ACC_top5 0.7086 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=6.99324 loss_avg=6.96634 acc=0.29688 acc_top1_avg=0.29567 acc_top5_avg=0.71334 lr=0.01000 gn=2.88138 time=57.35it/s +epoch=7 global_step=2800 loss=7.36128 loss_avg=7.11683 acc=0.24219 acc_top1_avg=0.27505 acc_top5_avg=0.71156 lr=0.01000 gn=3.42432 time=54.66it/s +epoch=7 global_step=2850 loss=6.65770 loss_avg=7.10423 acc=0.32812 acc_top1_avg=0.27669 acc_top5_avg=0.70914 lr=0.01000 gn=2.91573 time=53.83it/s +epoch=7 global_step=2900 loss=7.25819 loss_avg=7.08888 acc=0.26562 acc_top1_avg=0.27809 acc_top5_avg=0.70945 lr=0.01000 gn=3.18259 time=53.53it/s +epoch=7 global_step=2950 loss=7.28855 loss_avg=7.08813 acc=0.28125 acc_top1_avg=0.27839 acc_top5_avg=0.70819 lr=0.01000 gn=4.18287 time=60.74it/s +epoch=7 global_step=3000 loss=7.05837 loss_avg=7.10864 acc=0.26562 acc_top1_avg=0.27608 acc_top5_avg=0.70752 lr=0.01000 gn=2.90320 time=57.86it/s +epoch=7 global_step=3050 loss=7.59955 loss_avg=7.11756 acc=0.21094 acc_top1_avg=0.27543 acc_top5_avg=0.70634 lr=0.01000 gn=2.93194 time=57.87it/s +epoch=7 global_step=3100 loss=7.11808 loss_avg=7.10740 acc=0.28906 acc_top1_avg=0.27688 acc_top5_avg=0.70663 lr=0.01000 gn=3.53656 time=56.57it/s +====================Eval==================== +epoch=7 global_step=3128 loss=3.43567 test_loss_avg=4.34536 acc=0.14062 test_acc_avg=0.10455 test_acc_top5_avg=0.60605 time=236.07it/s +epoch=7 global_step=3128 loss=5.42254 test_loss_avg=3.78639 acc=0.00000 test_acc_avg=0.22251 test_acc_top5_avg=0.71667 time=507.05it/s +curr_acc 0.2225 +BEST_ACC 0.2697 +curr_acc_top5 0.7167 +BEST_ACC_top5 0.7119 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=7.27921 loss_avg=6.98517 acc=0.26562 acc_top1_avg=0.29261 acc_top5_avg=0.71378 lr=0.01000 gn=3.24975 time=48.55it/s +epoch=8 global_step=3200 loss=7.12991 loss_avg=7.04713 acc=0.28125 acc_top1_avg=0.28440 acc_top5_avg=0.71322 lr=0.01000 gn=3.08551 time=60.26it/s +epoch=8 global_step=3250 loss=6.96127 loss_avg=7.03644 acc=0.28125 acc_top1_avg=0.28528 acc_top5_avg=0.71087 lr=0.01000 gn=2.73851 time=56.27it/s +epoch=8 global_step=3300 loss=7.01645 loss_avg=7.07051 acc=0.30469 acc_top1_avg=0.28116 acc_top5_avg=0.70976 lr=0.01000 gn=3.24718 time=64.78it/s +epoch=8 global_step=3350 loss=7.14033 loss_avg=7.07765 acc=0.27344 acc_top1_avg=0.28086 acc_top5_avg=0.70844 lr=0.01000 gn=3.01958 time=61.08it/s +epoch=8 global_step=3400 loss=7.35175 loss_avg=7.08697 acc=0.25000 acc_top1_avg=0.27941 acc_top5_avg=0.70686 lr=0.01000 gn=3.43072 time=55.11it/s +epoch=8 global_step=3450 loss=6.74799 loss_avg=7.09031 acc=0.33594 acc_top1_avg=0.27899 acc_top5_avg=0.70771 lr=0.01000 gn=3.28207 time=60.44it/s +epoch=8 global_step=3500 loss=6.63647 loss_avg=7.08295 acc=0.30469 acc_top1_avg=0.27909 acc_top5_avg=0.70693 lr=0.01000 gn=2.83702 time=56.72it/s +====================Eval==================== +epoch=8 global_step=3519 loss=4.52363 test_loss_avg=3.65892 acc=0.00000 test_acc_avg=0.22005 test_acc_top5_avg=0.90278 time=148.75it/s +epoch=8 global_step=3519 loss=0.03935 test_loss_avg=3.48247 acc=1.00000 test_acc_avg=0.21576 test_acc_top5_avg=0.73288 time=241.26it/s +epoch=8 global_step=3519 loss=5.53737 test_loss_avg=3.59245 acc=0.00000 test_acc_avg=0.21470 test_acc_top5_avg=0.76533 time=823.87it/s +curr_acc 0.2147 +BEST_ACC 0.2697 +curr_acc_top5 0.7653 +BEST_ACC_top5 0.7167 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=6.74485 loss_avg=7.06618 acc=0.32031 acc_top1_avg=0.28301 acc_top5_avg=0.71522 lr=0.01000 gn=3.44705 time=53.40it/s +epoch=9 global_step=3600 loss=7.15615 loss_avg=7.06298 acc=0.26562 acc_top1_avg=0.28308 acc_top5_avg=0.71508 lr=0.01000 gn=3.14574 time=62.81it/s +epoch=9 global_step=3650 loss=6.68053 loss_avg=7.03836 acc=0.34375 acc_top1_avg=0.28525 acc_top5_avg=0.71022 lr=0.01000 gn=3.79181 time=58.81it/s +epoch=9 global_step=3700 loss=6.80522 loss_avg=7.06893 acc=0.31250 acc_top1_avg=0.28147 acc_top5_avg=0.70805 lr=0.01000 gn=2.73224 time=56.40it/s +epoch=9 global_step=3750 loss=6.80295 loss_avg=7.08200 acc=0.31250 acc_top1_avg=0.27986 acc_top5_avg=0.70918 lr=0.01000 gn=3.48296 time=64.19it/s +epoch=9 global_step=3800 loss=6.77532 loss_avg=7.06780 acc=0.33594 acc_top1_avg=0.28125 acc_top5_avg=0.71041 lr=0.01000 gn=3.56025 time=59.85it/s +epoch=9 global_step=3850 loss=7.08661 loss_avg=7.07564 acc=0.28906 acc_top1_avg=0.28038 acc_top5_avg=0.71073 lr=0.01000 gn=3.00639 time=53.41it/s +epoch=9 global_step=3900 loss=7.14937 loss_avg=7.07563 acc=0.25000 acc_top1_avg=0.28035 acc_top5_avg=0.71055 lr=0.01000 gn=2.96335 time=57.54it/s +====================Eval==================== +epoch=9 global_step=3910 loss=5.50569 test_loss_avg=4.32493 acc=0.00000 test_acc_avg=0.10196 test_acc_top5_avg=0.52764 time=223.02it/s +epoch=9 global_step=3910 loss=5.54204 test_loss_avg=3.53078 acc=0.00000 test_acc_avg=0.24555 test_acc_top5_avg=0.70787 time=692.93it/s +curr_acc 0.2455 +BEST_ACC 0.2697 +curr_acc_top5 0.7079 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=6.59972 loss_avg=6.97450 acc=0.33594 acc_top1_avg=0.29492 acc_top5_avg=0.70840 lr=0.01000 gn=3.18616 time=54.06it/s +epoch=10 global_step=4000 loss=7.52871 loss_avg=7.02042 acc=0.21875 acc_top1_avg=0.28828 acc_top5_avg=0.70911 lr=0.01000 gn=3.65865 time=57.95it/s +epoch=10 global_step=4050 loss=7.28588 loss_avg=7.03099 acc=0.26562 acc_top1_avg=0.28744 acc_top5_avg=0.70748 lr=0.01000 gn=3.62476 time=55.32it/s +epoch=10 global_step=4100 loss=7.54288 loss_avg=7.03417 acc=0.24219 acc_top1_avg=0.28655 acc_top5_avg=0.70674 lr=0.01000 gn=3.22468 time=59.02it/s +epoch=10 global_step=4150 loss=7.63417 loss_avg=7.05147 acc=0.22656 acc_top1_avg=0.28431 acc_top5_avg=0.70788 lr=0.01000 gn=3.52783 time=64.31it/s +epoch=10 global_step=4200 loss=6.71793 loss_avg=7.07350 acc=0.32031 acc_top1_avg=0.28144 acc_top5_avg=0.70671 lr=0.01000 gn=3.60674 time=56.11it/s +epoch=10 global_step=4250 loss=7.41277 loss_avg=7.08014 acc=0.24219 acc_top1_avg=0.28116 acc_top5_avg=0.70595 lr=0.01000 gn=3.34262 time=58.76it/s +epoch=10 global_step=4300 loss=7.39491 loss_avg=7.08316 acc=0.25781 acc_top1_avg=0.28041 acc_top5_avg=0.70607 lr=0.01000 gn=3.98690 time=64.69it/s +====================Eval==================== +epoch=10 global_step=4301 loss=1.08984 test_loss_avg=3.34651 acc=0.64062 test_acc_avg=0.16172 test_acc_top5_avg=0.88203 time=239.85it/s +epoch=10 global_step=4301 loss=0.50818 test_loss_avg=3.36288 acc=0.85156 test_acc_avg=0.18945 test_acc_top5_avg=0.63411 time=229.54it/s +epoch=10 global_step=4301 loss=5.34507 test_loss_avg=3.21006 acc=0.00000 test_acc_avg=0.25237 test_acc_top5_avg=0.69709 time=505.03it/s +curr_acc 0.2524 +BEST_ACC 0.2697 +curr_acc_top5 0.6971 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=6.96217 loss_avg=7.09649 acc=0.30469 acc_top1_avg=0.28077 acc_top5_avg=0.70265 lr=0.01000 gn=3.68974 time=64.67it/s +epoch=11 global_step=4400 loss=6.96397 loss_avg=7.06170 acc=0.28125 acc_top1_avg=0.28267 acc_top5_avg=0.70352 lr=0.01000 gn=2.81639 time=63.12it/s +epoch=11 global_step=4450 loss=7.79225 loss_avg=7.08754 acc=0.19531 acc_top1_avg=0.27999 acc_top5_avg=0.70590 lr=0.01000 gn=3.97101 time=57.68it/s +epoch=11 global_step=4500 loss=7.49371 loss_avg=7.09393 acc=0.21875 acc_top1_avg=0.27866 acc_top5_avg=0.70693 lr=0.01000 gn=2.62128 time=58.52it/s +epoch=11 global_step=4550 loss=7.50926 loss_avg=7.08722 acc=0.23438 acc_top1_avg=0.27927 acc_top5_avg=0.70764 lr=0.01000 gn=3.34703 time=54.17it/s +epoch=11 global_step=4600 loss=6.98361 loss_avg=7.08368 acc=0.29688 acc_top1_avg=0.27979 acc_top5_avg=0.70770 lr=0.01000 gn=2.98953 time=29.65it/s +epoch=11 global_step=4650 loss=6.85018 loss_avg=7.07620 acc=0.28906 acc_top1_avg=0.28078 acc_top5_avg=0.70794 lr=0.01000 gn=3.52929 time=59.10it/s +====================Eval==================== +epoch=11 global_step=4692 loss=4.88296 test_loss_avg=3.60290 acc=0.00000 test_acc_avg=0.18070 test_acc_top5_avg=0.66935 time=240.47it/s +epoch=11 global_step=4692 loss=5.07151 test_loss_avg=3.34062 acc=0.00000 test_acc_avg=0.26919 test_acc_top5_avg=0.69749 time=492.75it/s +curr_acc 0.2692 +BEST_ACC 0.2697 +curr_acc_top5 0.6975 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=7.01855 loss_avg=7.11743 acc=0.27344 acc_top1_avg=0.27637 acc_top5_avg=0.71973 lr=0.01000 gn=3.88333 time=57.25it/s +epoch=12 global_step=4750 loss=6.69836 loss_avg=7.07844 acc=0.30469 acc_top1_avg=0.28152 acc_top5_avg=0.70353 lr=0.01000 gn=3.87061 time=60.42it/s +epoch=12 global_step=4800 loss=7.46647 loss_avg=7.06104 acc=0.23438 acc_top1_avg=0.28277 acc_top5_avg=0.70819 lr=0.01000 gn=3.81538 time=55.76it/s +epoch=12 global_step=4850 loss=7.14635 loss_avg=7.07357 acc=0.27344 acc_top1_avg=0.28209 acc_top5_avg=0.70847 lr=0.01000 gn=3.40315 time=56.68it/s +epoch=12 global_step=4900 loss=7.14029 loss_avg=7.07704 acc=0.28906 acc_top1_avg=0.28148 acc_top5_avg=0.70737 lr=0.01000 gn=2.33381 time=56.78it/s +epoch=12 global_step=4950 loss=7.29400 loss_avg=7.08983 acc=0.26562 acc_top1_avg=0.27977 acc_top5_avg=0.70685 lr=0.01000 gn=2.68274 time=57.56it/s +epoch=12 global_step=5000 loss=7.47452 loss_avg=7.07343 acc=0.25000 acc_top1_avg=0.28163 acc_top5_avg=0.70746 lr=0.01000 gn=3.33493 time=58.12it/s +epoch=12 global_step=5050 loss=6.91273 loss_avg=7.07187 acc=0.29688 acc_top1_avg=0.28171 acc_top5_avg=0.70749 lr=0.01000 gn=3.06049 time=55.13it/s +====================Eval==================== +epoch=12 global_step=5083 loss=5.37155 test_loss_avg=5.48829 acc=0.03125 test_acc_avg=0.03516 test_acc_top5_avg=0.97266 time=242.92it/s +epoch=12 global_step=5083 loss=4.65747 test_loss_avg=4.46918 acc=0.00000 test_acc_avg=0.08233 test_acc_top5_avg=0.61779 time=159.62it/s +epoch=12 global_step=5083 loss=5.20441 test_loss_avg=3.77191 acc=0.00000 test_acc_avg=0.22577 test_acc_top5_avg=0.72449 time=830.06it/s +curr_acc 0.2258 +BEST_ACC 0.2697 +curr_acc_top5 0.7245 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=6.64764 loss_avg=6.94928 acc=0.35938 acc_top1_avg=0.29320 acc_top5_avg=0.72426 lr=0.01000 gn=4.31069 time=55.13it/s +epoch=13 global_step=5150 loss=7.43165 loss_avg=6.99056 acc=0.23438 acc_top1_avg=0.28988 acc_top5_avg=0.71747 lr=0.01000 gn=3.30006 time=56.46it/s +epoch=13 global_step=5200 loss=6.82881 loss_avg=7.02762 acc=0.31250 acc_top1_avg=0.28619 acc_top5_avg=0.71327 lr=0.01000 gn=3.18911 time=56.28it/s +epoch=13 global_step=5250 loss=7.05407 loss_avg=7.02890 acc=0.28125 acc_top1_avg=0.28504 acc_top5_avg=0.71328 lr=0.01000 gn=3.31716 time=56.00it/s +epoch=13 global_step=5300 loss=7.00209 loss_avg=7.02116 acc=0.30469 acc_top1_avg=0.28647 acc_top5_avg=0.71339 lr=0.01000 gn=4.35896 time=56.52it/s +epoch=13 global_step=5350 loss=6.62459 loss_avg=7.05074 acc=0.34375 acc_top1_avg=0.28368 acc_top5_avg=0.71176 lr=0.01000 gn=4.33783 time=64.98it/s +epoch=13 global_step=5400 loss=7.05506 loss_avg=7.04827 acc=0.28906 acc_top1_avg=0.28413 acc_top5_avg=0.71081 lr=0.01000 gn=3.58500 time=53.43it/s +epoch=13 global_step=5450 loss=7.21706 loss_avg=7.05142 acc=0.25781 acc_top1_avg=0.28359 acc_top5_avg=0.70857 lr=0.01000 gn=3.95741 time=61.97it/s +====================Eval==================== +epoch=13 global_step=5474 loss=4.08113 test_loss_avg=2.59264 acc=0.00000 test_acc_avg=0.31216 test_acc_top5_avg=0.85292 time=241.32it/s +epoch=13 global_step=5474 loss=4.99676 test_loss_avg=2.82654 acc=0.00000 test_acc_avg=0.31978 test_acc_top5_avg=0.66856 time=240.29it/s +epoch=13 global_step=5474 loss=4.92677 test_loss_avg=2.99104 acc=0.00000 test_acc_avg=0.29549 test_acc_top5_avg=0.67128 time=497.72it/s +curr_acc 0.2955 +BEST_ACC 0.2697 +curr_acc_top5 0.6713 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=7.15786 loss_avg=7.09575 acc=0.28125 acc_top1_avg=0.28095 acc_top5_avg=0.70162 lr=0.01000 gn=3.64548 time=58.74it/s +epoch=14 global_step=5550 loss=7.18031 loss_avg=7.03523 acc=0.28125 acc_top1_avg=0.28660 acc_top5_avg=0.70929 lr=0.01000 gn=3.88325 time=59.87it/s +epoch=14 global_step=5600 loss=7.62019 loss_avg=7.05256 acc=0.23438 acc_top1_avg=0.28472 acc_top5_avg=0.70641 lr=0.01000 gn=3.45632 time=52.98it/s +epoch=14 global_step=5650 loss=7.27692 loss_avg=7.05664 acc=0.26562 acc_top1_avg=0.28414 acc_top5_avg=0.70854 lr=0.01000 gn=3.54338 time=59.23it/s +epoch=14 global_step=5700 loss=7.71913 loss_avg=7.05774 acc=0.21875 acc_top1_avg=0.28350 acc_top5_avg=0.70921 lr=0.01000 gn=4.10866 time=46.64it/s +epoch=14 global_step=5750 loss=7.25281 loss_avg=7.05844 acc=0.25781 acc_top1_avg=0.28343 acc_top5_avg=0.70947 lr=0.01000 gn=4.19493 time=55.89it/s +epoch=14 global_step=5800 loss=6.56561 loss_avg=7.05738 acc=0.35156 acc_top1_avg=0.28369 acc_top5_avg=0.71015 lr=0.01000 gn=4.93563 time=53.49it/s +epoch=14 global_step=5850 loss=6.59354 loss_avg=7.06199 acc=0.33594 acc_top1_avg=0.28306 acc_top5_avg=0.70952 lr=0.01000 gn=3.90251 time=45.61it/s +====================Eval==================== +epoch=14 global_step=5865 loss=2.39672 test_loss_avg=3.90642 acc=0.22656 test_acc_avg=0.15430 test_acc_top5_avg=0.57706 time=237.14it/s +epoch=14 global_step=5865 loss=5.08712 test_loss_avg=3.34121 acc=0.00000 test_acc_avg=0.27255 test_acc_top5_avg=0.71232 time=505.22it/s +curr_acc 0.2725 +BEST_ACC 0.2955 +curr_acc_top5 0.7123 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=7.08291 loss_avg=7.04725 acc=0.27344 acc_top1_avg=0.28437 acc_top5_avg=0.70759 lr=0.01000 gn=2.98894 time=52.96it/s +epoch=15 global_step=5950 loss=6.58248 loss_avg=7.02928 acc=0.32812 acc_top1_avg=0.28621 acc_top5_avg=0.71048 lr=0.01000 gn=2.92308 time=60.31it/s +epoch=15 global_step=6000 loss=7.61085 loss_avg=7.02392 acc=0.19531 acc_top1_avg=0.28565 acc_top5_avg=0.71128 lr=0.01000 gn=2.99044 time=55.67it/s +epoch=15 global_step=6050 loss=6.49888 loss_avg=7.03388 acc=0.35156 acc_top1_avg=0.28518 acc_top5_avg=0.70950 lr=0.01000 gn=3.94042 time=56.60it/s +epoch=15 global_step=6100 loss=6.76995 loss_avg=7.04355 acc=0.28906 acc_top1_avg=0.28354 acc_top5_avg=0.70805 lr=0.01000 gn=2.54365 time=52.35it/s +epoch=15 global_step=6150 loss=7.21655 loss_avg=7.04490 acc=0.28125 acc_top1_avg=0.28311 acc_top5_avg=0.70869 lr=0.01000 gn=3.08620 time=57.51it/s +epoch=15 global_step=6200 loss=6.77954 loss_avg=7.04607 acc=0.31250 acc_top1_avg=0.28337 acc_top5_avg=0.70914 lr=0.01000 gn=3.02745 time=57.61it/s +epoch=15 global_step=6250 loss=7.02415 loss_avg=7.04580 acc=0.28906 acc_top1_avg=0.28330 acc_top5_avg=0.70933 lr=0.01000 gn=2.45446 time=53.72it/s +====================Eval==================== +epoch=15 global_step=6256 loss=2.42102 test_loss_avg=4.00048 acc=0.37500 test_acc_avg=0.19896 test_acc_top5_avg=0.97552 time=236.07it/s +epoch=15 global_step=6256 loss=0.01616 test_loss_avg=4.20963 acc=1.00000 test_acc_avg=0.15613 test_acc_top5_avg=0.62151 time=235.60it/s +epoch=15 global_step=6256 loss=5.99461 test_loss_avg=4.11249 acc=0.00000 test_acc_avg=0.19571 test_acc_top5_avg=0.68790 time=496.72it/s +curr_acc 0.1957 +BEST_ACC 0.2955 +curr_acc_top5 0.6879 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=7.14659 loss_avg=7.00430 acc=0.28906 acc_top1_avg=0.28853 acc_top5_avg=0.71822 lr=0.01000 gn=4.04161 time=54.95it/s +epoch=16 global_step=6350 loss=6.99287 loss_avg=7.02688 acc=0.28906 acc_top1_avg=0.28615 acc_top5_avg=0.71617 lr=0.01000 gn=4.65933 time=52.93it/s +epoch=16 global_step=6400 loss=7.61240 loss_avg=7.05890 acc=0.23438 acc_top1_avg=0.28185 acc_top5_avg=0.71061 lr=0.01000 gn=3.42222 time=52.43it/s +epoch=16 global_step=6450 loss=6.63350 loss_avg=7.06210 acc=0.32031 acc_top1_avg=0.28202 acc_top5_avg=0.70925 lr=0.01000 gn=3.25696 time=59.20it/s +epoch=16 global_step=6500 loss=7.07619 loss_avg=7.03830 acc=0.26562 acc_top1_avg=0.28512 acc_top5_avg=0.71043 lr=0.01000 gn=3.75896 time=53.48it/s +epoch=16 global_step=6550 loss=6.69014 loss_avg=7.04135 acc=0.32031 acc_top1_avg=0.28449 acc_top5_avg=0.70897 lr=0.01000 gn=2.81305 time=62.70it/s +epoch=16 global_step=6600 loss=7.94298 loss_avg=7.06129 acc=0.17188 acc_top1_avg=0.28216 acc_top5_avg=0.70735 lr=0.01000 gn=2.87265 time=59.94it/s +====================Eval==================== +epoch=16 global_step=6647 loss=5.06639 test_loss_avg=3.91106 acc=0.00000 test_acc_avg=0.16428 test_acc_top5_avg=0.62500 time=238.94it/s +epoch=16 global_step=6647 loss=4.80532 test_loss_avg=3.30303 acc=0.00000 test_acc_avg=0.28125 test_acc_top5_avg=0.70827 time=854.59it/s +curr_acc 0.2812 +BEST_ACC 0.2955 +curr_acc_top5 0.7083 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=6.90780 loss_avg=6.96062 acc=0.31250 acc_top1_avg=0.29948 acc_top5_avg=0.70312 lr=0.01000 gn=4.03420 time=56.60it/s +epoch=17 global_step=6700 loss=7.22654 loss_avg=7.06390 acc=0.27344 acc_top1_avg=0.28213 acc_top5_avg=0.70607 lr=0.01000 gn=3.50384 time=57.43it/s +epoch=17 global_step=6750 loss=6.63415 loss_avg=7.03439 acc=0.33594 acc_top1_avg=0.28595 acc_top5_avg=0.70639 lr=0.01000 gn=3.29928 time=60.07it/s +epoch=17 global_step=6800 loss=7.01724 loss_avg=7.04939 acc=0.26562 acc_top1_avg=0.28406 acc_top5_avg=0.70731 lr=0.01000 gn=3.36097 time=55.70it/s +epoch=17 global_step=6850 loss=6.89899 loss_avg=7.05673 acc=0.32812 acc_top1_avg=0.28267 acc_top5_avg=0.70755 lr=0.01000 gn=4.56790 time=60.92it/s +epoch=17 global_step=6900 loss=6.48082 loss_avg=7.04902 acc=0.36719 acc_top1_avg=0.28357 acc_top5_avg=0.70967 lr=0.01000 gn=3.96025 time=63.50it/s +epoch=17 global_step=6950 loss=7.31457 loss_avg=7.03980 acc=0.24219 acc_top1_avg=0.28432 acc_top5_avg=0.70973 lr=0.01000 gn=2.77041 time=58.36it/s +epoch=17 global_step=7000 loss=6.63156 loss_avg=7.04021 acc=0.33594 acc_top1_avg=0.28410 acc_top5_avg=0.71052 lr=0.01000 gn=3.60873 time=62.84it/s +====================Eval==================== +epoch=17 global_step=7038 loss=3.89554 test_loss_avg=3.91034 acc=0.00781 test_acc_avg=0.01897 test_acc_top5_avg=0.94643 time=236.47it/s +epoch=17 global_step=7038 loss=1.02847 test_loss_avg=3.67217 acc=0.73438 test_acc_avg=0.17379 test_acc_top5_avg=0.65296 time=228.82it/s +epoch=17 global_step=7038 loss=5.23461 test_loss_avg=3.30805 acc=0.00000 test_acc_avg=0.26830 test_acc_top5_avg=0.71578 time=613.56it/s +curr_acc 0.2683 +BEST_ACC 0.2955 +curr_acc_top5 0.7158 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=7.03090 loss_avg=7.02616 acc=0.26562 acc_top1_avg=0.28385 acc_top5_avg=0.72331 lr=0.01000 gn=3.45846 time=55.50it/s +epoch=18 global_step=7100 loss=6.95807 loss_avg=7.04142 acc=0.28906 acc_top1_avg=0.28553 acc_top5_avg=0.71724 lr=0.01000 gn=3.72355 time=64.14it/s +epoch=18 global_step=7150 loss=6.65936 loss_avg=7.06546 acc=0.34375 acc_top1_avg=0.28306 acc_top5_avg=0.70912 lr=0.01000 gn=3.27221 time=47.01it/s +epoch=18 global_step=7200 loss=7.13396 loss_avg=7.06462 acc=0.28906 acc_top1_avg=0.28323 acc_top5_avg=0.71012 lr=0.01000 gn=3.83120 time=45.38it/s +epoch=18 global_step=7250 loss=6.79084 loss_avg=7.06406 acc=0.32812 acc_top1_avg=0.28353 acc_top5_avg=0.70921 lr=0.01000 gn=5.01028 time=52.61it/s +epoch=18 global_step=7300 loss=6.86810 loss_avg=7.04340 acc=0.31250 acc_top1_avg=0.28632 acc_top5_avg=0.70969 lr=0.01000 gn=3.06446 time=55.19it/s +epoch=18 global_step=7350 loss=6.90664 loss_avg=7.02984 acc=0.28906 acc_top1_avg=0.28736 acc_top5_avg=0.71199 lr=0.01000 gn=2.85016 time=51.81it/s +epoch=18 global_step=7400 loss=7.13916 loss_avg=7.02862 acc=0.28125 acc_top1_avg=0.28731 acc_top5_avg=0.71184 lr=0.01000 gn=3.42773 time=60.47it/s +====================Eval==================== +epoch=18 global_step=7429 loss=5.87413 test_loss_avg=4.69791 acc=0.00000 test_acc_avg=0.11077 test_acc_top5_avg=0.56808 time=159.12it/s +epoch=18 global_step=7429 loss=7.94825 test_loss_avg=3.99191 acc=0.00000 test_acc_avg=0.26242 test_acc_top5_avg=0.67668 time=241.29it/s +epoch=18 global_step=7429 loss=8.10046 test_loss_avg=4.04392 acc=0.00000 test_acc_avg=0.25910 test_acc_top5_avg=0.68078 time=505.52it/s +curr_acc 0.2591 +BEST_ACC 0.2955 +curr_acc_top5 0.6808 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=6.71442 loss_avg=6.95717 acc=0.32031 acc_top1_avg=0.29092 acc_top5_avg=0.71801 lr=0.01000 gn=4.31190 time=52.95it/s +epoch=19 global_step=7500 loss=7.80219 loss_avg=6.96743 acc=0.20312 acc_top1_avg=0.29104 acc_top5_avg=0.70830 lr=0.01000 gn=4.04125 time=63.83it/s +epoch=19 global_step=7550 loss=6.96782 loss_avg=6.97925 acc=0.29688 acc_top1_avg=0.29093 acc_top5_avg=0.70965 lr=0.01000 gn=3.86066 time=55.61it/s +epoch=19 global_step=7600 loss=6.88662 loss_avg=7.01563 acc=0.31250 acc_top1_avg=0.28733 acc_top5_avg=0.71034 lr=0.01000 gn=3.79648 time=59.09it/s +epoch=19 global_step=7650 loss=7.01084 loss_avg=7.00415 acc=0.28906 acc_top1_avg=0.28793 acc_top5_avg=0.71129 lr=0.01000 gn=2.78579 time=55.35it/s +epoch=19 global_step=7700 loss=6.64218 loss_avg=7.02815 acc=0.32031 acc_top1_avg=0.28543 acc_top5_avg=0.71036 lr=0.01000 gn=3.05244 time=45.77it/s +epoch=19 global_step=7750 loss=7.26044 loss_avg=7.02627 acc=0.25781 acc_top1_avg=0.28612 acc_top5_avg=0.71023 lr=0.01000 gn=3.14624 time=54.72it/s +epoch=19 global_step=7800 loss=6.77498 loss_avg=7.02956 acc=0.33594 acc_top1_avg=0.28584 acc_top5_avg=0.70961 lr=0.01000 gn=3.57965 time=61.24it/s +====================Eval==================== +epoch=19 global_step=7820 loss=4.05229 test_loss_avg=3.36333 acc=0.00000 test_acc_avg=0.21269 test_acc_top5_avg=0.65322 time=238.61it/s +epoch=19 global_step=7820 loss=5.30606 test_loss_avg=3.08390 acc=0.00000 test_acc_avg=0.28728 test_acc_top5_avg=0.73853 time=807.22it/s +curr_acc 0.2873 +BEST_ACC 0.2955 +curr_acc_top5 0.7385 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=6.84710 loss_avg=7.06044 acc=0.29688 acc_top1_avg=0.28255 acc_top5_avg=0.71745 lr=0.01000 gn=3.71288 time=56.73it/s +epoch=20 global_step=7900 loss=7.04635 loss_avg=7.09855 acc=0.28906 acc_top1_avg=0.27930 acc_top5_avg=0.70732 lr=0.01000 gn=3.94944 time=55.05it/s +epoch=20 global_step=7950 loss=7.02289 loss_avg=7.06451 acc=0.26562 acc_top1_avg=0.28227 acc_top5_avg=0.70980 lr=0.01000 gn=4.14134 time=62.21it/s +epoch=20 global_step=8000 loss=7.01907 loss_avg=7.07195 acc=0.28906 acc_top1_avg=0.28186 acc_top5_avg=0.70803 lr=0.01000 gn=3.24219 time=58.49it/s +epoch=20 global_step=8050 loss=6.70742 loss_avg=7.03251 acc=0.30469 acc_top1_avg=0.28584 acc_top5_avg=0.71012 lr=0.01000 gn=3.89332 time=53.46it/s +epoch=20 global_step=8100 loss=7.24747 loss_avg=7.02943 acc=0.26562 acc_top1_avg=0.28630 acc_top5_avg=0.71077 lr=0.01000 gn=3.93731 time=48.02it/s +epoch=20 global_step=8150 loss=6.61845 loss_avg=7.03137 acc=0.30469 acc_top1_avg=0.28558 acc_top5_avg=0.70945 lr=0.01000 gn=3.78091 time=56.26it/s +epoch=20 global_step=8200 loss=7.04513 loss_avg=7.03035 acc=0.30469 acc_top1_avg=0.28577 acc_top5_avg=0.71028 lr=0.01000 gn=3.60249 time=55.57it/s +====================Eval==================== +epoch=20 global_step=8211 loss=4.65448 test_loss_avg=2.79432 acc=0.00000 test_acc_avg=0.31016 test_acc_top5_avg=0.90312 time=241.16it/s +epoch=20 global_step=8211 loss=0.14439 test_loss_avg=3.06325 acc=0.95312 test_acc_avg=0.30446 test_acc_top5_avg=0.71652 time=239.80it/s +epoch=20 global_step=8211 loss=4.91472 test_loss_avg=3.25903 acc=0.00000 test_acc_avg=0.27354 test_acc_top5_avg=0.73151 time=849.22it/s +curr_acc 0.2735 +BEST_ACC 0.2955 +curr_acc_top5 0.7315 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=7.16665 loss_avg=7.09860 acc=0.28125 acc_top1_avg=0.27925 acc_top5_avg=0.70092 lr=0.01000 gn=3.78089 time=56.43it/s +epoch=21 global_step=8300 loss=7.29290 loss_avg=7.09189 acc=0.25781 acc_top1_avg=0.27949 acc_top5_avg=0.70523 lr=0.01000 gn=3.46149 time=60.04it/s +epoch=21 global_step=8350 loss=7.32844 loss_avg=7.08537 acc=0.25781 acc_top1_avg=0.27996 acc_top5_avg=0.70807 lr=0.01000 gn=3.21625 time=56.95it/s +epoch=21 global_step=8400 loss=6.78553 loss_avg=7.08246 acc=0.30469 acc_top1_avg=0.28038 acc_top5_avg=0.70937 lr=0.01000 gn=3.70938 time=61.78it/s +epoch=21 global_step=8450 loss=6.75312 loss_avg=7.06729 acc=0.32031 acc_top1_avg=0.28145 acc_top5_avg=0.70992 lr=0.01000 gn=3.47966 time=53.16it/s +epoch=21 global_step=8500 loss=7.25442 loss_avg=7.06432 acc=0.26562 acc_top1_avg=0.28136 acc_top5_avg=0.71059 lr=0.01000 gn=3.12053 time=52.26it/s +epoch=21 global_step=8550 loss=7.05693 loss_avg=7.05496 acc=0.28125 acc_top1_avg=0.28245 acc_top5_avg=0.71048 lr=0.01000 gn=4.04718 time=56.90it/s +epoch=21 global_step=8600 loss=7.11981 loss_avg=7.05030 acc=0.26562 acc_top1_avg=0.28254 acc_top5_avg=0.71042 lr=0.01000 gn=2.90127 time=63.73it/s +====================Eval==================== +epoch=21 global_step=8602 loss=3.50005 test_loss_avg=4.44804 acc=0.14062 test_acc_avg=0.12919 test_acc_top5_avg=0.59851 time=232.33it/s +epoch=21 global_step=8602 loss=6.41886 test_loss_avg=3.95000 acc=0.00000 test_acc_avg=0.22785 test_acc_top5_avg=0.72152 time=502.73it/s +curr_acc 0.2278 +BEST_ACC 0.2955 +curr_acc_top5 0.7215 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=6.62014 loss_avg=7.05044 acc=0.32812 acc_top1_avg=0.28288 acc_top5_avg=0.70443 lr=0.01000 gn=3.60396 time=53.14it/s +epoch=22 global_step=8700 loss=7.15817 loss_avg=7.02472 acc=0.28125 acc_top1_avg=0.28635 acc_top5_avg=0.70823 lr=0.01000 gn=3.79314 time=55.86it/s +epoch=22 global_step=8750 loss=7.10647 loss_avg=7.01043 acc=0.27344 acc_top1_avg=0.28774 acc_top5_avg=0.70735 lr=0.01000 gn=3.99786 time=53.09it/s +epoch=22 global_step=8800 loss=7.45779 loss_avg=7.01263 acc=0.24219 acc_top1_avg=0.28744 acc_top5_avg=0.70857 lr=0.01000 gn=4.54866 time=61.85it/s +epoch=22 global_step=8850 loss=7.38396 loss_avg=7.02759 acc=0.24219 acc_top1_avg=0.28566 acc_top5_avg=0.70880 lr=0.01000 gn=3.83321 time=55.95it/s +epoch=22 global_step=8900 loss=6.87064 loss_avg=7.03368 acc=0.31250 acc_top1_avg=0.28521 acc_top5_avg=0.70790 lr=0.01000 gn=3.64529 time=55.03it/s +epoch=22 global_step=8950 loss=6.93292 loss_avg=7.03309 acc=0.28906 acc_top1_avg=0.28480 acc_top5_avg=0.70804 lr=0.01000 gn=2.46606 time=55.65it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.70735 test_loss_avg=3.64398 acc=0.76562 test_acc_avg=0.25716 test_acc_top5_avg=0.91536 time=230.71it/s +epoch=22 global_step=8993 loss=0.98845 test_loss_avg=3.57443 acc=0.74219 test_acc_avg=0.22757 test_acc_top5_avg=0.61240 time=179.37it/s +epoch=22 global_step=8993 loss=5.51628 test_loss_avg=3.43913 acc=0.00000 test_acc_avg=0.28095 test_acc_top5_avg=0.68928 time=826.46it/s +curr_acc 0.2810 +BEST_ACC 0.2955 +curr_acc_top5 0.6893 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=7.72486 loss_avg=7.27663 acc=0.22656 acc_top1_avg=0.25335 acc_top5_avg=0.70647 lr=0.01000 gn=3.70734 time=53.01it/s +epoch=23 global_step=9050 loss=6.77246 loss_avg=7.00454 acc=0.33594 acc_top1_avg=0.28934 acc_top5_avg=0.71450 lr=0.01000 gn=3.84526 time=62.15it/s +epoch=23 global_step=9100 loss=7.19507 loss_avg=7.03177 acc=0.24219 acc_top1_avg=0.28570 acc_top5_avg=0.71013 lr=0.01000 gn=3.43739 time=55.54it/s +epoch=23 global_step=9150 loss=7.06745 loss_avg=7.03690 acc=0.27344 acc_top1_avg=0.28424 acc_top5_avg=0.70880 lr=0.01000 gn=4.43866 time=63.51it/s +epoch=23 global_step=9200 loss=6.56343 loss_avg=7.01892 acc=0.33594 acc_top1_avg=0.28638 acc_top5_avg=0.71165 lr=0.01000 gn=3.79805 time=56.18it/s +epoch=23 global_step=9250 loss=7.16302 loss_avg=7.00034 acc=0.25000 acc_top1_avg=0.28842 acc_top5_avg=0.71413 lr=0.01000 gn=2.96025 time=60.69it/s +epoch=23 global_step=9300 loss=6.64840 loss_avg=7.01249 acc=0.32031 acc_top1_avg=0.28715 acc_top5_avg=0.71280 lr=0.01000 gn=3.60659 time=53.09it/s +epoch=23 global_step=9350 loss=7.23476 loss_avg=7.02187 acc=0.25781 acc_top1_avg=0.28600 acc_top5_avg=0.71127 lr=0.01000 gn=3.53120 time=59.58it/s +====================Eval==================== +epoch=23 global_step=9384 loss=5.59928 test_loss_avg=3.94862 acc=0.00000 test_acc_avg=0.15436 test_acc_top5_avg=0.63163 time=231.59it/s +epoch=23 global_step=9384 loss=5.58560 test_loss_avg=3.49001 acc=0.00000 test_acc_avg=0.27166 test_acc_top5_avg=0.74258 time=534.44it/s +curr_acc 0.2717 +BEST_ACC 0.2955 +curr_acc_top5 0.7426 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=6.65998 loss_avg=7.08725 acc=0.32812 acc_top1_avg=0.27734 acc_top5_avg=0.71143 lr=0.01000 gn=3.45018 time=63.71it/s +epoch=24 global_step=9450 loss=7.09334 loss_avg=7.00725 acc=0.30469 acc_top1_avg=0.28741 acc_top5_avg=0.71745 lr=0.01000 gn=4.11819 time=58.01it/s +epoch=24 global_step=9500 loss=6.31649 loss_avg=7.02196 acc=0.35938 acc_top1_avg=0.28442 acc_top5_avg=0.71080 lr=0.01000 gn=3.59683 time=54.53it/s +epoch=24 global_step=9550 loss=6.98005 loss_avg=7.00471 acc=0.29688 acc_top1_avg=0.28662 acc_top5_avg=0.71291 lr=0.01000 gn=3.93911 time=63.94it/s +epoch=24 global_step=9600 loss=7.30434 loss_avg=7.01473 acc=0.25781 acc_top1_avg=0.28559 acc_top5_avg=0.71228 lr=0.01000 gn=3.37173 time=62.18it/s +epoch=24 global_step=9650 loss=6.83485 loss_avg=7.00806 acc=0.28906 acc_top1_avg=0.28718 acc_top5_avg=0.71285 lr=0.01000 gn=4.35073 time=63.89it/s +epoch=24 global_step=9700 loss=6.97719 loss_avg=7.01053 acc=0.29688 acc_top1_avg=0.28691 acc_top5_avg=0.71198 lr=0.01000 gn=3.43918 time=57.06it/s +epoch=24 global_step=9750 loss=7.01819 loss_avg=7.02058 acc=0.28906 acc_top1_avg=0.28612 acc_top5_avg=0.71168 lr=0.01000 gn=4.13657 time=57.09it/s +====================Eval==================== +epoch=24 global_step=9775 loss=4.48597 test_loss_avg=4.29160 acc=0.00000 test_acc_avg=0.00391 test_acc_top5_avg=0.95898 time=219.20it/s +epoch=24 global_step=9775 loss=4.24244 test_loss_avg=3.60164 acc=0.00000 test_acc_avg=0.18171 test_acc_top5_avg=0.63093 time=237.87it/s +epoch=24 global_step=9775 loss=5.40062 test_loss_avg=3.15774 acc=0.00000 test_acc_avg=0.29826 test_acc_top5_avg=0.71539 time=577.73it/s +curr_acc 0.2983 +BEST_ACC 0.2955 +curr_acc_top5 0.7154 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=8.10622 loss_avg=7.02628 acc=0.17969 acc_top1_avg=0.28844 acc_top5_avg=0.70594 lr=0.01000 gn=3.60674 time=53.35it/s +epoch=25 global_step=9850 loss=7.06001 loss_avg=6.99357 acc=0.26562 acc_top1_avg=0.29083 acc_top5_avg=0.71281 lr=0.01000 gn=3.82264 time=58.44it/s +epoch=25 global_step=9900 loss=6.57229 loss_avg=7.00255 acc=0.33594 acc_top1_avg=0.28950 acc_top5_avg=0.71113 lr=0.01000 gn=3.58958 time=52.97it/s +epoch=25 global_step=9950 loss=7.19567 loss_avg=6.99433 acc=0.25781 acc_top1_avg=0.29080 acc_top5_avg=0.71210 lr=0.01000 gn=3.99216 time=48.34it/s +epoch=25 global_step=10000 loss=7.38065 loss_avg=6.99208 acc=0.23438 acc_top1_avg=0.29097 acc_top5_avg=0.71444 lr=0.01000 gn=3.59167 time=55.87it/s +epoch=25 global_step=10050 loss=6.74602 loss_avg=6.99925 acc=0.32031 acc_top1_avg=0.29043 acc_top5_avg=0.71267 lr=0.01000 gn=3.33962 time=56.34it/s +epoch=25 global_step=10100 loss=6.89199 loss_avg=7.00712 acc=0.31250 acc_top1_avg=0.28901 acc_top5_avg=0.71358 lr=0.01000 gn=4.18037 time=55.39it/s +epoch=25 global_step=10150 loss=7.24440 loss_avg=7.01363 acc=0.24219 acc_top1_avg=0.28821 acc_top5_avg=0.71169 lr=0.01000 gn=2.86654 time=55.43it/s +====================Eval==================== +epoch=25 global_step=10166 loss=4.59690 test_loss_avg=2.54060 acc=0.00000 test_acc_avg=0.34719 test_acc_top5_avg=0.80188 time=132.17it/s +epoch=25 global_step=10166 loss=5.47899 test_loss_avg=3.00583 acc=0.00000 test_acc_avg=0.31292 test_acc_top5_avg=0.69448 time=241.57it/s +epoch=25 global_step=10166 loss=5.18114 test_loss_avg=3.12388 acc=0.00000 test_acc_avg=0.29707 test_acc_top5_avg=0.70233 time=791.08it/s +curr_acc 0.2971 +BEST_ACC 0.2983 +curr_acc_top5 0.7023 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=6.96949 loss_avg=6.96489 acc=0.29688 acc_top1_avg=0.29297 acc_top5_avg=0.71921 lr=0.01000 gn=4.43921 time=54.88it/s +epoch=26 global_step=10250 loss=6.69454 loss_avg=7.00078 acc=0.31250 acc_top1_avg=0.29009 acc_top5_avg=0.71838 lr=0.01000 gn=3.93090 time=52.91it/s +epoch=26 global_step=10300 loss=6.65786 loss_avg=6.98110 acc=0.32812 acc_top1_avg=0.29180 acc_top5_avg=0.71968 lr=0.01000 gn=3.94688 time=57.45it/s +epoch=26 global_step=10350 loss=6.97884 loss_avg=6.98510 acc=0.28125 acc_top1_avg=0.29182 acc_top5_avg=0.71850 lr=0.01000 gn=3.71441 time=62.70it/s +epoch=26 global_step=10400 loss=7.34013 loss_avg=7.00110 acc=0.25781 acc_top1_avg=0.28990 acc_top5_avg=0.71645 lr=0.01000 gn=4.31499 time=60.57it/s +epoch=26 global_step=10450 loss=7.28439 loss_avg=6.99885 acc=0.27344 acc_top1_avg=0.29033 acc_top5_avg=0.71413 lr=0.01000 gn=3.80615 time=58.99it/s +epoch=26 global_step=10500 loss=7.04097 loss_avg=7.00622 acc=0.28906 acc_top1_avg=0.28916 acc_top5_avg=0.71307 lr=0.01000 gn=3.74620 time=54.92it/s +epoch=26 global_step=10550 loss=6.90920 loss_avg=7.01018 acc=0.30469 acc_top1_avg=0.28857 acc_top5_avg=0.71360 lr=0.01000 gn=3.23206 time=57.55it/s +====================Eval==================== +epoch=26 global_step=10557 loss=3.14251 test_loss_avg=3.41332 acc=0.07812 test_acc_avg=0.17629 test_acc_top5_avg=0.60190 time=231.19it/s +epoch=26 global_step=10557 loss=4.83858 test_loss_avg=3.02865 acc=0.00000 test_acc_avg=0.27670 test_acc_top5_avg=0.71044 time=599.44it/s +curr_acc 0.2767 +BEST_ACC 0.2983 +curr_acc_top5 0.7104 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=7.28078 loss_avg=7.02473 acc=0.24219 acc_top1_avg=0.28525 acc_top5_avg=0.70603 lr=0.01000 gn=2.94948 time=56.32it/s +epoch=27 global_step=10650 loss=6.76813 loss_avg=7.00597 acc=0.31250 acc_top1_avg=0.28780 acc_top5_avg=0.70724 lr=0.01000 gn=3.51227 time=62.41it/s +epoch=27 global_step=10700 loss=7.30002 loss_avg=7.00423 acc=0.25781 acc_top1_avg=0.28802 acc_top5_avg=0.70810 lr=0.01000 gn=4.37666 time=60.25it/s +epoch=27 global_step=10750 loss=7.62293 loss_avg=7.01097 acc=0.20312 acc_top1_avg=0.28732 acc_top5_avg=0.71094 lr=0.01000 gn=3.53027 time=55.95it/s +epoch=27 global_step=10800 loss=7.33406 loss_avg=7.00671 acc=0.24219 acc_top1_avg=0.28839 acc_top5_avg=0.71155 lr=0.01000 gn=3.33931 time=52.78it/s +epoch=27 global_step=10850 loss=6.82934 loss_avg=7.00916 acc=0.30469 acc_top1_avg=0.28792 acc_top5_avg=0.71142 lr=0.01000 gn=2.69411 time=55.90it/s +epoch=27 global_step=10900 loss=7.27562 loss_avg=7.02059 acc=0.25000 acc_top1_avg=0.28672 acc_top5_avg=0.70900 lr=0.01000 gn=3.39114 time=60.82it/s +====================Eval==================== +epoch=27 global_step=10948 loss=4.60752 test_loss_avg=3.05953 acc=0.00000 test_acc_avg=0.35662 test_acc_top5_avg=0.92601 time=235.71it/s +epoch=27 global_step=10948 loss=0.10215 test_loss_avg=3.32385 acc=0.96875 test_acc_avg=0.27962 test_acc_top5_avg=0.66266 time=225.93it/s +epoch=27 global_step=10948 loss=4.99414 test_loss_avg=3.39598 acc=0.00000 test_acc_avg=0.27759 test_acc_top5_avg=0.69887 time=457.39it/s +curr_acc 0.2776 +BEST_ACC 0.2983 +curr_acc_top5 0.6989 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=6.31385 loss_avg=6.60264 acc=0.37500 acc_top1_avg=0.33984 acc_top5_avg=0.75000 lr=0.01000 gn=3.37189 time=61.71it/s +epoch=28 global_step=11000 loss=6.84262 loss_avg=6.99572 acc=0.30469 acc_top1_avg=0.28816 acc_top5_avg=0.70613 lr=0.01000 gn=3.40262 time=59.56it/s +epoch=28 global_step=11050 loss=7.27211 loss_avg=6.99136 acc=0.25000 acc_top1_avg=0.28891 acc_top5_avg=0.71009 lr=0.01000 gn=3.54144 time=62.15it/s +epoch=28 global_step=11100 loss=7.27946 loss_avg=7.00702 acc=0.26562 acc_top1_avg=0.28773 acc_top5_avg=0.71063 lr=0.01000 gn=3.48910 time=56.11it/s +epoch=28 global_step=11150 loss=7.00638 loss_avg=7.01888 acc=0.28906 acc_top1_avg=0.28724 acc_top5_avg=0.70931 lr=0.01000 gn=3.85397 time=57.14it/s +epoch=28 global_step=11200 loss=7.69233 loss_avg=7.01936 acc=0.21094 acc_top1_avg=0.28714 acc_top5_avg=0.70827 lr=0.01000 gn=2.84215 time=59.86it/s +epoch=28 global_step=11250 loss=7.06602 loss_avg=7.01948 acc=0.27344 acc_top1_avg=0.28658 acc_top5_avg=0.70721 lr=0.01000 gn=2.73395 time=60.08it/s +epoch=28 global_step=11300 loss=6.65943 loss_avg=7.01571 acc=0.32031 acc_top1_avg=0.28693 acc_top5_avg=0.70785 lr=0.01000 gn=2.96763 time=58.37it/s +====================Eval==================== +epoch=28 global_step=11339 loss=5.20188 test_loss_avg=4.03271 acc=0.00000 test_acc_avg=0.16242 test_acc_top5_avg=0.51933 time=238.71it/s +epoch=28 global_step=11339 loss=5.09153 test_loss_avg=3.26899 acc=0.00000 test_acc_avg=0.29816 test_acc_top5_avg=0.71588 time=647.47it/s +curr_acc 0.2982 +BEST_ACC 0.2983 +curr_acc_top5 0.7159 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=6.88953 loss_avg=6.81220 acc=0.30469 acc_top1_avg=0.31250 acc_top5_avg=0.73651 lr=0.01000 gn=3.65765 time=56.82it/s +epoch=29 global_step=11400 loss=7.48668 loss_avg=6.99569 acc=0.24219 acc_top1_avg=0.28804 acc_top5_avg=0.70786 lr=0.01000 gn=4.21931 time=58.51it/s +epoch=29 global_step=11450 loss=6.79098 loss_avg=7.00341 acc=0.32031 acc_top1_avg=0.28667 acc_top5_avg=0.70932 lr=0.01000 gn=2.95456 time=60.23it/s +epoch=29 global_step=11500 loss=7.23538 loss_avg=6.96682 acc=0.26562 acc_top1_avg=0.29125 acc_top5_avg=0.71584 lr=0.01000 gn=3.52150 time=57.11it/s +epoch=29 global_step=11550 loss=7.33403 loss_avg=6.97678 acc=0.26562 acc_top1_avg=0.29088 acc_top5_avg=0.71560 lr=0.01000 gn=3.45081 time=55.69it/s +epoch=29 global_step=11600 loss=6.69259 loss_avg=6.98978 acc=0.32812 acc_top1_avg=0.29026 acc_top5_avg=0.71594 lr=0.01000 gn=4.54910 time=37.76it/s +epoch=29 global_step=11650 loss=7.18334 loss_avg=6.98680 acc=0.25781 acc_top1_avg=0.29019 acc_top5_avg=0.71518 lr=0.01000 gn=4.35691 time=64.10it/s +epoch=29 global_step=11700 loss=6.75295 loss_avg=6.99998 acc=0.32812 acc_top1_avg=0.28908 acc_top5_avg=0.71403 lr=0.01000 gn=3.53396 time=54.08it/s +====================Eval==================== +epoch=29 global_step=11730 loss=1.59532 test_loss_avg=4.08961 acc=0.61719 test_acc_avg=0.08681 test_acc_top5_avg=0.91319 time=243.39it/s +epoch=29 global_step=11730 loss=0.78322 test_loss_avg=3.94648 acc=0.81250 test_acc_avg=0.19160 test_acc_top5_avg=0.61970 time=235.90it/s +epoch=29 global_step=11730 loss=5.91705 test_loss_avg=3.60786 acc=0.00000 test_acc_avg=0.27591 test_acc_top5_avg=0.71114 time=645.08it/s +curr_acc 0.2759 +BEST_ACC 0.2983 +curr_acc_top5 0.7111 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=7.35774 loss_avg=6.97944 acc=0.24219 acc_top1_avg=0.28711 acc_top5_avg=0.70234 lr=0.01000 gn=3.58013 time=57.48it/s +epoch=30 global_step=11800 loss=7.12864 loss_avg=7.01132 acc=0.28125 acc_top1_avg=0.28583 acc_top5_avg=0.70268 lr=0.01000 gn=3.52244 time=60.66it/s +epoch=30 global_step=11850 loss=7.52862 loss_avg=7.01397 acc=0.23438 acc_top1_avg=0.28574 acc_top5_avg=0.70918 lr=0.01000 gn=3.80618 time=57.00it/s +epoch=30 global_step=11900 loss=7.38484 loss_avg=7.00597 acc=0.23438 acc_top1_avg=0.28667 acc_top5_avg=0.71383 lr=0.01000 gn=2.85332 time=56.89it/s +epoch=30 global_step=11950 loss=7.21588 loss_avg=7.00595 acc=0.25781 acc_top1_avg=0.28697 acc_top5_avg=0.71207 lr=0.01000 gn=2.64488 time=60.57it/s +epoch=30 global_step=12000 loss=7.34249 loss_avg=7.00646 acc=0.25000 acc_top1_avg=0.28675 acc_top5_avg=0.71134 lr=0.01000 gn=3.37625 time=52.61it/s +epoch=30 global_step=12050 loss=7.22341 loss_avg=7.01927 acc=0.26562 acc_top1_avg=0.28525 acc_top5_avg=0.71096 lr=0.01000 gn=3.36944 time=33.10it/s +epoch=30 global_step=12100 loss=7.56303 loss_avg=7.01882 acc=0.23438 acc_top1_avg=0.28543 acc_top5_avg=0.71130 lr=0.01000 gn=3.01030 time=61.53it/s +====================Eval==================== +epoch=30 global_step=12121 loss=5.22587 test_loss_avg=3.65721 acc=0.00000 test_acc_avg=0.20625 test_acc_top5_avg=0.63594 time=238.35it/s +epoch=30 global_step=12121 loss=5.19024 test_loss_avg=3.36212 acc=0.00000 test_acc_avg=0.27621 test_acc_top5_avg=0.70352 time=505.58it/s +curr_acc 0.2762 +BEST_ACC 0.2983 +curr_acc_top5 0.7035 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=7.21511 loss_avg=7.00285 acc=0.26562 acc_top1_avg=0.29041 acc_top5_avg=0.71040 lr=0.01000 gn=3.58299 time=59.39it/s +epoch=31 global_step=12200 loss=7.45128 loss_avg=6.97752 acc=0.24219 acc_top1_avg=0.29134 acc_top5_avg=0.71766 lr=0.01000 gn=4.13431 time=63.83it/s +epoch=31 global_step=12250 loss=6.69220 loss_avg=6.99360 acc=0.33594 acc_top1_avg=0.29009 acc_top5_avg=0.71245 lr=0.01000 gn=4.43649 time=58.84it/s +epoch=31 global_step=12300 loss=6.99760 loss_avg=6.99608 acc=0.28906 acc_top1_avg=0.29020 acc_top5_avg=0.71085 lr=0.01000 gn=3.74166 time=58.38it/s +epoch=31 global_step=12350 loss=6.94330 loss_avg=7.00622 acc=0.29688 acc_top1_avg=0.28855 acc_top5_avg=0.71333 lr=0.01000 gn=3.45036 time=60.95it/s +epoch=31 global_step=12400 loss=7.13229 loss_avg=7.01063 acc=0.26562 acc_top1_avg=0.28791 acc_top5_avg=0.71279 lr=0.01000 gn=3.20506 time=57.61it/s +epoch=31 global_step=12450 loss=7.15852 loss_avg=7.01159 acc=0.26562 acc_top1_avg=0.28769 acc_top5_avg=0.71326 lr=0.01000 gn=2.71706 time=56.50it/s +epoch=31 global_step=12500 loss=7.09604 loss_avg=7.01392 acc=0.28125 acc_top1_avg=0.28758 acc_top5_avg=0.71351 lr=0.01000 gn=4.29275 time=56.42it/s +====================Eval==================== +epoch=31 global_step=12512 loss=4.47330 test_loss_avg=4.47330 acc=0.00781 test_acc_avg=0.00781 test_acc_top5_avg=0.79688 time=197.14it/s +epoch=31 global_step=12512 loss=4.31540 test_loss_avg=3.72398 acc=0.00000 test_acc_avg=0.19608 test_acc_top5_avg=0.55086 time=238.68it/s +epoch=31 global_step=12512 loss=6.11704 test_loss_avg=3.39125 acc=0.00000 test_acc_avg=0.30083 test_acc_top5_avg=0.67197 time=583.11it/s +curr_acc 0.3008 +BEST_ACC 0.2983 +curr_acc_top5 0.6720 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=6.50335 loss_avg=6.99428 acc=0.36719 acc_top1_avg=0.29153 acc_top5_avg=0.70744 lr=0.01000 gn=3.92545 time=55.66it/s +epoch=32 global_step=12600 loss=6.66407 loss_avg=7.02583 acc=0.33594 acc_top1_avg=0.28613 acc_top5_avg=0.70987 lr=0.01000 gn=3.99428 time=60.74it/s +epoch=32 global_step=12650 loss=7.01474 loss_avg=7.01962 acc=0.28906 acc_top1_avg=0.28748 acc_top5_avg=0.71088 lr=0.01000 gn=4.06239 time=53.58it/s +epoch=32 global_step=12700 loss=7.28243 loss_avg=7.01402 acc=0.25000 acc_top1_avg=0.28819 acc_top5_avg=0.71331 lr=0.01000 gn=3.62720 time=53.87it/s +epoch=32 global_step=12750 loss=7.03894 loss_avg=7.02713 acc=0.28125 acc_top1_avg=0.28726 acc_top5_avg=0.71169 lr=0.01000 gn=4.40217 time=60.09it/s +epoch=32 global_step=12800 loss=7.11398 loss_avg=7.01575 acc=0.28125 acc_top1_avg=0.28857 acc_top5_avg=0.71308 lr=0.01000 gn=4.15643 time=62.54it/s +epoch=32 global_step=12850 loss=7.19694 loss_avg=7.01837 acc=0.28125 acc_top1_avg=0.28788 acc_top5_avg=0.71274 lr=0.01000 gn=3.18132 time=52.08it/s +epoch=32 global_step=12900 loss=7.73362 loss_avg=7.01948 acc=0.21094 acc_top1_avg=0.28783 acc_top5_avg=0.71207 lr=0.01000 gn=3.51395 time=56.67it/s +====================Eval==================== +epoch=32 global_step=12903 loss=4.89411 test_loss_avg=4.00390 acc=0.00000 test_acc_avg=0.13885 test_acc_top5_avg=0.74751 time=242.46it/s +epoch=32 global_step=12903 loss=5.61299 test_loss_avg=3.68738 acc=0.00000 test_acc_avg=0.25152 test_acc_top5_avg=0.66070 time=239.74it/s +epoch=32 global_step=12903 loss=5.82468 test_loss_avg=3.86180 acc=0.00000 test_acc_avg=0.22923 test_acc_top5_avg=0.68938 time=791.68it/s +curr_acc 0.2292 +BEST_ACC 0.3008 +curr_acc_top5 0.6894 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=7.29319 loss_avg=6.92818 acc=0.25000 acc_top1_avg=0.29870 acc_top5_avg=0.71609 lr=0.01000 gn=3.35172 time=58.70it/s +epoch=33 global_step=13000 loss=7.03888 loss_avg=7.03857 acc=0.28906 acc_top1_avg=0.28657 acc_top5_avg=0.70909 lr=0.01000 gn=3.80860 time=55.24it/s +epoch=33 global_step=13050 loss=7.22381 loss_avg=7.01545 acc=0.26562 acc_top1_avg=0.28917 acc_top5_avg=0.70823 lr=0.01000 gn=2.59435 time=56.78it/s +epoch=33 global_step=13100 loss=7.08762 loss_avg=7.02488 acc=0.26562 acc_top1_avg=0.28736 acc_top5_avg=0.70773 lr=0.01000 gn=3.09075 time=56.83it/s +epoch=33 global_step=13150 loss=7.17111 loss_avg=7.01585 acc=0.27344 acc_top1_avg=0.28818 acc_top5_avg=0.70876 lr=0.01000 gn=4.06464 time=55.92it/s +epoch=33 global_step=13200 loss=7.54121 loss_avg=7.00588 acc=0.21875 acc_top1_avg=0.28912 acc_top5_avg=0.70875 lr=0.01000 gn=4.45590 time=41.11it/s +epoch=33 global_step=13250 loss=7.34865 loss_avg=7.00701 acc=0.23438 acc_top1_avg=0.28868 acc_top5_avg=0.70970 lr=0.01000 gn=3.44871 time=51.32it/s +====================Eval==================== +epoch=33 global_step=13294 loss=3.20998 test_loss_avg=3.77456 acc=0.12500 test_acc_avg=0.15570 test_acc_top5_avg=0.59993 time=237.40it/s +epoch=33 global_step=13294 loss=4.68098 test_loss_avg=3.28927 acc=0.00000 test_acc_avg=0.27008 test_acc_top5_avg=0.72913 time=513.38it/s +curr_acc 0.2701 +BEST_ACC 0.3008 +curr_acc_top5 0.7291 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=6.62422 loss_avg=6.78578 acc=0.33594 acc_top1_avg=0.32552 acc_top5_avg=0.71875 lr=0.01000 gn=3.44615 time=60.11it/s +epoch=34 global_step=13350 loss=7.11185 loss_avg=6.84763 acc=0.28125 acc_top1_avg=0.30636 acc_top5_avg=0.72028 lr=0.01000 gn=3.44943 time=60.23it/s +epoch=34 global_step=13400 loss=7.21332 loss_avg=6.90052 acc=0.24219 acc_top1_avg=0.29872 acc_top5_avg=0.72022 lr=0.01000 gn=4.52053 time=56.24it/s +epoch=34 global_step=13450 loss=6.89974 loss_avg=6.93433 acc=0.28125 acc_top1_avg=0.29622 acc_top5_avg=0.71760 lr=0.01000 gn=3.93979 time=59.36it/s +epoch=34 global_step=13500 loss=7.14349 loss_avg=6.96209 acc=0.28906 acc_top1_avg=0.29361 acc_top5_avg=0.71632 lr=0.01000 gn=3.27579 time=62.92it/s +epoch=34 global_step=13550 loss=6.78373 loss_avg=6.96654 acc=0.31250 acc_top1_avg=0.29300 acc_top5_avg=0.71487 lr=0.01000 gn=3.45527 time=60.07it/s +epoch=34 global_step=13600 loss=6.30998 loss_avg=6.96761 acc=0.35156 acc_top1_avg=0.29284 acc_top5_avg=0.71441 lr=0.01000 gn=3.02048 time=60.30it/s +epoch=34 global_step=13650 loss=7.29857 loss_avg=6.98527 acc=0.25781 acc_top1_avg=0.29099 acc_top5_avg=0.71250 lr=0.01000 gn=4.17578 time=64.20it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.75070 test_loss_avg=2.48185 acc=0.78125 test_acc_avg=0.36272 test_acc_top5_avg=0.98158 time=233.03it/s +epoch=34 global_step=13685 loss=0.09941 test_loss_avg=3.29631 acc=0.96875 test_acc_avg=0.26672 test_acc_top5_avg=0.64551 time=241.11it/s +epoch=34 global_step=13685 loss=5.44216 test_loss_avg=3.26509 acc=0.00000 test_acc_avg=0.29391 test_acc_top5_avg=0.70332 time=819.68it/s +curr_acc 0.2939 +BEST_ACC 0.3008 +curr_acc_top5 0.7033 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=6.83888 loss_avg=6.87197 acc=0.31250 acc_top1_avg=0.30677 acc_top5_avg=0.72135 lr=0.01000 gn=3.16816 time=54.36it/s +epoch=35 global_step=13750 loss=6.40692 loss_avg=6.93239 acc=0.36719 acc_top1_avg=0.29724 acc_top5_avg=0.71514 lr=0.01000 gn=4.31992 time=60.48it/s +epoch=35 global_step=13800 loss=7.25630 loss_avg=6.97754 acc=0.27344 acc_top1_avg=0.29178 acc_top5_avg=0.70856 lr=0.01000 gn=3.66744 time=58.30it/s +epoch=35 global_step=13850 loss=6.75138 loss_avg=6.99169 acc=0.31250 acc_top1_avg=0.29096 acc_top5_avg=0.70653 lr=0.01000 gn=4.68678 time=54.64it/s +epoch=35 global_step=13900 loss=7.07154 loss_avg=6.99172 acc=0.28906 acc_top1_avg=0.29070 acc_top5_avg=0.70908 lr=0.01000 gn=4.50205 time=55.16it/s +epoch=35 global_step=13950 loss=6.80668 loss_avg=6.99546 acc=0.32031 acc_top1_avg=0.29080 acc_top5_avg=0.71088 lr=0.01000 gn=3.97489 time=63.81it/s +epoch=35 global_step=14000 loss=6.72112 loss_avg=7.00204 acc=0.33594 acc_top1_avg=0.28973 acc_top5_avg=0.71009 lr=0.01000 gn=3.62127 time=63.69it/s +epoch=35 global_step=14050 loss=6.95520 loss_avg=6.99973 acc=0.28906 acc_top1_avg=0.28977 acc_top5_avg=0.71062 lr=0.01000 gn=3.38195 time=63.37it/s +====================Eval==================== +epoch=35 global_step=14076 loss=5.62032 test_loss_avg=3.86745 acc=0.00000 test_acc_avg=0.15223 test_acc_top5_avg=0.51987 time=239.24it/s +epoch=35 global_step=14076 loss=5.09176 test_loss_avg=3.16691 acc=0.00000 test_acc_avg=0.30034 test_acc_top5_avg=0.69660 time=502.13it/s +curr_acc 0.3003 +BEST_ACC 0.3008 +curr_acc_top5 0.6966 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=7.17906 loss_avg=6.92355 acc=0.26562 acc_top1_avg=0.29720 acc_top5_avg=0.70605 lr=0.01000 gn=3.43096 time=59.22it/s +epoch=36 global_step=14150 loss=7.36194 loss_avg=6.96595 acc=0.24219 acc_top1_avg=0.29307 acc_top5_avg=0.71094 lr=0.01000 gn=3.60109 time=60.08it/s +epoch=36 global_step=14200 loss=6.69771 loss_avg=6.97693 acc=0.34375 acc_top1_avg=0.29234 acc_top5_avg=0.71384 lr=0.01000 gn=3.68286 time=59.52it/s +epoch=36 global_step=14250 loss=7.56335 loss_avg=6.97657 acc=0.21875 acc_top1_avg=0.29252 acc_top5_avg=0.71026 lr=0.01000 gn=3.55329 time=62.04it/s +epoch=36 global_step=14300 loss=7.26018 loss_avg=6.98256 acc=0.25781 acc_top1_avg=0.29098 acc_top5_avg=0.71205 lr=0.01000 gn=3.18441 time=56.89it/s +epoch=36 global_step=14350 loss=6.90622 loss_avg=6.99387 acc=0.30469 acc_top1_avg=0.28949 acc_top5_avg=0.71231 lr=0.01000 gn=3.28936 time=55.95it/s +epoch=36 global_step=14400 loss=7.25209 loss_avg=6.99061 acc=0.25000 acc_top1_avg=0.28976 acc_top5_avg=0.71149 lr=0.01000 gn=3.78018 time=58.78it/s +epoch=36 global_step=14450 loss=7.10069 loss_avg=6.99411 acc=0.28125 acc_top1_avg=0.28933 acc_top5_avg=0.71065 lr=0.01000 gn=3.75355 time=61.59it/s +====================Eval==================== +epoch=36 global_step=14467 loss=3.07242 test_loss_avg=3.13045 acc=0.11719 test_acc_avg=0.10677 test_acc_top5_avg=0.96484 time=240.91it/s +epoch=36 global_step=14467 loss=1.47915 test_loss_avg=3.50270 acc=0.58594 test_acc_avg=0.20396 test_acc_top5_avg=0.71136 time=234.55it/s +epoch=36 global_step=14467 loss=5.29840 test_loss_avg=3.20903 acc=0.00000 test_acc_avg=0.28402 test_acc_top5_avg=0.76592 time=833.53it/s +curr_acc 0.2840 +BEST_ACC 0.3008 +curr_acc_top5 0.7659 +BEST_ACC_top5 0.7653 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=6.59697 loss_avg=6.95900 acc=0.32812 acc_top1_avg=0.29403 acc_top5_avg=0.71757 lr=0.01000 gn=3.24193 time=60.61it/s +epoch=37 global_step=14550 loss=6.64810 loss_avg=6.94033 acc=0.32812 acc_top1_avg=0.29735 acc_top5_avg=0.71536 lr=0.01000 gn=5.09429 time=52.82it/s +epoch=37 global_step=14600 loss=7.35331 loss_avg=6.99245 acc=0.25000 acc_top1_avg=0.29118 acc_top5_avg=0.70824 lr=0.01000 gn=3.71857 time=53.55it/s +epoch=37 global_step=14650 loss=6.79510 loss_avg=6.98874 acc=0.30469 acc_top1_avg=0.29120 acc_top5_avg=0.70880 lr=0.01000 gn=3.02419 time=58.42it/s +epoch=37 global_step=14700 loss=6.34713 loss_avg=6.98361 acc=0.36719 acc_top1_avg=0.29158 acc_top5_avg=0.70906 lr=0.01000 gn=3.54938 time=56.44it/s +epoch=37 global_step=14750 loss=6.93118 loss_avg=6.99899 acc=0.28906 acc_top1_avg=0.28950 acc_top5_avg=0.70812 lr=0.01000 gn=3.45346 time=58.75it/s +epoch=37 global_step=14800 loss=7.62794 loss_avg=7.00523 acc=0.21875 acc_top1_avg=0.28880 acc_top5_avg=0.70824 lr=0.01000 gn=3.19296 time=56.05it/s +epoch=37 global_step=14850 loss=6.78903 loss_avg=7.01142 acc=0.32812 acc_top1_avg=0.28784 acc_top5_avg=0.70857 lr=0.01000 gn=4.32844 time=42.61it/s +====================Eval==================== +epoch=37 global_step=14858 loss=5.41233 test_loss_avg=3.69785 acc=0.00000 test_acc_avg=0.26100 test_acc_top5_avg=0.66117 time=237.79it/s +epoch=37 global_step=14858 loss=5.80939 test_loss_avg=3.55513 acc=0.00000 test_acc_avg=0.25913 test_acc_top5_avg=0.68679 time=239.09it/s +epoch=37 global_step=14858 loss=6.02199 test_loss_avg=3.61423 acc=0.00000 test_acc_avg=0.25257 test_acc_top5_avg=0.68631 time=501.35it/s +curr_acc 0.2526 +BEST_ACC 0.3008 +curr_acc_top5 0.6863 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=6.21813 loss_avg=6.88680 acc=0.38281 acc_top1_avg=0.30283 acc_top5_avg=0.71801 lr=0.01000 gn=4.51133 time=62.41it/s +epoch=38 global_step=14950 loss=7.55136 loss_avg=6.92124 acc=0.21875 acc_top1_avg=0.29883 acc_top5_avg=0.71578 lr=0.01000 gn=3.46805 time=53.20it/s +epoch=38 global_step=15000 loss=7.02115 loss_avg=6.96537 acc=0.28125 acc_top1_avg=0.29357 acc_top5_avg=0.71330 lr=0.01000 gn=3.34740 time=61.59it/s +epoch=38 global_step=15050 loss=7.06056 loss_avg=6.99031 acc=0.28906 acc_top1_avg=0.29069 acc_top5_avg=0.71220 lr=0.01000 gn=3.89043 time=59.84it/s +epoch=38 global_step=15100 loss=7.03460 loss_avg=6.99160 acc=0.27344 acc_top1_avg=0.28987 acc_top5_avg=0.71242 lr=0.01000 gn=3.33003 time=56.14it/s +epoch=38 global_step=15150 loss=6.96644 loss_avg=6.98081 acc=0.28125 acc_top1_avg=0.29061 acc_top5_avg=0.71273 lr=0.01000 gn=4.42957 time=61.38it/s +epoch=38 global_step=15200 loss=7.04183 loss_avg=6.99907 acc=0.29688 acc_top1_avg=0.28842 acc_top5_avg=0.71224 lr=0.01000 gn=4.36488 time=30.39it/s +====================Eval==================== +epoch=38 global_step=15249 loss=4.75270 test_loss_avg=3.52474 acc=0.00000 test_acc_avg=0.17057 test_acc_top5_avg=0.64225 time=226.83it/s +epoch=38 global_step=15249 loss=4.64616 test_loss_avg=3.14485 acc=0.00000 test_acc_avg=0.28056 test_acc_top5_avg=0.75079 time=806.60it/s +curr_acc 0.2806 +BEST_ACC 0.3008 +curr_acc_top5 0.7508 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=6.47560 loss_avg=6.47560 acc=0.34375 acc_top1_avg=0.34375 acc_top5_avg=0.75781 lr=0.01000 gn=3.23439 time=44.11it/s +epoch=39 global_step=15300 loss=6.50981 loss_avg=6.94977 acc=0.34375 acc_top1_avg=0.29335 acc_top5_avg=0.71170 lr=0.01000 gn=3.82884 time=57.61it/s +epoch=39 global_step=15350 loss=7.56192 loss_avg=6.96148 acc=0.24219 acc_top1_avg=0.29254 acc_top5_avg=0.71225 lr=0.01000 gn=4.01322 time=50.80it/s +epoch=39 global_step=15400 loss=7.08244 loss_avg=7.00244 acc=0.27344 acc_top1_avg=0.28767 acc_top5_avg=0.70664 lr=0.01000 gn=3.33177 time=53.46it/s +epoch=39 global_step=15450 loss=7.24540 loss_avg=7.00198 acc=0.27344 acc_top1_avg=0.28751 acc_top5_avg=0.71105 lr=0.01000 gn=3.86226 time=54.58it/s +epoch=39 global_step=15500 loss=7.09703 loss_avg=6.98809 acc=0.26562 acc_top1_avg=0.28962 acc_top5_avg=0.71081 lr=0.01000 gn=3.72030 time=64.41it/s +epoch=39 global_step=15550 loss=7.05131 loss_avg=6.99245 acc=0.27344 acc_top1_avg=0.28930 acc_top5_avg=0.71000 lr=0.01000 gn=3.23727 time=60.55it/s +epoch=39 global_step=15600 loss=6.98375 loss_avg=6.99509 acc=0.29688 acc_top1_avg=0.28877 acc_top5_avg=0.71158 lr=0.01000 gn=3.52966 time=56.03it/s +====================Eval==================== +epoch=39 global_step=15640 loss=4.66897 test_loss_avg=2.80403 acc=0.00000 test_acc_avg=0.35074 test_acc_top5_avg=0.89268 time=242.22it/s +epoch=39 global_step=15640 loss=0.22546 test_loss_avg=3.19417 acc=0.92188 test_acc_avg=0.29223 test_acc_top5_avg=0.68127 time=232.78it/s +epoch=39 global_step=15640 loss=6.87185 test_loss_avg=3.48227 acc=0.00000 test_acc_avg=0.27097 test_acc_top5_avg=0.71252 time=492.58it/s +curr_acc 0.2710 +BEST_ACC 0.3008 +curr_acc_top5 0.7125 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=7.28395 loss_avg=7.04864 acc=0.25000 acc_top1_avg=0.27422 acc_top5_avg=0.69922 lr=0.00100 gn=2.51473 time=62.80it/s +epoch=40 global_step=15700 loss=7.14964 loss_avg=6.98238 acc=0.28125 acc_top1_avg=0.28932 acc_top5_avg=0.70755 lr=0.00100 gn=3.80581 time=56.68it/s +epoch=40 global_step=15750 loss=6.96125 loss_avg=6.96151 acc=0.29688 acc_top1_avg=0.29183 acc_top5_avg=0.70930 lr=0.00100 gn=4.06826 time=63.14it/s +epoch=40 global_step=15800 loss=7.44712 loss_avg=6.93548 acc=0.25000 acc_top1_avg=0.29600 acc_top5_avg=0.71079 lr=0.00100 gn=3.99975 time=64.27it/s +epoch=40 global_step=15850 loss=6.67710 loss_avg=6.90844 acc=0.33594 acc_top1_avg=0.29900 acc_top5_avg=0.71243 lr=0.00100 gn=3.76660 time=48.16it/s +epoch=40 global_step=15900 loss=6.77379 loss_avg=6.89288 acc=0.32031 acc_top1_avg=0.30051 acc_top5_avg=0.71337 lr=0.00100 gn=3.35799 time=17.15it/s +epoch=40 global_step=15950 loss=6.86635 loss_avg=6.87438 acc=0.28906 acc_top1_avg=0.30262 acc_top5_avg=0.71618 lr=0.00100 gn=4.33744 time=53.03it/s +epoch=40 global_step=16000 loss=6.98982 loss_avg=6.86907 acc=0.30469 acc_top1_avg=0.30308 acc_top5_avg=0.71730 lr=0.00100 gn=3.59871 time=58.91it/s +====================Eval==================== +epoch=40 global_step=16031 loss=2.33758 test_loss_avg=3.54168 acc=0.33594 test_acc_avg=0.19648 test_acc_top5_avg=0.57520 time=233.39it/s +epoch=40 global_step=16031 loss=5.89223 test_loss_avg=3.12887 acc=0.00000 test_acc_avg=0.31082 test_acc_top5_avg=0.76009 time=396.10it/s +curr_acc 0.3108 +BEST_ACC 0.3008 +curr_acc_top5 0.7601 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=6.83051 loss_avg=6.84596 acc=0.30469 acc_top1_avg=0.30633 acc_top5_avg=0.71135 lr=0.00100 gn=4.49918 time=61.81it/s +epoch=41 global_step=16100 loss=6.81268 loss_avg=6.79571 acc=0.30469 acc_top1_avg=0.31295 acc_top5_avg=0.72113 lr=0.00100 gn=3.69367 time=56.10it/s +epoch=41 global_step=16150 loss=6.67384 loss_avg=6.79093 acc=0.32031 acc_top1_avg=0.31224 acc_top5_avg=0.72354 lr=0.00100 gn=4.88196 time=57.17it/s +epoch=41 global_step=16200 loss=7.29776 loss_avg=6.81578 acc=0.25000 acc_top1_avg=0.30973 acc_top5_avg=0.72397 lr=0.00100 gn=4.19598 time=61.12it/s +epoch=41 global_step=16250 loss=6.68687 loss_avg=6.80865 acc=0.32812 acc_top1_avg=0.30947 acc_top5_avg=0.72546 lr=0.00100 gn=3.64742 time=58.51it/s +epoch=41 global_step=16300 loss=7.28833 loss_avg=6.81392 acc=0.26562 acc_top1_avg=0.30872 acc_top5_avg=0.72627 lr=0.00100 gn=3.67417 time=56.89it/s +epoch=41 global_step=16350 loss=6.69928 loss_avg=6.80798 acc=0.31250 acc_top1_avg=0.30963 acc_top5_avg=0.72666 lr=0.00100 gn=4.83364 time=55.11it/s +epoch=41 global_step=16400 loss=6.40362 loss_avg=6.81391 acc=0.33594 acc_top1_avg=0.30884 acc_top5_avg=0.72673 lr=0.00100 gn=4.57025 time=63.60it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.46205 test_loss_avg=2.06464 acc=0.88281 test_acc_avg=0.41051 test_acc_top5_avg=0.97159 time=236.87it/s +epoch=41 global_step=16422 loss=0.26461 test_loss_avg=3.12800 acc=0.92969 test_acc_avg=0.28458 test_acc_top5_avg=0.69288 time=211.69it/s +epoch=41 global_step=16422 loss=6.01989 test_loss_avg=3.07391 acc=0.00000 test_acc_avg=0.33376 test_acc_top5_avg=0.75732 time=497.43it/s +curr_acc 0.3338 +BEST_ACC 0.3108 +curr_acc_top5 0.7573 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=7.11767 loss_avg=6.81369 acc=0.27344 acc_top1_avg=0.31027 acc_top5_avg=0.72879 lr=0.00100 gn=4.17468 time=56.71it/s +epoch=42 global_step=16500 loss=7.29904 loss_avg=6.76261 acc=0.24219 acc_top1_avg=0.31550 acc_top5_avg=0.73127 lr=0.00100 gn=4.93923 time=56.26it/s +epoch=42 global_step=16550 loss=6.81352 loss_avg=6.75262 acc=0.33594 acc_top1_avg=0.31720 acc_top5_avg=0.73181 lr=0.00100 gn=4.69840 time=55.81it/s +epoch=42 global_step=16600 loss=6.61078 loss_avg=6.75149 acc=0.35156 acc_top1_avg=0.31746 acc_top5_avg=0.73139 lr=0.00100 gn=4.57264 time=57.10it/s +epoch=42 global_step=16650 loss=6.83863 loss_avg=6.76920 acc=0.32031 acc_top1_avg=0.31507 acc_top5_avg=0.72797 lr=0.00100 gn=5.07561 time=62.72it/s +epoch=42 global_step=16700 loss=6.83111 loss_avg=6.75327 acc=0.30469 acc_top1_avg=0.31725 acc_top5_avg=0.72918 lr=0.00100 gn=4.10349 time=60.57it/s +epoch=42 global_step=16750 loss=6.84693 loss_avg=6.75664 acc=0.31250 acc_top1_avg=0.31681 acc_top5_avg=0.72966 lr=0.00100 gn=3.56543 time=55.83it/s +epoch=42 global_step=16800 loss=6.78852 loss_avg=6.76896 acc=0.29688 acc_top1_avg=0.31515 acc_top5_avg=0.72939 lr=0.00100 gn=3.98714 time=55.73it/s +====================Eval==================== +epoch=42 global_step=16813 loss=5.11127 test_loss_avg=3.37385 acc=0.00000 test_acc_avg=0.25293 test_acc_top5_avg=0.69336 time=238.31it/s +epoch=42 global_step=16813 loss=6.18940 test_loss_avg=3.15836 acc=0.00000 test_acc_avg=0.32545 test_acc_top5_avg=0.76642 time=719.68it/s +curr_acc 0.3255 +BEST_ACC 0.3338 +curr_acc_top5 0.7664 +BEST_ACC_top5 0.7659 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=7.03212 loss_avg=6.74643 acc=0.29688 acc_top1_avg=0.31715 acc_top5_avg=0.73269 lr=0.00100 gn=4.00439 time=60.95it/s +epoch=43 global_step=16900 loss=7.52700 loss_avg=6.75354 acc=0.22656 acc_top1_avg=0.31501 acc_top5_avg=0.72899 lr=0.00100 gn=4.24924 time=56.51it/s +epoch=43 global_step=16950 loss=6.53702 loss_avg=6.75936 acc=0.34375 acc_top1_avg=0.31569 acc_top5_avg=0.72993 lr=0.00100 gn=4.37073 time=48.47it/s +epoch=43 global_step=17000 loss=6.73957 loss_avg=6.75073 acc=0.29688 acc_top1_avg=0.31605 acc_top5_avg=0.73053 lr=0.00100 gn=4.12213 time=55.67it/s +epoch=43 global_step=17050 loss=7.35524 loss_avg=6.75315 acc=0.23438 acc_top1_avg=0.31632 acc_top5_avg=0.72890 lr=0.00100 gn=4.16546 time=55.88it/s +epoch=43 global_step=17100 loss=7.14185 loss_avg=6.75832 acc=0.27344 acc_top1_avg=0.31566 acc_top5_avg=0.73005 lr=0.00100 gn=5.46453 time=61.66it/s +epoch=43 global_step=17150 loss=6.55478 loss_avg=6.74688 acc=0.35156 acc_top1_avg=0.31677 acc_top5_avg=0.73053 lr=0.00100 gn=4.91705 time=61.59it/s +epoch=43 global_step=17200 loss=7.24162 loss_avg=6.74565 acc=0.25000 acc_top1_avg=0.31680 acc_top5_avg=0.73001 lr=0.00100 gn=3.58390 time=57.20it/s +====================Eval==================== +epoch=43 global_step=17204 loss=2.31954 test_loss_avg=2.22116 acc=0.35156 test_acc_avg=0.36979 test_acc_top5_avg=0.98438 time=233.72it/s +epoch=43 global_step=17204 loss=4.58677 test_loss_avg=3.42099 acc=0.00000 test_acc_avg=0.23320 test_acc_top5_avg=0.66981 time=220.75it/s +epoch=43 global_step=17204 loss=5.75079 test_loss_avg=3.04394 acc=0.00000 test_acc_avg=0.34138 test_acc_top5_avg=0.76899 time=511.13it/s +curr_acc 0.3414 +BEST_ACC 0.3338 +curr_acc_top5 0.7690 +BEST_ACC_top5 0.7664 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=6.73228 loss_avg=6.69750 acc=0.32812 acc_top1_avg=0.32065 acc_top5_avg=0.72486 lr=0.00100 gn=4.90418 time=56.68it/s +epoch=44 global_step=17300 loss=6.09651 loss_avg=6.72457 acc=0.39844 acc_top1_avg=0.31925 acc_top5_avg=0.73063 lr=0.00100 gn=5.75022 time=50.32it/s +epoch=44 global_step=17350 loss=6.64879 loss_avg=6.71695 acc=0.33594 acc_top1_avg=0.31946 acc_top5_avg=0.73122 lr=0.00100 gn=4.85642 time=63.17it/s +epoch=44 global_step=17400 loss=6.35319 loss_avg=6.74157 acc=0.33594 acc_top1_avg=0.31684 acc_top5_avg=0.73063 lr=0.00100 gn=5.13980 time=54.12it/s +epoch=44 global_step=17450 loss=6.95136 loss_avg=6.74385 acc=0.27344 acc_top1_avg=0.31682 acc_top5_avg=0.72961 lr=0.00100 gn=6.13044 time=60.63it/s +epoch=44 global_step=17500 loss=6.73871 loss_avg=6.72526 acc=0.32812 acc_top1_avg=0.31928 acc_top5_avg=0.73200 lr=0.00100 gn=4.60462 time=51.59it/s +epoch=44 global_step=17550 loss=6.46713 loss_avg=6.72269 acc=0.34375 acc_top1_avg=0.31930 acc_top5_avg=0.73261 lr=0.00100 gn=5.72342 time=57.89it/s +====================Eval==================== +epoch=44 global_step=17595 loss=5.00245 test_loss_avg=2.65378 acc=0.00000 test_acc_avg=0.35938 test_acc_top5_avg=0.85970 time=220.23it/s +epoch=44 global_step=17595 loss=5.38546 test_loss_avg=2.89419 acc=0.00000 test_acc_avg=0.36856 test_acc_top5_avg=0.76351 time=256.02it/s +epoch=44 global_step=17595 loss=5.43355 test_loss_avg=3.04944 acc=0.00000 test_acc_avg=0.34523 test_acc_top5_avg=0.77373 time=798.76it/s +curr_acc 0.3452 +BEST_ACC 0.3414 +curr_acc_top5 0.7737 +BEST_ACC_top5 0.7690 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=6.72415 loss_avg=6.85109 acc=0.32812 acc_top1_avg=0.30938 acc_top5_avg=0.74219 lr=0.00100 gn=6.00767 time=53.38it/s +epoch=45 global_step=17650 loss=6.39284 loss_avg=6.72677 acc=0.37500 acc_top1_avg=0.31690 acc_top5_avg=0.73182 lr=0.00100 gn=4.26607 time=56.35it/s +epoch=45 global_step=17700 loss=6.60155 loss_avg=6.70179 acc=0.32031 acc_top1_avg=0.32091 acc_top5_avg=0.73118 lr=0.00100 gn=4.61553 time=55.94it/s +epoch=45 global_step=17750 loss=6.73896 loss_avg=6.67465 acc=0.32031 acc_top1_avg=0.32429 acc_top5_avg=0.73317 lr=0.00100 gn=5.70642 time=61.91it/s +epoch=45 global_step=17800 loss=6.47557 loss_avg=6.69758 acc=0.34375 acc_top1_avg=0.32149 acc_top5_avg=0.73125 lr=0.00100 gn=5.00050 time=55.74it/s +epoch=45 global_step=17850 loss=7.02667 loss_avg=6.70775 acc=0.27344 acc_top1_avg=0.32028 acc_top5_avg=0.73116 lr=0.00100 gn=6.57980 time=55.49it/s +epoch=45 global_step=17900 loss=7.10228 loss_avg=6.72019 acc=0.26562 acc_top1_avg=0.31890 acc_top5_avg=0.73089 lr=0.00100 gn=5.54168 time=47.37it/s +epoch=45 global_step=17950 loss=6.87775 loss_avg=6.70705 acc=0.30469 acc_top1_avg=0.32044 acc_top5_avg=0.73118 lr=0.00100 gn=5.06355 time=54.24it/s +====================Eval==================== +epoch=45 global_step=17986 loss=1.81117 test_loss_avg=3.53300 acc=0.42969 test_acc_avg=0.22066 test_acc_top5_avg=0.64965 time=239.70it/s +epoch=45 global_step=17986 loss=6.04986 test_loss_avg=3.19566 acc=0.00000 test_acc_avg=0.32229 test_acc_top5_avg=0.78095 time=508.46it/s +curr_acc 0.3223 +BEST_ACC 0.3452 +curr_acc_top5 0.7810 +BEST_ACC_top5 0.7737 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=5.64484 loss_avg=6.66498 acc=0.45312 acc_top1_avg=0.32757 acc_top5_avg=0.72154 lr=0.00100 gn=6.38644 time=53.94it/s +epoch=46 global_step=18050 loss=6.45666 loss_avg=6.67796 acc=0.35156 acc_top1_avg=0.32642 acc_top5_avg=0.73401 lr=0.00100 gn=5.22518 time=50.30it/s +epoch=46 global_step=18100 loss=7.05807 loss_avg=6.67088 acc=0.27344 acc_top1_avg=0.32614 acc_top5_avg=0.73677 lr=0.00100 gn=6.00625 time=54.76it/s +epoch=46 global_step=18150 loss=6.72695 loss_avg=6.68845 acc=0.32812 acc_top1_avg=0.32441 acc_top5_avg=0.73452 lr=0.00100 gn=6.40365 time=54.16it/s +epoch=46 global_step=18200 loss=6.52319 loss_avg=6.67916 acc=0.32031 acc_top1_avg=0.32502 acc_top5_avg=0.73416 lr=0.00100 gn=5.82390 time=51.60it/s +epoch=46 global_step=18250 loss=5.96030 loss_avg=6.69231 acc=0.42969 acc_top1_avg=0.32339 acc_top5_avg=0.73396 lr=0.00100 gn=6.45311 time=58.13it/s +epoch=46 global_step=18300 loss=6.49929 loss_avg=6.68968 acc=0.34375 acc_top1_avg=0.32362 acc_top5_avg=0.73233 lr=0.00100 gn=5.16583 time=49.59it/s +epoch=46 global_step=18350 loss=7.09265 loss_avg=6.68769 acc=0.28125 acc_top1_avg=0.32377 acc_top5_avg=0.73399 lr=0.00100 gn=5.87900 time=54.93it/s +====================Eval==================== +epoch=46 global_step=18377 loss=2.25296 test_loss_avg=1.92762 acc=0.45312 test_acc_avg=0.50195 test_acc_top5_avg=0.97705 time=238.38it/s +epoch=46 global_step=18377 loss=0.05736 test_loss_avg=3.03489 acc=0.96875 test_acc_avg=0.33073 test_acc_top5_avg=0.74751 time=234.87it/s +epoch=46 global_step=18377 loss=5.95622 test_loss_avg=3.17454 acc=0.00000 test_acc_avg=0.33000 test_acc_top5_avg=0.78491 time=834.02it/s +curr_acc 0.3300 +BEST_ACC 0.3452 +curr_acc_top5 0.7849 +BEST_ACC_top5 0.7810 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=6.85157 loss_avg=6.58939 acc=0.30469 acc_top1_avg=0.33186 acc_top5_avg=0.73947 lr=0.00100 gn=4.58388 time=55.34it/s +epoch=47 global_step=18450 loss=6.29096 loss_avg=6.64970 acc=0.36719 acc_top1_avg=0.32545 acc_top5_avg=0.74005 lr=0.00100 gn=5.54427 time=55.37it/s +epoch=47 global_step=18500 loss=6.35904 loss_avg=6.66228 acc=0.34375 acc_top1_avg=0.32393 acc_top5_avg=0.73825 lr=0.00100 gn=5.88946 time=56.18it/s +epoch=47 global_step=18550 loss=6.65002 loss_avg=6.64938 acc=0.31250 acc_top1_avg=0.32587 acc_top5_avg=0.73632 lr=0.00100 gn=6.26741 time=52.68it/s +epoch=47 global_step=18600 loss=6.51484 loss_avg=6.65465 acc=0.35156 acc_top1_avg=0.32602 acc_top5_avg=0.73476 lr=0.00100 gn=7.35384 time=50.56it/s +epoch=47 global_step=18650 loss=6.17462 loss_avg=6.64643 acc=0.38281 acc_top1_avg=0.32718 acc_top5_avg=0.73546 lr=0.00100 gn=7.47993 time=60.42it/s +epoch=47 global_step=18700 loss=6.72862 loss_avg=6.64841 acc=0.32812 acc_top1_avg=0.32776 acc_top5_avg=0.73505 lr=0.00100 gn=6.29994 time=62.62it/s +epoch=47 global_step=18750 loss=6.49655 loss_avg=6.66669 acc=0.35156 acc_top1_avg=0.32578 acc_top5_avg=0.73360 lr=0.00100 gn=8.81363 time=54.05it/s +====================Eval==================== +epoch=47 global_step=18768 loss=5.20636 test_loss_avg=3.47884 acc=0.00000 test_acc_avg=0.24198 test_acc_top5_avg=0.65604 time=234.00it/s +epoch=47 global_step=18768 loss=5.22975 test_loss_avg=2.96623 acc=0.00000 test_acc_avg=0.35275 test_acc_top5_avg=0.78699 time=492.06it/s +curr_acc 0.3527 +BEST_ACC 0.3452 +curr_acc_top5 0.7870 +BEST_ACC_top5 0.7849 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=6.59980 loss_avg=6.66271 acc=0.32812 acc_top1_avg=0.32642 acc_top5_avg=0.72998 lr=0.00100 gn=5.41350 time=59.32it/s +epoch=48 global_step=18850 loss=6.81888 loss_avg=6.63573 acc=0.31250 acc_top1_avg=0.32870 acc_top5_avg=0.74009 lr=0.00100 gn=5.88537 time=55.77it/s +epoch=48 global_step=18900 loss=7.22527 loss_avg=6.67901 acc=0.26562 acc_top1_avg=0.32410 acc_top5_avg=0.73509 lr=0.00100 gn=6.33349 time=62.65it/s +epoch=48 global_step=18950 loss=6.04227 loss_avg=6.65759 acc=0.41406 acc_top1_avg=0.32667 acc_top5_avg=0.73704 lr=0.00100 gn=8.86902 time=54.51it/s +epoch=48 global_step=19000 loss=6.32395 loss_avg=6.66703 acc=0.35938 acc_top1_avg=0.32604 acc_top5_avg=0.73673 lr=0.00100 gn=7.50749 time=56.29it/s +epoch=48 global_step=19050 loss=6.61852 loss_avg=6.65052 acc=0.32031 acc_top1_avg=0.32860 acc_top5_avg=0.73709 lr=0.00100 gn=6.86037 time=54.19it/s +epoch=48 global_step=19100 loss=7.23220 loss_avg=6.64512 acc=0.27344 acc_top1_avg=0.32914 acc_top5_avg=0.73696 lr=0.00100 gn=8.14658 time=53.40it/s +epoch=48 global_step=19150 loss=6.58415 loss_avg=6.64307 acc=0.34375 acc_top1_avg=0.32972 acc_top5_avg=0.73611 lr=0.00100 gn=7.25270 time=55.39it/s +====================Eval==================== +epoch=48 global_step=19159 loss=2.44696 test_loss_avg=2.80596 acc=0.36719 test_acc_avg=0.26562 test_acc_top5_avg=0.97070 time=243.15it/s +epoch=48 global_step=19159 loss=0.31339 test_loss_avg=3.32256 acc=0.93750 test_acc_avg=0.27492 test_acc_top5_avg=0.70838 time=241.50it/s +epoch=48 global_step=19159 loss=5.77001 test_loss_avg=3.09327 acc=0.00000 test_acc_avg=0.35018 test_acc_top5_avg=0.77809 time=854.24it/s +curr_acc 0.3502 +BEST_ACC 0.3527 +curr_acc_top5 0.7781 +BEST_ACC_top5 0.7870 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=6.58825 loss_avg=6.61527 acc=0.33594 acc_top1_avg=0.33098 acc_top5_avg=0.73476 lr=0.00100 gn=6.94941 time=53.48it/s +epoch=49 global_step=19250 loss=6.42014 loss_avg=6.61861 acc=0.35938 acc_top1_avg=0.33001 acc_top5_avg=0.73438 lr=0.00100 gn=7.10955 time=57.35it/s +epoch=49 global_step=19300 loss=6.32700 loss_avg=6.64084 acc=0.38281 acc_top1_avg=0.32862 acc_top5_avg=0.73493 lr=0.00100 gn=7.08906 time=56.81it/s +epoch=49 global_step=19350 loss=6.89653 loss_avg=6.63519 acc=0.29688 acc_top1_avg=0.32976 acc_top5_avg=0.73425 lr=0.00100 gn=8.53850 time=60.64it/s +epoch=49 global_step=19400 loss=6.84680 loss_avg=6.65540 acc=0.29688 acc_top1_avg=0.32751 acc_top5_avg=0.73311 lr=0.00100 gn=7.47481 time=56.03it/s +epoch=49 global_step=19450 loss=6.85648 loss_avg=6.64426 acc=0.30469 acc_top1_avg=0.32847 acc_top5_avg=0.73429 lr=0.00100 gn=6.72523 time=55.18it/s +epoch=49 global_step=19500 loss=7.01004 loss_avg=6.64371 acc=0.27344 acc_top1_avg=0.32833 acc_top5_avg=0.73403 lr=0.00100 gn=4.95266 time=56.85it/s +epoch=49 global_step=19550 loss=6.62353 loss_avg=6.63521 acc=0.32500 acc_top1_avg=0.32956 acc_top5_avg=0.73512 lr=0.00100 gn=10.26630 time=78.20it/s +====================Eval==================== +epoch=49 global_step=19550 loss=5.41432 test_loss_avg=3.00470 acc=0.00000 test_acc_avg=0.32651 test_acc_top5_avg=0.80011 time=239.73it/s +epoch=49 global_step=19550 loss=5.79846 test_loss_avg=3.04987 acc=0.00000 test_acc_avg=0.34494 test_acc_top5_avg=0.78758 time=490.22it/s +epoch=49 global_step=19550 loss=5.79846 test_loss_avg=3.04987 acc=0.00000 test_acc_avg=0.34494 test_acc_top5_avg=0.78758 time=490.22it/s +curr_acc 0.3449 +BEST_ACC 0.3527 +curr_acc_top5 0.7876 +BEST_ACC_top5 0.7870 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=6.50849 loss_avg=6.73468 acc=0.34375 acc_top1_avg=0.32172 acc_top5_avg=0.71828 lr=0.00100 gn=6.38689 time=59.73it/s +epoch=50 global_step=19650 loss=6.11349 loss_avg=6.63219 acc=0.39844 acc_top1_avg=0.33227 acc_top5_avg=0.73109 lr=0.00100 gn=7.60716 time=56.68it/s +epoch=50 global_step=19700 loss=6.70479 loss_avg=6.64048 acc=0.32031 acc_top1_avg=0.33109 acc_top5_avg=0.73193 lr=0.00100 gn=6.78826 time=56.68it/s +epoch=50 global_step=19750 loss=6.62696 loss_avg=6.62784 acc=0.31250 acc_top1_avg=0.33160 acc_top5_avg=0.73410 lr=0.00100 gn=8.00963 time=53.27it/s +epoch=50 global_step=19800 loss=6.90224 loss_avg=6.61589 acc=0.28125 acc_top1_avg=0.33325 acc_top5_avg=0.73366 lr=0.00100 gn=7.82273 time=56.94it/s +epoch=50 global_step=19850 loss=6.95209 loss_avg=6.61330 acc=0.28906 acc_top1_avg=0.33354 acc_top5_avg=0.73294 lr=0.00100 gn=9.96402 time=55.11it/s +epoch=50 global_step=19900 loss=6.05542 loss_avg=6.61236 acc=0.40625 acc_top1_avg=0.33306 acc_top5_avg=0.73290 lr=0.00100 gn=8.34781 time=62.94it/s +====================Eval==================== +epoch=50 global_step=19941 loss=4.39493 test_loss_avg=3.31617 acc=0.00000 test_acc_avg=0.26438 test_acc_top5_avg=0.69469 time=240.15it/s +epoch=50 global_step=19941 loss=5.87082 test_loss_avg=3.01616 acc=0.00000 test_acc_avg=0.35255 test_acc_top5_avg=0.79252 time=749.12it/s +curr_acc 0.3526 +BEST_ACC 0.3527 +curr_acc_top5 0.7925 +BEST_ACC_top5 0.7876 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=7.42738 loss_avg=6.73996 acc=0.25000 acc_top1_avg=0.32292 acc_top5_avg=0.71615 lr=0.00100 gn=6.96036 time=54.80it/s +epoch=51 global_step=20000 loss=6.49780 loss_avg=6.59211 acc=0.33594 acc_top1_avg=0.33647 acc_top5_avg=0.73914 lr=0.00100 gn=7.66200 time=54.43it/s +epoch=51 global_step=20050 loss=6.63396 loss_avg=6.56895 acc=0.34375 acc_top1_avg=0.33866 acc_top5_avg=0.74147 lr=0.00100 gn=7.72087 time=52.80it/s +epoch=51 global_step=20100 loss=6.68412 loss_avg=6.55445 acc=0.32812 acc_top1_avg=0.34056 acc_top5_avg=0.74003 lr=0.00100 gn=10.37627 time=50.64it/s +epoch=51 global_step=20150 loss=7.06447 loss_avg=6.56580 acc=0.29688 acc_top1_avg=0.33900 acc_top5_avg=0.73666 lr=0.00100 gn=8.69184 time=53.53it/s +epoch=51 global_step=20200 loss=6.30549 loss_avg=6.57383 acc=0.37500 acc_top1_avg=0.33799 acc_top5_avg=0.73667 lr=0.00100 gn=7.77856 time=51.60it/s +epoch=51 global_step=20250 loss=7.26708 loss_avg=6.58204 acc=0.28125 acc_top1_avg=0.33713 acc_top5_avg=0.73518 lr=0.00100 gn=8.90169 time=58.46it/s +epoch=51 global_step=20300 loss=6.65342 loss_avg=6.58616 acc=0.31250 acc_top1_avg=0.33631 acc_top5_avg=0.73636 lr=0.00100 gn=8.74655 time=52.17it/s +====================Eval==================== +epoch=51 global_step=20332 loss=4.47111 test_loss_avg=2.44257 acc=0.00000 test_acc_avg=0.40997 test_acc_top5_avg=0.90476 time=238.49it/s +epoch=51 global_step=20332 loss=4.08700 test_loss_avg=2.80471 acc=0.31250 test_acc_avg=0.38281 test_acc_top5_avg=0.76342 time=241.80it/s +epoch=51 global_step=20332 loss=6.17664 test_loss_avg=3.09704 acc=0.00000 test_acc_avg=0.34405 test_acc_top5_avg=0.78214 time=677.05it/s +curr_acc 0.3440 +BEST_ACC 0.3527 +curr_acc_top5 0.7821 +BEST_ACC_top5 0.7925 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=6.95941 loss_avg=6.56583 acc=0.28906 acc_top1_avg=0.33767 acc_top5_avg=0.73134 lr=0.00100 gn=8.55072 time=57.19it/s +epoch=52 global_step=20400 loss=6.44837 loss_avg=6.56103 acc=0.35938 acc_top1_avg=0.33824 acc_top5_avg=0.73506 lr=0.00100 gn=8.43692 time=59.67it/s +epoch=52 global_step=20450 loss=5.93032 loss_avg=6.53175 acc=0.38281 acc_top1_avg=0.34090 acc_top5_avg=0.73947 lr=0.00100 gn=8.33355 time=52.55it/s +epoch=52 global_step=20500 loss=7.11972 loss_avg=6.54570 acc=0.25000 acc_top1_avg=0.33980 acc_top5_avg=0.73921 lr=0.00100 gn=8.35881 time=58.80it/s +epoch=52 global_step=20550 loss=6.50652 loss_avg=6.55937 acc=0.35938 acc_top1_avg=0.33791 acc_top5_avg=0.73846 lr=0.00100 gn=8.40199 time=55.04it/s +epoch=52 global_step=20600 loss=6.78850 loss_avg=6.55838 acc=0.30469 acc_top1_avg=0.33804 acc_top5_avg=0.73747 lr=0.00100 gn=9.83079 time=54.90it/s +epoch=52 global_step=20650 loss=6.97177 loss_avg=6.55718 acc=0.27344 acc_top1_avg=0.33849 acc_top5_avg=0.73691 lr=0.00100 gn=10.25137 time=47.93it/s +epoch=52 global_step=20700 loss=6.59300 loss_avg=6.56457 acc=0.32812 acc_top1_avg=0.33759 acc_top5_avg=0.73716 lr=0.00100 gn=9.76650 time=55.56it/s +====================Eval==================== +epoch=52 global_step=20723 loss=1.42072 test_loss_avg=3.36373 acc=0.57812 test_acc_avg=0.27269 test_acc_top5_avg=0.64546 time=239.16it/s +epoch=52 global_step=20723 loss=5.28793 test_loss_avg=2.90525 acc=0.00000 test_acc_avg=0.36669 test_acc_top5_avg=0.78491 time=821.93it/s +curr_acc 0.3667 +BEST_ACC 0.3527 +curr_acc_top5 0.7849 +BEST_ACC_top5 0.7925 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=5.91094 loss_avg=6.45862 acc=0.40625 acc_top1_avg=0.34664 acc_top5_avg=0.74248 lr=0.00100 gn=10.45165 time=54.43it/s +epoch=53 global_step=20800 loss=6.53526 loss_avg=6.50896 acc=0.33594 acc_top1_avg=0.34659 acc_top5_avg=0.73417 lr=0.00100 gn=11.71692 time=56.14it/s +epoch=53 global_step=20850 loss=6.66272 loss_avg=6.51401 acc=0.32031 acc_top1_avg=0.34504 acc_top5_avg=0.73893 lr=0.00100 gn=8.54168 time=53.39it/s +epoch=53 global_step=20900 loss=5.86974 loss_avg=6.55159 acc=0.42188 acc_top1_avg=0.34057 acc_top5_avg=0.73574 lr=0.00100 gn=9.67192 time=63.02it/s +epoch=53 global_step=20950 loss=6.31702 loss_avg=6.53818 acc=0.37500 acc_top1_avg=0.34196 acc_top5_avg=0.73393 lr=0.00100 gn=8.98257 time=53.28it/s +epoch=53 global_step=21000 loss=6.43665 loss_avg=6.54039 acc=0.35156 acc_top1_avg=0.34132 acc_top5_avg=0.73344 lr=0.00100 gn=10.56071 time=52.33it/s +epoch=53 global_step=21050 loss=7.42087 loss_avg=6.55692 acc=0.24219 acc_top1_avg=0.33945 acc_top5_avg=0.73347 lr=0.00100 gn=8.74375 time=60.22it/s +epoch=53 global_step=21100 loss=6.44557 loss_avg=6.54999 acc=0.35156 acc_top1_avg=0.34085 acc_top5_avg=0.73454 lr=0.00100 gn=10.13091 time=57.55it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.49981 test_loss_avg=1.52153 acc=0.86719 test_acc_avg=0.56250 test_acc_top5_avg=0.99099 time=237.62it/s +epoch=53 global_step=21114 loss=0.22574 test_loss_avg=3.11747 acc=0.94531 test_acc_avg=0.31300 test_acc_top5_avg=0.75124 time=238.96it/s +epoch=53 global_step=21114 loss=5.40158 test_loss_avg=3.05518 acc=0.00000 test_acc_avg=0.33890 test_acc_top5_avg=0.79391 time=834.52it/s +curr_acc 0.3389 +BEST_ACC 0.3667 +curr_acc_top5 0.7939 +BEST_ACC_top5 0.7925 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=6.73855 loss_avg=6.54776 acc=0.29688 acc_top1_avg=0.33963 acc_top5_avg=0.73438 lr=0.00100 gn=11.16765 time=56.33it/s +epoch=54 global_step=21200 loss=5.77781 loss_avg=6.51055 acc=0.43750 acc_top1_avg=0.34475 acc_top5_avg=0.73910 lr=0.00100 gn=11.78376 time=55.28it/s +epoch=54 global_step=21250 loss=6.39860 loss_avg=6.49126 acc=0.36719 acc_top1_avg=0.34616 acc_top5_avg=0.73874 lr=0.00100 gn=13.92697 time=51.83it/s +epoch=54 global_step=21300 loss=6.06883 loss_avg=6.48871 acc=0.38281 acc_top1_avg=0.34661 acc_top5_avg=0.73769 lr=0.00100 gn=12.65927 time=49.00it/s +epoch=54 global_step=21350 loss=6.73358 loss_avg=6.52123 acc=0.31250 acc_top1_avg=0.34223 acc_top5_avg=0.73577 lr=0.00100 gn=11.24719 time=57.62it/s +epoch=54 global_step=21400 loss=6.56536 loss_avg=6.52957 acc=0.33594 acc_top1_avg=0.34173 acc_top5_avg=0.73700 lr=0.00100 gn=9.06357 time=53.20it/s +epoch=54 global_step=21450 loss=6.64614 loss_avg=6.53527 acc=0.33594 acc_top1_avg=0.34101 acc_top5_avg=0.73621 lr=0.00100 gn=11.79548 time=61.46it/s +epoch=54 global_step=21500 loss=6.53517 loss_avg=6.53262 acc=0.33594 acc_top1_avg=0.34126 acc_top5_avg=0.73605 lr=0.00100 gn=8.24223 time=62.98it/s +====================Eval==================== +epoch=54 global_step=21505 loss=5.10947 test_loss_avg=3.29397 acc=0.00000 test_acc_avg=0.28424 test_acc_top5_avg=0.72197 time=212.14it/s +epoch=54 global_step=21505 loss=5.87969 test_loss_avg=3.00749 acc=0.00000 test_acc_avg=0.35957 test_acc_top5_avg=0.79183 time=479.51it/s +curr_acc 0.3596 +BEST_ACC 0.3667 +curr_acc_top5 0.7918 +BEST_ACC_top5 0.7939 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=6.64962 loss_avg=6.47662 acc=0.32812 acc_top1_avg=0.34983 acc_top5_avg=0.74809 lr=0.00100 gn=12.46948 time=59.91it/s +epoch=55 global_step=21600 loss=6.46145 loss_avg=6.45546 acc=0.32812 acc_top1_avg=0.35164 acc_top5_avg=0.74211 lr=0.00100 gn=10.29002 time=55.49it/s +epoch=55 global_step=21650 loss=6.59622 loss_avg=6.51784 acc=0.33594 acc_top1_avg=0.34353 acc_top5_avg=0.73594 lr=0.00100 gn=12.17098 time=54.80it/s +epoch=55 global_step=21700 loss=6.69437 loss_avg=6.51310 acc=0.31250 acc_top1_avg=0.34347 acc_top5_avg=0.73646 lr=0.00100 gn=9.36288 time=54.43it/s +epoch=55 global_step=21750 loss=6.03077 loss_avg=6.49803 acc=0.39844 acc_top1_avg=0.34528 acc_top5_avg=0.73591 lr=0.00100 gn=10.94884 time=58.76it/s +epoch=55 global_step=21800 loss=6.85700 loss_avg=6.51067 acc=0.32031 acc_top1_avg=0.34396 acc_top5_avg=0.73615 lr=0.00100 gn=10.99575 time=56.93it/s +epoch=55 global_step=21850 loss=6.52465 loss_avg=6.51858 acc=0.33594 acc_top1_avg=0.34309 acc_top5_avg=0.73451 lr=0.00100 gn=10.00484 time=59.80it/s +====================Eval==================== +epoch=55 global_step=21896 loss=1.96447 test_loss_avg=2.06642 acc=0.40625 test_acc_avg=0.40937 test_acc_top5_avg=0.96875 time=248.43it/s +epoch=55 global_step=21896 loss=3.09109 test_loss_avg=3.35023 acc=0.27344 test_acc_avg=0.25767 test_acc_top5_avg=0.70639 time=237.97it/s +epoch=55 global_step=21896 loss=6.00747 test_loss_avg=2.98030 acc=0.00000 test_acc_avg=0.36135 test_acc_top5_avg=0.78877 time=495.96it/s +curr_acc 0.3614 +BEST_ACC 0.3667 +curr_acc_top5 0.7888 +BEST_ACC_top5 0.7939 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=7.27935 loss_avg=6.58670 acc=0.25781 acc_top1_avg=0.32812 acc_top5_avg=0.72266 lr=0.00100 gn=10.31055 time=51.48it/s +epoch=56 global_step=21950 loss=6.37836 loss_avg=6.48576 acc=0.36719 acc_top1_avg=0.34520 acc_top5_avg=0.73582 lr=0.00100 gn=12.28437 time=55.43it/s +epoch=56 global_step=22000 loss=6.49201 loss_avg=6.43874 acc=0.37500 acc_top1_avg=0.35329 acc_top5_avg=0.74264 lr=0.00100 gn=13.18606 time=51.94it/s +epoch=56 global_step=22050 loss=6.56693 loss_avg=6.45890 acc=0.34375 acc_top1_avg=0.35080 acc_top5_avg=0.73869 lr=0.00100 gn=11.99021 time=59.54it/s +epoch=56 global_step=22100 loss=6.08418 loss_avg=6.46700 acc=0.36719 acc_top1_avg=0.34953 acc_top5_avg=0.73809 lr=0.00100 gn=10.26266 time=56.83it/s +epoch=56 global_step=22150 loss=6.06594 loss_avg=6.47738 acc=0.39844 acc_top1_avg=0.34799 acc_top5_avg=0.73570 lr=0.00100 gn=15.38351 time=56.01it/s +epoch=56 global_step=22200 loss=6.24010 loss_avg=6.49197 acc=0.37500 acc_top1_avg=0.34622 acc_top5_avg=0.73533 lr=0.00100 gn=12.32412 time=63.42it/s +epoch=56 global_step=22250 loss=6.39286 loss_avg=6.50480 acc=0.35156 acc_top1_avg=0.34505 acc_top5_avg=0.73497 lr=0.00100 gn=12.32308 time=63.13it/s +====================Eval==================== +epoch=56 global_step=22287 loss=5.16629 test_loss_avg=2.59059 acc=0.00000 test_acc_avg=0.40625 test_acc_top5_avg=0.85877 time=237.66it/s +epoch=56 global_step=22287 loss=5.32753 test_loss_avg=2.88262 acc=0.00000 test_acc_avg=0.37079 test_acc_top5_avg=0.78937 time=253.14it/s +epoch=56 global_step=22287 loss=5.91699 test_loss_avg=2.97879 acc=0.00000 test_acc_avg=0.35670 test_acc_top5_avg=0.79569 time=807.68it/s +curr_acc 0.3567 +BEST_ACC 0.3667 +curr_acc_top5 0.7957 +BEST_ACC_top5 0.7939 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=6.29079 loss_avg=6.62002 acc=0.35156 acc_top1_avg=0.33173 acc_top5_avg=0.73978 lr=0.00100 gn=13.52760 time=55.92it/s +epoch=57 global_step=22350 loss=6.73275 loss_avg=6.47664 acc=0.32812 acc_top1_avg=0.34846 acc_top5_avg=0.73599 lr=0.00100 gn=12.71207 time=54.31it/s +epoch=57 global_step=22400 loss=6.22038 loss_avg=6.44484 acc=0.37500 acc_top1_avg=0.35205 acc_top5_avg=0.73977 lr=0.00100 gn=12.26195 time=54.84it/s +epoch=57 global_step=22450 loss=6.66718 loss_avg=6.43808 acc=0.33594 acc_top1_avg=0.35219 acc_top5_avg=0.73787 lr=0.00100 gn=11.33250 time=59.96it/s +epoch=57 global_step=22500 loss=6.60791 loss_avg=6.46365 acc=0.33594 acc_top1_avg=0.34903 acc_top5_avg=0.73698 lr=0.00100 gn=13.44848 time=54.35it/s +epoch=57 global_step=22550 loss=6.66540 loss_avg=6.46124 acc=0.32031 acc_top1_avg=0.34904 acc_top5_avg=0.73746 lr=0.00100 gn=10.78721 time=55.29it/s +epoch=57 global_step=22600 loss=6.67471 loss_avg=6.46710 acc=0.32812 acc_top1_avg=0.34887 acc_top5_avg=0.73705 lr=0.00100 gn=12.16381 time=60.62it/s +epoch=57 global_step=22650 loss=6.98988 loss_avg=6.47394 acc=0.30469 acc_top1_avg=0.34836 acc_top5_avg=0.73648 lr=0.00100 gn=12.66297 time=57.62it/s +====================Eval==================== +epoch=57 global_step=22678 loss=2.37004 test_loss_avg=3.44244 acc=0.35938 test_acc_avg=0.25449 test_acc_top5_avg=0.68767 time=236.01it/s +epoch=57 global_step=22678 loss=6.56292 test_loss_avg=3.18663 acc=0.00000 test_acc_avg=0.33485 test_acc_top5_avg=0.79411 time=854.24it/s +curr_acc 0.3348 +BEST_ACC 0.3667 +curr_acc_top5 0.7941 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=6.47058 loss_avg=6.35353 acc=0.33594 acc_top1_avg=0.36293 acc_top5_avg=0.73935 lr=0.00100 gn=10.76315 time=59.63it/s +epoch=58 global_step=22750 loss=7.06974 loss_avg=6.37132 acc=0.28125 acc_top1_avg=0.36198 acc_top5_avg=0.73633 lr=0.00100 gn=14.73392 time=46.52it/s +epoch=58 global_step=22800 loss=6.85733 loss_avg=6.39164 acc=0.29688 acc_top1_avg=0.35950 acc_top5_avg=0.73655 lr=0.00100 gn=12.71948 time=54.42it/s +epoch=58 global_step=22850 loss=6.07449 loss_avg=6.41512 acc=0.42188 acc_top1_avg=0.35697 acc_top5_avg=0.73596 lr=0.00100 gn=15.45072 time=54.59it/s +epoch=58 global_step=22900 loss=6.16922 loss_avg=6.43250 acc=0.38281 acc_top1_avg=0.35515 acc_top5_avg=0.73455 lr=0.00100 gn=11.65071 time=58.89it/s +epoch=58 global_step=22950 loss=7.08978 loss_avg=6.44785 acc=0.28125 acc_top1_avg=0.35308 acc_top5_avg=0.73394 lr=0.00100 gn=17.29837 time=51.59it/s +epoch=58 global_step=23000 loss=6.20033 loss_avg=6.44920 acc=0.38281 acc_top1_avg=0.35292 acc_top5_avg=0.73399 lr=0.00100 gn=16.45798 time=60.86it/s +epoch=58 global_step=23050 loss=6.32054 loss_avg=6.44843 acc=0.37500 acc_top1_avg=0.35276 acc_top5_avg=0.73526 lr=0.00100 gn=15.45154 time=52.41it/s +====================Eval==================== +epoch=58 global_step=23069 loss=4.82743 test_loss_avg=2.05711 acc=0.00000 test_acc_avg=0.48047 test_acc_top5_avg=0.93533 time=149.64it/s +epoch=58 global_step=23069 loss=0.12442 test_loss_avg=2.90405 acc=0.97656 test_acc_avg=0.36156 test_acc_top5_avg=0.76195 time=243.85it/s +epoch=58 global_step=23069 loss=6.59319 test_loss_avg=3.16730 acc=0.00000 test_acc_avg=0.33979 test_acc_top5_avg=0.79074 time=486.24it/s +curr_acc 0.3398 +BEST_ACC 0.3667 +curr_acc_top5 0.7907 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=6.14617 loss_avg=6.37558 acc=0.42188 acc_top1_avg=0.36064 acc_top5_avg=0.73311 lr=0.00100 gn=15.86473 time=55.86it/s +epoch=59 global_step=23150 loss=6.62071 loss_avg=6.42408 acc=0.33594 acc_top1_avg=0.35465 acc_top5_avg=0.72820 lr=0.00100 gn=13.87082 time=57.35it/s +epoch=59 global_step=23200 loss=5.88607 loss_avg=6.41917 acc=0.42969 acc_top1_avg=0.35526 acc_top5_avg=0.73503 lr=0.00100 gn=16.37494 time=53.78it/s +epoch=59 global_step=23250 loss=7.09570 loss_avg=6.41847 acc=0.29688 acc_top1_avg=0.35584 acc_top5_avg=0.73360 lr=0.00100 gn=19.11691 time=57.54it/s +epoch=59 global_step=23300 loss=6.37932 loss_avg=6.41765 acc=0.35938 acc_top1_avg=0.35589 acc_top5_avg=0.73333 lr=0.00100 gn=12.79776 time=52.63it/s +epoch=59 global_step=23350 loss=6.47387 loss_avg=6.43595 acc=0.34375 acc_top1_avg=0.35376 acc_top5_avg=0.73290 lr=0.00100 gn=14.21635 time=52.50it/s +epoch=59 global_step=23400 loss=6.41014 loss_avg=6.43729 acc=0.33594 acc_top1_avg=0.35355 acc_top5_avg=0.73423 lr=0.00100 gn=12.03658 time=55.63it/s +epoch=59 global_step=23450 loss=6.18331 loss_avg=6.44443 acc=0.38281 acc_top1_avg=0.35228 acc_top5_avg=0.73358 lr=0.00100 gn=16.46258 time=58.78it/s +====================Eval==================== +epoch=59 global_step=23460 loss=5.32876 test_loss_avg=3.55637 acc=0.00000 test_acc_avg=0.26663 test_acc_top5_avg=0.65024 time=233.89it/s +epoch=59 global_step=23460 loss=6.51750 test_loss_avg=3.01780 acc=0.00000 test_acc_avg=0.37174 test_acc_top5_avg=0.79341 time=508.03it/s +curr_acc 0.3717 +BEST_ACC 0.3667 +curr_acc_top5 0.7934 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=6.77997 loss_avg=6.33325 acc=0.32812 acc_top1_avg=0.36641 acc_top5_avg=0.74316 lr=0.00100 gn=17.20928 time=58.11it/s +epoch=60 global_step=23550 loss=6.98074 loss_avg=6.41725 acc=0.29688 acc_top1_avg=0.35660 acc_top5_avg=0.74036 lr=0.00100 gn=12.21010 time=54.91it/s +epoch=60 global_step=23600 loss=6.48407 loss_avg=6.42280 acc=0.34375 acc_top1_avg=0.35502 acc_top5_avg=0.73867 lr=0.00100 gn=14.84971 time=55.31it/s +epoch=60 global_step=23650 loss=6.15964 loss_avg=6.41893 acc=0.39844 acc_top1_avg=0.35543 acc_top5_avg=0.73779 lr=0.00100 gn=12.60927 time=63.12it/s +epoch=60 global_step=23700 loss=6.48763 loss_avg=6.42504 acc=0.33594 acc_top1_avg=0.35518 acc_top5_avg=0.73675 lr=0.00100 gn=13.66232 time=59.33it/s +epoch=60 global_step=23750 loss=6.21709 loss_avg=6.43151 acc=0.36719 acc_top1_avg=0.35485 acc_top5_avg=0.73712 lr=0.00100 gn=12.51421 time=59.10it/s +epoch=60 global_step=23800 loss=5.97878 loss_avg=6.43377 acc=0.41406 acc_top1_avg=0.35466 acc_top5_avg=0.73706 lr=0.00100 gn=14.48197 time=62.59it/s +epoch=60 global_step=23850 loss=6.63459 loss_avg=6.42982 acc=0.35156 acc_top1_avg=0.35537 acc_top5_avg=0.73762 lr=0.00100 gn=16.77779 time=63.57it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.43592 test_loss_avg=2.36499 acc=0.83594 test_acc_avg=0.37734 test_acc_top5_avg=0.97266 time=235.49it/s +epoch=60 global_step=23851 loss=0.25211 test_loss_avg=3.29925 acc=0.87500 test_acc_avg=0.28503 test_acc_top5_avg=0.73984 time=223.59it/s +epoch=60 global_step=23851 loss=5.42063 test_loss_avg=3.10775 acc=0.00000 test_acc_avg=0.34049 test_acc_top5_avg=0.79559 time=590.41it/s +curr_acc 0.3405 +BEST_ACC 0.3717 +curr_acc_top5 0.7956 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=6.35378 loss_avg=6.37535 acc=0.38281 acc_top1_avg=0.36336 acc_top5_avg=0.74809 lr=0.00100 gn=16.87480 time=55.01it/s +epoch=61 global_step=23950 loss=6.05392 loss_avg=6.37070 acc=0.42188 acc_top1_avg=0.36332 acc_top5_avg=0.74471 lr=0.00100 gn=15.07853 time=52.79it/s +epoch=61 global_step=24000 loss=5.88786 loss_avg=6.40137 acc=0.42188 acc_top1_avg=0.35958 acc_top5_avg=0.74035 lr=0.00100 gn=18.46149 time=56.20it/s +epoch=61 global_step=24050 loss=6.35157 loss_avg=6.41646 acc=0.35938 acc_top1_avg=0.35780 acc_top5_avg=0.73685 lr=0.00100 gn=17.63033 time=62.45it/s +epoch=61 global_step=24100 loss=6.89180 loss_avg=6.40104 acc=0.31250 acc_top1_avg=0.35985 acc_top5_avg=0.73776 lr=0.00100 gn=16.20845 time=60.85it/s +epoch=61 global_step=24150 loss=6.37440 loss_avg=6.40020 acc=0.38281 acc_top1_avg=0.35938 acc_top5_avg=0.73850 lr=0.00100 gn=17.85637 time=57.17it/s +epoch=61 global_step=24200 loss=6.19686 loss_avg=6.40358 acc=0.36719 acc_top1_avg=0.35868 acc_top5_avg=0.73684 lr=0.00100 gn=14.08038 time=62.87it/s +====================Eval==================== +epoch=61 global_step=24242 loss=5.25015 test_loss_avg=3.24923 acc=0.00000 test_acc_avg=0.27117 test_acc_top5_avg=0.74068 time=241.16it/s +epoch=61 global_step=24242 loss=6.17733 test_loss_avg=3.08384 acc=0.00000 test_acc_avg=0.33890 test_acc_top5_avg=0.78501 time=841.22it/s +curr_acc 0.3389 +BEST_ACC 0.3717 +curr_acc_top5 0.7850 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=6.56616 loss_avg=6.56065 acc=0.32812 acc_top1_avg=0.33496 acc_top5_avg=0.73828 lr=0.00100 gn=15.82135 time=60.41it/s +epoch=62 global_step=24300 loss=6.90749 loss_avg=6.34838 acc=0.31250 acc_top1_avg=0.36288 acc_top5_avg=0.73478 lr=0.00100 gn=17.41408 time=56.04it/s +epoch=62 global_step=24350 loss=6.79170 loss_avg=6.37108 acc=0.32031 acc_top1_avg=0.36053 acc_top5_avg=0.73488 lr=0.00100 gn=17.88008 time=55.12it/s +epoch=62 global_step=24400 loss=6.19936 loss_avg=6.37099 acc=0.37500 acc_top1_avg=0.36111 acc_top5_avg=0.73423 lr=0.00100 gn=15.86541 time=53.56it/s +epoch=62 global_step=24450 loss=6.77868 loss_avg=6.37927 acc=0.31250 acc_top1_avg=0.36080 acc_top5_avg=0.73558 lr=0.00100 gn=14.77442 time=59.86it/s +epoch=62 global_step=24500 loss=6.66682 loss_avg=6.39305 acc=0.32812 acc_top1_avg=0.35919 acc_top5_avg=0.73580 lr=0.00100 gn=14.75297 time=57.72it/s +epoch=62 global_step=24550 loss=6.64924 loss_avg=6.40536 acc=0.31250 acc_top1_avg=0.35795 acc_top5_avg=0.73503 lr=0.00100 gn=17.10187 time=59.91it/s +epoch=62 global_step=24600 loss=5.51510 loss_avg=6.38619 acc=0.46094 acc_top1_avg=0.36025 acc_top5_avg=0.73614 lr=0.00100 gn=20.07286 time=58.45it/s +====================Eval==================== +epoch=62 global_step=24633 loss=2.73824 test_loss_avg=2.76196 acc=0.31250 test_acc_avg=0.27344 test_acc_top5_avg=0.96094 time=123.69it/s +epoch=62 global_step=24633 loss=4.69893 test_loss_avg=3.50773 acc=0.00000 test_acc_avg=0.23948 test_acc_top5_avg=0.70403 time=230.43it/s +epoch=62 global_step=24633 loss=6.67207 test_loss_avg=3.15468 acc=0.00000 test_acc_avg=0.33910 test_acc_top5_avg=0.79292 time=641.23it/s +curr_acc 0.3391 +BEST_ACC 0.3717 +curr_acc_top5 0.7929 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=6.44375 loss_avg=6.36051 acc=0.34375 acc_top1_avg=0.36259 acc_top5_avg=0.73346 lr=0.00100 gn=19.21092 time=59.79it/s +epoch=63 global_step=24700 loss=5.65023 loss_avg=6.31433 acc=0.45312 acc_top1_avg=0.36847 acc_top5_avg=0.73822 lr=0.00100 gn=17.27178 time=48.68it/s +epoch=63 global_step=24750 loss=6.11112 loss_avg=6.30886 acc=0.37500 acc_top1_avg=0.36839 acc_top5_avg=0.73558 lr=0.00100 gn=16.38841 time=56.81it/s +epoch=63 global_step=24800 loss=5.86800 loss_avg=6.31146 acc=0.44531 acc_top1_avg=0.36887 acc_top5_avg=0.73690 lr=0.00100 gn=21.09329 time=58.97it/s +epoch=63 global_step=24850 loss=6.16718 loss_avg=6.33162 acc=0.37500 acc_top1_avg=0.36676 acc_top5_avg=0.73578 lr=0.00100 gn=19.89255 time=53.94it/s +epoch=63 global_step=24900 loss=7.10739 loss_avg=6.34719 acc=0.28906 acc_top1_avg=0.36508 acc_top5_avg=0.73476 lr=0.00100 gn=19.33546 time=52.24it/s +epoch=63 global_step=24950 loss=6.12299 loss_avg=6.35870 acc=0.39062 acc_top1_avg=0.36364 acc_top5_avg=0.73332 lr=0.00100 gn=15.16042 time=58.63it/s +epoch=63 global_step=25000 loss=7.04198 loss_avg=6.35863 acc=0.28906 acc_top1_avg=0.36357 acc_top5_avg=0.73320 lr=0.00100 gn=17.29154 time=51.65it/s +====================Eval==================== +epoch=63 global_step=25024 loss=4.56820 test_loss_avg=2.43157 acc=0.00000 test_acc_avg=0.42833 test_acc_top5_avg=0.86651 time=186.55it/s +epoch=63 global_step=25024 loss=5.42388 test_loss_avg=2.89971 acc=0.00000 test_acc_avg=0.37147 test_acc_top5_avg=0.78521 time=253.72it/s +epoch=63 global_step=25024 loss=5.88636 test_loss_avg=3.10846 acc=0.00000 test_acc_avg=0.34326 test_acc_top5_avg=0.79875 time=830.06it/s +curr_acc 0.3433 +BEST_ACC 0.3717 +curr_acc_top5 0.7988 +BEST_ACC_top5 0.7957 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=5.49823 loss_avg=6.25888 acc=0.44531 acc_top1_avg=0.37861 acc_top5_avg=0.73528 lr=0.00100 gn=20.72358 time=51.65it/s +epoch=64 global_step=25100 loss=5.94409 loss_avg=6.31047 acc=0.39062 acc_top1_avg=0.37079 acc_top5_avg=0.73417 lr=0.00100 gn=21.09261 time=60.44it/s +epoch=64 global_step=25150 loss=6.26976 loss_avg=6.28283 acc=0.38281 acc_top1_avg=0.37277 acc_top5_avg=0.73568 lr=0.00100 gn=19.33382 time=51.19it/s +epoch=64 global_step=25200 loss=5.78512 loss_avg=6.29764 acc=0.42969 acc_top1_avg=0.37043 acc_top5_avg=0.73806 lr=0.00100 gn=18.39981 time=54.99it/s +epoch=64 global_step=25250 loss=6.45587 loss_avg=6.32532 acc=0.34375 acc_top1_avg=0.36746 acc_top5_avg=0.73690 lr=0.00100 gn=21.56074 time=55.40it/s +epoch=64 global_step=25300 loss=6.22097 loss_avg=6.34213 acc=0.36719 acc_top1_avg=0.36523 acc_top5_avg=0.73588 lr=0.00100 gn=14.94748 time=50.49it/s +epoch=64 global_step=25350 loss=6.67128 loss_avg=6.35437 acc=0.32812 acc_top1_avg=0.36354 acc_top5_avg=0.73584 lr=0.00100 gn=19.42071 time=63.38it/s +epoch=64 global_step=25400 loss=6.75103 loss_avg=6.34812 acc=0.30469 acc_top1_avg=0.36384 acc_top5_avg=0.73649 lr=0.00100 gn=17.86638 time=53.12it/s +====================Eval==================== +epoch=64 global_step=25415 loss=1.72113 test_loss_avg=3.34051 acc=0.50781 test_acc_avg=0.26935 test_acc_top5_avg=0.66584 time=234.03it/s +epoch=64 global_step=25415 loss=6.18954 test_loss_avg=3.05451 acc=0.00000 test_acc_avg=0.34850 test_acc_top5_avg=0.79341 time=854.24it/s +curr_acc 0.3485 +BEST_ACC 0.3717 +curr_acc_top5 0.7934 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=6.47881 loss_avg=6.32441 acc=0.33594 acc_top1_avg=0.36741 acc_top5_avg=0.73549 lr=0.00100 gn=20.68833 time=53.48it/s +epoch=65 global_step=25500 loss=6.11210 loss_avg=6.33550 acc=0.39844 acc_top1_avg=0.36654 acc_top5_avg=0.72932 lr=0.00100 gn=19.37874 time=61.72it/s +epoch=65 global_step=25550 loss=5.90733 loss_avg=6.33993 acc=0.42969 acc_top1_avg=0.36580 acc_top5_avg=0.73015 lr=0.00100 gn=18.83585 time=52.27it/s +epoch=65 global_step=25600 loss=6.11123 loss_avg=6.31663 acc=0.39062 acc_top1_avg=0.36888 acc_top5_avg=0.73366 lr=0.00100 gn=17.45907 time=51.65it/s +epoch=65 global_step=25650 loss=6.39792 loss_avg=6.33204 acc=0.34375 acc_top1_avg=0.36705 acc_top5_avg=0.73358 lr=0.00100 gn=21.93425 time=54.88it/s +epoch=65 global_step=25700 loss=6.67772 loss_avg=6.33750 acc=0.32031 acc_top1_avg=0.36664 acc_top5_avg=0.73339 lr=0.00100 gn=22.71442 time=52.01it/s +epoch=65 global_step=25750 loss=6.51426 loss_avg=6.33159 acc=0.33594 acc_top1_avg=0.36712 acc_top5_avg=0.73524 lr=0.00100 gn=16.92567 time=60.60it/s +epoch=65 global_step=25800 loss=6.46845 loss_avg=6.33467 acc=0.34375 acc_top1_avg=0.36670 acc_top5_avg=0.73440 lr=0.00100 gn=20.41108 time=54.14it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.85706 test_loss_avg=1.36365 acc=0.81250 test_acc_avg=0.60938 test_acc_top5_avg=0.98490 time=240.43it/s +epoch=65 global_step=25806 loss=0.12870 test_loss_avg=2.99299 acc=0.96094 test_acc_avg=0.33762 test_acc_top5_avg=0.74459 time=237.95it/s +epoch=65 global_step=25806 loss=6.62637 test_loss_avg=3.11309 acc=0.00000 test_acc_avg=0.34217 test_acc_top5_avg=0.78530 time=652.61it/s +curr_acc 0.3422 +BEST_ACC 0.3717 +curr_acc_top5 0.7853 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=6.23542 loss_avg=6.32886 acc=0.38281 acc_top1_avg=0.36737 acc_top5_avg=0.73526 lr=0.00100 gn=24.80653 time=63.06it/s +epoch=66 global_step=25900 loss=6.11916 loss_avg=6.29392 acc=0.38281 acc_top1_avg=0.37209 acc_top5_avg=0.73546 lr=0.00100 gn=15.43428 time=54.11it/s +epoch=66 global_step=25950 loss=6.40350 loss_avg=6.30213 acc=0.35156 acc_top1_avg=0.37104 acc_top5_avg=0.73465 lr=0.00100 gn=17.03391 time=31.40it/s +epoch=66 global_step=26000 loss=6.06193 loss_avg=6.30758 acc=0.37500 acc_top1_avg=0.36976 acc_top5_avg=0.73393 lr=0.00100 gn=15.81898 time=61.34it/s +epoch=66 global_step=26050 loss=5.43976 loss_avg=6.31798 acc=0.48438 acc_top1_avg=0.36856 acc_top5_avg=0.73245 lr=0.00100 gn=16.75561 time=52.71it/s +epoch=66 global_step=26100 loss=6.41589 loss_avg=6.32962 acc=0.35938 acc_top1_avg=0.36663 acc_top5_avg=0.73220 lr=0.00100 gn=24.19998 time=59.10it/s +epoch=66 global_step=26150 loss=6.39642 loss_avg=6.32187 acc=0.37500 acc_top1_avg=0.36719 acc_top5_avg=0.73376 lr=0.00100 gn=23.21491 time=41.58it/s +====================Eval==================== +epoch=66 global_step=26197 loss=5.13459 test_loss_avg=3.45296 acc=0.00000 test_acc_avg=0.26476 test_acc_top5_avg=0.65690 time=236.06it/s +epoch=66 global_step=26197 loss=6.00274 test_loss_avg=3.08371 acc=0.00000 test_acc_avg=0.35077 test_acc_top5_avg=0.78066 time=819.04it/s +curr_acc 0.3508 +BEST_ACC 0.3717 +curr_acc_top5 0.7807 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=6.43101 loss_avg=6.23176 acc=0.36719 acc_top1_avg=0.38021 acc_top5_avg=0.73698 lr=0.00100 gn=21.31900 time=52.22it/s +epoch=67 global_step=26250 loss=6.75660 loss_avg=6.28111 acc=0.31250 acc_top1_avg=0.37308 acc_top5_avg=0.73187 lr=0.00100 gn=20.70743 time=60.84it/s +epoch=67 global_step=26300 loss=6.19193 loss_avg=6.27152 acc=0.36719 acc_top1_avg=0.37288 acc_top5_avg=0.73597 lr=0.00100 gn=15.98972 time=59.52it/s +epoch=67 global_step=26350 loss=5.92455 loss_avg=6.25570 acc=0.39062 acc_top1_avg=0.37485 acc_top5_avg=0.73524 lr=0.00100 gn=22.49915 time=54.77it/s +epoch=67 global_step=26400 loss=6.54539 loss_avg=6.26485 acc=0.33594 acc_top1_avg=0.37331 acc_top5_avg=0.73707 lr=0.00100 gn=22.89798 time=52.39it/s +epoch=67 global_step=26450 loss=6.88350 loss_avg=6.28161 acc=0.28906 acc_top1_avg=0.37126 acc_top5_avg=0.73558 lr=0.00100 gn=21.24301 time=62.86it/s +epoch=67 global_step=26500 loss=6.26544 loss_avg=6.28903 acc=0.38281 acc_top1_avg=0.37041 acc_top5_avg=0.73633 lr=0.00100 gn=23.38567 time=55.38it/s +epoch=67 global_step=26550 loss=5.94260 loss_avg=6.29604 acc=0.42969 acc_top1_avg=0.36998 acc_top5_avg=0.73557 lr=0.00100 gn=18.98637 time=60.48it/s +====================Eval==================== +epoch=67 global_step=26588 loss=2.90386 test_loss_avg=2.68629 acc=0.26562 test_acc_avg=0.28125 test_acc_top5_avg=0.96205 time=228.16it/s +epoch=67 global_step=26588 loss=0.40840 test_loss_avg=3.41492 acc=0.89062 test_acc_avg=0.24863 test_acc_top5_avg=0.70134 time=240.78it/s +epoch=67 global_step=26588 loss=6.51789 test_loss_avg=3.13373 acc=0.00000 test_acc_avg=0.33861 test_acc_top5_avg=0.77947 time=492.64it/s +curr_acc 0.3386 +BEST_ACC 0.3717 +curr_acc_top5 0.7795 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=5.98496 loss_avg=6.15646 acc=0.39062 acc_top1_avg=0.39128 acc_top5_avg=0.73372 lr=0.00100 gn=22.70278 time=55.00it/s +epoch=68 global_step=26650 loss=6.42405 loss_avg=6.18807 acc=0.34375 acc_top1_avg=0.38420 acc_top5_avg=0.73160 lr=0.00100 gn=21.30778 time=60.46it/s +epoch=68 global_step=26700 loss=5.86896 loss_avg=6.22547 acc=0.42188 acc_top1_avg=0.37905 acc_top5_avg=0.73647 lr=0.00100 gn=16.67532 time=54.86it/s +epoch=68 global_step=26750 loss=6.47791 loss_avg=6.24596 acc=0.36719 acc_top1_avg=0.37640 acc_top5_avg=0.73630 lr=0.00100 gn=24.72419 time=52.69it/s +epoch=68 global_step=26800 loss=6.30524 loss_avg=6.25948 acc=0.39062 acc_top1_avg=0.37496 acc_top5_avg=0.73603 lr=0.00100 gn=21.78031 time=52.61it/s +epoch=68 global_step=26850 loss=6.31611 loss_avg=6.25976 acc=0.36719 acc_top1_avg=0.37476 acc_top5_avg=0.73575 lr=0.00100 gn=28.82715 time=61.30it/s +epoch=68 global_step=26900 loss=6.87360 loss_avg=6.26853 acc=0.31250 acc_top1_avg=0.37360 acc_top5_avg=0.73440 lr=0.00100 gn=20.41192 time=56.01it/s +epoch=68 global_step=26950 loss=7.06151 loss_avg=6.26995 acc=0.28125 acc_top1_avg=0.37373 acc_top5_avg=0.73435 lr=0.00100 gn=17.00639 time=56.32it/s +====================Eval==================== +epoch=68 global_step=26979 loss=5.08400 test_loss_avg=2.85019 acc=0.00000 test_acc_avg=0.34487 test_acc_top5_avg=0.79632 time=148.11it/s +epoch=68 global_step=26979 loss=5.52259 test_loss_avg=2.96303 acc=0.00000 test_acc_avg=0.36208 test_acc_top5_avg=0.77244 time=239.43it/s +epoch=68 global_step=26979 loss=5.89711 test_loss_avg=3.00017 acc=0.00000 test_acc_avg=0.35750 test_acc_top5_avg=0.77532 time=499.56it/s +curr_acc 0.3575 +BEST_ACC 0.3717 +curr_acc_top5 0.7753 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=5.69569 loss_avg=6.22102 acc=0.42969 acc_top1_avg=0.37984 acc_top5_avg=0.73251 lr=0.00100 gn=23.29532 time=54.68it/s +epoch=69 global_step=27050 loss=6.37699 loss_avg=6.22595 acc=0.38281 acc_top1_avg=0.37808 acc_top5_avg=0.73779 lr=0.00100 gn=22.05954 time=52.32it/s +epoch=69 global_step=27100 loss=6.74853 loss_avg=6.18894 acc=0.31250 acc_top1_avg=0.38294 acc_top5_avg=0.73948 lr=0.00100 gn=24.07907 time=58.65it/s +epoch=69 global_step=27150 loss=6.55444 loss_avg=6.19737 acc=0.35156 acc_top1_avg=0.38231 acc_top5_avg=0.73853 lr=0.00100 gn=22.23647 time=58.08it/s +epoch=69 global_step=27200 loss=5.77999 loss_avg=6.21026 acc=0.42188 acc_top1_avg=0.38097 acc_top5_avg=0.73936 lr=0.00100 gn=20.64453 time=49.08it/s +epoch=69 global_step=27250 loss=5.74659 loss_avg=6.23029 acc=0.42188 acc_top1_avg=0.37875 acc_top5_avg=0.73792 lr=0.00100 gn=17.48153 time=63.20it/s +epoch=69 global_step=27300 loss=6.47970 loss_avg=6.23315 acc=0.35156 acc_top1_avg=0.37889 acc_top5_avg=0.73747 lr=0.00100 gn=20.25869 time=54.91it/s +epoch=69 global_step=27350 loss=6.44510 loss_avg=6.25005 acc=0.34375 acc_top1_avg=0.37687 acc_top5_avg=0.73606 lr=0.00100 gn=22.34614 time=52.41it/s +====================Eval==================== +epoch=69 global_step=27370 loss=4.52297 test_loss_avg=3.38253 acc=0.00000 test_acc_avg=0.25973 test_acc_top5_avg=0.68463 time=241.76it/s +epoch=69 global_step=27370 loss=5.30910 test_loss_avg=3.06443 acc=0.00000 test_acc_avg=0.33989 test_acc_top5_avg=0.78540 time=828.59it/s +curr_acc 0.3399 +BEST_ACC 0.3717 +curr_acc_top5 0.7854 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=6.16065 loss_avg=6.12302 acc=0.38281 acc_top1_avg=0.38932 acc_top5_avg=0.75078 lr=0.00100 gn=19.16741 time=62.68it/s +epoch=70 global_step=27450 loss=6.04144 loss_avg=6.25543 acc=0.39844 acc_top1_avg=0.37520 acc_top5_avg=0.73633 lr=0.00100 gn=19.60560 time=53.68it/s +epoch=70 global_step=27500 loss=6.39046 loss_avg=6.24375 acc=0.38281 acc_top1_avg=0.37692 acc_top5_avg=0.73413 lr=0.00100 gn=26.90642 time=55.72it/s +epoch=70 global_step=27550 loss=6.52504 loss_avg=6.23672 acc=0.35938 acc_top1_avg=0.37756 acc_top5_avg=0.73368 lr=0.00100 gn=22.94701 time=52.73it/s +epoch=70 global_step=27600 loss=5.38040 loss_avg=6.20899 acc=0.49219 acc_top1_avg=0.38074 acc_top5_avg=0.73393 lr=0.00100 gn=20.13202 time=55.84it/s +epoch=70 global_step=27650 loss=6.28049 loss_avg=6.22190 acc=0.35156 acc_top1_avg=0.37899 acc_top5_avg=0.73518 lr=0.00100 gn=20.09738 time=45.77it/s +epoch=70 global_step=27700 loss=6.15187 loss_avg=6.23059 acc=0.38281 acc_top1_avg=0.37789 acc_top5_avg=0.73525 lr=0.00100 gn=21.75453 time=62.53it/s +epoch=70 global_step=27750 loss=6.94248 loss_avg=6.23648 acc=0.28906 acc_top1_avg=0.37724 acc_top5_avg=0.73581 lr=0.00100 gn=21.48847 time=54.44it/s +====================Eval==================== +epoch=70 global_step=27761 loss=4.62563 test_loss_avg=2.39991 acc=0.00000 test_acc_avg=0.40703 test_acc_top5_avg=0.86602 time=238.16it/s +epoch=70 global_step=27761 loss=0.09016 test_loss_avg=2.96923 acc=0.97656 test_acc_avg=0.36451 test_acc_top5_avg=0.75882 time=239.47it/s +epoch=70 global_step=27761 loss=6.13175 test_loss_avg=3.24397 acc=0.00000 test_acc_avg=0.32694 test_acc_top5_avg=0.78283 time=719.31it/s +curr_acc 0.3269 +BEST_ACC 0.3717 +curr_acc_top5 0.7828 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=6.36461 loss_avg=6.15661 acc=0.35938 acc_top1_avg=0.38582 acc_top5_avg=0.74339 lr=0.00100 gn=20.17258 time=62.15it/s +epoch=71 global_step=27850 loss=6.18629 loss_avg=6.21650 acc=0.38281 acc_top1_avg=0.38158 acc_top5_avg=0.73692 lr=0.00100 gn=22.03995 time=62.27it/s +epoch=71 global_step=27900 loss=6.16140 loss_avg=6.23926 acc=0.37500 acc_top1_avg=0.37871 acc_top5_avg=0.73370 lr=0.00100 gn=18.78483 time=59.55it/s +epoch=71 global_step=27950 loss=6.44019 loss_avg=6.23873 acc=0.35156 acc_top1_avg=0.37756 acc_top5_avg=0.73454 lr=0.00100 gn=24.17115 time=55.43it/s +epoch=71 global_step=28000 loss=5.52112 loss_avg=6.22951 acc=0.43750 acc_top1_avg=0.37905 acc_top5_avg=0.73653 lr=0.00100 gn=18.83208 time=55.63it/s +epoch=71 global_step=28050 loss=6.25036 loss_avg=6.22783 acc=0.38281 acc_top1_avg=0.37933 acc_top5_avg=0.73786 lr=0.00100 gn=19.42694 time=54.14it/s +epoch=71 global_step=28100 loss=5.98077 loss_avg=6.22281 acc=0.39062 acc_top1_avg=0.38030 acc_top5_avg=0.73650 lr=0.00100 gn=27.51698 time=57.14it/s +epoch=71 global_step=28150 loss=6.35754 loss_avg=6.23000 acc=0.35938 acc_top1_avg=0.37916 acc_top5_avg=0.73492 lr=0.00100 gn=19.67490 time=55.34it/s +====================Eval==================== +epoch=71 global_step=28152 loss=2.14747 test_loss_avg=3.50808 acc=0.42969 test_acc_avg=0.24333 test_acc_top5_avg=0.63567 time=82.07it/s +epoch=71 global_step=28152 loss=6.10978 test_loss_avg=3.12035 acc=0.00000 test_acc_avg=0.33871 test_acc_top5_avg=0.78422 time=682.67it/s +curr_acc 0.3387 +BEST_ACC 0.3717 +curr_acc_top5 0.7842 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=5.31268 loss_avg=6.16671 acc=0.50781 acc_top1_avg=0.38509 acc_top5_avg=0.73747 lr=0.00100 gn=28.37758 time=52.54it/s +epoch=72 global_step=28250 loss=6.54801 loss_avg=6.18535 acc=0.36719 acc_top1_avg=0.38441 acc_top5_avg=0.73868 lr=0.00100 gn=24.31726 time=52.36it/s +epoch=72 global_step=28300 loss=6.24392 loss_avg=6.17461 acc=0.37500 acc_top1_avg=0.38477 acc_top5_avg=0.73786 lr=0.00100 gn=21.73419 time=62.59it/s +epoch=72 global_step=28350 loss=5.73507 loss_avg=6.16934 acc=0.43750 acc_top1_avg=0.38490 acc_top5_avg=0.73793 lr=0.00100 gn=22.43032 time=55.48it/s +epoch=72 global_step=28400 loss=6.81486 loss_avg=6.19813 acc=0.31250 acc_top1_avg=0.38152 acc_top5_avg=0.73649 lr=0.00100 gn=18.95909 time=60.52it/s +epoch=72 global_step=28450 loss=5.92283 loss_avg=6.19427 acc=0.40625 acc_top1_avg=0.38158 acc_top5_avg=0.73611 lr=0.00100 gn=26.03669 time=54.41it/s +epoch=72 global_step=28500 loss=6.54753 loss_avg=6.21097 acc=0.32031 acc_top1_avg=0.37983 acc_top5_avg=0.73424 lr=0.00100 gn=22.00283 time=51.95it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.84377 test_loss_avg=1.62157 acc=0.77344 test_acc_avg=0.53776 test_acc_top5_avg=0.98698 time=230.37it/s +epoch=72 global_step=28543 loss=0.44403 test_loss_avg=3.05551 acc=0.87500 test_acc_avg=0.32031 test_acc_top5_avg=0.70779 time=232.53it/s +epoch=72 global_step=28543 loss=5.98272 test_loss_avg=3.02697 acc=0.00000 test_acc_avg=0.35255 test_acc_top5_avg=0.76543 time=500.10it/s +curr_acc 0.3526 +BEST_ACC 0.3717 +curr_acc_top5 0.7654 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=6.18424 loss_avg=6.12848 acc=0.36719 acc_top1_avg=0.38616 acc_top5_avg=0.73438 lr=0.00100 gn=28.55015 time=53.77it/s +epoch=73 global_step=28600 loss=6.78651 loss_avg=6.12865 acc=0.32031 acc_top1_avg=0.38884 acc_top5_avg=0.73396 lr=0.00100 gn=27.32465 time=53.46it/s +epoch=73 global_step=28650 loss=6.10935 loss_avg=6.17559 acc=0.37500 acc_top1_avg=0.38383 acc_top5_avg=0.73299 lr=0.00100 gn=26.01063 time=59.99it/s +epoch=73 global_step=28700 loss=5.88093 loss_avg=6.19678 acc=0.42188 acc_top1_avg=0.38142 acc_top5_avg=0.73164 lr=0.00100 gn=26.13009 time=50.40it/s +epoch=73 global_step=28750 loss=7.31944 loss_avg=6.19321 acc=0.25781 acc_top1_avg=0.38228 acc_top5_avg=0.73226 lr=0.00100 gn=17.63664 time=58.00it/s +epoch=73 global_step=28800 loss=6.44274 loss_avg=6.19479 acc=0.35938 acc_top1_avg=0.38208 acc_top5_avg=0.73298 lr=0.00100 gn=29.68285 time=61.37it/s +epoch=73 global_step=28850 loss=7.02225 loss_avg=6.19788 acc=0.28906 acc_top1_avg=0.38177 acc_top5_avg=0.73280 lr=0.00100 gn=23.69202 time=62.90it/s +epoch=73 global_step=28900 loss=6.29755 loss_avg=6.20181 acc=0.37500 acc_top1_avg=0.38139 acc_top5_avg=0.73300 lr=0.00100 gn=25.52703 time=57.62it/s +====================Eval==================== +epoch=73 global_step=28934 loss=4.97586 test_loss_avg=3.15369 acc=0.00000 test_acc_avg=0.29995 test_acc_top5_avg=0.72846 time=229.25it/s +epoch=73 global_step=28934 loss=5.37759 test_loss_avg=3.03800 acc=0.00000 test_acc_avg=0.34039 test_acc_top5_avg=0.78135 time=484.55it/s +curr_acc 0.3404 +BEST_ACC 0.3717 +curr_acc_top5 0.7813 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=6.64469 loss_avg=6.22778 acc=0.33594 acc_top1_avg=0.37842 acc_top5_avg=0.73291 lr=0.00100 gn=27.00166 time=55.48it/s +epoch=74 global_step=29000 loss=5.89615 loss_avg=6.17871 acc=0.42188 acc_top1_avg=0.38317 acc_top5_avg=0.73639 lr=0.00100 gn=24.59219 time=49.37it/s +epoch=74 global_step=29050 loss=6.77705 loss_avg=6.14707 acc=0.33594 acc_top1_avg=0.38578 acc_top5_avg=0.73592 lr=0.00100 gn=23.37605 time=58.09it/s +epoch=74 global_step=29100 loss=5.90696 loss_avg=6.11971 acc=0.42188 acc_top1_avg=0.38926 acc_top5_avg=0.73565 lr=0.00100 gn=21.61516 time=60.58it/s +epoch=74 global_step=29150 loss=6.58461 loss_avg=6.14419 acc=0.35938 acc_top1_avg=0.38596 acc_top5_avg=0.73517 lr=0.00100 gn=29.27886 time=52.04it/s +epoch=74 global_step=29200 loss=6.68029 loss_avg=6.15840 acc=0.34375 acc_top1_avg=0.38466 acc_top5_avg=0.73482 lr=0.00100 gn=25.97439 time=55.52it/s +epoch=74 global_step=29250 loss=6.22359 loss_avg=6.17062 acc=0.37500 acc_top1_avg=0.38360 acc_top5_avg=0.73487 lr=0.00100 gn=28.12575 time=60.59it/s +epoch=74 global_step=29300 loss=6.49000 loss_avg=6.18002 acc=0.33594 acc_top1_avg=0.38279 acc_top5_avg=0.73578 lr=0.00100 gn=18.86953 time=53.52it/s +====================Eval==================== +epoch=74 global_step=29325 loss=2.49809 test_loss_avg=2.50601 acc=0.31250 test_acc_avg=0.32227 test_acc_top5_avg=0.97852 time=253.88it/s +epoch=74 global_step=29325 loss=4.29676 test_loss_avg=3.59694 acc=0.00000 test_acc_avg=0.21166 test_acc_top5_avg=0.66797 time=234.91it/s +epoch=74 global_step=29325 loss=6.18913 test_loss_avg=3.16044 acc=0.00000 test_acc_avg=0.32882 test_acc_top5_avg=0.76562 time=556.35it/s +curr_acc 0.3288 +BEST_ACC 0.3717 +curr_acc_top5 0.7656 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=5.41792 loss_avg=6.14678 acc=0.46875 acc_top1_avg=0.38812 acc_top5_avg=0.73031 lr=0.00100 gn=20.65479 time=56.71it/s +epoch=75 global_step=29400 loss=6.37487 loss_avg=6.13374 acc=0.38281 acc_top1_avg=0.39052 acc_top5_avg=0.73792 lr=0.00100 gn=31.70378 time=56.54it/s +epoch=75 global_step=29450 loss=7.12291 loss_avg=6.13955 acc=0.28125 acc_top1_avg=0.38931 acc_top5_avg=0.73725 lr=0.00100 gn=31.57190 time=53.90it/s +epoch=75 global_step=29500 loss=6.03539 loss_avg=6.15982 acc=0.39062 acc_top1_avg=0.38746 acc_top5_avg=0.74076 lr=0.00100 gn=19.28693 time=56.31it/s +epoch=75 global_step=29550 loss=6.15742 loss_avg=6.15561 acc=0.37500 acc_top1_avg=0.38778 acc_top5_avg=0.73906 lr=0.00100 gn=27.85446 time=56.10it/s +epoch=75 global_step=29600 loss=6.26301 loss_avg=6.15937 acc=0.36719 acc_top1_avg=0.38719 acc_top5_avg=0.73761 lr=0.00100 gn=23.94634 time=43.75it/s +epoch=75 global_step=29650 loss=6.60985 loss_avg=6.16554 acc=0.32812 acc_top1_avg=0.38678 acc_top5_avg=0.73810 lr=0.00100 gn=22.64872 time=57.98it/s +epoch=75 global_step=29700 loss=6.38441 loss_avg=6.17624 acc=0.35156 acc_top1_avg=0.38508 acc_top5_avg=0.73583 lr=0.00100 gn=27.40636 time=59.03it/s +====================Eval==================== +epoch=75 global_step=29716 loss=5.52451 test_loss_avg=2.79328 acc=0.00000 test_acc_avg=0.36750 test_acc_top5_avg=0.81031 time=228.76it/s +epoch=75 global_step=29716 loss=5.91182 test_loss_avg=3.00824 acc=0.00000 test_acc_avg=0.36135 test_acc_top5_avg=0.76469 time=239.67it/s +epoch=75 global_step=29716 loss=6.14378 test_loss_avg=3.15195 acc=0.00000 test_acc_avg=0.34306 test_acc_top5_avg=0.77551 time=786.92it/s +curr_acc 0.3431 +BEST_ACC 0.3717 +curr_acc_top5 0.7755 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=6.02034 loss_avg=6.16148 acc=0.41406 acc_top1_avg=0.38511 acc_top5_avg=0.74127 lr=0.00100 gn=26.77989 time=60.54it/s +epoch=76 global_step=29800 loss=5.81067 loss_avg=6.12755 acc=0.42188 acc_top1_avg=0.38746 acc_top5_avg=0.73754 lr=0.00100 gn=32.18280 time=62.37it/s +epoch=76 global_step=29850 loss=6.29726 loss_avg=6.13372 acc=0.35938 acc_top1_avg=0.38719 acc_top5_avg=0.73805 lr=0.00100 gn=23.37159 time=55.22it/s +epoch=76 global_step=29900 loss=6.05660 loss_avg=6.14612 acc=0.39844 acc_top1_avg=0.38638 acc_top5_avg=0.73713 lr=0.00100 gn=25.80994 time=55.93it/s +epoch=76 global_step=29950 loss=6.84150 loss_avg=6.14943 acc=0.30469 acc_top1_avg=0.38675 acc_top5_avg=0.73738 lr=0.00100 gn=22.40021 time=63.11it/s +epoch=76 global_step=30000 loss=5.83395 loss_avg=6.15244 acc=0.42969 acc_top1_avg=0.38658 acc_top5_avg=0.73707 lr=0.00100 gn=33.72897 time=55.78it/s +epoch=76 global_step=30050 loss=6.35797 loss_avg=6.15243 acc=0.35938 acc_top1_avg=0.38686 acc_top5_avg=0.73723 lr=0.00100 gn=26.79592 time=55.90it/s +epoch=76 global_step=30100 loss=5.88755 loss_avg=6.15459 acc=0.41406 acc_top1_avg=0.38641 acc_top5_avg=0.73564 lr=0.00100 gn=25.97477 time=56.42it/s +====================Eval==================== +epoch=76 global_step=30107 loss=2.03230 test_loss_avg=3.50929 acc=0.45312 test_acc_avg=0.23794 test_acc_top5_avg=0.64776 time=239.80it/s +epoch=76 global_step=30107 loss=5.58172 test_loss_avg=3.16222 acc=0.00000 test_acc_avg=0.32892 test_acc_top5_avg=0.77255 time=818.08it/s +curr_acc 0.3289 +BEST_ACC 0.3717 +curr_acc_top5 0.7725 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=6.48064 loss_avg=6.12292 acc=0.35156 acc_top1_avg=0.39044 acc_top5_avg=0.72929 lr=0.00100 gn=29.34086 time=54.97it/s +epoch=77 global_step=30200 loss=5.99090 loss_avg=6.10940 acc=0.39844 acc_top1_avg=0.39138 acc_top5_avg=0.73370 lr=0.00100 gn=29.30682 time=62.55it/s +epoch=77 global_step=30250 loss=6.37970 loss_avg=6.07755 acc=0.36719 acc_top1_avg=0.39598 acc_top5_avg=0.73629 lr=0.00100 gn=26.46609 time=52.87it/s +epoch=77 global_step=30300 loss=5.67961 loss_avg=6.08825 acc=0.44531 acc_top1_avg=0.39451 acc_top5_avg=0.73510 lr=0.00100 gn=26.36012 time=62.92it/s +epoch=77 global_step=30350 loss=6.22571 loss_avg=6.11168 acc=0.37500 acc_top1_avg=0.39140 acc_top5_avg=0.73421 lr=0.00100 gn=24.62663 time=54.56it/s +epoch=77 global_step=30400 loss=5.80242 loss_avg=6.10958 acc=0.42969 acc_top1_avg=0.39097 acc_top5_avg=0.73461 lr=0.00100 gn=32.35227 time=50.65it/s +epoch=77 global_step=30450 loss=5.94018 loss_avg=6.11944 acc=0.42969 acc_top1_avg=0.39049 acc_top5_avg=0.73401 lr=0.00100 gn=30.42838 time=57.11it/s +====================Eval==================== +epoch=77 global_step=30498 loss=4.61186 test_loss_avg=1.89196 acc=0.00000 test_acc_avg=0.50046 test_acc_top5_avg=0.93061 time=235.17it/s +epoch=77 global_step=30498 loss=0.12782 test_loss_avg=3.02477 acc=0.97656 test_acc_avg=0.33431 test_acc_top5_avg=0.74335 time=243.74it/s +epoch=77 global_step=30498 loss=5.80040 test_loss_avg=3.18369 acc=0.00000 test_acc_avg=0.32298 test_acc_top5_avg=0.77907 time=666.40it/s +curr_acc 0.3230 +BEST_ACC 0.3717 +curr_acc_top5 0.7791 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=6.13580 loss_avg=6.36064 acc=0.40625 acc_top1_avg=0.36719 acc_top5_avg=0.72266 lr=0.00100 gn=30.23291 time=51.59it/s +epoch=78 global_step=30550 loss=6.51660 loss_avg=6.02802 acc=0.35156 acc_top1_avg=0.40445 acc_top5_avg=0.73257 lr=0.00100 gn=23.04909 time=52.09it/s +epoch=78 global_step=30600 loss=6.34407 loss_avg=6.10883 acc=0.35938 acc_top1_avg=0.39262 acc_top5_avg=0.73039 lr=0.00100 gn=25.05287 time=51.34it/s +epoch=78 global_step=30650 loss=6.08751 loss_avg=6.10007 acc=0.39062 acc_top1_avg=0.39458 acc_top5_avg=0.73412 lr=0.00100 gn=27.93931 time=53.64it/s +epoch=78 global_step=30700 loss=5.79754 loss_avg=6.09884 acc=0.40625 acc_top1_avg=0.39411 acc_top5_avg=0.73283 lr=0.00100 gn=28.11876 time=60.89it/s +epoch=78 global_step=30750 loss=5.64451 loss_avg=6.11220 acc=0.45312 acc_top1_avg=0.39187 acc_top5_avg=0.73419 lr=0.00100 gn=32.11694 time=53.96it/s +epoch=78 global_step=30800 loss=6.22185 loss_avg=6.10806 acc=0.37500 acc_top1_avg=0.39254 acc_top5_avg=0.73334 lr=0.00100 gn=27.06484 time=63.45it/s +epoch=78 global_step=30850 loss=5.64968 loss_avg=6.10942 acc=0.44531 acc_top1_avg=0.39256 acc_top5_avg=0.73513 lr=0.00100 gn=27.79371 time=58.72it/s +====================Eval==================== +epoch=78 global_step=30889 loss=4.65531 test_loss_avg=3.49080 acc=0.00000 test_acc_avg=0.24630 test_acc_top5_avg=0.63672 time=226.47it/s +epoch=78 global_step=30889 loss=5.94461 test_loss_avg=3.02157 acc=0.00000 test_acc_avg=0.34474 test_acc_top5_avg=0.77967 time=625.46it/s +curr_acc 0.3447 +BEST_ACC 0.3717 +curr_acc_top5 0.7797 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=5.75340 loss_avg=5.99600 acc=0.42969 acc_top1_avg=0.40270 acc_top5_avg=0.73580 lr=0.00100 gn=27.41296 time=57.09it/s +epoch=79 global_step=30950 loss=6.25870 loss_avg=6.00933 acc=0.36719 acc_top1_avg=0.40228 acc_top5_avg=0.73566 lr=0.00100 gn=22.07391 time=52.08it/s +epoch=79 global_step=31000 loss=5.84021 loss_avg=6.03391 acc=0.42188 acc_top1_avg=0.39949 acc_top5_avg=0.73276 lr=0.00100 gn=27.48890 time=57.72it/s +epoch=79 global_step=31050 loss=6.91114 loss_avg=6.04329 acc=0.28906 acc_top1_avg=0.39858 acc_top5_avg=0.73253 lr=0.00100 gn=30.13569 time=61.34it/s +epoch=79 global_step=31100 loss=6.37344 loss_avg=6.06742 acc=0.37500 acc_top1_avg=0.39636 acc_top5_avg=0.73160 lr=0.00100 gn=29.22891 time=51.02it/s +epoch=79 global_step=31150 loss=5.83267 loss_avg=6.08538 acc=0.42188 acc_top1_avg=0.39458 acc_top5_avg=0.73213 lr=0.00100 gn=31.65080 time=54.15it/s +epoch=79 global_step=31200 loss=6.08208 loss_avg=6.08825 acc=0.40625 acc_top1_avg=0.39452 acc_top5_avg=0.73221 lr=0.00100 gn=27.66462 time=60.95it/s +epoch=79 global_step=31250 loss=5.79180 loss_avg=6.09583 acc=0.42969 acc_top1_avg=0.39365 acc_top5_avg=0.73280 lr=0.00100 gn=28.27798 time=53.48it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.92496 test_loss_avg=2.01002 acc=0.76562 test_acc_avg=0.43837 test_acc_top5_avg=0.94878 time=240.66it/s +epoch=79 global_step=31280 loss=0.51367 test_loss_avg=3.39401 acc=0.86719 test_acc_avg=0.24987 test_acc_top5_avg=0.69386 time=236.06it/s +epoch=79 global_step=31280 loss=6.23142 test_loss_avg=3.19591 acc=0.00000 test_acc_avg=0.32130 test_acc_top5_avg=0.76622 time=494.61it/s +curr_acc 0.3213 +BEST_ACC 0.3717 +curr_acc_top5 0.7662 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=6.26647 loss_avg=5.95100 acc=0.37500 acc_top1_avg=0.41250 acc_top5_avg=0.72383 lr=0.00010 gn=28.02182 time=59.38it/s +epoch=80 global_step=31350 loss=5.63124 loss_avg=5.96612 acc=0.42188 acc_top1_avg=0.41027 acc_top5_avg=0.73248 lr=0.00010 gn=28.70914 time=60.22it/s +epoch=80 global_step=31400 loss=5.97622 loss_avg=5.94953 acc=0.40625 acc_top1_avg=0.41113 acc_top5_avg=0.73398 lr=0.00010 gn=24.83417 time=55.83it/s +epoch=80 global_step=31450 loss=5.91679 loss_avg=5.99924 acc=0.40625 acc_top1_avg=0.40551 acc_top5_avg=0.72817 lr=0.00010 gn=29.91590 time=55.51it/s +epoch=80 global_step=31500 loss=6.19728 loss_avg=5.97584 acc=0.37500 acc_top1_avg=0.40742 acc_top5_avg=0.73168 lr=0.00010 gn=21.44191 time=61.83it/s +epoch=80 global_step=31550 loss=6.29990 loss_avg=5.98340 acc=0.35156 acc_top1_avg=0.40692 acc_top5_avg=0.73226 lr=0.00010 gn=33.23221 time=53.59it/s +epoch=80 global_step=31600 loss=6.25915 loss_avg=5.97949 acc=0.36719 acc_top1_avg=0.40750 acc_top5_avg=0.73289 lr=0.00010 gn=25.48065 time=58.36it/s +epoch=80 global_step=31650 loss=6.19717 loss_avg=5.96686 acc=0.36719 acc_top1_avg=0.40872 acc_top5_avg=0.73418 lr=0.00010 gn=40.22677 time=51.71it/s +====================Eval==================== +epoch=80 global_step=31671 loss=4.95929 test_loss_avg=3.01144 acc=0.00000 test_acc_avg=0.32396 test_acc_top5_avg=0.76250 time=228.68it/s +epoch=80 global_step=31671 loss=5.95780 test_loss_avg=3.00475 acc=0.00000 test_acc_avg=0.34771 test_acc_top5_avg=0.76780 time=518.26it/s +curr_acc 0.3477 +BEST_ACC 0.3717 +curr_acc_top5 0.7678 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=6.29957 loss_avg=5.93000 acc=0.37500 acc_top1_avg=0.41002 acc_top5_avg=0.73249 lr=0.00010 gn=32.24027 time=57.74it/s +epoch=81 global_step=31750 loss=6.17544 loss_avg=5.95626 acc=0.39844 acc_top1_avg=0.40852 acc_top5_avg=0.73101 lr=0.00010 gn=32.49288 time=63.28it/s +epoch=81 global_step=31800 loss=6.11447 loss_avg=5.91754 acc=0.38281 acc_top1_avg=0.41382 acc_top5_avg=0.73219 lr=0.00010 gn=28.09629 time=55.67it/s +epoch=81 global_step=31850 loss=5.94939 loss_avg=5.90998 acc=0.42188 acc_top1_avg=0.41472 acc_top5_avg=0.73224 lr=0.00010 gn=37.55984 time=56.38it/s +epoch=81 global_step=31900 loss=5.91088 loss_avg=5.89007 acc=0.41406 acc_top1_avg=0.41693 acc_top5_avg=0.73311 lr=0.00010 gn=26.54809 time=57.83it/s +epoch=81 global_step=31950 loss=5.67584 loss_avg=5.87857 acc=0.45312 acc_top1_avg=0.41812 acc_top5_avg=0.73351 lr=0.00010 gn=32.30808 time=54.77it/s +epoch=81 global_step=32000 loss=6.23911 loss_avg=5.89815 acc=0.37500 acc_top1_avg=0.41572 acc_top5_avg=0.73388 lr=0.00010 gn=33.85190 time=57.02it/s +epoch=81 global_step=32050 loss=6.04654 loss_avg=5.89600 acc=0.39062 acc_top1_avg=0.41563 acc_top5_avg=0.73291 lr=0.00010 gn=30.00027 time=54.54it/s +====================Eval==================== +epoch=81 global_step=32062 loss=2.34981 test_loss_avg=2.34981 acc=0.32812 test_acc_avg=0.32812 test_acc_top5_avg=0.98438 time=218.28it/s +epoch=81 global_step=32062 loss=4.62780 test_loss_avg=3.38374 acc=0.00000 test_acc_avg=0.24311 test_acc_top5_avg=0.66468 time=237.70it/s +epoch=81 global_step=32062 loss=6.05527 test_loss_avg=3.05257 acc=0.00000 test_acc_avg=0.33762 test_acc_top5_avg=0.76741 time=767.77it/s +curr_acc 0.3376 +BEST_ACC 0.3717 +curr_acc_top5 0.7674 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=6.30653 loss_avg=5.90417 acc=0.38281 acc_top1_avg=0.41262 acc_top5_avg=0.73294 lr=0.00010 gn=35.05620 time=41.91it/s +epoch=82 global_step=32150 loss=5.71589 loss_avg=5.86071 acc=0.42188 acc_top1_avg=0.41752 acc_top5_avg=0.73065 lr=0.00010 gn=29.47967 time=57.85it/s +epoch=82 global_step=32200 loss=5.84825 loss_avg=5.88042 acc=0.41406 acc_top1_avg=0.41621 acc_top5_avg=0.73307 lr=0.00010 gn=26.95554 time=53.11it/s +epoch=82 global_step=32250 loss=5.96168 loss_avg=5.88283 acc=0.42188 acc_top1_avg=0.41656 acc_top5_avg=0.73043 lr=0.00010 gn=36.14377 time=54.31it/s +epoch=82 global_step=32300 loss=6.45263 loss_avg=5.86718 acc=0.36719 acc_top1_avg=0.41863 acc_top5_avg=0.73293 lr=0.00010 gn=39.66377 time=56.96it/s +epoch=82 global_step=32350 loss=6.16078 loss_avg=5.88253 acc=0.39062 acc_top1_avg=0.41699 acc_top5_avg=0.73296 lr=0.00010 gn=28.16438 time=56.98it/s +epoch=82 global_step=32400 loss=6.15641 loss_avg=5.88714 acc=0.39062 acc_top1_avg=0.41663 acc_top5_avg=0.73345 lr=0.00010 gn=30.72486 time=55.53it/s +epoch=82 global_step=32450 loss=5.73261 loss_avg=5.87216 acc=0.43750 acc_top1_avg=0.41833 acc_top5_avg=0.73303 lr=0.00010 gn=27.69329 time=60.79it/s +====================Eval==================== +epoch=82 global_step=32453 loss=4.83916 test_loss_avg=2.38872 acc=0.00000 test_acc_avg=0.42045 test_acc_top5_avg=0.86754 time=236.95it/s +epoch=82 global_step=32453 loss=5.26976 test_loss_avg=2.80051 acc=0.00000 test_acc_avg=0.37543 test_acc_top5_avg=0.74826 time=239.59it/s +epoch=82 global_step=32453 loss=6.23532 test_loss_avg=3.04417 acc=0.00000 test_acc_avg=0.34217 test_acc_top5_avg=0.76671 time=636.85it/s +curr_acc 0.3422 +BEST_ACC 0.3717 +curr_acc_top5 0.7667 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=5.57853 loss_avg=5.81520 acc=0.44531 acc_top1_avg=0.42487 acc_top5_avg=0.74102 lr=0.00010 gn=28.75878 time=48.56it/s +epoch=83 global_step=32550 loss=6.36664 loss_avg=5.85605 acc=0.35938 acc_top1_avg=0.42034 acc_top5_avg=0.73800 lr=0.00010 gn=32.05199 time=47.43it/s +epoch=83 global_step=32600 loss=6.22255 loss_avg=5.89745 acc=0.36719 acc_top1_avg=0.41513 acc_top5_avg=0.73512 lr=0.00010 gn=30.92787 time=56.05it/s +epoch=83 global_step=32650 loss=5.95217 loss_avg=5.87731 acc=0.40625 acc_top1_avg=0.41755 acc_top5_avg=0.73275 lr=0.00010 gn=25.75031 time=54.86it/s +epoch=83 global_step=32700 loss=5.38690 loss_avg=5.86302 acc=0.50000 acc_top1_avg=0.41969 acc_top5_avg=0.73431 lr=0.00010 gn=36.66786 time=57.47it/s +epoch=83 global_step=32750 loss=5.98408 loss_avg=5.87095 acc=0.39062 acc_top1_avg=0.41864 acc_top5_avg=0.73280 lr=0.00010 gn=27.14196 time=56.69it/s +epoch=83 global_step=32800 loss=4.85529 loss_avg=5.86485 acc=0.53125 acc_top1_avg=0.41935 acc_top5_avg=0.73237 lr=0.00010 gn=35.97845 time=62.45it/s +====================Eval==================== +epoch=83 global_step=32844 loss=2.40724 test_loss_avg=3.37977 acc=0.32031 test_acc_avg=0.24727 test_acc_top5_avg=0.62736 time=217.46it/s +epoch=83 global_step=32844 loss=5.96187 test_loss_avg=3.02186 acc=0.00000 test_acc_avg=0.33851 test_acc_top5_avg=0.76592 time=485.90it/s +curr_acc 0.3385 +BEST_ACC 0.3717 +curr_acc_top5 0.7659 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=5.98364 loss_avg=5.84534 acc=0.39844 acc_top1_avg=0.42057 acc_top5_avg=0.73438 lr=0.00010 gn=31.76550 time=53.08it/s +epoch=84 global_step=32900 loss=5.79601 loss_avg=5.81272 acc=0.42188 acc_top1_avg=0.42648 acc_top5_avg=0.73689 lr=0.00010 gn=30.33132 time=54.42it/s +epoch=84 global_step=32950 loss=5.52542 loss_avg=5.79847 acc=0.44531 acc_top1_avg=0.42925 acc_top5_avg=0.73762 lr=0.00010 gn=29.18233 time=51.30it/s +epoch=84 global_step=33000 loss=5.24203 loss_avg=5.79981 acc=0.50000 acc_top1_avg=0.42889 acc_top5_avg=0.73788 lr=0.00010 gn=32.84230 time=59.00it/s +epoch=84 global_step=33050 loss=6.10215 loss_avg=5.81761 acc=0.38281 acc_top1_avg=0.42620 acc_top5_avg=0.73760 lr=0.00010 gn=31.36463 time=52.48it/s +epoch=84 global_step=33100 loss=5.51503 loss_avg=5.83029 acc=0.43750 acc_top1_avg=0.42456 acc_top5_avg=0.73541 lr=0.00010 gn=35.40725 time=50.90it/s +epoch=84 global_step=33150 loss=5.93846 loss_avg=5.82305 acc=0.40625 acc_top1_avg=0.42522 acc_top5_avg=0.73499 lr=0.00010 gn=31.66149 time=56.72it/s +epoch=84 global_step=33200 loss=5.73125 loss_avg=5.83636 acc=0.43750 acc_top1_avg=0.42385 acc_top5_avg=0.73459 lr=0.00010 gn=35.81335 time=56.18it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.85992 test_loss_avg=1.59273 acc=0.75000 test_acc_avg=0.54408 test_acc_top5_avg=0.98214 time=190.13it/s +epoch=84 global_step=33235 loss=0.14974 test_loss_avg=3.00531 acc=0.96094 test_acc_avg=0.32324 test_acc_top5_avg=0.71655 time=237.02it/s +epoch=84 global_step=33235 loss=5.98447 test_loss_avg=3.03993 acc=0.00000 test_acc_avg=0.33792 test_acc_top5_avg=0.76493 time=503.70it/s +curr_acc 0.3379 +BEST_ACC 0.3717 +curr_acc_top5 0.7649 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=5.41179 loss_avg=5.65082 acc=0.46875 acc_top1_avg=0.44167 acc_top5_avg=0.75833 lr=0.00010 gn=27.68814 time=61.12it/s +epoch=85 global_step=33300 loss=5.27947 loss_avg=5.77561 acc=0.50000 acc_top1_avg=0.43137 acc_top5_avg=0.73954 lr=0.00010 gn=41.76321 time=51.46it/s +epoch=85 global_step=33350 loss=5.36259 loss_avg=5.81298 acc=0.47656 acc_top1_avg=0.42683 acc_top5_avg=0.73832 lr=0.00010 gn=30.82566 time=60.20it/s +epoch=85 global_step=33400 loss=4.98511 loss_avg=5.80656 acc=0.52344 acc_top1_avg=0.42675 acc_top5_avg=0.73684 lr=0.00010 gn=28.85757 time=58.23it/s +epoch=85 global_step=33450 loss=5.40279 loss_avg=5.81312 acc=0.46094 acc_top1_avg=0.42587 acc_top5_avg=0.73677 lr=0.00010 gn=32.18489 time=59.29it/s +epoch=85 global_step=33500 loss=5.63670 loss_avg=5.82902 acc=0.43750 acc_top1_avg=0.42373 acc_top5_avg=0.73511 lr=0.00010 gn=29.41911 time=48.40it/s +epoch=85 global_step=33550 loss=5.48711 loss_avg=5.83543 acc=0.46875 acc_top1_avg=0.42250 acc_top5_avg=0.73368 lr=0.00010 gn=30.37157 time=59.44it/s +epoch=85 global_step=33600 loss=5.32780 loss_avg=5.82416 acc=0.48438 acc_top1_avg=0.42406 acc_top5_avg=0.73438 lr=0.00010 gn=31.96360 time=51.14it/s +====================Eval==================== +epoch=85 global_step=33626 loss=5.05289 test_loss_avg=3.37160 acc=0.00000 test_acc_avg=0.25022 test_acc_top5_avg=0.67299 time=233.82it/s +epoch=85 global_step=33626 loss=5.95216 test_loss_avg=3.04597 acc=0.00000 test_acc_avg=0.33129 test_acc_top5_avg=0.77037 time=482.88it/s +curr_acc 0.3313 +BEST_ACC 0.3717 +curr_acc_top5 0.7704 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=6.03655 loss_avg=5.79114 acc=0.41406 acc_top1_avg=0.42904 acc_top5_avg=0.72168 lr=0.00010 gn=35.24714 time=60.08it/s +epoch=86 global_step=33700 loss=5.72381 loss_avg=5.82429 acc=0.42969 acc_top1_avg=0.42620 acc_top5_avg=0.73237 lr=0.00010 gn=35.02220 time=57.21it/s +epoch=86 global_step=33750 loss=6.11452 loss_avg=5.86640 acc=0.38281 acc_top1_avg=0.42061 acc_top5_avg=0.72915 lr=0.00010 gn=36.08445 time=59.04it/s +epoch=86 global_step=33800 loss=5.72742 loss_avg=5.84535 acc=0.42188 acc_top1_avg=0.42223 acc_top5_avg=0.73110 lr=0.00010 gn=29.70918 time=57.95it/s +epoch=86 global_step=33850 loss=4.99226 loss_avg=5.83292 acc=0.52344 acc_top1_avg=0.42372 acc_top5_avg=0.73169 lr=0.00010 gn=31.74776 time=53.05it/s +epoch=86 global_step=33900 loss=5.44556 loss_avg=5.82643 acc=0.46875 acc_top1_avg=0.42470 acc_top5_avg=0.73244 lr=0.00010 gn=33.64606 time=51.30it/s +epoch=86 global_step=33950 loss=5.72332 loss_avg=5.81858 acc=0.45312 acc_top1_avg=0.42547 acc_top5_avg=0.73327 lr=0.00010 gn=32.51639 time=52.81it/s +epoch=86 global_step=34000 loss=5.38407 loss_avg=5.80868 acc=0.47656 acc_top1_avg=0.42649 acc_top5_avg=0.73325 lr=0.00010 gn=38.27862 time=54.33it/s +====================Eval==================== +epoch=86 global_step=34017 loss=2.01078 test_loss_avg=2.01586 acc=0.40625 test_acc_avg=0.42057 test_acc_top5_avg=0.97656 time=234.24it/s +epoch=86 global_step=34017 loss=0.42014 test_loss_avg=3.36755 acc=0.88281 test_acc_avg=0.24149 test_acc_top5_avg=0.67969 time=162.32it/s +epoch=86 global_step=34017 loss=6.28323 test_loss_avg=3.04404 acc=0.00000 test_acc_avg=0.33712 test_acc_top5_avg=0.76632 time=517.11it/s +curr_acc 0.3371 +BEST_ACC 0.3717 +curr_acc_top5 0.7663 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=6.30878 loss_avg=5.74257 acc=0.36719 acc_top1_avg=0.43371 acc_top5_avg=0.73982 lr=0.00010 gn=28.87471 time=62.69it/s +epoch=87 global_step=34100 loss=5.80493 loss_avg=5.77840 acc=0.42969 acc_top1_avg=0.43082 acc_top5_avg=0.73221 lr=0.00010 gn=30.40106 time=52.98it/s +epoch=87 global_step=34150 loss=5.78932 loss_avg=5.76558 acc=0.44531 acc_top1_avg=0.43292 acc_top5_avg=0.73396 lr=0.00010 gn=35.76700 time=53.41it/s +epoch=87 global_step=34200 loss=5.51361 loss_avg=5.79483 acc=0.45312 acc_top1_avg=0.42947 acc_top5_avg=0.73224 lr=0.00010 gn=30.40090 time=54.27it/s +epoch=87 global_step=34250 loss=5.79943 loss_avg=5.77638 acc=0.43750 acc_top1_avg=0.43100 acc_top5_avg=0.73491 lr=0.00010 gn=35.55069 time=62.70it/s +epoch=87 global_step=34300 loss=5.65618 loss_avg=5.79664 acc=0.43750 acc_top1_avg=0.42795 acc_top5_avg=0.73498 lr=0.00010 gn=36.37980 time=53.12it/s +epoch=87 global_step=34350 loss=6.30449 loss_avg=5.79547 acc=0.38281 acc_top1_avg=0.42814 acc_top5_avg=0.73466 lr=0.00010 gn=44.50681 time=54.56it/s +epoch=87 global_step=34400 loss=5.53233 loss_avg=5.78975 acc=0.46875 acc_top1_avg=0.42865 acc_top5_avg=0.73503 lr=0.00010 gn=34.36242 time=55.26it/s +====================Eval==================== +epoch=87 global_step=34408 loss=5.14046 test_loss_avg=2.88940 acc=0.00000 test_acc_avg=0.33738 test_acc_top5_avg=0.79253 time=246.85it/s +epoch=87 global_step=34408 loss=5.16444 test_loss_avg=2.96872 acc=0.00000 test_acc_avg=0.34385 test_acc_top5_avg=0.75923 time=239.28it/s +epoch=87 global_step=34408 loss=6.09273 test_loss_avg=3.03943 acc=0.00000 test_acc_avg=0.33515 test_acc_top5_avg=0.76464 time=502.01it/s +curr_acc 0.3351 +BEST_ACC 0.3717 +curr_acc_top5 0.7646 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=5.48681 loss_avg=5.77809 acc=0.45312 acc_top1_avg=0.42690 acc_top5_avg=0.73958 lr=0.00010 gn=36.40267 time=54.53it/s +epoch=88 global_step=34500 loss=5.85188 loss_avg=5.79571 acc=0.40625 acc_top1_avg=0.42595 acc_top5_avg=0.73412 lr=0.00010 gn=37.74536 time=59.76it/s +epoch=88 global_step=34550 loss=5.90832 loss_avg=5.79433 acc=0.41406 acc_top1_avg=0.42661 acc_top5_avg=0.73415 lr=0.00010 gn=28.31645 time=55.22it/s +epoch=88 global_step=34600 loss=5.72604 loss_avg=5.79915 acc=0.44531 acc_top1_avg=0.42733 acc_top5_avg=0.73478 lr=0.00010 gn=34.24765 time=60.09it/s +epoch=88 global_step=34650 loss=5.88491 loss_avg=5.80082 acc=0.40625 acc_top1_avg=0.42633 acc_top5_avg=0.73412 lr=0.00010 gn=35.59951 time=54.51it/s +epoch=88 global_step=34700 loss=5.44238 loss_avg=5.79655 acc=0.48438 acc_top1_avg=0.42688 acc_top5_avg=0.73253 lr=0.00010 gn=38.41957 time=50.02it/s +epoch=88 global_step=34750 loss=5.89443 loss_avg=5.79407 acc=0.41406 acc_top1_avg=0.42713 acc_top5_avg=0.73305 lr=0.00010 gn=29.41692 time=58.20it/s +====================Eval==================== +epoch=88 global_step=34799 loss=4.71205 test_loss_avg=3.32447 acc=0.00000 test_acc_avg=0.25326 test_acc_top5_avg=0.64616 time=237.01it/s +epoch=88 global_step=34799 loss=6.21464 test_loss_avg=3.05437 acc=0.00000 test_acc_avg=0.33485 test_acc_top5_avg=0.76345 time=639.67it/s +curr_acc 0.3348 +BEST_ACC 0.3717 +curr_acc_top5 0.7634 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=5.73856 loss_avg=5.73856 acc=0.43750 acc_top1_avg=0.43750 acc_top5_avg=0.75000 lr=0.00010 gn=32.51108 time=41.64it/s +epoch=89 global_step=34850 loss=5.83036 loss_avg=5.80616 acc=0.42188 acc_top1_avg=0.42831 acc_top5_avg=0.72564 lr=0.00010 gn=28.81101 time=58.96it/s +epoch=89 global_step=34900 loss=5.24062 loss_avg=5.81164 acc=0.48438 acc_top1_avg=0.42590 acc_top5_avg=0.72935 lr=0.00010 gn=25.76565 time=60.53it/s +epoch=89 global_step=34950 loss=6.08586 loss_avg=5.80726 acc=0.39062 acc_top1_avg=0.42684 acc_top5_avg=0.72998 lr=0.00010 gn=24.21486 time=56.71it/s +epoch=89 global_step=35000 loss=6.34514 loss_avg=5.79491 acc=0.35156 acc_top1_avg=0.42790 acc_top5_avg=0.73064 lr=0.00010 gn=35.72873 time=54.58it/s +epoch=89 global_step=35050 loss=6.05784 loss_avg=5.77481 acc=0.39844 acc_top1_avg=0.43050 acc_top5_avg=0.73213 lr=0.00010 gn=27.11584 time=37.75it/s +epoch=89 global_step=35100 loss=6.08615 loss_avg=5.78939 acc=0.39062 acc_top1_avg=0.42865 acc_top5_avg=0.73157 lr=0.00010 gn=32.76326 time=60.20it/s +epoch=89 global_step=35150 loss=5.90513 loss_avg=5.78557 acc=0.40625 acc_top1_avg=0.42851 acc_top5_avg=0.73206 lr=0.00010 gn=35.52059 time=50.50it/s +====================Eval==================== +epoch=89 global_step=35190 loss=4.59304 test_loss_avg=2.05097 acc=0.00000 test_acc_avg=0.47410 test_acc_top5_avg=0.90625 time=227.27it/s +epoch=89 global_step=35190 loss=0.24429 test_loss_avg=2.80069 acc=0.89844 test_acc_avg=0.36787 test_acc_top5_avg=0.73766 time=240.75it/s +epoch=89 global_step=35190 loss=5.98920 test_loss_avg=3.03471 acc=0.00000 test_acc_avg=0.33683 test_acc_top5_avg=0.76543 time=831.05it/s +curr_acc 0.3368 +BEST_ACC 0.3717 +curr_acc_top5 0.7654 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=5.35108 loss_avg=5.80979 acc=0.48438 acc_top1_avg=0.43516 acc_top5_avg=0.70781 lr=0.00010 gn=35.26125 time=59.51it/s +epoch=90 global_step=35250 loss=5.39635 loss_avg=5.78093 acc=0.49219 acc_top1_avg=0.43307 acc_top5_avg=0.72695 lr=0.00010 gn=37.00641 time=60.28it/s +epoch=90 global_step=35300 loss=5.78764 loss_avg=5.76896 acc=0.42969 acc_top1_avg=0.43196 acc_top5_avg=0.73175 lr=0.00010 gn=34.28207 time=55.17it/s +epoch=90 global_step=35350 loss=5.65478 loss_avg=5.78125 acc=0.46875 acc_top1_avg=0.43091 acc_top5_avg=0.72881 lr=0.00010 gn=37.42088 time=56.93it/s +epoch=90 global_step=35400 loss=5.20985 loss_avg=5.79538 acc=0.48438 acc_top1_avg=0.42861 acc_top5_avg=0.72827 lr=0.00010 gn=27.56641 time=56.60it/s +epoch=90 global_step=35450 loss=5.82237 loss_avg=5.79451 acc=0.43750 acc_top1_avg=0.42843 acc_top5_avg=0.72858 lr=0.00010 gn=35.37887 time=54.87it/s +epoch=90 global_step=35500 loss=5.20708 loss_avg=5.78939 acc=0.48438 acc_top1_avg=0.42908 acc_top5_avg=0.72941 lr=0.00010 gn=29.49513 time=56.54it/s +epoch=90 global_step=35550 loss=5.76033 loss_avg=5.78783 acc=0.42188 acc_top1_avg=0.42888 acc_top5_avg=0.72921 lr=0.00010 gn=37.53370 time=54.37it/s +====================Eval==================== +epoch=90 global_step=35581 loss=2.48951 test_loss_avg=3.52718 acc=0.37500 test_acc_avg=0.23047 test_acc_top5_avg=0.59922 time=234.98it/s +epoch=90 global_step=35581 loss=6.25040 test_loss_avg=3.08277 acc=0.00000 test_acc_avg=0.33297 test_acc_top5_avg=0.76305 time=510.07it/s +curr_acc 0.3330 +BEST_ACC 0.3717 +curr_acc_top5 0.7631 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=6.02454 loss_avg=5.62018 acc=0.40625 acc_top1_avg=0.44737 acc_top5_avg=0.73602 lr=0.00010 gn=35.38321 time=54.60it/s +epoch=91 global_step=35650 loss=5.88283 loss_avg=5.77776 acc=0.42188 acc_top1_avg=0.42923 acc_top5_avg=0.73268 lr=0.00010 gn=35.86940 time=62.68it/s +epoch=91 global_step=35700 loss=5.54491 loss_avg=5.78092 acc=0.46094 acc_top1_avg=0.43028 acc_top5_avg=0.73444 lr=0.00010 gn=32.82752 time=56.83it/s +epoch=91 global_step=35750 loss=5.20175 loss_avg=5.76566 acc=0.48438 acc_top1_avg=0.43144 acc_top5_avg=0.73567 lr=0.00010 gn=30.91060 time=61.47it/s +epoch=91 global_step=35800 loss=5.67947 loss_avg=5.76504 acc=0.42969 acc_top1_avg=0.43076 acc_top5_avg=0.73334 lr=0.00010 gn=35.02931 time=58.10it/s +epoch=91 global_step=35850 loss=6.04054 loss_avg=5.77887 acc=0.40625 acc_top1_avg=0.42922 acc_top5_avg=0.73214 lr=0.00010 gn=28.25457 time=40.99it/s +epoch=91 global_step=35900 loss=5.93408 loss_avg=5.77653 acc=0.41406 acc_top1_avg=0.42952 acc_top5_avg=0.73357 lr=0.00010 gn=36.38661 time=52.26it/s +epoch=91 global_step=35950 loss=5.39128 loss_avg=5.77288 acc=0.46875 acc_top1_avg=0.43003 acc_top5_avg=0.73319 lr=0.00010 gn=34.50625 time=50.95it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.74217 test_loss_avg=1.67856 acc=0.79688 test_acc_avg=0.52699 test_acc_top5_avg=0.98509 time=222.89it/s +epoch=91 global_step=35972 loss=0.38989 test_loss_avg=3.14456 acc=0.87500 test_acc_avg=0.29495 test_acc_top5_avg=0.70069 time=239.95it/s +epoch=91 global_step=35972 loss=6.13387 test_loss_avg=3.04677 acc=0.00000 test_acc_avg=0.33821 test_acc_top5_avg=0.76266 time=512.69it/s +curr_acc 0.3382 +BEST_ACC 0.3717 +curr_acc_top5 0.7627 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=5.55370 loss_avg=5.68414 acc=0.46875 acc_top1_avg=0.44364 acc_top5_avg=0.73996 lr=0.00010 gn=35.30940 time=54.21it/s +epoch=92 global_step=36050 loss=5.57747 loss_avg=5.74273 acc=0.47656 acc_top1_avg=0.43450 acc_top5_avg=0.73488 lr=0.00010 gn=41.12605 time=53.49it/s +epoch=92 global_step=36100 loss=5.30030 loss_avg=5.72833 acc=0.46875 acc_top1_avg=0.43604 acc_top5_avg=0.73529 lr=0.00010 gn=29.07954 time=62.36it/s +epoch=92 global_step=36150 loss=5.93368 loss_avg=5.75098 acc=0.39844 acc_top1_avg=0.43250 acc_top5_avg=0.73328 lr=0.00010 gn=29.08472 time=59.52it/s +epoch=92 global_step=36200 loss=5.85289 loss_avg=5.74248 acc=0.43750 acc_top1_avg=0.43339 acc_top5_avg=0.73331 lr=0.00010 gn=32.15727 time=62.98it/s +epoch=92 global_step=36250 loss=5.82912 loss_avg=5.76023 acc=0.42188 acc_top1_avg=0.43163 acc_top5_avg=0.73291 lr=0.00010 gn=33.63838 time=58.44it/s +epoch=92 global_step=36300 loss=5.46810 loss_avg=5.76158 acc=0.45312 acc_top1_avg=0.43126 acc_top5_avg=0.73335 lr=0.00010 gn=30.48249 time=54.66it/s +epoch=92 global_step=36350 loss=5.94996 loss_avg=5.75240 acc=0.40625 acc_top1_avg=0.43233 acc_top5_avg=0.73398 lr=0.00010 gn=25.51612 time=55.92it/s +====================Eval==================== +epoch=92 global_step=36363 loss=4.77413 test_loss_avg=3.21013 acc=0.00000 test_acc_avg=0.27344 test_acc_top5_avg=0.70630 time=234.63it/s +epoch=92 global_step=36363 loss=6.36999 test_loss_avg=3.05721 acc=0.00000 test_acc_avg=0.33297 test_acc_top5_avg=0.75939 time=801.66it/s +curr_acc 0.3330 +BEST_ACC 0.3717 +curr_acc_top5 0.7594 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=6.10118 loss_avg=5.77310 acc=0.41406 acc_top1_avg=0.42927 acc_top5_avg=0.72149 lr=0.00010 gn=34.97556 time=49.24it/s +epoch=93 global_step=36450 loss=6.11760 loss_avg=5.76401 acc=0.38281 acc_top1_avg=0.43032 acc_top5_avg=0.72647 lr=0.00010 gn=35.06062 time=54.89it/s +epoch=93 global_step=36500 loss=5.46724 loss_avg=5.75545 acc=0.47656 acc_top1_avg=0.43237 acc_top5_avg=0.73112 lr=0.00010 gn=32.64839 time=56.13it/s +epoch=93 global_step=36550 loss=5.72382 loss_avg=5.74883 acc=0.42188 acc_top1_avg=0.43303 acc_top5_avg=0.73446 lr=0.00010 gn=34.29974 time=59.63it/s +epoch=93 global_step=36600 loss=5.56578 loss_avg=5.73049 acc=0.46875 acc_top1_avg=0.43486 acc_top5_avg=0.73527 lr=0.00010 gn=33.27591 time=49.87it/s +epoch=93 global_step=36650 loss=6.15780 loss_avg=5.73176 acc=0.38281 acc_top1_avg=0.43432 acc_top5_avg=0.73557 lr=0.00010 gn=33.30550 time=55.17it/s +epoch=93 global_step=36700 loss=5.69479 loss_avg=5.73996 acc=0.43750 acc_top1_avg=0.43351 acc_top5_avg=0.73475 lr=0.00010 gn=34.14593 time=60.51it/s +epoch=93 global_step=36750 loss=5.67141 loss_avg=5.75333 acc=0.42969 acc_top1_avg=0.43211 acc_top5_avg=0.73377 lr=0.00010 gn=33.78805 time=58.23it/s +====================Eval==================== +epoch=93 global_step=36754 loss=2.29354 test_loss_avg=2.26240 acc=0.35156 test_acc_avg=0.36458 test_acc_top5_avg=0.96875 time=245.57it/s +epoch=93 global_step=36754 loss=4.57995 test_loss_avg=3.45212 acc=0.00000 test_acc_avg=0.22907 test_acc_top5_avg=0.66274 time=240.46it/s +epoch=93 global_step=36754 loss=6.24573 test_loss_avg=3.06061 acc=0.00000 test_acc_avg=0.33436 test_acc_top5_avg=0.76058 time=860.02it/s +curr_acc 0.3344 +BEST_ACC 0.3717 +curr_acc_top5 0.7606 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=5.94080 loss_avg=5.75664 acc=0.42188 acc_top1_avg=0.42986 acc_top5_avg=0.72673 lr=0.00010 gn=34.12614 time=43.67it/s +epoch=94 global_step=36850 loss=5.74426 loss_avg=5.74609 acc=0.44531 acc_top1_avg=0.43197 acc_top5_avg=0.73356 lr=0.00010 gn=42.55508 time=55.00it/s +epoch=94 global_step=36900 loss=6.06750 loss_avg=5.74321 acc=0.39062 acc_top1_avg=0.43231 acc_top5_avg=0.73432 lr=0.00010 gn=38.00724 time=59.85it/s +epoch=94 global_step=36950 loss=6.29805 loss_avg=5.74058 acc=0.37500 acc_top1_avg=0.43300 acc_top5_avg=0.73489 lr=0.00010 gn=32.52089 time=60.18it/s +epoch=94 global_step=37000 loss=5.59892 loss_avg=5.74046 acc=0.42969 acc_top1_avg=0.43258 acc_top5_avg=0.73463 lr=0.00010 gn=24.93806 time=54.84it/s +epoch=94 global_step=37050 loss=6.01393 loss_avg=5.72600 acc=0.40625 acc_top1_avg=0.43420 acc_top5_avg=0.73424 lr=0.00010 gn=36.88946 time=59.28it/s +epoch=94 global_step=37100 loss=6.06264 loss_avg=5.73589 acc=0.39844 acc_top1_avg=0.43344 acc_top5_avg=0.73388 lr=0.00010 gn=35.54027 time=59.74it/s +====================Eval==================== +epoch=94 global_step=37145 loss=4.80038 test_loss_avg=2.70613 acc=0.00000 test_acc_avg=0.34603 test_acc_top5_avg=0.81478 time=57.83it/s +epoch=94 global_step=37145 loss=5.41556 test_loss_avg=2.90370 acc=0.00000 test_acc_avg=0.34840 test_acc_top5_avg=0.74514 time=241.80it/s +epoch=94 global_step=37145 loss=6.26525 test_loss_avg=3.07324 acc=0.00000 test_acc_avg=0.32634 test_acc_top5_avg=0.75920 time=499.08it/s +curr_acc 0.3263 +BEST_ACC 0.3717 +curr_acc_top5 0.7592 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=6.19680 loss_avg=5.82408 acc=0.38281 acc_top1_avg=0.42188 acc_top5_avg=0.73750 lr=0.00010 gn=36.73536 time=55.43it/s +epoch=95 global_step=37200 loss=5.96713 loss_avg=5.86865 acc=0.40625 acc_top1_avg=0.41719 acc_top5_avg=0.72528 lr=0.00010 gn=40.69721 time=56.24it/s +epoch=95 global_step=37250 loss=5.36618 loss_avg=5.84339 acc=0.47656 acc_top1_avg=0.42054 acc_top5_avg=0.72708 lr=0.00010 gn=37.72095 time=56.98it/s +epoch=95 global_step=37300 loss=5.91241 loss_avg=5.77021 acc=0.39844 acc_top1_avg=0.42853 acc_top5_avg=0.72969 lr=0.00010 gn=34.02434 time=55.26it/s +epoch=95 global_step=37350 loss=5.80406 loss_avg=5.75567 acc=0.40625 acc_top1_avg=0.43041 acc_top5_avg=0.73072 lr=0.00010 gn=30.47085 time=62.58it/s +epoch=95 global_step=37400 loss=5.74914 loss_avg=5.74314 acc=0.42188 acc_top1_avg=0.43177 acc_top5_avg=0.73287 lr=0.00010 gn=38.61146 time=54.31it/s +epoch=95 global_step=37450 loss=5.77804 loss_avg=5.73411 acc=0.42188 acc_top1_avg=0.43276 acc_top5_avg=0.73230 lr=0.00010 gn=35.73483 time=29.77it/s +epoch=95 global_step=37500 loss=5.00458 loss_avg=5.72543 acc=0.49219 acc_top1_avg=0.43402 acc_top5_avg=0.73272 lr=0.00010 gn=28.96778 time=57.80it/s +====================Eval==================== +epoch=95 global_step=37536 loss=1.92006 test_loss_avg=3.35221 acc=0.43750 test_acc_avg=0.24670 test_acc_top5_avg=0.63542 time=240.15it/s +epoch=95 global_step=37536 loss=6.29099 test_loss_avg=3.08120 acc=0.00000 test_acc_avg=0.32961 test_acc_top5_avg=0.76365 time=832.53it/s +curr_acc 0.3296 +BEST_ACC 0.3717 +curr_acc_top5 0.7636 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=5.93249 loss_avg=5.69980 acc=0.41406 acc_top1_avg=0.43862 acc_top5_avg=0.73661 lr=0.00010 gn=36.33936 time=52.16it/s +epoch=96 global_step=37600 loss=5.96120 loss_avg=5.70480 acc=0.42188 acc_top1_avg=0.43799 acc_top5_avg=0.73792 lr=0.00010 gn=38.00861 time=55.01it/s +epoch=96 global_step=37650 loss=5.82156 loss_avg=5.71391 acc=0.42188 acc_top1_avg=0.43688 acc_top5_avg=0.73451 lr=0.00010 gn=30.65538 time=55.82it/s +epoch=96 global_step=37700 loss=5.47788 loss_avg=5.72583 acc=0.46875 acc_top1_avg=0.43626 acc_top5_avg=0.73318 lr=0.00010 gn=36.84707 time=59.01it/s +epoch=96 global_step=37750 loss=5.98426 loss_avg=5.71293 acc=0.38281 acc_top1_avg=0.43728 acc_top5_avg=0.73364 lr=0.00010 gn=36.00301 time=61.44it/s +epoch=96 global_step=37800 loss=5.64998 loss_avg=5.72062 acc=0.43750 acc_top1_avg=0.43617 acc_top5_avg=0.73396 lr=0.00010 gn=38.56283 time=58.04it/s +epoch=96 global_step=37850 loss=6.22845 loss_avg=5.72186 acc=0.35938 acc_top1_avg=0.43566 acc_top5_avg=0.73482 lr=0.00010 gn=29.70645 time=60.15it/s +epoch=96 global_step=37900 loss=5.90101 loss_avg=5.71767 acc=0.41406 acc_top1_avg=0.43619 acc_top5_avg=0.73468 lr=0.00010 gn=30.12667 time=56.86it/s +====================Eval==================== +epoch=96 global_step=37927 loss=2.40086 test_loss_avg=1.70961 acc=0.44531 test_acc_avg=0.52881 test_acc_top5_avg=0.97021 time=192.68it/s +epoch=96 global_step=37927 loss=0.17382 test_loss_avg=2.98217 acc=0.95312 test_acc_avg=0.33037 test_acc_top5_avg=0.72100 time=243.40it/s +epoch=96 global_step=37927 loss=6.18678 test_loss_avg=3.09287 acc=0.00000 test_acc_avg=0.32803 test_acc_top5_avg=0.76177 time=641.33it/s +curr_acc 0.3280 +BEST_ACC 0.3717 +curr_acc_top5 0.7618 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=5.82874 loss_avg=5.74360 acc=0.42188 acc_top1_avg=0.42833 acc_top5_avg=0.72317 lr=0.00010 gn=37.41960 time=61.13it/s +epoch=97 global_step=38000 loss=5.34390 loss_avg=5.68663 acc=0.46875 acc_top1_avg=0.43750 acc_top5_avg=0.72785 lr=0.00010 gn=38.58734 time=58.46it/s +epoch=97 global_step=38050 loss=6.00698 loss_avg=5.71815 acc=0.40625 acc_top1_avg=0.43413 acc_top5_avg=0.72796 lr=0.00010 gn=30.95021 time=57.48it/s +epoch=97 global_step=38100 loss=5.78677 loss_avg=5.71828 acc=0.43750 acc_top1_avg=0.43551 acc_top5_avg=0.72950 lr=0.00010 gn=36.08648 time=54.37it/s +epoch=97 global_step=38150 loss=5.49453 loss_avg=5.72725 acc=0.46094 acc_top1_avg=0.43424 acc_top5_avg=0.72937 lr=0.00010 gn=31.14848 time=53.47it/s +epoch=97 global_step=38200 loss=5.43356 loss_avg=5.72638 acc=0.47656 acc_top1_avg=0.43461 acc_top5_avg=0.73011 lr=0.00010 gn=32.50629 time=53.11it/s +epoch=97 global_step=38250 loss=5.75008 loss_avg=5.71227 acc=0.41406 acc_top1_avg=0.43619 acc_top5_avg=0.73198 lr=0.00010 gn=31.88318 time=63.09it/s +epoch=97 global_step=38300 loss=5.56890 loss_avg=5.71244 acc=0.43750 acc_top1_avg=0.43601 acc_top5_avg=0.73280 lr=0.00010 gn=33.42151 time=54.28it/s +====================Eval==================== +epoch=97 global_step=38318 loss=5.25592 test_loss_avg=3.44409 acc=0.00000 test_acc_avg=0.23628 test_acc_top5_avg=0.61212 time=236.07it/s +epoch=97 global_step=38318 loss=6.10536 test_loss_avg=3.03950 acc=0.00000 test_acc_avg=0.33307 test_acc_top5_avg=0.75811 time=580.85it/s +curr_acc 0.3331 +BEST_ACC 0.3717 +curr_acc_top5 0.7581 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=5.22052 loss_avg=5.73659 acc=0.49219 acc_top1_avg=0.42920 acc_top5_avg=0.73267 lr=0.00010 gn=35.28691 time=51.81it/s +epoch=98 global_step=38400 loss=5.36109 loss_avg=5.72374 acc=0.45312 acc_top1_avg=0.43397 acc_top5_avg=0.73428 lr=0.00010 gn=33.07537 time=63.09it/s +epoch=98 global_step=38450 loss=5.79359 loss_avg=5.70813 acc=0.43750 acc_top1_avg=0.43685 acc_top5_avg=0.73627 lr=0.00010 gn=34.90624 time=57.35it/s +epoch=98 global_step=38500 loss=5.57347 loss_avg=5.72071 acc=0.45312 acc_top1_avg=0.43557 acc_top5_avg=0.73339 lr=0.00010 gn=32.37738 time=59.14it/s +epoch=98 global_step=38550 loss=6.25436 loss_avg=5.71676 acc=0.36719 acc_top1_avg=0.43609 acc_top5_avg=0.73414 lr=0.00010 gn=32.49280 time=55.05it/s +epoch=98 global_step=38600 loss=5.65768 loss_avg=5.72082 acc=0.44531 acc_top1_avg=0.43611 acc_top5_avg=0.73271 lr=0.00010 gn=42.62705 time=62.21it/s +epoch=98 global_step=38650 loss=5.83173 loss_avg=5.70891 acc=0.43750 acc_top1_avg=0.43734 acc_top5_avg=0.73461 lr=0.00010 gn=40.53945 time=59.47it/s +epoch=98 global_step=38700 loss=5.65223 loss_avg=5.71262 acc=0.45312 acc_top1_avg=0.43685 acc_top5_avg=0.73468 lr=0.00010 gn=38.75646 time=57.37it/s +====================Eval==================== +epoch=98 global_step=38709 loss=1.74035 test_loss_avg=2.12144 acc=0.52344 test_acc_avg=0.39160 test_acc_top5_avg=0.96777 time=137.86it/s +epoch=98 global_step=38709 loss=0.47800 test_loss_avg=3.28130 acc=0.87500 test_acc_avg=0.26145 test_acc_top5_avg=0.68602 time=240.24it/s +epoch=98 global_step=38709 loss=6.11874 test_loss_avg=3.04748 acc=0.00000 test_acc_avg=0.33337 test_acc_top5_avg=0.76177 time=499.20it/s +curr_acc 0.3334 +BEST_ACC 0.3717 +curr_acc_top5 0.7618 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=4.97221 loss_avg=5.72424 acc=0.50781 acc_top1_avg=0.43579 acc_top5_avg=0.73590 lr=0.00010 gn=39.94226 time=52.63it/s +epoch=99 global_step=38800 loss=6.07302 loss_avg=5.73739 acc=0.39844 acc_top1_avg=0.43355 acc_top5_avg=0.73609 lr=0.00010 gn=35.45257 time=48.80it/s +epoch=99 global_step=38850 loss=5.93602 loss_avg=5.71994 acc=0.40625 acc_top1_avg=0.43539 acc_top5_avg=0.73631 lr=0.00010 gn=31.11278 time=57.56it/s +epoch=99 global_step=38900 loss=6.10571 loss_avg=5.71254 acc=0.39062 acc_top1_avg=0.43599 acc_top5_avg=0.73495 lr=0.00010 gn=31.99716 time=53.39it/s +epoch=99 global_step=38950 loss=5.41450 loss_avg=5.70680 acc=0.47656 acc_top1_avg=0.43672 acc_top5_avg=0.73502 lr=0.00010 gn=33.72422 time=54.21it/s +epoch=99 global_step=39000 loss=5.57902 loss_avg=5.69038 acc=0.44531 acc_top1_avg=0.43906 acc_top5_avg=0.73518 lr=0.00010 gn=29.69391 time=60.51it/s +epoch=99 global_step=39050 loss=5.72325 loss_avg=5.69153 acc=0.46094 acc_top1_avg=0.43901 acc_top5_avg=0.73433 lr=0.00010 gn=45.24824 time=50.87it/s +epoch=99 global_step=39100 loss=6.52902 loss_avg=5.69658 acc=0.33750 acc_top1_avg=0.43886 acc_top5_avg=0.73359 lr=0.00010 gn=40.57541 time=85.85it/s +====================Eval==================== +epoch=99 global_step=39100 loss=4.84624 test_loss_avg=3.03774 acc=0.00000 test_acc_avg=0.30603 test_acc_top5_avg=0.75485 time=224.03it/s +epoch=99 global_step=39100 loss=5.97640 test_loss_avg=3.06528 acc=0.00000 test_acc_avg=0.32981 test_acc_top5_avg=0.75732 time=858.61it/s +epoch=99 global_step=39100 loss=5.97640 test_loss_avg=3.06528 acc=0.00000 test_acc_avg=0.32981 test_acc_top5_avg=0.75732 time=858.61it/s +curr_acc 0.3298 +BEST_ACC 0.3717 +curr_acc_top5 0.7573 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=6.19297 loss_avg=5.70471 acc=0.37500 acc_top1_avg=0.43641 acc_top5_avg=0.73578 lr=0.00010 gn=31.10287 time=54.80it/s +epoch=100 global_step=39200 loss=5.77992 loss_avg=5.63752 acc=0.43750 acc_top1_avg=0.44391 acc_top5_avg=0.73547 lr=0.00010 gn=35.51474 time=51.97it/s +epoch=100 global_step=39250 loss=5.97429 loss_avg=5.65778 acc=0.41406 acc_top1_avg=0.44146 acc_top5_avg=0.73354 lr=0.00010 gn=34.55478 time=57.29it/s +epoch=100 global_step=39300 loss=5.63695 loss_avg=5.66401 acc=0.46094 acc_top1_avg=0.44125 acc_top5_avg=0.73195 lr=0.00010 gn=45.91236 time=55.21it/s +epoch=100 global_step=39350 loss=5.36675 loss_avg=5.68240 acc=0.49219 acc_top1_avg=0.43931 acc_top5_avg=0.73225 lr=0.00010 gn=42.30986 time=54.90it/s +epoch=100 global_step=39400 loss=6.23241 loss_avg=5.69296 acc=0.37500 acc_top1_avg=0.43807 acc_top5_avg=0.73307 lr=0.00010 gn=34.32920 time=52.39it/s +epoch=100 global_step=39450 loss=5.82043 loss_avg=5.69542 acc=0.42188 acc_top1_avg=0.43815 acc_top5_avg=0.73237 lr=0.00010 gn=40.53235 time=62.45it/s +====================Eval==================== +epoch=100 global_step=39491 loss=4.58492 test_loss_avg=3.42176 acc=0.00000 test_acc_avg=0.23656 test_acc_top5_avg=0.64828 time=201.42it/s +epoch=100 global_step=39491 loss=6.12357 test_loss_avg=3.11388 acc=0.00000 test_acc_avg=0.32902 test_acc_top5_avg=0.75979 time=850.08it/s +curr_acc 0.3290 +BEST_ACC 0.3717 +curr_acc_top5 0.7598 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=6.05114 loss_avg=5.66753 acc=0.38281 acc_top1_avg=0.43924 acc_top5_avg=0.71007 lr=0.00010 gn=32.85080 time=55.83it/s +epoch=101 global_step=39550 loss=6.09256 loss_avg=5.67327 acc=0.41406 acc_top1_avg=0.44187 acc_top5_avg=0.73477 lr=0.00010 gn=42.72366 time=48.42it/s +epoch=101 global_step=39600 loss=5.27230 loss_avg=5.68809 acc=0.47656 acc_top1_avg=0.43994 acc_top5_avg=0.73645 lr=0.00010 gn=36.56418 time=60.10it/s +epoch=101 global_step=39650 loss=5.55117 loss_avg=5.66665 acc=0.45312 acc_top1_avg=0.44168 acc_top5_avg=0.73772 lr=0.00010 gn=35.70050 time=59.13it/s +epoch=101 global_step=39700 loss=5.89098 loss_avg=5.66444 acc=0.40625 acc_top1_avg=0.44195 acc_top5_avg=0.73594 lr=0.00010 gn=40.26041 time=55.53it/s +epoch=101 global_step=39750 loss=5.81774 loss_avg=5.68050 acc=0.43750 acc_top1_avg=0.44073 acc_top5_avg=0.73332 lr=0.00010 gn=40.79131 time=48.86it/s +epoch=101 global_step=39800 loss=5.78714 loss_avg=5.68984 acc=0.40625 acc_top1_avg=0.43957 acc_top5_avg=0.73180 lr=0.00010 gn=35.93467 time=60.67it/s +epoch=101 global_step=39850 loss=5.57537 loss_avg=5.69251 acc=0.46094 acc_top1_avg=0.43931 acc_top5_avg=0.73250 lr=0.00010 gn=42.63351 time=55.70it/s +====================Eval==================== +epoch=101 global_step=39882 loss=4.61635 test_loss_avg=2.29634 acc=0.00000 test_acc_avg=0.42857 test_acc_top5_avg=0.86161 time=241.02it/s +epoch=101 global_step=39882 loss=3.72354 test_loss_avg=2.80160 acc=0.29688 test_acc_avg=0.36917 test_acc_top5_avg=0.73404 time=250.95it/s +epoch=101 global_step=39882 loss=6.14125 test_loss_avg=3.07147 acc=0.00000 test_acc_avg=0.33178 test_acc_top5_avg=0.75702 time=799.22it/s +curr_acc 0.3318 +BEST_ACC 0.3717 +curr_acc_top5 0.7570 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=5.88309 loss_avg=5.60816 acc=0.42188 acc_top1_avg=0.45009 acc_top5_avg=0.74349 lr=0.00010 gn=36.62932 time=54.13it/s +epoch=102 global_step=39950 loss=5.15012 loss_avg=5.66745 acc=0.50781 acc_top1_avg=0.44221 acc_top5_avg=0.74092 lr=0.00010 gn=39.02649 time=53.83it/s +epoch=102 global_step=40000 loss=6.42861 loss_avg=5.70289 acc=0.35938 acc_top1_avg=0.43863 acc_top5_avg=0.72974 lr=0.00010 gn=37.62785 time=55.39it/s +epoch=102 global_step=40050 loss=5.50039 loss_avg=5.70286 acc=0.44531 acc_top1_avg=0.43857 acc_top5_avg=0.73233 lr=0.00010 gn=33.44348 time=58.26it/s +epoch=102 global_step=40100 loss=5.49642 loss_avg=5.69417 acc=0.45312 acc_top1_avg=0.43965 acc_top5_avg=0.73280 lr=0.00010 gn=37.92702 time=57.10it/s +epoch=102 global_step=40150 loss=6.34947 loss_avg=5.69160 acc=0.35938 acc_top1_avg=0.44009 acc_top5_avg=0.73333 lr=0.00010 gn=24.56009 time=54.03it/s +epoch=102 global_step=40200 loss=4.83619 loss_avg=5.68452 acc=0.52344 acc_top1_avg=0.44094 acc_top5_avg=0.73391 lr=0.00010 gn=38.45089 time=44.24it/s +epoch=102 global_step=40250 loss=5.67662 loss_avg=5.68805 acc=0.44531 acc_top1_avg=0.44020 acc_top5_avg=0.73376 lr=0.00010 gn=38.17054 time=54.37it/s +====================Eval==================== +epoch=102 global_step=40273 loss=2.08309 test_loss_avg=3.44041 acc=0.45312 test_acc_avg=0.23549 test_acc_top5_avg=0.61068 time=187.37it/s +epoch=102 global_step=40273 loss=5.93649 test_loss_avg=3.07637 acc=0.00000 test_acc_avg=0.32694 test_acc_top5_avg=0.75771 time=566.64it/s +curr_acc 0.3269 +BEST_ACC 0.3717 +curr_acc_top5 0.7577 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=6.02537 loss_avg=5.73313 acc=0.40625 acc_top1_avg=0.43345 acc_top5_avg=0.73119 lr=0.00010 gn=38.81236 time=55.66it/s +epoch=103 global_step=40350 loss=5.48196 loss_avg=5.62633 acc=0.46875 acc_top1_avg=0.44663 acc_top5_avg=0.73935 lr=0.00010 gn=29.99566 time=50.70it/s +epoch=103 global_step=40400 loss=5.64817 loss_avg=5.66952 acc=0.45312 acc_top1_avg=0.44199 acc_top5_avg=0.73628 lr=0.00010 gn=38.98165 time=59.66it/s +epoch=103 global_step=40450 loss=5.33020 loss_avg=5.64924 acc=0.48438 acc_top1_avg=0.44425 acc_top5_avg=0.73685 lr=0.00010 gn=34.12963 time=53.98it/s +epoch=103 global_step=40500 loss=6.18517 loss_avg=5.66426 acc=0.37500 acc_top1_avg=0.44225 acc_top5_avg=0.73654 lr=0.00010 gn=37.80423 time=56.89it/s +epoch=103 global_step=40550 loss=5.44464 loss_avg=5.66784 acc=0.46875 acc_top1_avg=0.44196 acc_top5_avg=0.73547 lr=0.00010 gn=37.38680 time=57.61it/s +epoch=103 global_step=40600 loss=5.71820 loss_avg=5.67673 acc=0.42969 acc_top1_avg=0.44108 acc_top5_avg=0.73607 lr=0.00010 gn=37.67520 time=61.29it/s +epoch=103 global_step=40650 loss=5.75479 loss_avg=5.67981 acc=0.43750 acc_top1_avg=0.44075 acc_top5_avg=0.73525 lr=0.00010 gn=45.07012 time=54.74it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.90158 test_loss_avg=1.63517 acc=0.73438 test_acc_avg=0.53666 test_acc_top5_avg=0.97296 time=239.91it/s +epoch=103 global_step=40664 loss=0.20816 test_loss_avg=3.07985 acc=0.94531 test_acc_avg=0.30642 test_acc_top5_avg=0.70486 time=234.32it/s +epoch=103 global_step=40664 loss=6.01113 test_loss_avg=3.06758 acc=0.00000 test_acc_avg=0.33169 test_acc_top5_avg=0.75880 time=845.97it/s +curr_acc 0.3317 +BEST_ACC 0.3717 +curr_acc_top5 0.7588 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=4.93926 loss_avg=5.59117 acc=0.51562 acc_top1_avg=0.45030 acc_top5_avg=0.74110 lr=0.00010 gn=27.71168 time=56.69it/s +epoch=104 global_step=40750 loss=5.59701 loss_avg=5.62743 acc=0.44531 acc_top1_avg=0.44622 acc_top5_avg=0.72947 lr=0.00010 gn=40.10633 time=63.14it/s +epoch=104 global_step=40800 loss=6.10779 loss_avg=5.64534 acc=0.39844 acc_top1_avg=0.44451 acc_top5_avg=0.72984 lr=0.00010 gn=31.93100 time=55.66it/s +epoch=104 global_step=40850 loss=5.91210 loss_avg=5.67049 acc=0.42188 acc_top1_avg=0.44094 acc_top5_avg=0.73080 lr=0.00010 gn=42.72038 time=54.74it/s +epoch=104 global_step=40900 loss=5.73087 loss_avg=5.69446 acc=0.43750 acc_top1_avg=0.43793 acc_top5_avg=0.72928 lr=0.00010 gn=39.44349 time=48.82it/s +epoch=104 global_step=40950 loss=5.72790 loss_avg=5.69212 acc=0.43750 acc_top1_avg=0.43794 acc_top5_avg=0.72979 lr=0.00010 gn=41.37329 time=62.86it/s +epoch=104 global_step=41000 loss=5.67211 loss_avg=5.67805 acc=0.42969 acc_top1_avg=0.43957 acc_top5_avg=0.73121 lr=0.00010 gn=40.27404 time=56.07it/s +epoch=104 global_step=41050 loss=5.17246 loss_avg=5.68040 acc=0.49219 acc_top1_avg=0.43924 acc_top5_avg=0.73233 lr=0.00010 gn=40.26599 time=63.39it/s +====================Eval==================== +epoch=104 global_step=41055 loss=5.07974 test_loss_avg=3.36678 acc=0.00000 test_acc_avg=0.24839 test_acc_top5_avg=0.66176 time=233.56it/s +epoch=104 global_step=41055 loss=6.20957 test_loss_avg=3.08875 acc=0.00000 test_acc_avg=0.33149 test_acc_top5_avg=0.75564 time=797.24it/s +curr_acc 0.3315 +BEST_ACC 0.3717 +curr_acc_top5 0.7556 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=5.66152 loss_avg=5.62551 acc=0.42969 acc_top1_avg=0.44462 acc_top5_avg=0.74271 lr=0.00010 gn=40.36313 time=31.45it/s +epoch=105 global_step=41150 loss=5.91089 loss_avg=5.62300 acc=0.41406 acc_top1_avg=0.44778 acc_top5_avg=0.73873 lr=0.00010 gn=41.53011 time=54.85it/s +epoch=105 global_step=41200 loss=5.71579 loss_avg=5.62371 acc=0.43750 acc_top1_avg=0.44682 acc_top5_avg=0.73675 lr=0.00010 gn=41.82606 time=54.93it/s +epoch=105 global_step=41250 loss=6.60280 loss_avg=5.63946 acc=0.33594 acc_top1_avg=0.44447 acc_top5_avg=0.73554 lr=0.00010 gn=43.30085 time=55.51it/s +epoch=105 global_step=41300 loss=5.32591 loss_avg=5.62352 acc=0.47656 acc_top1_avg=0.44668 acc_top5_avg=0.73629 lr=0.00010 gn=36.53524 time=54.79it/s +epoch=105 global_step=41350 loss=5.91735 loss_avg=5.62640 acc=0.41406 acc_top1_avg=0.44627 acc_top5_avg=0.73623 lr=0.00010 gn=39.22068 time=58.96it/s +epoch=105 global_step=41400 loss=5.85640 loss_avg=5.65015 acc=0.42969 acc_top1_avg=0.44364 acc_top5_avg=0.73424 lr=0.00010 gn=36.96809 time=55.24it/s +====================Eval==================== +epoch=105 global_step=41446 loss=2.15389 test_loss_avg=2.19948 acc=0.36719 test_acc_avg=0.37188 test_acc_top5_avg=0.96562 time=184.96it/s +epoch=105 global_step=41446 loss=3.39186 test_loss_avg=3.46905 acc=0.25000 test_acc_avg=0.22429 test_acc_top5_avg=0.66207 time=237.09it/s +epoch=105 global_step=41446 loss=6.03725 test_loss_avg=3.06863 acc=0.00000 test_acc_avg=0.33218 test_acc_top5_avg=0.75643 time=702.09it/s +curr_acc 0.3322 +BEST_ACC 0.3717 +curr_acc_top5 0.7564 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=5.36360 loss_avg=5.91701 acc=0.47656 acc_top1_avg=0.41797 acc_top5_avg=0.74609 lr=0.00010 gn=34.80879 time=51.65it/s +epoch=106 global_step=41500 loss=5.39878 loss_avg=5.60754 acc=0.47656 acc_top1_avg=0.44936 acc_top5_avg=0.73626 lr=0.00010 gn=34.43197 time=54.77it/s +epoch=106 global_step=41550 loss=5.73969 loss_avg=5.64230 acc=0.42969 acc_top1_avg=0.44449 acc_top5_avg=0.73400 lr=0.00010 gn=33.56431 time=55.83it/s +epoch=106 global_step=41600 loss=6.04300 loss_avg=5.64667 acc=0.39062 acc_top1_avg=0.44349 acc_top5_avg=0.73605 lr=0.00010 gn=39.28776 time=60.58it/s +epoch=106 global_step=41650 loss=5.82047 loss_avg=5.66407 acc=0.42969 acc_top1_avg=0.44160 acc_top5_avg=0.73411 lr=0.00010 gn=42.29436 time=63.26it/s +epoch=106 global_step=41700 loss=5.95706 loss_avg=5.64531 acc=0.40625 acc_top1_avg=0.44365 acc_top5_avg=0.73551 lr=0.00010 gn=29.40669 time=54.74it/s +epoch=106 global_step=41750 loss=5.27744 loss_avg=5.64740 acc=0.47656 acc_top1_avg=0.44336 acc_top5_avg=0.73463 lr=0.00010 gn=32.88920 time=32.28it/s +epoch=106 global_step=41800 loss=5.61328 loss_avg=5.64469 acc=0.44531 acc_top1_avg=0.44379 acc_top5_avg=0.73506 lr=0.00010 gn=38.04100 time=54.56it/s +====================Eval==================== +epoch=106 global_step=41837 loss=5.24175 test_loss_avg=2.91156 acc=0.00000 test_acc_avg=0.32422 test_acc_top5_avg=0.77674 time=238.31it/s +epoch=106 global_step=41837 loss=5.44975 test_loss_avg=3.01954 acc=0.00000 test_acc_avg=0.34077 test_acc_top5_avg=0.74794 time=244.81it/s +epoch=106 global_step=41837 loss=6.07428 test_loss_avg=3.11894 acc=0.00000 test_acc_avg=0.32783 test_acc_top5_avg=0.75603 time=552.10it/s +curr_acc 0.3278 +BEST_ACC 0.3717 +curr_acc_top5 0.7560 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=5.68647 loss_avg=5.61074 acc=0.42188 acc_top1_avg=0.44832 acc_top5_avg=0.71514 lr=0.00010 gn=38.47760 time=60.27it/s +epoch=107 global_step=41900 loss=5.94074 loss_avg=5.72404 acc=0.42188 acc_top1_avg=0.43614 acc_top5_avg=0.71937 lr=0.00010 gn=38.89816 time=55.51it/s +epoch=107 global_step=41950 loss=5.77101 loss_avg=5.68426 acc=0.42969 acc_top1_avg=0.44020 acc_top5_avg=0.72981 lr=0.00010 gn=34.45215 time=55.81it/s +epoch=107 global_step=42000 loss=5.58286 loss_avg=5.69216 acc=0.44531 acc_top1_avg=0.43865 acc_top5_avg=0.72925 lr=0.00010 gn=37.95244 time=63.01it/s +epoch=107 global_step=42050 loss=6.17291 loss_avg=5.67781 acc=0.38281 acc_top1_avg=0.44029 acc_top5_avg=0.73008 lr=0.00010 gn=39.91817 time=57.05it/s +epoch=107 global_step=42100 loss=5.77059 loss_avg=5.65825 acc=0.42969 acc_top1_avg=0.44297 acc_top5_avg=0.73170 lr=0.00010 gn=35.34082 time=58.37it/s +epoch=107 global_step=42150 loss=5.55994 loss_avg=5.63754 acc=0.45312 acc_top1_avg=0.44539 acc_top5_avg=0.73360 lr=0.00010 gn=36.70159 time=62.91it/s +epoch=107 global_step=42200 loss=5.15563 loss_avg=5.64459 acc=0.46875 acc_top1_avg=0.44462 acc_top5_avg=0.73289 lr=0.00010 gn=37.36026 time=58.60it/s +====================Eval==================== +epoch=107 global_step=42228 loss=2.36378 test_loss_avg=3.32865 acc=0.39062 test_acc_avg=0.25349 test_acc_top5_avg=0.63414 time=234.19it/s +epoch=107 global_step=42228 loss=6.17762 test_loss_avg=3.09582 acc=0.00000 test_acc_avg=0.32882 test_acc_top5_avg=0.75554 time=816.33it/s +curr_acc 0.3288 +BEST_ACC 0.3717 +curr_acc_top5 0.7555 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=4.91863 loss_avg=5.43026 acc=0.53125 acc_top1_avg=0.47195 acc_top5_avg=0.74432 lr=0.00010 gn=35.40083 time=57.56it/s +epoch=108 global_step=42300 loss=5.67004 loss_avg=5.52607 acc=0.45312 acc_top1_avg=0.45996 acc_top5_avg=0.74392 lr=0.00010 gn=44.56844 time=59.21it/s +epoch=108 global_step=42350 loss=5.66261 loss_avg=5.56361 acc=0.45312 acc_top1_avg=0.45447 acc_top5_avg=0.73783 lr=0.00010 gn=40.63121 time=55.72it/s +epoch=108 global_step=42400 loss=5.34833 loss_avg=5.58324 acc=0.48438 acc_top1_avg=0.45190 acc_top5_avg=0.73833 lr=0.00010 gn=38.27077 time=57.14it/s +epoch=108 global_step=42450 loss=5.73644 loss_avg=5.60469 acc=0.42188 acc_top1_avg=0.44911 acc_top5_avg=0.73599 lr=0.00010 gn=35.11173 time=58.75it/s +epoch=108 global_step=42500 loss=5.44525 loss_avg=5.63161 acc=0.47656 acc_top1_avg=0.44580 acc_top5_avg=0.73475 lr=0.00010 gn=40.99338 time=63.48it/s +epoch=108 global_step=42550 loss=5.63839 loss_avg=5.64510 acc=0.42969 acc_top1_avg=0.44386 acc_top5_avg=0.73372 lr=0.00010 gn=31.02009 time=56.51it/s +epoch=108 global_step=42600 loss=5.04826 loss_avg=5.64527 acc=0.51562 acc_top1_avg=0.44342 acc_top5_avg=0.73423 lr=0.00010 gn=40.64072 time=58.62it/s +====================Eval==================== +epoch=108 global_step=42619 loss=4.66294 test_loss_avg=2.12144 acc=0.00000 test_acc_avg=0.45009 test_acc_top5_avg=0.90017 time=241.80it/s +epoch=108 global_step=42619 loss=0.17273 test_loss_avg=2.94785 acc=0.95312 test_acc_avg=0.34214 test_acc_top5_avg=0.71714 time=231.23it/s +epoch=108 global_step=42619 loss=6.43298 test_loss_avg=3.15480 acc=0.00000 test_acc_avg=0.32170 test_acc_top5_avg=0.75148 time=665.34it/s +curr_acc 0.3217 +BEST_ACC 0.3717 +curr_acc_top5 0.7515 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=5.04864 loss_avg=5.45454 acc=0.50781 acc_top1_avg=0.46749 acc_top5_avg=0.74395 lr=0.00010 gn=29.33536 time=54.48it/s +epoch=109 global_step=42700 loss=5.35773 loss_avg=5.60177 acc=0.46875 acc_top1_avg=0.44994 acc_top5_avg=0.73206 lr=0.00010 gn=32.85094 time=55.30it/s +epoch=109 global_step=42750 loss=5.16358 loss_avg=5.62263 acc=0.50000 acc_top1_avg=0.44698 acc_top5_avg=0.73127 lr=0.00010 gn=35.35948 time=51.79it/s +epoch=109 global_step=42800 loss=5.45280 loss_avg=5.60771 acc=0.46094 acc_top1_avg=0.44846 acc_top5_avg=0.73097 lr=0.00010 gn=40.64882 time=54.61it/s +epoch=109 global_step=42850 loss=5.30656 loss_avg=5.63051 acc=0.48438 acc_top1_avg=0.44694 acc_top5_avg=0.73140 lr=0.00010 gn=39.21050 time=60.63it/s +epoch=109 global_step=42900 loss=5.56907 loss_avg=5.62167 acc=0.45312 acc_top1_avg=0.44762 acc_top5_avg=0.73260 lr=0.00010 gn=38.41092 time=55.14it/s +epoch=109 global_step=42950 loss=5.57945 loss_avg=5.63239 acc=0.43750 acc_top1_avg=0.44619 acc_top5_avg=0.73305 lr=0.00010 gn=29.65299 time=62.56it/s +epoch=109 global_step=43000 loss=5.85424 loss_avg=5.63016 acc=0.42188 acc_top1_avg=0.44658 acc_top5_avg=0.73355 lr=0.00010 gn=38.58111 time=56.06it/s +====================Eval==================== +epoch=109 global_step=43010 loss=4.98834 test_loss_avg=3.55683 acc=0.00000 test_acc_avg=0.22276 test_acc_top5_avg=0.58794 time=232.65it/s +epoch=109 global_step=43010 loss=6.11087 test_loss_avg=3.09531 acc=0.00000 test_acc_avg=0.32902 test_acc_top5_avg=0.75752 time=841.05it/s +curr_acc 0.3290 +BEST_ACC 0.3717 +curr_acc_top5 0.7575 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=5.04735 loss_avg=5.63477 acc=0.50781 acc_top1_avg=0.44590 acc_top5_avg=0.73340 lr=0.00010 gn=40.44977 time=47.75it/s +epoch=110 global_step=43100 loss=6.06219 loss_avg=5.57443 acc=0.39844 acc_top1_avg=0.45304 acc_top5_avg=0.73854 lr=0.00010 gn=41.54393 time=62.97it/s +epoch=110 global_step=43150 loss=5.44658 loss_avg=5.61452 acc=0.47656 acc_top1_avg=0.44844 acc_top5_avg=0.73281 lr=0.00010 gn=42.46488 time=54.94it/s +epoch=110 global_step=43200 loss=5.24163 loss_avg=5.63096 acc=0.50000 acc_top1_avg=0.44597 acc_top5_avg=0.73154 lr=0.00010 gn=47.81819 time=55.00it/s +epoch=110 global_step=43250 loss=5.50875 loss_avg=5.61803 acc=0.46875 acc_top1_avg=0.44736 acc_top5_avg=0.73122 lr=0.00010 gn=43.37657 time=58.52it/s +epoch=110 global_step=43300 loss=5.60344 loss_avg=5.61703 acc=0.42969 acc_top1_avg=0.44782 acc_top5_avg=0.73227 lr=0.00010 gn=36.17195 time=57.39it/s +epoch=110 global_step=43350 loss=5.27635 loss_avg=5.63681 acc=0.49219 acc_top1_avg=0.44554 acc_top5_avg=0.73088 lr=0.00010 gn=37.43599 time=59.00it/s +epoch=110 global_step=43400 loss=5.06156 loss_avg=5.63102 acc=0.52344 acc_top1_avg=0.44617 acc_top5_avg=0.73223 lr=0.00010 gn=45.05579 time=61.35it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.79373 test_loss_avg=1.89817 acc=0.78906 test_acc_avg=0.46250 test_acc_top5_avg=0.97109 time=229.76it/s +epoch=110 global_step=43401 loss=0.35450 test_loss_avg=3.22227 acc=0.86719 test_acc_avg=0.27552 test_acc_top5_avg=0.67969 time=239.62it/s +epoch=110 global_step=43401 loss=6.33835 test_loss_avg=3.08767 acc=0.00000 test_acc_avg=0.32951 test_acc_top5_avg=0.75010 time=506.50it/s +curr_acc 0.3295 +BEST_ACC 0.3717 +curr_acc_top5 0.7501 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=5.43407 loss_avg=5.59942 acc=0.46875 acc_top1_avg=0.45057 acc_top5_avg=0.74426 lr=0.00010 gn=39.32430 time=49.17it/s +epoch=111 global_step=43500 loss=5.64768 loss_avg=5.59147 acc=0.45312 acc_top1_avg=0.45107 acc_top5_avg=0.74140 lr=0.00010 gn=35.51467 time=60.82it/s +epoch=111 global_step=43550 loss=6.06976 loss_avg=5.61707 acc=0.39844 acc_top1_avg=0.44746 acc_top5_avg=0.73784 lr=0.00010 gn=33.01467 time=59.87it/s +epoch=111 global_step=43600 loss=6.07371 loss_avg=5.62919 acc=0.37500 acc_top1_avg=0.44633 acc_top5_avg=0.73653 lr=0.00010 gn=27.27769 time=56.06it/s +epoch=111 global_step=43650 loss=5.86285 loss_avg=5.62803 acc=0.42969 acc_top1_avg=0.44688 acc_top5_avg=0.73538 lr=0.00010 gn=38.26795 time=40.75it/s +epoch=111 global_step=43700 loss=6.48084 loss_avg=5.62559 acc=0.35938 acc_top1_avg=0.44685 acc_top5_avg=0.73461 lr=0.00010 gn=44.49354 time=54.88it/s +epoch=111 global_step=43750 loss=5.76868 loss_avg=5.62344 acc=0.40625 acc_top1_avg=0.44704 acc_top5_avg=0.73317 lr=0.00010 gn=36.56029 time=41.17it/s +====================Eval==================== +epoch=111 global_step=43792 loss=4.94964 test_loss_avg=3.20973 acc=0.00000 test_acc_avg=0.27495 test_acc_top5_avg=0.72933 time=248.30it/s +epoch=111 global_step=43792 loss=6.27130 test_loss_avg=3.08696 acc=0.00000 test_acc_avg=0.32793 test_acc_top5_avg=0.75316 time=502.07it/s +curr_acc 0.3279 +BEST_ACC 0.3717 +curr_acc_top5 0.7532 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=5.31791 loss_avg=5.38099 acc=0.48438 acc_top1_avg=0.48047 acc_top5_avg=0.73926 lr=0.00010 gn=42.81516 time=54.12it/s +epoch=112 global_step=43850 loss=5.91390 loss_avg=5.60358 acc=0.39844 acc_top1_avg=0.44747 acc_top5_avg=0.73613 lr=0.00010 gn=34.36306 time=54.44it/s +epoch=112 global_step=43900 loss=5.76238 loss_avg=5.60564 acc=0.44531 acc_top1_avg=0.44850 acc_top5_avg=0.73524 lr=0.00010 gn=36.85936 time=56.09it/s +epoch=112 global_step=43950 loss=6.16632 loss_avg=5.60607 acc=0.39062 acc_top1_avg=0.44877 acc_top5_avg=0.73512 lr=0.00010 gn=40.92236 time=56.65it/s +epoch=112 global_step=44000 loss=6.39489 loss_avg=5.59627 acc=0.35938 acc_top1_avg=0.45012 acc_top5_avg=0.73595 lr=0.00010 gn=45.18442 time=59.80it/s +epoch=112 global_step=44050 loss=5.54582 loss_avg=5.61440 acc=0.45312 acc_top1_avg=0.44783 acc_top5_avg=0.73265 lr=0.00010 gn=41.02176 time=51.44it/s +epoch=112 global_step=44100 loss=5.04575 loss_avg=5.60875 acc=0.50000 acc_top1_avg=0.44889 acc_top5_avg=0.73349 lr=0.00010 gn=39.50810 time=50.82it/s +epoch=112 global_step=44150 loss=5.63144 loss_avg=5.61327 acc=0.45312 acc_top1_avg=0.44813 acc_top5_avg=0.73413 lr=0.00010 gn=44.99767 time=53.46it/s +====================Eval==================== +epoch=112 global_step=44183 loss=2.26863 test_loss_avg=2.34214 acc=0.39062 test_acc_avg=0.36328 test_acc_top5_avg=0.96484 time=244.91it/s +epoch=112 global_step=44183 loss=4.60444 test_loss_avg=3.46535 acc=0.00000 test_acc_avg=0.22521 test_acc_top5_avg=0.64799 time=218.92it/s +epoch=112 global_step=44183 loss=6.43398 test_loss_avg=3.10299 acc=0.00000 test_acc_avg=0.32842 test_acc_top5_avg=0.75188 time=836.02it/s +curr_acc 0.3284 +BEST_ACC 0.3717 +curr_acc_top5 0.7519 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=5.85658 loss_avg=5.60759 acc=0.43750 acc_top1_avg=0.44991 acc_top5_avg=0.72564 lr=0.00010 gn=41.72761 time=60.96it/s +epoch=113 global_step=44250 loss=5.44988 loss_avg=5.55681 acc=0.46875 acc_top1_avg=0.45499 acc_top5_avg=0.73659 lr=0.00010 gn=44.97966 time=56.11it/s +epoch=113 global_step=44300 loss=6.29187 loss_avg=5.61659 acc=0.38281 acc_top1_avg=0.44738 acc_top5_avg=0.73491 lr=0.00010 gn=38.35697 time=56.50it/s +epoch=113 global_step=44350 loss=5.82240 loss_avg=5.61799 acc=0.42188 acc_top1_avg=0.44672 acc_top5_avg=0.73372 lr=0.00010 gn=32.12649 time=56.88it/s +epoch=113 global_step=44400 loss=5.89425 loss_avg=5.61597 acc=0.41406 acc_top1_avg=0.44679 acc_top5_avg=0.73355 lr=0.00010 gn=39.22214 time=56.70it/s +epoch=113 global_step=44450 loss=5.04883 loss_avg=5.61109 acc=0.52344 acc_top1_avg=0.44713 acc_top5_avg=0.73461 lr=0.00010 gn=46.15007 time=53.94it/s +epoch=113 global_step=44500 loss=5.71259 loss_avg=5.60714 acc=0.41406 acc_top1_avg=0.44768 acc_top5_avg=0.73425 lr=0.00010 gn=38.37631 time=56.42it/s +epoch=113 global_step=44550 loss=5.04252 loss_avg=5.60512 acc=0.50781 acc_top1_avg=0.44797 acc_top5_avg=0.73425 lr=0.00010 gn=29.98555 time=51.99it/s +====================Eval==================== +epoch=113 global_step=44574 loss=4.62832 test_loss_avg=2.61232 acc=0.00000 test_acc_avg=0.36413 test_acc_top5_avg=0.81929 time=244.51it/s +epoch=113 global_step=44574 loss=5.18050 test_loss_avg=2.91001 acc=0.00000 test_acc_avg=0.35146 test_acc_top5_avg=0.73555 time=241.79it/s +epoch=113 global_step=44574 loss=6.33913 test_loss_avg=3.11385 acc=0.00000 test_acc_avg=0.32476 test_acc_top5_avg=0.75267 time=499.44it/s +curr_acc 0.3248 +BEST_ACC 0.3717 +curr_acc_top5 0.7527 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=5.59290 loss_avg=5.59595 acc=0.46094 acc_top1_avg=0.44952 acc_top5_avg=0.72656 lr=0.00010 gn=38.85609 time=60.34it/s +epoch=114 global_step=44650 loss=4.69710 loss_avg=5.58881 acc=0.55469 acc_top1_avg=0.45107 acc_top5_avg=0.73057 lr=0.00010 gn=40.19402 time=54.68it/s +epoch=114 global_step=44700 loss=5.77230 loss_avg=5.59381 acc=0.42188 acc_top1_avg=0.44990 acc_top5_avg=0.73239 lr=0.00010 gn=30.12936 time=55.62it/s +epoch=114 global_step=44750 loss=5.87632 loss_avg=5.59586 acc=0.43750 acc_top1_avg=0.44997 acc_top5_avg=0.73256 lr=0.00010 gn=37.89823 time=55.73it/s +epoch=114 global_step=44800 loss=5.52204 loss_avg=5.61120 acc=0.46875 acc_top1_avg=0.44835 acc_top5_avg=0.73272 lr=0.00010 gn=40.19768 time=56.05it/s +epoch=114 global_step=44850 loss=5.47590 loss_avg=5.60847 acc=0.45312 acc_top1_avg=0.44809 acc_top5_avg=0.73268 lr=0.00010 gn=37.26776 time=55.26it/s +epoch=114 global_step=44900 loss=5.45720 loss_avg=5.61716 acc=0.47656 acc_top1_avg=0.44737 acc_top5_avg=0.73267 lr=0.00010 gn=43.44226 time=60.13it/s +epoch=114 global_step=44950 loss=5.06696 loss_avg=5.60376 acc=0.51562 acc_top1_avg=0.44934 acc_top5_avg=0.73325 lr=0.00010 gn=39.47570 time=45.64it/s +====================Eval==================== +epoch=114 global_step=44965 loss=2.00115 test_loss_avg=3.39300 acc=0.42969 test_acc_avg=0.23828 test_acc_top5_avg=0.60831 time=240.80it/s +epoch=114 global_step=44965 loss=5.98040 test_loss_avg=3.05942 acc=0.00000 test_acc_avg=0.32723 test_acc_top5_avg=0.74604 time=788.85it/s +curr_acc 0.3272 +BEST_ACC 0.3717 +curr_acc_top5 0.7460 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=5.45233 loss_avg=5.51790 acc=0.46875 acc_top1_avg=0.45982 acc_top5_avg=0.73951 lr=0.00010 gn=37.82234 time=57.41it/s +epoch=115 global_step=45050 loss=5.14233 loss_avg=5.58510 acc=0.50000 acc_top1_avg=0.45138 acc_top5_avg=0.72858 lr=0.00010 gn=50.80069 time=51.54it/s +epoch=115 global_step=45100 loss=5.18475 loss_avg=5.59004 acc=0.48438 acc_top1_avg=0.45174 acc_top5_avg=0.72946 lr=0.00010 gn=44.22164 time=55.07it/s +epoch=115 global_step=45150 loss=5.18856 loss_avg=5.59596 acc=0.50000 acc_top1_avg=0.45122 acc_top5_avg=0.73155 lr=0.00010 gn=42.74853 time=57.10it/s +epoch=115 global_step=45200 loss=6.00417 loss_avg=5.58118 acc=0.41406 acc_top1_avg=0.45289 acc_top5_avg=0.73368 lr=0.00010 gn=39.64108 time=55.22it/s +epoch=115 global_step=45250 loss=5.82890 loss_avg=5.57586 acc=0.43750 acc_top1_avg=0.45329 acc_top5_avg=0.73342 lr=0.00010 gn=37.62131 time=55.43it/s +epoch=115 global_step=45300 loss=5.06251 loss_avg=5.58683 acc=0.50781 acc_top1_avg=0.45236 acc_top5_avg=0.73186 lr=0.00010 gn=45.84326 time=60.97it/s +epoch=115 global_step=45350 loss=5.35522 loss_avg=5.59596 acc=0.50781 acc_top1_avg=0.45110 acc_top5_avg=0.73127 lr=0.00010 gn=40.55156 time=54.81it/s +====================Eval==================== +epoch=115 global_step=45356 loss=1.21917 test_loss_avg=1.51653 acc=0.67188 test_acc_avg=0.55885 test_acc_top5_avg=0.97760 time=222.34it/s +epoch=115 global_step=45356 loss=0.15051 test_loss_avg=2.96347 acc=0.93750 test_acc_avg=0.32452 test_acc_top5_avg=0.70421 time=229.60it/s +epoch=115 global_step=45356 loss=6.10086 test_loss_avg=3.04510 acc=0.00000 test_acc_avg=0.32961 test_acc_top5_avg=0.75049 time=834.02it/s +curr_acc 0.3296 +BEST_ACC 0.3717 +curr_acc_top5 0.7505 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=5.26701 loss_avg=5.58580 acc=0.50000 acc_top1_avg=0.45312 acc_top5_avg=0.73331 lr=0.00010 gn=48.16245 time=58.43it/s +epoch=116 global_step=45450 loss=5.60241 loss_avg=5.56973 acc=0.44531 acc_top1_avg=0.45445 acc_top5_avg=0.73230 lr=0.00010 gn=39.08634 time=62.40it/s +epoch=116 global_step=45500 loss=5.08225 loss_avg=5.55778 acc=0.51562 acc_top1_avg=0.45475 acc_top5_avg=0.73345 lr=0.00010 gn=42.83164 time=55.42it/s +epoch=116 global_step=45550 loss=6.48583 loss_avg=5.56519 acc=0.34375 acc_top1_avg=0.45377 acc_top5_avg=0.73550 lr=0.00010 gn=36.00345 time=62.92it/s +epoch=116 global_step=45600 loss=5.63648 loss_avg=5.57692 acc=0.45312 acc_top1_avg=0.45194 acc_top5_avg=0.73415 lr=0.00010 gn=46.21197 time=58.29it/s +epoch=116 global_step=45650 loss=5.88173 loss_avg=5.59389 acc=0.42969 acc_top1_avg=0.44994 acc_top5_avg=0.73294 lr=0.00010 gn=41.05370 time=56.30it/s +epoch=116 global_step=45700 loss=5.77688 loss_avg=5.58558 acc=0.41406 acc_top1_avg=0.45076 acc_top5_avg=0.73285 lr=0.00010 gn=42.86349 time=51.27it/s +====================Eval==================== +epoch=116 global_step=45747 loss=4.84608 test_loss_avg=3.41443 acc=0.00000 test_acc_avg=0.23850 test_acc_top5_avg=0.61914 time=240.18it/s +epoch=116 global_step=45747 loss=6.09149 test_loss_avg=3.09028 acc=0.00000 test_acc_avg=0.32367 test_acc_top5_avg=0.75059 time=494.20it/s +curr_acc 0.3237 +BEST_ACC 0.3717 +curr_acc_top5 0.7506 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=4.58490 loss_avg=5.06880 acc=0.55469 acc_top1_avg=0.50521 acc_top5_avg=0.75521 lr=0.00010 gn=39.68017 time=41.66it/s +epoch=117 global_step=45800 loss=6.14781 loss_avg=5.53741 acc=0.42188 acc_top1_avg=0.45799 acc_top5_avg=0.74278 lr=0.00010 gn=50.19765 time=54.23it/s +epoch=117 global_step=45850 loss=5.71778 loss_avg=5.52222 acc=0.42969 acc_top1_avg=0.45919 acc_top5_avg=0.73923 lr=0.00010 gn=48.03089 time=53.39it/s +epoch=117 global_step=45900 loss=5.60929 loss_avg=5.54892 acc=0.44531 acc_top1_avg=0.45655 acc_top5_avg=0.73570 lr=0.00010 gn=45.91842 time=55.59it/s +epoch=117 global_step=45950 loss=5.22578 loss_avg=5.54474 acc=0.46875 acc_top1_avg=0.45667 acc_top5_avg=0.73553 lr=0.00010 gn=37.83241 time=62.98it/s +epoch=117 global_step=46000 loss=5.53449 loss_avg=5.55937 acc=0.46094 acc_top1_avg=0.45458 acc_top5_avg=0.73570 lr=0.00010 gn=43.81716 time=57.13it/s +epoch=117 global_step=46050 loss=5.89273 loss_avg=5.56618 acc=0.39844 acc_top1_avg=0.45351 acc_top5_avg=0.73520 lr=0.00010 gn=40.41422 time=62.21it/s +epoch=117 global_step=46100 loss=5.91997 loss_avg=5.57466 acc=0.40625 acc_top1_avg=0.45231 acc_top5_avg=0.73415 lr=0.00010 gn=43.18190 time=60.56it/s +====================Eval==================== +epoch=117 global_step=46138 loss=2.33498 test_loss_avg=2.27055 acc=0.35938 test_acc_avg=0.35491 test_acc_top5_avg=0.96205 time=228.86it/s +epoch=117 global_step=46138 loss=0.43814 test_loss_avg=3.36519 acc=0.88281 test_acc_avg=0.24452 test_acc_top5_avg=0.66379 time=240.40it/s +epoch=117 global_step=46138 loss=6.06810 test_loss_avg=3.06487 acc=0.00000 test_acc_avg=0.33119 test_acc_top5_avg=0.75000 time=837.02it/s +curr_acc 0.3312 +BEST_ACC 0.3717 +curr_acc_top5 0.7500 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=5.54206 loss_avg=5.48253 acc=0.46094 acc_top1_avg=0.46419 acc_top5_avg=0.73177 lr=0.00010 gn=35.00223 time=51.06it/s +epoch=118 global_step=46200 loss=4.97219 loss_avg=5.53842 acc=0.50781 acc_top1_avg=0.45514 acc_top5_avg=0.72959 lr=0.00010 gn=43.38390 time=57.34it/s +epoch=118 global_step=46250 loss=4.98685 loss_avg=5.54681 acc=0.50781 acc_top1_avg=0.45431 acc_top5_avg=0.73061 lr=0.00010 gn=47.14128 time=61.72it/s +epoch=118 global_step=46300 loss=5.90813 loss_avg=5.56293 acc=0.40625 acc_top1_avg=0.45259 acc_top5_avg=0.73143 lr=0.00010 gn=33.65077 time=59.86it/s +epoch=118 global_step=46350 loss=4.94997 loss_avg=5.57817 acc=0.52344 acc_top1_avg=0.45125 acc_top5_avg=0.73176 lr=0.00010 gn=35.73488 time=60.50it/s +epoch=118 global_step=46400 loss=5.53327 loss_avg=5.59155 acc=0.46094 acc_top1_avg=0.44946 acc_top5_avg=0.73193 lr=0.00010 gn=33.40589 time=54.65it/s +epoch=118 global_step=46450 loss=6.35945 loss_avg=5.57816 acc=0.35156 acc_top1_avg=0.45135 acc_top5_avg=0.73317 lr=0.00010 gn=47.07315 time=55.76it/s +epoch=118 global_step=46500 loss=5.72281 loss_avg=5.59066 acc=0.45312 acc_top1_avg=0.45010 acc_top5_avg=0.73334 lr=0.00010 gn=40.95353 time=61.99it/s +====================Eval==================== +epoch=118 global_step=46529 loss=4.55292 test_loss_avg=3.01750 acc=0.00000 test_acc_avg=0.29771 test_acc_top5_avg=0.75140 time=149.54it/s +epoch=118 global_step=46529 loss=5.36590 test_loss_avg=3.02540 acc=0.00000 test_acc_avg=0.32903 test_acc_top5_avg=0.74960 time=243.36it/s +epoch=118 global_step=46529 loss=5.98132 test_loss_avg=3.06281 acc=0.00000 test_acc_avg=0.32486 test_acc_top5_avg=0.75277 time=830.39it/s +curr_acc 0.3249 +BEST_ACC 0.3717 +curr_acc_top5 0.7528 +BEST_ACC_top5 0.7988 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=5.11542 loss_avg=5.52322 acc=0.50000 acc_top1_avg=0.46094 acc_top5_avg=0.73400 lr=0.00010 gn=29.99741 time=61.10it/s +epoch=119 global_step=46600 loss=5.43083 loss_avg=5.57716 acc=0.47656 acc_top1_avg=0.45246 acc_top5_avg=0.73261 lr=0.00010 gn=41.32102 time=60.64it/s +epoch=119 global_step=46650 loss=5.43050 loss_avg=5.56817 acc=0.46875 acc_top1_avg=0.45377 acc_top5_avg=0.73199 lr=0.00010 gn=38.55539 time=56.88it/s +epoch=119 global_step=46700 loss=5.34538 loss_avg=5.58024 acc=0.46875 acc_top1_avg=0.45198 acc_top5_avg=0.73186 lr=0.00010 gn=43.80471 time=56.40it/s +epoch=119 global_step=46750 loss=5.42081 loss_avg=5.56952 acc=0.46875 acc_top1_avg=0.45344 acc_top5_avg=0.73194 lr=0.00010 gn=36.89603 time=62.38it/s +epoch=119 global_step=46800 loss=4.67582 loss_avg=5.56385 acc=0.55469 acc_top1_avg=0.45431 acc_top5_avg=0.73259 lr=0.00010 gn=45.74430 time=46.74it/s +epoch=119 global_step=46850 loss=5.35826 loss_avg=5.55947 acc=0.46875 acc_top1_avg=0.45522 acc_top5_avg=0.73321 lr=0.00010 gn=41.42847 time=60.23it/s +epoch=119 global_step=46900 loss=5.69548 loss_avg=5.56700 acc=0.44531 acc_top1_avg=0.45426 acc_top5_avg=0.73277 lr=0.00010 gn=43.70719 time=53.57it/s +====================Eval==================== +epoch=119 global_step=46920 loss=4.51506 test_loss_avg=3.35705 acc=0.00000 test_acc_avg=0.24665 test_acc_top5_avg=0.63903 time=234.12it/s +epoch=119 global_step=46920 loss=6.16793 test_loss_avg=3.08204 acc=0.00000 test_acc_avg=0.33040 test_acc_top5_avg=0.75000 time=517.50it/s +curr_acc 0.3304 +BEST_ACC 0.3717 +curr_acc_top5 0.7500 +BEST_ACC_top5 0.7988 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_2_2.log b/other_methods/sceloss/sceloss_results/out_2_2.log new file mode 100644 index 0000000..aa4102b --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_2_2.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.2__noise_amount__0.2.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.06378 loss_avg=7.35347 acc=0.26562 acc_top1_avg=0.25922 acc_top5_avg=0.72813 lr=0.01000 gn=8.05618 time=52.87it/s +epoch=0 global_step=100 loss=6.25783 loss_avg=6.93917 acc=0.36719 acc_top1_avg=0.30281 acc_top5_avg=0.76078 lr=0.01000 gn=7.56135 time=63.42it/s +epoch=0 global_step=150 loss=5.75778 loss_avg=6.68851 acc=0.42969 acc_top1_avg=0.32797 acc_top5_avg=0.77750 lr=0.01000 gn=7.78524 time=62.45it/s +epoch=0 global_step=200 loss=5.54210 loss_avg=6.52695 acc=0.46875 acc_top1_avg=0.34535 acc_top5_avg=0.78672 lr=0.01000 gn=7.47884 time=63.91it/s +epoch=0 global_step=250 loss=5.88118 loss_avg=6.41687 acc=0.43750 acc_top1_avg=0.35675 acc_top5_avg=0.79356 lr=0.01000 gn=6.97103 time=63.25it/s +epoch=0 global_step=300 loss=5.85390 loss_avg=6.30206 acc=0.42188 acc_top1_avg=0.36859 acc_top5_avg=0.80109 lr=0.01000 gn=6.29373 time=58.49it/s +epoch=0 global_step=350 loss=5.60967 loss_avg=6.20809 acc=0.45312 acc_top1_avg=0.37853 acc_top5_avg=0.80636 lr=0.01000 gn=6.33202 time=61.34it/s +====================Eval==================== +epoch=0 global_step=391 loss=1.32154 test_loss_avg=2.21370 acc=0.66406 test_acc_avg=0.48234 test_acc_top5_avg=0.93172 time=250.72it/s +epoch=0 global_step=391 loss=2.17700 test_loss_avg=2.02295 acc=0.50000 test_acc_avg=0.52116 test_acc_top5_avg=0.93147 time=30.34it/s +curr_acc 0.5212 +BEST_ACC 0.0000 +curr_acc_top5 0.9315 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=5.70168 loss_avg=5.36650 acc=0.46094 acc_top1_avg=0.46962 acc_top5_avg=0.85764 lr=0.01000 gn=5.73864 time=63.17it/s +epoch=1 global_step=450 loss=5.54572 loss_avg=5.49596 acc=0.46875 acc_top1_avg=0.45127 acc_top5_avg=0.84719 lr=0.01000 gn=6.23089 time=62.95it/s +epoch=1 global_step=500 loss=4.83469 loss_avg=5.41008 acc=0.53125 acc_top1_avg=0.46230 acc_top5_avg=0.85257 lr=0.01000 gn=5.47064 time=61.09it/s +epoch=1 global_step=550 loss=5.44777 loss_avg=5.31981 acc=0.45312 acc_top1_avg=0.47268 acc_top5_avg=0.85333 lr=0.01000 gn=6.39539 time=60.23it/s +epoch=1 global_step=600 loss=4.65942 loss_avg=5.24113 acc=0.54688 acc_top1_avg=0.48131 acc_top5_avg=0.85463 lr=0.01000 gn=7.85120 time=65.22it/s +epoch=1 global_step=650 loss=4.90281 loss_avg=5.19366 acc=0.49219 acc_top1_avg=0.48646 acc_top5_avg=0.85600 lr=0.01000 gn=6.79801 time=44.49it/s +epoch=1 global_step=700 loss=4.12864 loss_avg=5.13366 acc=0.60156 acc_top1_avg=0.49259 acc_top5_avg=0.85912 lr=0.01000 gn=6.07972 time=56.05it/s +epoch=1 global_step=750 loss=4.65507 loss_avg=5.08381 acc=0.54688 acc_top1_avg=0.49758 acc_top5_avg=0.86220 lr=0.01000 gn=6.07791 time=57.49it/s +====================Eval==================== +epoch=1 global_step=782 loss=3.18486 test_loss_avg=1.48016 acc=0.27344 test_acc_avg=0.64583 test_acc_top5_avg=0.97024 time=251.05it/s +epoch=1 global_step=782 loss=0.94599 test_loss_avg=1.56293 acc=0.70312 test_acc_avg=0.63820 test_acc_top5_avg=0.95588 time=252.29it/s +epoch=1 global_step=782 loss=1.48550 test_loss_avg=1.51759 acc=0.62500 test_acc_avg=0.64498 test_acc_top5_avg=0.95807 time=553.19it/s +curr_acc 0.6450 +BEST_ACC 0.5212 +curr_acc_top5 0.9581 +BEST_ACC_top5 0.9315 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=5.41555 loss_avg=4.69213 acc=0.44531 acc_top1_avg=0.53776 acc_top5_avg=0.88281 lr=0.01000 gn=6.15418 time=58.36it/s +epoch=2 global_step=850 loss=4.15883 loss_avg=4.61833 acc=0.58594 acc_top1_avg=0.54630 acc_top5_avg=0.87914 lr=0.01000 gn=5.41176 time=54.51it/s +epoch=2 global_step=900 loss=4.64613 loss_avg=4.56841 acc=0.55469 acc_top1_avg=0.55303 acc_top5_avg=0.88182 lr=0.01000 gn=6.80856 time=53.60it/s +epoch=2 global_step=950 loss=4.39798 loss_avg=4.56545 acc=0.57031 acc_top1_avg=0.55246 acc_top5_avg=0.88114 lr=0.01000 gn=4.85017 time=66.41it/s +epoch=2 global_step=1000 loss=4.66378 loss_avg=4.54160 acc=0.53906 acc_top1_avg=0.55530 acc_top5_avg=0.88073 lr=0.01000 gn=7.31617 time=64.31it/s +epoch=2 global_step=1050 loss=3.76913 loss_avg=4.53208 acc=0.66406 acc_top1_avg=0.55606 acc_top5_avg=0.88130 lr=0.01000 gn=6.61891 time=63.12it/s +epoch=2 global_step=1100 loss=4.01098 loss_avg=4.51957 acc=0.61719 acc_top1_avg=0.55732 acc_top5_avg=0.88298 lr=0.01000 gn=6.35131 time=55.86it/s +epoch=2 global_step=1150 loss=5.47152 loss_avg=4.51508 acc=0.46875 acc_top1_avg=0.55772 acc_top5_avg=0.88224 lr=0.01000 gn=6.57651 time=55.02it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.65228 test_loss_avg=1.48694 acc=0.59375 test_acc_avg=0.62612 test_acc_top5_avg=0.93397 time=244.11it/s +epoch=2 global_step=1173 loss=0.34708 test_loss_avg=1.20107 acc=0.87500 test_acc_avg=0.69057 test_acc_top5_avg=0.94729 time=882.27it/s +curr_acc 0.6906 +BEST_ACC 0.6450 +curr_acc_top5 0.9473 +BEST_ACC_top5 0.9581 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=4.21028 loss_avg=4.30261 acc=0.57812 acc_top1_avg=0.57986 acc_top5_avg=0.89265 lr=0.01000 gn=7.87900 time=65.30it/s +epoch=3 global_step=1250 loss=4.57012 loss_avg=4.28373 acc=0.55469 acc_top1_avg=0.58228 acc_top5_avg=0.88748 lr=0.01000 gn=6.95990 time=54.90it/s +epoch=3 global_step=1300 loss=4.18814 loss_avg=4.25530 acc=0.59375 acc_top1_avg=0.58428 acc_top5_avg=0.88736 lr=0.01000 gn=5.91401 time=56.56it/s +epoch=3 global_step=1350 loss=4.72102 loss_avg=4.25278 acc=0.57031 acc_top1_avg=0.58501 acc_top5_avg=0.88723 lr=0.01000 gn=7.49058 time=64.85it/s +epoch=3 global_step=1400 loss=4.69943 loss_avg=4.26180 acc=0.53125 acc_top1_avg=0.58401 acc_top5_avg=0.88732 lr=0.01000 gn=5.66327 time=62.45it/s +epoch=3 global_step=1450 loss=4.38912 loss_avg=4.26870 acc=0.58594 acc_top1_avg=0.58334 acc_top5_avg=0.88843 lr=0.01000 gn=6.55415 time=63.89it/s +epoch=3 global_step=1500 loss=4.29068 loss_avg=4.26025 acc=0.57812 acc_top1_avg=0.58417 acc_top5_avg=0.88812 lr=0.01000 gn=5.11696 time=64.99it/s +epoch=3 global_step=1550 loss=4.61727 loss_avg=4.24360 acc=0.53906 acc_top1_avg=0.58606 acc_top5_avg=0.88899 lr=0.01000 gn=4.11288 time=57.70it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.41557 test_loss_avg=0.89103 acc=0.89062 test_acc_avg=0.76442 test_acc_top5_avg=0.98918 time=253.16it/s +epoch=3 global_step=1564 loss=0.15887 test_loss_avg=1.35851 acc=0.95312 test_acc_avg=0.67684 test_acc_top5_avg=0.95697 time=249.13it/s +epoch=3 global_step=1564 loss=0.64016 test_loss_avg=1.19675 acc=0.75000 test_acc_avg=0.70936 test_acc_top5_avg=0.96400 time=891.08it/s +curr_acc 0.7094 +BEST_ACC 0.6906 +curr_acc_top5 0.9640 +BEST_ACC_top5 0.9581 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.07773 loss_avg=4.14255 acc=0.62500 acc_top1_avg=0.59766 acc_top5_avg=0.88433 lr=0.01000 gn=7.98149 time=63.73it/s +epoch=4 global_step=1650 loss=3.87171 loss_avg=4.09917 acc=0.64062 acc_top1_avg=0.60147 acc_top5_avg=0.88699 lr=0.01000 gn=7.43547 time=63.10it/s +epoch=4 global_step=1700 loss=4.10506 loss_avg=4.06844 acc=0.58594 acc_top1_avg=0.60357 acc_top5_avg=0.88948 lr=0.01000 gn=7.26067 time=61.57it/s +epoch=4 global_step=1750 loss=4.57684 loss_avg=4.08281 acc=0.54688 acc_top1_avg=0.60160 acc_top5_avg=0.89025 lr=0.01000 gn=7.18388 time=55.61it/s +epoch=4 global_step=1800 loss=3.91040 loss_avg=4.08960 acc=0.61719 acc_top1_avg=0.60087 acc_top5_avg=0.89016 lr=0.01000 gn=6.97205 time=64.79it/s +epoch=4 global_step=1850 loss=4.01692 loss_avg=4.07932 acc=0.61719 acc_top1_avg=0.60255 acc_top5_avg=0.89093 lr=0.01000 gn=6.47258 time=62.32it/s +epoch=4 global_step=1900 loss=3.82992 loss_avg=4.07058 acc=0.61719 acc_top1_avg=0.60328 acc_top5_avg=0.89137 lr=0.01000 gn=6.94804 time=58.54it/s +epoch=4 global_step=1950 loss=4.22553 loss_avg=4.06387 acc=0.59375 acc_top1_avg=0.60425 acc_top5_avg=0.89241 lr=0.01000 gn=6.91561 time=56.48it/s +====================Eval==================== +epoch=4 global_step=1955 loss=1.42183 test_loss_avg=1.48900 acc=0.64844 test_acc_avg=0.64913 test_acc_top5_avg=0.96875 time=245.34it/s +epoch=4 global_step=1955 loss=0.49452 test_loss_avg=1.46548 acc=0.75000 test_acc_avg=0.66584 test_acc_top5_avg=0.97132 time=778.02it/s +curr_acc 0.6658 +BEST_ACC 0.7094 +curr_acc_top5 0.9713 +BEST_ACC_top5 0.9640 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=3.81174 loss_avg=3.89649 acc=0.64844 acc_top1_avg=0.62240 acc_top5_avg=0.89844 lr=0.01000 gn=7.08804 time=62.31it/s +epoch=5 global_step=2050 loss=3.20184 loss_avg=3.95776 acc=0.69531 acc_top1_avg=0.61620 acc_top5_avg=0.89984 lr=0.01000 gn=7.21460 time=57.40it/s +epoch=5 global_step=2100 loss=4.43415 loss_avg=3.92845 acc=0.56250 acc_top1_avg=0.61880 acc_top5_avg=0.89892 lr=0.01000 gn=7.12619 time=53.36it/s +epoch=5 global_step=2150 loss=4.03386 loss_avg=3.93343 acc=0.59375 acc_top1_avg=0.61787 acc_top5_avg=0.89820 lr=0.01000 gn=6.60340 time=56.58it/s +epoch=5 global_step=2200 loss=4.41255 loss_avg=3.93053 acc=0.56250 acc_top1_avg=0.61827 acc_top5_avg=0.89723 lr=0.01000 gn=6.02545 time=55.82it/s +epoch=5 global_step=2250 loss=3.47248 loss_avg=3.92199 acc=0.67188 acc_top1_avg=0.61920 acc_top5_avg=0.89799 lr=0.01000 gn=6.35897 time=56.62it/s +epoch=5 global_step=2300 loss=4.70575 loss_avg=3.93997 acc=0.54688 acc_top1_avg=0.61719 acc_top5_avg=0.89796 lr=0.01000 gn=6.82218 time=63.42it/s +====================Eval==================== +epoch=5 global_step=2346 loss=1.05552 test_loss_avg=1.18478 acc=0.71875 test_acc_avg=0.69063 test_acc_top5_avg=0.98125 time=260.00it/s +epoch=5 global_step=2346 loss=1.44785 test_loss_avg=1.36791 acc=0.61719 test_acc_avg=0.66733 test_acc_top5_avg=0.97713 time=243.56it/s +epoch=5 global_step=2346 loss=0.28723 test_loss_avg=1.17383 acc=0.93750 test_acc_avg=0.70777 test_acc_top5_avg=0.97716 time=685.23it/s +curr_acc 0.7078 +BEST_ACC 0.7094 +curr_acc_top5 0.9772 +BEST_ACC_top5 0.9713 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=3.91863 loss_avg=3.61988 acc=0.64062 acc_top1_avg=0.66602 acc_top5_avg=0.91016 lr=0.01000 gn=7.59898 time=55.24it/s +epoch=6 global_step=2400 loss=3.64013 loss_avg=3.85033 acc=0.67188 acc_top1_avg=0.62775 acc_top5_avg=0.89714 lr=0.01000 gn=6.71516 time=53.40it/s +epoch=6 global_step=2450 loss=3.88938 loss_avg=3.79261 acc=0.61719 acc_top1_avg=0.63477 acc_top5_avg=0.90032 lr=0.01000 gn=7.04575 time=57.16it/s +epoch=6 global_step=2500 loss=3.57668 loss_avg=3.81655 acc=0.64844 acc_top1_avg=0.63185 acc_top5_avg=0.90168 lr=0.01000 gn=7.55744 time=52.61it/s +epoch=6 global_step=2550 loss=3.53229 loss_avg=3.83668 acc=0.65625 acc_top1_avg=0.62940 acc_top5_avg=0.90200 lr=0.01000 gn=7.22746 time=63.76it/s +epoch=6 global_step=2600 loss=3.69188 loss_avg=3.83425 acc=0.64844 acc_top1_avg=0.62964 acc_top5_avg=0.90173 lr=0.01000 gn=7.95920 time=55.15it/s +epoch=6 global_step=2650 loss=3.25915 loss_avg=3.84146 acc=0.70312 acc_top1_avg=0.62945 acc_top5_avg=0.90083 lr=0.01000 gn=6.48625 time=58.59it/s +epoch=6 global_step=2700 loss=3.52269 loss_avg=3.83600 acc=0.65625 acc_top1_avg=0.62966 acc_top5_avg=0.90133 lr=0.01000 gn=7.16810 time=65.08it/s +====================Eval==================== +epoch=6 global_step=2737 loss=0.69064 test_loss_avg=1.30739 acc=0.79688 test_acc_avg=0.68690 test_acc_top5_avg=0.97296 time=248.85it/s +epoch=6 global_step=2737 loss=0.45289 test_loss_avg=1.01078 acc=0.89062 test_acc_avg=0.75031 test_acc_top5_avg=0.98109 time=244.78it/s +epoch=6 global_step=2737 loss=0.24195 test_loss_avg=0.98637 acc=0.93750 test_acc_avg=0.75554 test_acc_top5_avg=0.98180 time=544.15it/s +curr_acc 0.7555 +BEST_ACC 0.7094 +curr_acc_top5 0.9818 +BEST_ACC_top5 0.9772 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=3.35701 loss_avg=3.66807 acc=0.69531 acc_top1_avg=0.65325 acc_top5_avg=0.90565 lr=0.01000 gn=6.15368 time=55.76it/s +epoch=7 global_step=2800 loss=3.34710 loss_avg=3.69129 acc=0.67969 acc_top1_avg=0.64559 acc_top5_avg=0.89869 lr=0.01000 gn=6.53770 time=63.03it/s +epoch=7 global_step=2850 loss=3.94606 loss_avg=3.72203 acc=0.60938 acc_top1_avg=0.64208 acc_top5_avg=0.89899 lr=0.01000 gn=6.16572 time=56.80it/s +epoch=7 global_step=2900 loss=3.58661 loss_avg=3.73396 acc=0.64844 acc_top1_avg=0.64034 acc_top5_avg=0.89829 lr=0.01000 gn=7.51537 time=57.62it/s +epoch=7 global_step=2950 loss=4.22200 loss_avg=3.75257 acc=0.57812 acc_top1_avg=0.63765 acc_top5_avg=0.89935 lr=0.01000 gn=7.52899 time=55.52it/s +epoch=7 global_step=3000 loss=3.84781 loss_avg=3.75090 acc=0.61719 acc_top1_avg=0.63739 acc_top5_avg=0.90197 lr=0.01000 gn=8.81054 time=55.68it/s +epoch=7 global_step=3050 loss=4.16506 loss_avg=3.75382 acc=0.57812 acc_top1_avg=0.63765 acc_top5_avg=0.90183 lr=0.01000 gn=7.74693 time=56.79it/s +epoch=7 global_step=3100 loss=3.99773 loss_avg=3.77031 acc=0.61719 acc_top1_avg=0.63544 acc_top5_avg=0.90201 lr=0.01000 gn=7.08381 time=58.21it/s +====================Eval==================== +epoch=7 global_step=3128 loss=0.54475 test_loss_avg=1.13654 acc=0.83594 test_acc_avg=0.71725 test_acc_top5_avg=0.97207 time=246.42it/s +epoch=7 global_step=3128 loss=1.24439 test_loss_avg=1.10095 acc=0.68750 test_acc_avg=0.72597 test_acc_top5_avg=0.97191 time=305.71it/s +curr_acc 0.7260 +BEST_ACC 0.7555 +curr_acc_top5 0.9719 +BEST_ACC_top5 0.9818 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=3.74416 loss_avg=3.58780 acc=0.61719 acc_top1_avg=0.65625 acc_top5_avg=0.90874 lr=0.01000 gn=9.09952 time=59.01it/s +epoch=8 global_step=3200 loss=3.93441 loss_avg=3.60168 acc=0.64062 acc_top1_avg=0.65408 acc_top5_avg=0.90441 lr=0.01000 gn=6.71249 time=54.57it/s +epoch=8 global_step=3250 loss=3.57198 loss_avg=3.64454 acc=0.67969 acc_top1_avg=0.64882 acc_top5_avg=0.90375 lr=0.01000 gn=7.63919 time=62.90it/s +epoch=8 global_step=3300 loss=3.85188 loss_avg=3.65363 acc=0.63281 acc_top1_avg=0.64839 acc_top5_avg=0.90348 lr=0.01000 gn=6.79128 time=57.10it/s +epoch=8 global_step=3350 loss=3.65495 loss_avg=3.69013 acc=0.61719 acc_top1_avg=0.64393 acc_top5_avg=0.90301 lr=0.01000 gn=8.46968 time=64.77it/s +epoch=8 global_step=3400 loss=4.14712 loss_avg=3.70011 acc=0.59375 acc_top1_avg=0.64203 acc_top5_avg=0.90326 lr=0.01000 gn=8.56457 time=57.48it/s +epoch=8 global_step=3450 loss=4.09943 loss_avg=3.70097 acc=0.60156 acc_top1_avg=0.64198 acc_top5_avg=0.90419 lr=0.01000 gn=7.95887 time=60.01it/s +epoch=8 global_step=3500 loss=4.34486 loss_avg=3.70828 acc=0.58594 acc_top1_avg=0.64113 acc_top5_avg=0.90470 lr=0.01000 gn=7.76057 time=52.77it/s +====================Eval==================== +epoch=8 global_step=3519 loss=1.79010 test_loss_avg=0.54602 acc=0.62500 test_acc_avg=0.85851 test_acc_top5_avg=0.99262 time=256.71it/s +epoch=8 global_step=3519 loss=0.41391 test_loss_avg=0.89356 acc=0.89062 test_acc_avg=0.77493 test_acc_top5_avg=0.97381 time=251.91it/s +epoch=8 global_step=3519 loss=0.81575 test_loss_avg=0.87715 acc=0.75000 test_acc_avg=0.77601 test_acc_top5_avg=0.97577 time=850.77it/s +curr_acc 0.7760 +BEST_ACC 0.7555 +curr_acc_top5 0.9758 +BEST_ACC_top5 0.9818 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=3.18976 loss_avg=3.62153 acc=0.69531 acc_top1_avg=0.64945 acc_top5_avg=0.91205 lr=0.01000 gn=8.39235 time=61.10it/s +epoch=9 global_step=3600 loss=4.53256 loss_avg=3.65494 acc=0.53125 acc_top1_avg=0.64689 acc_top5_avg=0.91107 lr=0.01000 gn=6.65519 time=57.83it/s +epoch=9 global_step=3650 loss=3.78342 loss_avg=3.69221 acc=0.64062 acc_top1_avg=0.64253 acc_top5_avg=0.90613 lr=0.01000 gn=9.59457 time=56.81it/s +epoch=9 global_step=3700 loss=4.08584 loss_avg=3.68206 acc=0.60938 acc_top1_avg=0.64378 acc_top5_avg=0.90647 lr=0.01000 gn=7.96083 time=56.69it/s +epoch=9 global_step=3750 loss=3.44553 loss_avg=3.67485 acc=0.69531 acc_top1_avg=0.64495 acc_top5_avg=0.90598 lr=0.01000 gn=8.09286 time=52.66it/s +epoch=9 global_step=3800 loss=3.55758 loss_avg=3.66701 acc=0.67969 acc_top1_avg=0.64624 acc_top5_avg=0.90575 lr=0.01000 gn=8.81750 time=59.88it/s +epoch=9 global_step=3850 loss=3.49772 loss_avg=3.66481 acc=0.65625 acc_top1_avg=0.64721 acc_top5_avg=0.90632 lr=0.01000 gn=8.29706 time=64.32it/s +epoch=9 global_step=3900 loss=3.75500 loss_avg=3.66204 acc=0.63281 acc_top1_avg=0.64792 acc_top5_avg=0.90598 lr=0.01000 gn=6.70513 time=54.82it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.81915 test_loss_avg=1.26657 acc=0.82031 test_acc_avg=0.67949 test_acc_top5_avg=0.96194 time=247.45it/s +epoch=9 global_step=3910 loss=0.13649 test_loss_avg=0.93933 acc=0.93750 test_acc_avg=0.75643 test_acc_top5_avg=0.97280 time=882.45it/s +curr_acc 0.7564 +BEST_ACC 0.7760 +curr_acc_top5 0.9728 +BEST_ACC_top5 0.9818 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=3.40777 loss_avg=3.55981 acc=0.67188 acc_top1_avg=0.65645 acc_top5_avg=0.90957 lr=0.01000 gn=8.60057 time=51.51it/s +epoch=10 global_step=4000 loss=3.68063 loss_avg=3.60752 acc=0.64844 acc_top1_avg=0.65182 acc_top5_avg=0.90347 lr=0.01000 gn=9.89774 time=56.64it/s +epoch=10 global_step=4050 loss=3.81881 loss_avg=3.60912 acc=0.63281 acc_top1_avg=0.65212 acc_top5_avg=0.90402 lr=0.01000 gn=8.60161 time=59.13it/s +epoch=10 global_step=4100 loss=4.01691 loss_avg=3.62435 acc=0.63281 acc_top1_avg=0.65107 acc_top5_avg=0.90407 lr=0.01000 gn=9.93404 time=63.23it/s +epoch=10 global_step=4150 loss=2.97470 loss_avg=3.63313 acc=0.71875 acc_top1_avg=0.65107 acc_top5_avg=0.90384 lr=0.01000 gn=7.39408 time=54.00it/s +epoch=10 global_step=4200 loss=4.06998 loss_avg=3.64234 acc=0.60938 acc_top1_avg=0.65011 acc_top5_avg=0.90536 lr=0.01000 gn=8.39850 time=58.41it/s +epoch=10 global_step=4250 loss=4.06267 loss_avg=3.63200 acc=0.60156 acc_top1_avg=0.65124 acc_top5_avg=0.90604 lr=0.01000 gn=8.35772 time=58.35it/s +epoch=10 global_step=4300 loss=3.62160 loss_avg=3.63450 acc=0.64844 acc_top1_avg=0.65072 acc_top5_avg=0.90679 lr=0.01000 gn=7.29153 time=60.71it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.12250 test_loss_avg=0.87705 acc=0.96875 test_acc_avg=0.77266 test_acc_top5_avg=0.98281 time=260.90it/s +epoch=10 global_step=4301 loss=0.73597 test_loss_avg=1.06974 acc=0.78125 test_acc_avg=0.74180 test_acc_top5_avg=0.98529 time=245.40it/s +epoch=10 global_step=4301 loss=0.82405 test_loss_avg=0.97114 acc=0.81250 test_acc_avg=0.76127 test_acc_top5_avg=0.98685 time=869.29it/s +curr_acc 0.7613 +BEST_ACC 0.7760 +curr_acc_top5 0.9868 +BEST_ACC_top5 0.9818 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=3.25433 loss_avg=3.70830 acc=0.68750 acc_top1_avg=0.64302 acc_top5_avg=0.90657 lr=0.01000 gn=10.69743 time=58.44it/s +epoch=11 global_step=4400 loss=3.42765 loss_avg=3.64859 acc=0.67188 acc_top1_avg=0.64883 acc_top5_avg=0.90672 lr=0.01000 gn=9.92027 time=56.04it/s +epoch=11 global_step=4450 loss=3.41935 loss_avg=3.65408 acc=0.67969 acc_top1_avg=0.64917 acc_top5_avg=0.90693 lr=0.01000 gn=7.70851 time=61.06it/s +epoch=11 global_step=4500 loss=3.58260 loss_avg=3.61454 acc=0.65625 acc_top1_avg=0.65342 acc_top5_avg=0.90900 lr=0.01000 gn=7.97397 time=57.32it/s +epoch=11 global_step=4550 loss=3.65537 loss_avg=3.61091 acc=0.64062 acc_top1_avg=0.65374 acc_top5_avg=0.90898 lr=0.01000 gn=7.76278 time=56.21it/s +epoch=11 global_step=4600 loss=3.50867 loss_avg=3.61269 acc=0.65625 acc_top1_avg=0.65348 acc_top5_avg=0.90850 lr=0.01000 gn=8.68009 time=62.28it/s +epoch=11 global_step=4650 loss=3.32300 loss_avg=3.61615 acc=0.68750 acc_top1_avg=0.65298 acc_top5_avg=0.90831 lr=0.01000 gn=7.78252 time=52.54it/s +====================Eval==================== +epoch=11 global_step=4692 loss=0.89156 test_loss_avg=1.06112 acc=0.77344 test_acc_avg=0.73992 test_acc_top5_avg=0.97732 time=242.91it/s +epoch=11 global_step=4692 loss=0.28122 test_loss_avg=0.84531 acc=0.93750 test_acc_avg=0.78738 test_acc_top5_avg=0.98141 time=880.05it/s +curr_acc 0.7874 +BEST_ACC 0.7760 +curr_acc_top5 0.9814 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=2.64580 loss_avg=3.60998 acc=0.75781 acc_top1_avg=0.64941 acc_top5_avg=0.89844 lr=0.01000 gn=9.02178 time=55.46it/s +epoch=12 global_step=4750 loss=3.69651 loss_avg=3.60592 acc=0.63281 acc_top1_avg=0.65396 acc_top5_avg=0.90638 lr=0.01000 gn=10.32621 time=54.59it/s +epoch=12 global_step=4800 loss=2.98587 loss_avg=3.55711 acc=0.71094 acc_top1_avg=0.65979 acc_top5_avg=0.90828 lr=0.01000 gn=8.23924 time=64.19it/s +epoch=12 global_step=4850 loss=2.66434 loss_avg=3.53683 acc=0.75781 acc_top1_avg=0.66055 acc_top5_avg=0.90857 lr=0.01000 gn=8.38508 time=58.72it/s +epoch=12 global_step=4900 loss=3.67647 loss_avg=3.54515 acc=0.64062 acc_top1_avg=0.65963 acc_top5_avg=0.90843 lr=0.01000 gn=10.53896 time=61.09it/s +epoch=12 global_step=4950 loss=3.34574 loss_avg=3.55943 acc=0.67969 acc_top1_avg=0.65864 acc_top5_avg=0.90758 lr=0.01000 gn=10.88434 time=59.67it/s +epoch=12 global_step=5000 loss=3.36145 loss_avg=3.55961 acc=0.68750 acc_top1_avg=0.65846 acc_top5_avg=0.90676 lr=0.01000 gn=10.13815 time=55.66it/s +epoch=12 global_step=5050 loss=4.28776 loss_avg=3.56078 acc=0.57031 acc_top1_avg=0.65850 acc_top5_avg=0.90725 lr=0.01000 gn=8.81441 time=57.54it/s +====================Eval==================== +epoch=12 global_step=5083 loss=1.76474 test_loss_avg=1.62704 acc=0.58594 test_acc_avg=0.61719 test_acc_top5_avg=0.96875 time=217.18it/s +epoch=12 global_step=5083 loss=0.89027 test_loss_avg=1.16979 acc=0.76562 test_acc_avg=0.71199 test_acc_top5_avg=0.97175 time=243.70it/s +epoch=12 global_step=5083 loss=1.88470 test_loss_avg=1.06572 acc=0.43750 test_acc_avg=0.73161 test_acc_top5_avg=0.97528 time=887.31it/s +curr_acc 0.7316 +BEST_ACC 0.7874 +curr_acc_top5 0.9753 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=3.43023 loss_avg=3.52308 acc=0.67188 acc_top1_avg=0.66774 acc_top5_avg=0.90993 lr=0.01000 gn=8.83256 time=58.92it/s +epoch=13 global_step=5150 loss=3.76855 loss_avg=3.57579 acc=0.66406 acc_top1_avg=0.66021 acc_top5_avg=0.90963 lr=0.01000 gn=10.43239 time=52.98it/s +epoch=13 global_step=5200 loss=4.38321 loss_avg=3.58837 acc=0.57031 acc_top1_avg=0.65779 acc_top5_avg=0.90972 lr=0.01000 gn=9.72249 time=62.93it/s +epoch=13 global_step=5250 loss=2.90364 loss_avg=3.58550 acc=0.73438 acc_top1_avg=0.65747 acc_top5_avg=0.90896 lr=0.01000 gn=7.67091 time=52.39it/s +epoch=13 global_step=5300 loss=2.67939 loss_avg=3.56957 acc=0.76562 acc_top1_avg=0.65899 acc_top5_avg=0.90949 lr=0.01000 gn=8.93273 time=55.39it/s +epoch=13 global_step=5350 loss=3.75960 loss_avg=3.56631 acc=0.63281 acc_top1_avg=0.65915 acc_top5_avg=0.90815 lr=0.01000 gn=9.73518 time=57.24it/s +epoch=13 global_step=5400 loss=2.99522 loss_avg=3.57232 acc=0.71875 acc_top1_avg=0.65820 acc_top5_avg=0.90773 lr=0.01000 gn=7.50405 time=62.69it/s +epoch=13 global_step=5450 loss=3.40747 loss_avg=3.57326 acc=0.66406 acc_top1_avg=0.65789 acc_top5_avg=0.90840 lr=0.01000 gn=8.35296 time=64.53it/s +====================Eval==================== +epoch=13 global_step=5474 loss=0.73408 test_loss_avg=0.98776 acc=0.81250 test_acc_avg=0.75917 test_acc_top5_avg=0.96807 time=251.40it/s +epoch=13 global_step=5474 loss=1.51184 test_loss_avg=1.11043 acc=0.61719 test_acc_avg=0.73170 test_acc_top5_avg=0.97485 time=252.50it/s +epoch=13 global_step=5474 loss=2.01901 test_loss_avg=1.13073 acc=0.56250 test_acc_avg=0.72725 test_acc_top5_avg=0.97597 time=728.94it/s +curr_acc 0.7273 +BEST_ACC 0.7874 +curr_acc_top5 0.9760 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=3.73441 loss_avg=3.64358 acc=0.64062 acc_top1_avg=0.65024 acc_top5_avg=0.90234 lr=0.01000 gn=8.49742 time=64.04it/s +epoch=14 global_step=5550 loss=4.27999 loss_avg=3.52859 acc=0.60156 acc_top1_avg=0.66139 acc_top5_avg=0.91067 lr=0.01000 gn=8.97989 time=54.99it/s +epoch=14 global_step=5600 loss=4.70674 loss_avg=3.55619 acc=0.52344 acc_top1_avg=0.65805 acc_top5_avg=0.90737 lr=0.01000 gn=7.18356 time=62.06it/s +epoch=14 global_step=5650 loss=3.99019 loss_avg=3.57573 acc=0.60938 acc_top1_avg=0.65532 acc_top5_avg=0.90523 lr=0.01000 gn=7.50503 time=58.36it/s +epoch=14 global_step=5700 loss=3.57969 loss_avg=3.58002 acc=0.65625 acc_top1_avg=0.65535 acc_top5_avg=0.90653 lr=0.01000 gn=9.11153 time=59.87it/s +epoch=14 global_step=5750 loss=3.79244 loss_avg=3.57411 acc=0.60938 acc_top1_avg=0.65670 acc_top5_avg=0.90631 lr=0.01000 gn=8.16655 time=55.78it/s +epoch=14 global_step=5800 loss=3.59983 loss_avg=3.56472 acc=0.67188 acc_top1_avg=0.65790 acc_top5_avg=0.90838 lr=0.01000 gn=7.38649 time=57.08it/s +epoch=14 global_step=5850 loss=3.83492 loss_avg=3.56577 acc=0.64844 acc_top1_avg=0.65800 acc_top5_avg=0.90804 lr=0.01000 gn=10.72888 time=55.48it/s +====================Eval==================== +epoch=14 global_step=5865 loss=0.92077 test_loss_avg=1.07950 acc=0.76562 test_acc_avg=0.72816 test_acc_top5_avg=0.97461 time=240.60it/s +epoch=14 global_step=5865 loss=0.73118 test_loss_avg=0.82378 acc=0.81250 test_acc_avg=0.78807 test_acc_top5_avg=0.98141 time=474.63it/s +curr_acc 0.7881 +BEST_ACC 0.7874 +curr_acc_top5 0.9814 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=3.25245 loss_avg=3.52563 acc=0.67969 acc_top1_avg=0.66183 acc_top5_avg=0.90960 lr=0.01000 gn=6.97746 time=60.52it/s +epoch=15 global_step=5950 loss=3.33190 loss_avg=3.51447 acc=0.67969 acc_top1_avg=0.66213 acc_top5_avg=0.91085 lr=0.01000 gn=7.74547 time=63.94it/s +epoch=15 global_step=6000 loss=3.53686 loss_avg=3.51802 acc=0.64844 acc_top1_avg=0.66291 acc_top5_avg=0.91227 lr=0.01000 gn=8.98778 time=58.73it/s +epoch=15 global_step=6050 loss=4.49530 loss_avg=3.51634 acc=0.54688 acc_top1_avg=0.66318 acc_top5_avg=0.91068 lr=0.01000 gn=10.51121 time=57.09it/s +epoch=15 global_step=6100 loss=2.86741 loss_avg=3.52422 acc=0.73438 acc_top1_avg=0.66280 acc_top5_avg=0.90984 lr=0.01000 gn=9.13132 time=51.24it/s +epoch=15 global_step=6150 loss=4.10981 loss_avg=3.54288 acc=0.58594 acc_top1_avg=0.66096 acc_top5_avg=0.90899 lr=0.01000 gn=9.06146 time=54.16it/s +epoch=15 global_step=6200 loss=3.72690 loss_avg=3.55199 acc=0.63281 acc_top1_avg=0.65956 acc_top5_avg=0.90896 lr=0.01000 gn=8.23100 time=58.46it/s +epoch=15 global_step=6250 loss=3.72156 loss_avg=3.55439 acc=0.64844 acc_top1_avg=0.65903 acc_top5_avg=0.90960 lr=0.01000 gn=9.00251 time=58.10it/s +====================Eval==================== +epoch=15 global_step=6256 loss=0.93953 test_loss_avg=1.44991 acc=0.75000 test_acc_avg=0.64375 test_acc_top5_avg=0.96094 time=234.20it/s +epoch=15 global_step=6256 loss=0.24888 test_loss_avg=1.12506 acc=0.92969 test_acc_avg=0.71082 test_acc_top5_avg=0.97572 time=238.61it/s +epoch=15 global_step=6256 loss=0.01417 test_loss_avg=0.97850 acc=1.00000 test_acc_avg=0.74614 test_acc_top5_avg=0.97864 time=470.90it/s +curr_acc 0.7461 +BEST_ACC 0.7881 +curr_acc_top5 0.9786 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=3.79613 loss_avg=3.47924 acc=0.64062 acc_top1_avg=0.66477 acc_top5_avg=0.91264 lr=0.01000 gn=8.77026 time=57.52it/s +epoch=16 global_step=6350 loss=3.34910 loss_avg=3.52173 acc=0.66406 acc_top1_avg=0.66215 acc_top5_avg=0.90841 lr=0.01000 gn=9.16877 time=62.86it/s +epoch=16 global_step=6400 loss=2.79483 loss_avg=3.55305 acc=0.75781 acc_top1_avg=0.65869 acc_top5_avg=0.90847 lr=0.01000 gn=8.37696 time=63.30it/s +epoch=16 global_step=6450 loss=3.46907 loss_avg=3.52749 acc=0.67188 acc_top1_avg=0.66205 acc_top5_avg=0.90859 lr=0.01000 gn=7.60296 time=63.20it/s +epoch=16 global_step=6500 loss=3.59868 loss_avg=3.51206 acc=0.63281 acc_top1_avg=0.66342 acc_top5_avg=0.90894 lr=0.01000 gn=12.01320 time=54.77it/s +epoch=16 global_step=6550 loss=3.61159 loss_avg=3.52707 acc=0.64844 acc_top1_avg=0.66247 acc_top5_avg=0.90814 lr=0.01000 gn=8.26056 time=60.12it/s +epoch=16 global_step=6600 loss=3.98536 loss_avg=3.52463 acc=0.60156 acc_top1_avg=0.66268 acc_top5_avg=0.90752 lr=0.01000 gn=8.89746 time=54.53it/s +====================Eval==================== +epoch=16 global_step=6647 loss=0.39048 test_loss_avg=0.75934 acc=0.85938 test_acc_avg=0.79427 test_acc_top5_avg=0.98676 time=238.07it/s +epoch=16 global_step=6647 loss=0.82972 test_loss_avg=0.84169 acc=0.75000 test_acc_avg=0.78105 test_acc_top5_avg=0.98042 time=744.07it/s +curr_acc 0.7811 +BEST_ACC 0.7881 +curr_acc_top5 0.9804 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=3.75883 loss_avg=3.66007 acc=0.63281 acc_top1_avg=0.64323 acc_top5_avg=0.90104 lr=0.01000 gn=8.11634 time=61.47it/s +epoch=17 global_step=6700 loss=3.98483 loss_avg=3.45900 acc=0.59375 acc_top1_avg=0.67129 acc_top5_avg=0.90817 lr=0.01000 gn=8.95704 time=50.32it/s +epoch=17 global_step=6750 loss=3.78155 loss_avg=3.47562 acc=0.60938 acc_top1_avg=0.66914 acc_top5_avg=0.90762 lr=0.01000 gn=7.94843 time=63.57it/s +epoch=17 global_step=6800 loss=3.24735 loss_avg=3.49147 acc=0.69531 acc_top1_avg=0.66667 acc_top5_avg=0.90753 lr=0.01000 gn=10.31205 time=55.59it/s +epoch=17 global_step=6850 loss=3.18645 loss_avg=3.51058 acc=0.70312 acc_top1_avg=0.66399 acc_top5_avg=0.90764 lr=0.01000 gn=9.87130 time=55.51it/s +epoch=17 global_step=6900 loss=3.54551 loss_avg=3.52485 acc=0.66406 acc_top1_avg=0.66215 acc_top5_avg=0.90884 lr=0.01000 gn=8.81150 time=55.79it/s +epoch=17 global_step=6950 loss=3.55165 loss_avg=3.52251 acc=0.65625 acc_top1_avg=0.66190 acc_top5_avg=0.90950 lr=0.01000 gn=9.06406 time=49.31it/s +epoch=17 global_step=7000 loss=4.06879 loss_avg=3.52537 acc=0.62500 acc_top1_avg=0.66196 acc_top5_avg=0.90915 lr=0.01000 gn=9.38101 time=52.77it/s +====================Eval==================== +epoch=17 global_step=7038 loss=0.45740 test_loss_avg=0.43806 acc=0.85938 test_acc_avg=0.87388 test_acc_top5_avg=0.99330 time=257.41it/s +epoch=17 global_step=7038 loss=0.64258 test_loss_avg=0.86644 acc=0.82031 test_acc_avg=0.77070 test_acc_top5_avg=0.98163 time=238.73it/s +epoch=17 global_step=7038 loss=0.57608 test_loss_avg=0.80042 acc=0.93750 test_acc_avg=0.78600 test_acc_top5_avg=0.98200 time=511.13it/s +curr_acc 0.7860 +BEST_ACC 0.7881 +curr_acc_top5 0.9820 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=3.11209 loss_avg=3.60848 acc=0.71094 acc_top1_avg=0.64909 acc_top5_avg=0.90690 lr=0.01000 gn=10.46345 time=55.76it/s +epoch=18 global_step=7100 loss=3.81131 loss_avg=3.53961 acc=0.64844 acc_top1_avg=0.65738 acc_top5_avg=0.91003 lr=0.01000 gn=11.07733 time=55.92it/s +epoch=18 global_step=7150 loss=3.26352 loss_avg=3.47093 acc=0.67969 acc_top1_avg=0.66553 acc_top5_avg=0.91211 lr=0.01000 gn=8.51863 time=63.62it/s +epoch=18 global_step=7200 loss=2.79736 loss_avg=3.47337 acc=0.74219 acc_top1_avg=0.66662 acc_top5_avg=0.91098 lr=0.01000 gn=9.45376 time=57.24it/s +epoch=18 global_step=7250 loss=3.42976 loss_avg=3.49037 acc=0.66406 acc_top1_avg=0.66495 acc_top5_avg=0.91122 lr=0.01000 gn=9.90411 time=57.88it/s +epoch=18 global_step=7300 loss=4.09425 loss_avg=3.48935 acc=0.59375 acc_top1_avg=0.66529 acc_top5_avg=0.91201 lr=0.01000 gn=9.10547 time=57.60it/s +epoch=18 global_step=7350 loss=3.15304 loss_avg=3.49821 acc=0.70312 acc_top1_avg=0.66406 acc_top5_avg=0.91186 lr=0.01000 gn=9.80108 time=59.65it/s +epoch=18 global_step=7400 loss=3.48907 loss_avg=3.50927 acc=0.67969 acc_top1_avg=0.66320 acc_top5_avg=0.91154 lr=0.01000 gn=11.24724 time=63.36it/s +====================Eval==================== +epoch=18 global_step=7429 loss=2.95025 test_loss_avg=1.03397 acc=0.34375 test_acc_avg=0.73884 test_acc_top5_avg=0.95257 time=245.09it/s +epoch=18 global_step=7429 loss=0.43560 test_loss_avg=0.90567 acc=0.85938 test_acc_avg=0.76472 test_acc_top5_avg=0.95954 time=240.87it/s +epoch=18 global_step=7429 loss=0.53991 test_loss_avg=0.90104 acc=0.93750 test_acc_avg=0.76691 test_acc_top5_avg=0.96005 time=504.12it/s +curr_acc 0.7669 +BEST_ACC 0.7881 +curr_acc_top5 0.9600 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=3.10751 loss_avg=3.49517 acc=0.67188 acc_top1_avg=0.66741 acc_top5_avg=0.90811 lr=0.01000 gn=8.00270 time=43.91it/s +epoch=19 global_step=7500 loss=3.27908 loss_avg=3.43396 acc=0.71094 acc_top1_avg=0.67331 acc_top5_avg=0.91186 lr=0.01000 gn=9.10839 time=62.62it/s +epoch=19 global_step=7550 loss=3.21985 loss_avg=3.44379 acc=0.70312 acc_top1_avg=0.67175 acc_top5_avg=0.90896 lr=0.01000 gn=9.34583 time=53.45it/s +epoch=19 global_step=7600 loss=3.30904 loss_avg=3.44960 acc=0.68750 acc_top1_avg=0.67174 acc_top5_avg=0.91082 lr=0.01000 gn=8.67197 time=62.33it/s +epoch=19 global_step=7650 loss=3.75084 loss_avg=3.47344 acc=0.63281 acc_top1_avg=0.66901 acc_top5_avg=0.91003 lr=0.01000 gn=8.76889 time=57.92it/s +epoch=19 global_step=7700 loss=3.92818 loss_avg=3.49750 acc=0.60938 acc_top1_avg=0.66602 acc_top5_avg=0.90893 lr=0.01000 gn=9.50564 time=53.53it/s +epoch=19 global_step=7750 loss=2.99673 loss_avg=3.50609 acc=0.71094 acc_top1_avg=0.66479 acc_top5_avg=0.90937 lr=0.01000 gn=7.28438 time=60.44it/s +epoch=19 global_step=7800 loss=3.67743 loss_avg=3.50233 acc=0.62500 acc_top1_avg=0.66482 acc_top5_avg=0.91042 lr=0.01000 gn=10.09494 time=53.44it/s +====================Eval==================== +epoch=19 global_step=7820 loss=0.21970 test_loss_avg=1.18810 acc=0.94531 test_acc_avg=0.71588 test_acc_top5_avg=0.97736 time=254.12it/s +epoch=19 global_step=7820 loss=0.91630 test_loss_avg=1.01196 acc=0.75000 test_acc_avg=0.75277 test_acc_top5_avg=0.97824 time=556.64it/s +curr_acc 0.7528 +BEST_ACC 0.7881 +curr_acc_top5 0.9782 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=3.56888 loss_avg=3.52202 acc=0.64844 acc_top1_avg=0.66302 acc_top5_avg=0.91120 lr=0.01000 gn=9.44683 time=55.33it/s +epoch=20 global_step=7900 loss=3.18420 loss_avg=3.44329 acc=0.68750 acc_top1_avg=0.67266 acc_top5_avg=0.91084 lr=0.01000 gn=9.66520 time=61.97it/s +epoch=20 global_step=7950 loss=3.53764 loss_avg=3.47510 acc=0.65625 acc_top1_avg=0.66887 acc_top5_avg=0.91244 lr=0.01000 gn=9.10547 time=53.86it/s +epoch=20 global_step=8000 loss=2.77089 loss_avg=3.48585 acc=0.74219 acc_top1_avg=0.66675 acc_top5_avg=0.91150 lr=0.01000 gn=6.14880 time=50.52it/s +epoch=20 global_step=8050 loss=3.66033 loss_avg=3.48096 acc=0.66406 acc_top1_avg=0.66726 acc_top5_avg=0.91270 lr=0.01000 gn=9.86889 time=57.14it/s +epoch=20 global_step=8100 loss=3.03445 loss_avg=3.48579 acc=0.72656 acc_top1_avg=0.66646 acc_top5_avg=0.91214 lr=0.01000 gn=9.34443 time=52.78it/s +epoch=20 global_step=8150 loss=3.39148 loss_avg=3.49736 acc=0.67188 acc_top1_avg=0.66525 acc_top5_avg=0.91155 lr=0.01000 gn=9.93676 time=54.95it/s +epoch=20 global_step=8200 loss=3.10181 loss_avg=3.50217 acc=0.72656 acc_top1_avg=0.66474 acc_top5_avg=0.91157 lr=0.01000 gn=10.03230 time=60.69it/s +====================Eval==================== +epoch=20 global_step=8211 loss=1.92667 test_loss_avg=0.82020 acc=0.51562 test_acc_avg=0.78398 test_acc_top5_avg=0.98828 time=242.14it/s +epoch=20 global_step=8211 loss=0.88172 test_loss_avg=0.91589 acc=0.74219 test_acc_avg=0.76306 test_acc_top5_avg=0.98103 time=259.29it/s +epoch=20 global_step=8211 loss=0.00816 test_loss_avg=0.84207 acc=1.00000 test_acc_avg=0.77996 test_acc_top5_avg=0.98230 time=699.63it/s +curr_acc 0.7800 +BEST_ACC 0.7881 +curr_acc_top5 0.9823 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=3.54100 loss_avg=3.28043 acc=0.64844 acc_top1_avg=0.68470 acc_top5_avg=0.92548 lr=0.01000 gn=8.63946 time=47.91it/s +epoch=21 global_step=8300 loss=3.44810 loss_avg=3.39698 acc=0.67188 acc_top1_avg=0.67521 acc_top5_avg=0.92003 lr=0.01000 gn=10.29910 time=56.26it/s +epoch=21 global_step=8350 loss=3.85470 loss_avg=3.40204 acc=0.64062 acc_top1_avg=0.67542 acc_top5_avg=0.91715 lr=0.01000 gn=7.72997 time=63.39it/s +epoch=21 global_step=8400 loss=4.22114 loss_avg=3.45727 acc=0.58594 acc_top1_avg=0.66902 acc_top5_avg=0.91555 lr=0.01000 gn=9.41485 time=63.38it/s +epoch=21 global_step=8450 loss=3.34710 loss_avg=3.45703 acc=0.66406 acc_top1_avg=0.66867 acc_top5_avg=0.91449 lr=0.01000 gn=8.42622 time=54.68it/s +epoch=21 global_step=8500 loss=3.57174 loss_avg=3.46844 acc=0.64844 acc_top1_avg=0.66741 acc_top5_avg=0.91374 lr=0.01000 gn=7.96921 time=62.14it/s +epoch=21 global_step=8550 loss=2.74835 loss_avg=3.48490 acc=0.74219 acc_top1_avg=0.66581 acc_top5_avg=0.91240 lr=0.01000 gn=9.71937 time=60.45it/s +epoch=21 global_step=8600 loss=3.63994 loss_avg=3.49290 acc=0.65625 acc_top1_avg=0.66487 acc_top5_avg=0.91209 lr=0.01000 gn=8.71001 time=57.61it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.39529 test_loss_avg=1.71516 acc=0.65625 test_acc_avg=0.64844 test_acc_top5_avg=0.94341 time=227.00it/s +epoch=21 global_step=8602 loss=1.83753 test_loss_avg=1.48472 acc=0.50000 test_acc_avg=0.67237 test_acc_top5_avg=0.95520 time=568.33it/s +curr_acc 0.6724 +BEST_ACC 0.7881 +curr_acc_top5 0.9552 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=3.16278 loss_avg=3.47862 acc=0.70312 acc_top1_avg=0.66585 acc_top5_avg=0.91178 lr=0.01000 gn=11.21309 time=60.41it/s +epoch=22 global_step=8700 loss=2.82483 loss_avg=3.47729 acc=0.73438 acc_top1_avg=0.66773 acc_top5_avg=0.91087 lr=0.01000 gn=9.38685 time=46.06it/s +epoch=22 global_step=8750 loss=3.73370 loss_avg=3.47607 acc=0.64844 acc_top1_avg=0.66707 acc_top5_avg=0.91253 lr=0.01000 gn=11.45948 time=51.11it/s +epoch=22 global_step=8800 loss=3.80419 loss_avg=3.48240 acc=0.63281 acc_top1_avg=0.66773 acc_top5_avg=0.91193 lr=0.01000 gn=12.42913 time=61.60it/s +epoch=22 global_step=8850 loss=3.30365 loss_avg=3.46687 acc=0.67969 acc_top1_avg=0.66913 acc_top5_avg=0.91327 lr=0.01000 gn=9.69818 time=61.42it/s +epoch=22 global_step=8900 loss=3.47848 loss_avg=3.48415 acc=0.64062 acc_top1_avg=0.66716 acc_top5_avg=0.91236 lr=0.01000 gn=8.78406 time=57.31it/s +epoch=22 global_step=8950 loss=3.17686 loss_avg=3.48418 acc=0.72656 acc_top1_avg=0.66689 acc_top5_avg=0.91303 lr=0.01000 gn=9.85983 time=61.17it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.84384 test_loss_avg=0.49512 acc=0.79688 test_acc_avg=0.87109 test_acc_top5_avg=0.99414 time=244.87it/s +epoch=22 global_step=8993 loss=0.79977 test_loss_avg=0.99320 acc=0.83594 test_acc_avg=0.75491 test_acc_top5_avg=0.97946 time=232.94it/s +epoch=22 global_step=8993 loss=0.21434 test_loss_avg=0.88635 acc=0.93750 test_acc_avg=0.77779 test_acc_top5_avg=0.98269 time=557.68it/s +curr_acc 0.7778 +BEST_ACC 0.7881 +curr_acc_top5 0.9827 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=3.88367 loss_avg=3.52957 acc=0.60938 acc_top1_avg=0.66741 acc_top5_avg=0.90737 lr=0.01000 gn=10.01704 time=55.18it/s +epoch=23 global_step=9050 loss=3.69960 loss_avg=3.43457 acc=0.64062 acc_top1_avg=0.67475 acc_top5_avg=0.91160 lr=0.01000 gn=8.66720 time=51.91it/s +epoch=23 global_step=9100 loss=3.87645 loss_avg=3.47704 acc=0.60156 acc_top1_avg=0.66808 acc_top5_avg=0.91341 lr=0.01000 gn=9.52106 time=58.44it/s +epoch=23 global_step=9150 loss=3.17098 loss_avg=3.48640 acc=0.69531 acc_top1_avg=0.66700 acc_top5_avg=0.91262 lr=0.01000 gn=8.50425 time=60.02it/s +epoch=23 global_step=9200 loss=3.12036 loss_avg=3.47282 acc=0.71875 acc_top1_avg=0.66893 acc_top5_avg=0.91316 lr=0.01000 gn=10.54569 time=60.24it/s +epoch=23 global_step=9250 loss=2.93829 loss_avg=3.47031 acc=0.71875 acc_top1_avg=0.66859 acc_top5_avg=0.91285 lr=0.01000 gn=8.38247 time=62.20it/s +epoch=23 global_step=9300 loss=3.04413 loss_avg=3.46790 acc=0.71094 acc_top1_avg=0.66846 acc_top5_avg=0.91282 lr=0.01000 gn=6.69225 time=59.03it/s +epoch=23 global_step=9350 loss=3.03706 loss_avg=3.47864 acc=0.71875 acc_top1_avg=0.66759 acc_top5_avg=0.91251 lr=0.01000 gn=9.92379 time=58.53it/s +====================Eval==================== +epoch=23 global_step=9384 loss=1.49482 test_loss_avg=0.87954 acc=0.66406 test_acc_avg=0.77225 test_acc_top5_avg=0.98319 time=242.89it/s +epoch=23 global_step=9384 loss=1.02449 test_loss_avg=0.90197 acc=0.81250 test_acc_avg=0.77156 test_acc_top5_avg=0.98319 time=560.81it/s +curr_acc 0.7716 +BEST_ACC 0.7881 +curr_acc_top5 0.9832 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=3.00104 loss_avg=3.37161 acc=0.71094 acc_top1_avg=0.67480 acc_top5_avg=0.91260 lr=0.01000 gn=8.71224 time=54.57it/s +epoch=24 global_step=9450 loss=3.29375 loss_avg=3.45196 acc=0.69531 acc_top1_avg=0.66939 acc_top5_avg=0.91205 lr=0.01000 gn=8.44518 time=59.84it/s +epoch=24 global_step=9500 loss=2.90497 loss_avg=3.40899 acc=0.73438 acc_top1_avg=0.67437 acc_top5_avg=0.91366 lr=0.01000 gn=7.65711 time=54.22it/s +epoch=24 global_step=9550 loss=4.00002 loss_avg=3.42389 acc=0.62500 acc_top1_avg=0.67206 acc_top5_avg=0.91124 lr=0.01000 gn=9.71788 time=56.37it/s +epoch=24 global_step=9600 loss=3.60078 loss_avg=3.44770 acc=0.66406 acc_top1_avg=0.66963 acc_top5_avg=0.91139 lr=0.01000 gn=8.28688 time=57.06it/s +epoch=24 global_step=9650 loss=3.54682 loss_avg=3.44153 acc=0.67188 acc_top1_avg=0.67047 acc_top5_avg=0.91177 lr=0.01000 gn=8.27314 time=57.47it/s +epoch=24 global_step=9700 loss=3.40864 loss_avg=3.46192 acc=0.67969 acc_top1_avg=0.66836 acc_top5_avg=0.91206 lr=0.01000 gn=9.90819 time=56.06it/s +epoch=24 global_step=9750 loss=3.84201 loss_avg=3.46921 acc=0.62500 acc_top1_avg=0.66720 acc_top5_avg=0.91235 lr=0.01000 gn=8.63590 time=51.70it/s +====================Eval==================== +epoch=24 global_step=9775 loss=0.52835 test_loss_avg=0.69328 acc=0.86719 test_acc_avg=0.82617 test_acc_top5_avg=0.98242 time=248.05it/s +epoch=24 global_step=9775 loss=0.48287 test_loss_avg=1.04926 acc=0.89062 test_acc_avg=0.76591 test_acc_top5_avg=0.98900 time=227.57it/s +epoch=24 global_step=9775 loss=0.70049 test_loss_avg=0.99376 acc=0.75000 test_acc_avg=0.77156 test_acc_top5_avg=0.98596 time=886.37it/s +curr_acc 0.7716 +BEST_ACC 0.7881 +curr_acc_top5 0.9860 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=2.87990 loss_avg=3.45051 acc=0.75000 acc_top1_avg=0.67188 acc_top5_avg=0.91687 lr=0.01000 gn=10.98351 time=59.97it/s +epoch=25 global_step=9850 loss=3.60421 loss_avg=3.43747 acc=0.65625 acc_top1_avg=0.67250 acc_top5_avg=0.91375 lr=0.01000 gn=9.62428 time=48.41it/s +epoch=25 global_step=9900 loss=4.04684 loss_avg=3.43455 acc=0.60938 acc_top1_avg=0.67212 acc_top5_avg=0.91369 lr=0.01000 gn=9.58564 time=55.33it/s +epoch=25 global_step=9950 loss=3.51728 loss_avg=3.45664 acc=0.66406 acc_top1_avg=0.66955 acc_top5_avg=0.91304 lr=0.01000 gn=10.09386 time=53.71it/s +epoch=25 global_step=10000 loss=3.69281 loss_avg=3.45442 acc=0.64062 acc_top1_avg=0.67045 acc_top5_avg=0.91278 lr=0.01000 gn=9.76209 time=59.02it/s +epoch=25 global_step=10050 loss=3.14249 loss_avg=3.46462 acc=0.69531 acc_top1_avg=0.66889 acc_top5_avg=0.91247 lr=0.01000 gn=9.47446 time=54.97it/s +epoch=25 global_step=10100 loss=2.95150 loss_avg=3.47530 acc=0.73438 acc_top1_avg=0.66774 acc_top5_avg=0.91221 lr=0.01000 gn=9.28204 time=55.17it/s +epoch=25 global_step=10150 loss=3.37027 loss_avg=3.46846 acc=0.68750 acc_top1_avg=0.66833 acc_top5_avg=0.91283 lr=0.01000 gn=9.65103 time=57.69it/s +====================Eval==================== +epoch=25 global_step=10166 loss=1.78944 test_loss_avg=0.75350 acc=0.58594 test_acc_avg=0.79406 test_acc_top5_avg=0.97469 time=234.02it/s +epoch=25 global_step=10166 loss=1.43341 test_loss_avg=0.77931 acc=0.67188 test_acc_avg=0.79333 test_acc_top5_avg=0.98021 time=241.48it/s +epoch=25 global_step=10166 loss=2.14529 test_loss_avg=0.83870 acc=0.50000 test_acc_avg=0.77996 test_acc_top5_avg=0.97953 time=740.00it/s +curr_acc 0.7800 +BEST_ACC 0.7881 +curr_acc_top5 0.9795 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=2.68383 loss_avg=3.39621 acc=0.75781 acc_top1_avg=0.67624 acc_top5_avg=0.91429 lr=0.01000 gn=8.21603 time=62.43it/s +epoch=26 global_step=10250 loss=4.28123 loss_avg=3.43503 acc=0.57812 acc_top1_avg=0.67104 acc_top5_avg=0.91285 lr=0.01000 gn=10.59183 time=54.68it/s +epoch=26 global_step=10300 loss=2.72934 loss_avg=3.47369 acc=0.75781 acc_top1_avg=0.66674 acc_top5_avg=0.91167 lr=0.01000 gn=10.73479 time=56.83it/s +epoch=26 global_step=10350 loss=3.38053 loss_avg=3.49583 acc=0.67969 acc_top1_avg=0.66453 acc_top5_avg=0.91041 lr=0.01000 gn=8.73725 time=61.01it/s +epoch=26 global_step=10400 loss=3.83976 loss_avg=3.47858 acc=0.60938 acc_top1_avg=0.66623 acc_top5_avg=0.90972 lr=0.01000 gn=7.98224 time=62.37it/s +epoch=26 global_step=10450 loss=3.16322 loss_avg=3.46302 acc=0.69531 acc_top1_avg=0.66824 acc_top5_avg=0.91216 lr=0.01000 gn=8.59396 time=53.35it/s +epoch=26 global_step=10500 loss=3.40150 loss_avg=3.46850 acc=0.66406 acc_top1_avg=0.66731 acc_top5_avg=0.91264 lr=0.01000 gn=9.35200 time=60.33it/s +epoch=26 global_step=10550 loss=4.00766 loss_avg=3.46569 acc=0.60938 acc_top1_avg=0.66764 acc_top5_avg=0.91288 lr=0.01000 gn=9.97810 time=59.59it/s +====================Eval==================== +epoch=26 global_step=10557 loss=1.77652 test_loss_avg=0.89122 acc=0.53125 test_acc_avg=0.76868 test_acc_top5_avg=0.97469 time=241.62it/s +epoch=26 global_step=10557 loss=1.14328 test_loss_avg=0.78468 acc=0.68750 test_acc_avg=0.78877 test_acc_top5_avg=0.98032 time=852.15it/s +curr_acc 0.7888 +BEST_ACC 0.7881 +curr_acc_top5 0.9803 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=3.24028 loss_avg=3.51318 acc=0.70312 acc_top1_avg=0.66443 acc_top5_avg=0.90734 lr=0.01000 gn=10.21940 time=54.37it/s +epoch=27 global_step=10650 loss=3.72551 loss_avg=3.44096 acc=0.64844 acc_top1_avg=0.67255 acc_top5_avg=0.90995 lr=0.01000 gn=8.45288 time=54.22it/s +epoch=27 global_step=10700 loss=3.39867 loss_avg=3.39461 acc=0.67969 acc_top1_avg=0.67772 acc_top5_avg=0.91297 lr=0.01000 gn=8.28149 time=57.81it/s +epoch=27 global_step=10750 loss=3.04655 loss_avg=3.41056 acc=0.71094 acc_top1_avg=0.67556 acc_top5_avg=0.91208 lr=0.01000 gn=10.52939 time=62.83it/s +epoch=27 global_step=10800 loss=3.33567 loss_avg=3.42487 acc=0.67188 acc_top1_avg=0.67393 acc_top5_avg=0.91197 lr=0.01000 gn=8.45234 time=54.90it/s +epoch=27 global_step=10850 loss=3.18217 loss_avg=3.42088 acc=0.70312 acc_top1_avg=0.67473 acc_top5_avg=0.91182 lr=0.01000 gn=8.72218 time=55.91it/s +epoch=27 global_step=10900 loss=3.90157 loss_avg=3.43111 acc=0.64062 acc_top1_avg=0.67367 acc_top5_avg=0.91183 lr=0.01000 gn=9.01366 time=60.85it/s +====================Eval==================== +epoch=27 global_step=10948 loss=1.32208 test_loss_avg=0.66930 acc=0.65625 test_acc_avg=0.81756 test_acc_top5_avg=0.99357 time=242.89it/s +epoch=27 global_step=10948 loss=0.17998 test_loss_avg=0.87702 acc=0.95312 test_acc_avg=0.77239 test_acc_top5_avg=0.98379 time=242.74it/s +epoch=27 global_step=10948 loss=0.23827 test_loss_avg=0.80019 acc=0.87500 test_acc_avg=0.78956 test_acc_top5_avg=0.98527 time=630.63it/s +curr_acc 0.7896 +BEST_ACC 0.7888 +curr_acc_top5 0.9853 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=3.55548 loss_avg=3.27812 acc=0.64844 acc_top1_avg=0.67578 acc_top5_avg=0.93359 lr=0.01000 gn=8.91614 time=51.14it/s +epoch=28 global_step=11000 loss=4.31507 loss_avg=3.40841 acc=0.59375 acc_top1_avg=0.67713 acc_top5_avg=0.90745 lr=0.01000 gn=10.00828 time=55.89it/s +epoch=28 global_step=11050 loss=3.35870 loss_avg=3.44200 acc=0.67969 acc_top1_avg=0.67165 acc_top5_avg=0.91039 lr=0.01000 gn=9.31371 time=58.70it/s +epoch=28 global_step=11100 loss=3.86895 loss_avg=3.42657 acc=0.62500 acc_top1_avg=0.67254 acc_top5_avg=0.91252 lr=0.01000 gn=9.34530 time=61.00it/s +epoch=28 global_step=11150 loss=3.36087 loss_avg=3.40356 acc=0.67188 acc_top1_avg=0.67462 acc_top5_avg=0.91368 lr=0.01000 gn=8.44664 time=63.12it/s +epoch=28 global_step=11200 loss=3.79895 loss_avg=3.42278 acc=0.60156 acc_top1_avg=0.67296 acc_top5_avg=0.91304 lr=0.01000 gn=11.41657 time=55.13it/s +epoch=28 global_step=11250 loss=3.16708 loss_avg=3.41952 acc=0.71094 acc_top1_avg=0.67366 acc_top5_avg=0.91305 lr=0.01000 gn=11.20750 time=56.95it/s +epoch=28 global_step=11300 loss=3.26711 loss_avg=3.42387 acc=0.69531 acc_top1_avg=0.67292 acc_top5_avg=0.91322 lr=0.01000 gn=11.27683 time=62.64it/s +====================Eval==================== +epoch=28 global_step=11339 loss=0.21365 test_loss_avg=1.47204 acc=0.92969 test_acc_avg=0.67064 test_acc_top5_avg=0.96690 time=240.06it/s +epoch=28 global_step=11339 loss=0.15756 test_loss_avg=1.16704 acc=0.93750 test_acc_avg=0.73536 test_acc_top5_avg=0.97607 time=872.36it/s +curr_acc 0.7354 +BEST_ACC 0.7896 +curr_acc_top5 0.9761 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=3.65737 loss_avg=3.52322 acc=0.64844 acc_top1_avg=0.65980 acc_top5_avg=0.91619 lr=0.01000 gn=8.65786 time=56.69it/s +epoch=29 global_step=11400 loss=3.89457 loss_avg=3.37014 acc=0.61719 acc_top1_avg=0.67508 acc_top5_avg=0.91432 lr=0.01000 gn=10.85385 time=59.01it/s +epoch=29 global_step=11450 loss=2.95076 loss_avg=3.40875 acc=0.72656 acc_top1_avg=0.67216 acc_top5_avg=0.91681 lr=0.01000 gn=8.77850 time=54.84it/s +epoch=29 global_step=11500 loss=2.77246 loss_avg=3.43347 acc=0.73438 acc_top1_avg=0.66974 acc_top5_avg=0.91503 lr=0.01000 gn=9.45882 time=59.11it/s +epoch=29 global_step=11550 loss=3.28355 loss_avg=3.45482 acc=0.69531 acc_top1_avg=0.66773 acc_top5_avg=0.91488 lr=0.01000 gn=10.05695 time=62.88it/s +epoch=29 global_step=11600 loss=3.30865 loss_avg=3.44325 acc=0.67188 acc_top1_avg=0.66963 acc_top5_avg=0.91418 lr=0.01000 gn=7.63078 time=52.16it/s +epoch=29 global_step=11650 loss=4.19154 loss_avg=3.44973 acc=0.59375 acc_top1_avg=0.66899 acc_top5_avg=0.91399 lr=0.01000 gn=10.20535 time=57.64it/s +epoch=29 global_step=11700 loss=3.28486 loss_avg=3.44519 acc=0.67969 acc_top1_avg=0.66949 acc_top5_avg=0.91359 lr=0.01000 gn=9.33977 time=58.13it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.16954 test_loss_avg=1.40838 acc=0.96094 test_acc_avg=0.62934 test_acc_top5_avg=0.98090 time=249.63it/s +epoch=29 global_step=11730 loss=1.03819 test_loss_avg=1.09216 acc=0.74219 test_acc_avg=0.72842 test_acc_top5_avg=0.96901 time=236.86it/s +epoch=29 global_step=11730 loss=0.45221 test_loss_avg=0.93138 acc=0.87500 test_acc_avg=0.76642 test_acc_top5_avg=0.97389 time=874.18it/s +curr_acc 0.7664 +BEST_ACC 0.7896 +curr_acc_top5 0.9739 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=3.58546 loss_avg=3.29993 acc=0.66406 acc_top1_avg=0.68945 acc_top5_avg=0.91406 lr=0.01000 gn=10.09260 time=54.01it/s +epoch=30 global_step=11800 loss=3.49879 loss_avg=3.39041 acc=0.66406 acc_top1_avg=0.67656 acc_top5_avg=0.91629 lr=0.01000 gn=11.26565 time=33.32it/s +epoch=30 global_step=11850 loss=3.05475 loss_avg=3.40015 acc=0.71094 acc_top1_avg=0.67500 acc_top5_avg=0.91445 lr=0.01000 gn=8.70897 time=52.26it/s +epoch=30 global_step=11900 loss=3.60115 loss_avg=3.41802 acc=0.67188 acc_top1_avg=0.67307 acc_top5_avg=0.91443 lr=0.01000 gn=8.90367 time=60.29it/s +epoch=30 global_step=11950 loss=3.15967 loss_avg=3.42967 acc=0.71094 acc_top1_avg=0.67166 acc_top5_avg=0.91484 lr=0.01000 gn=9.94943 time=60.72it/s +epoch=30 global_step=12000 loss=3.07402 loss_avg=3.42801 acc=0.71094 acc_top1_avg=0.67222 acc_top5_avg=0.91516 lr=0.01000 gn=11.55871 time=57.02it/s +epoch=30 global_step=12050 loss=3.38685 loss_avg=3.43632 acc=0.67969 acc_top1_avg=0.67207 acc_top5_avg=0.91404 lr=0.01000 gn=11.18793 time=63.26it/s +epoch=30 global_step=12100 loss=3.74336 loss_avg=3.44063 acc=0.62500 acc_top1_avg=0.67128 acc_top5_avg=0.91459 lr=0.01000 gn=10.14069 time=52.31it/s +====================Eval==================== +epoch=30 global_step=12121 loss=0.32964 test_loss_avg=1.25683 acc=0.90625 test_acc_avg=0.70521 test_acc_top5_avg=0.97943 time=237.06it/s +epoch=30 global_step=12121 loss=0.49110 test_loss_avg=1.15469 acc=0.81250 test_acc_avg=0.74219 test_acc_top5_avg=0.98220 time=875.45it/s +curr_acc 0.7422 +BEST_ACC 0.7896 +curr_acc_top5 0.9822 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=4.14243 loss_avg=3.38541 acc=0.60156 acc_top1_avg=0.67484 acc_top5_avg=0.91002 lr=0.01000 gn=11.23173 time=53.55it/s +epoch=31 global_step=12200 loss=3.69108 loss_avg=3.37586 acc=0.66406 acc_top1_avg=0.67464 acc_top5_avg=0.91377 lr=0.01000 gn=11.16009 time=61.02it/s +epoch=31 global_step=12250 loss=3.39108 loss_avg=3.35292 acc=0.65625 acc_top1_avg=0.67775 acc_top5_avg=0.91473 lr=0.01000 gn=8.44934 time=54.96it/s +epoch=31 global_step=12300 loss=3.31311 loss_avg=3.38108 acc=0.68750 acc_top1_avg=0.67576 acc_top5_avg=0.91354 lr=0.01000 gn=11.14504 time=54.58it/s +epoch=31 global_step=12350 loss=3.36357 loss_avg=3.39436 acc=0.67969 acc_top1_avg=0.67460 acc_top5_avg=0.91386 lr=0.01000 gn=9.82700 time=51.50it/s +epoch=31 global_step=12400 loss=3.19217 loss_avg=3.41105 acc=0.70312 acc_top1_avg=0.67300 acc_top5_avg=0.91252 lr=0.01000 gn=10.37055 time=52.49it/s +epoch=31 global_step=12450 loss=2.98728 loss_avg=3.42537 acc=0.73438 acc_top1_avg=0.67185 acc_top5_avg=0.91228 lr=0.01000 gn=10.35503 time=53.13it/s +epoch=31 global_step=12500 loss=3.96984 loss_avg=3.42433 acc=0.61719 acc_top1_avg=0.67231 acc_top5_avg=0.91274 lr=0.01000 gn=12.13200 time=58.90it/s +====================Eval==================== +epoch=31 global_step=12512 loss=1.62202 test_loss_avg=1.62202 acc=0.60938 test_acc_avg=0.60938 test_acc_top5_avg=0.90625 time=231.74it/s +epoch=31 global_step=12512 loss=1.12776 test_loss_avg=1.14399 acc=0.76562 test_acc_avg=0.72181 test_acc_top5_avg=0.97702 time=244.49it/s +epoch=31 global_step=12512 loss=0.37088 test_loss_avg=0.92099 acc=0.87500 test_acc_avg=0.77047 test_acc_top5_avg=0.98111 time=885.25it/s +curr_acc 0.7705 +BEST_ACC 0.7896 +curr_acc_top5 0.9811 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=3.87385 loss_avg=3.40780 acc=0.60938 acc_top1_avg=0.67270 acc_top5_avg=0.91057 lr=0.01000 gn=8.97514 time=52.72it/s +epoch=32 global_step=12600 loss=3.96304 loss_avg=3.46597 acc=0.62500 acc_top1_avg=0.66886 acc_top5_avg=0.90794 lr=0.01000 gn=9.05685 time=59.80it/s +epoch=32 global_step=12650 loss=3.20963 loss_avg=3.42700 acc=0.70312 acc_top1_avg=0.67188 acc_top5_avg=0.91044 lr=0.01000 gn=9.44952 time=55.27it/s +epoch=32 global_step=12700 loss=3.34801 loss_avg=3.41905 acc=0.67188 acc_top1_avg=0.67163 acc_top5_avg=0.91144 lr=0.01000 gn=8.88255 time=50.07it/s +epoch=32 global_step=12750 loss=3.30171 loss_avg=3.44085 acc=0.67188 acc_top1_avg=0.66941 acc_top5_avg=0.91196 lr=0.01000 gn=9.22063 time=53.14it/s +epoch=32 global_step=12800 loss=3.46003 loss_avg=3.43903 acc=0.67969 acc_top1_avg=0.66938 acc_top5_avg=0.91216 lr=0.01000 gn=10.74888 time=53.93it/s +epoch=32 global_step=12850 loss=4.74679 loss_avg=3.43531 acc=0.53125 acc_top1_avg=0.67040 acc_top5_avg=0.91240 lr=0.01000 gn=8.42054 time=54.10it/s +epoch=32 global_step=12900 loss=2.73903 loss_avg=3.43447 acc=0.73438 acc_top1_avg=0.67055 acc_top5_avg=0.91279 lr=0.01000 gn=9.51796 time=62.23it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.66043 test_loss_avg=1.25930 acc=0.66406 test_acc_avg=0.70206 test_acc_top5_avg=0.97692 time=236.67it/s +epoch=32 global_step=12903 loss=1.76595 test_loss_avg=1.24157 acc=0.56250 test_acc_avg=0.71105 test_acc_top5_avg=0.97418 time=216.12it/s +epoch=32 global_step=12903 loss=1.55576 test_loss_avg=1.27986 acc=0.68750 test_acc_avg=0.70174 test_acc_top5_avg=0.96964 time=516.41it/s +curr_acc 0.7017 +BEST_ACC 0.7896 +curr_acc_top5 0.9696 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=3.25226 loss_avg=3.28688 acc=0.69531 acc_top1_avg=0.68833 acc_top5_avg=0.91805 lr=0.01000 gn=11.16156 time=56.00it/s +epoch=33 global_step=13000 loss=2.84457 loss_avg=3.33643 acc=0.73438 acc_top1_avg=0.68267 acc_top5_avg=0.91720 lr=0.01000 gn=7.75162 time=63.60it/s +epoch=33 global_step=13050 loss=3.10982 loss_avg=3.36857 acc=0.71875 acc_top1_avg=0.67969 acc_top5_avg=0.91534 lr=0.01000 gn=10.87183 time=55.67it/s +epoch=33 global_step=13100 loss=3.33544 loss_avg=3.38794 acc=0.69531 acc_top1_avg=0.67814 acc_top5_avg=0.91442 lr=0.01000 gn=10.80572 time=60.29it/s +epoch=33 global_step=13150 loss=3.17290 loss_avg=3.40139 acc=0.67969 acc_top1_avg=0.67602 acc_top5_avg=0.91397 lr=0.01000 gn=8.41696 time=51.19it/s +epoch=33 global_step=13200 loss=3.59409 loss_avg=3.40872 acc=0.65625 acc_top1_avg=0.67501 acc_top5_avg=0.91362 lr=0.01000 gn=10.30181 time=56.31it/s +epoch=33 global_step=13250 loss=3.12372 loss_avg=3.42289 acc=0.71875 acc_top1_avg=0.67284 acc_top5_avg=0.91386 lr=0.01000 gn=10.42794 time=56.09it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.64053 test_loss_avg=1.42507 acc=0.83594 test_acc_avg=0.66642 test_acc_top5_avg=0.95658 time=235.75it/s +epoch=33 global_step=13294 loss=0.29869 test_loss_avg=1.17775 acc=0.93750 test_acc_avg=0.71499 test_acc_top5_avg=0.96885 time=511.69it/s +curr_acc 0.7150 +BEST_ACC 0.7896 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=3.57029 loss_avg=3.39967 acc=0.67969 acc_top1_avg=0.67318 acc_top5_avg=0.92057 lr=0.01000 gn=10.82026 time=59.32it/s +epoch=34 global_step=13350 loss=3.55487 loss_avg=3.32763 acc=0.64062 acc_top1_avg=0.68052 acc_top5_avg=0.91616 lr=0.01000 gn=11.44721 time=56.20it/s +epoch=34 global_step=13400 loss=3.15771 loss_avg=3.34157 acc=0.67969 acc_top1_avg=0.68124 acc_top5_avg=0.91532 lr=0.01000 gn=10.00074 time=58.82it/s +epoch=34 global_step=13450 loss=3.13869 loss_avg=3.35984 acc=0.70312 acc_top1_avg=0.67944 acc_top5_avg=0.91466 lr=0.01000 gn=10.50555 time=54.09it/s +epoch=34 global_step=13500 loss=4.20854 loss_avg=3.37669 acc=0.57031 acc_top1_avg=0.67768 acc_top5_avg=0.91528 lr=0.01000 gn=10.09324 time=49.58it/s +epoch=34 global_step=13550 loss=2.51426 loss_avg=3.39851 acc=0.75781 acc_top1_avg=0.67581 acc_top5_avg=0.91422 lr=0.01000 gn=8.09803 time=60.45it/s +epoch=34 global_step=13600 loss=3.85384 loss_avg=3.40132 acc=0.62500 acc_top1_avg=0.67553 acc_top5_avg=0.91427 lr=0.01000 gn=9.96362 time=61.56it/s +epoch=34 global_step=13650 loss=3.61704 loss_avg=3.41720 acc=0.66406 acc_top1_avg=0.67411 acc_top5_avg=0.91347 lr=0.01000 gn=10.55846 time=57.82it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.81510 test_loss_avg=0.58037 acc=0.74219 test_acc_avg=0.82254 test_acc_top5_avg=0.99107 time=239.59it/s +epoch=34 global_step=13685 loss=0.29411 test_loss_avg=0.80671 acc=0.89844 test_acc_avg=0.79126 test_acc_top5_avg=0.97656 time=242.63it/s +epoch=34 global_step=13685 loss=0.27829 test_loss_avg=0.73145 acc=0.81250 test_acc_avg=0.80568 test_acc_top5_avg=0.97983 time=748.72it/s +curr_acc 0.8057 +BEST_ACC 0.7896 +curr_acc_top5 0.9798 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=3.99666 loss_avg=3.38404 acc=0.60938 acc_top1_avg=0.67552 acc_top5_avg=0.91302 lr=0.01000 gn=9.54140 time=62.42it/s +epoch=35 global_step=13750 loss=3.38526 loss_avg=3.38036 acc=0.67188 acc_top1_avg=0.67644 acc_top5_avg=0.91178 lr=0.01000 gn=9.33605 time=46.32it/s +epoch=35 global_step=13800 loss=4.13154 loss_avg=3.38388 acc=0.58594 acc_top1_avg=0.67806 acc_top5_avg=0.91264 lr=0.01000 gn=10.96178 time=58.88it/s +epoch=35 global_step=13850 loss=2.88616 loss_avg=3.39466 acc=0.71875 acc_top1_avg=0.67590 acc_top5_avg=0.91378 lr=0.01000 gn=9.41481 time=60.59it/s +epoch=35 global_step=13900 loss=3.09318 loss_avg=3.41169 acc=0.71094 acc_top1_avg=0.67333 acc_top5_avg=0.91421 lr=0.01000 gn=9.97933 time=62.61it/s +epoch=35 global_step=13950 loss=3.43340 loss_avg=3.41249 acc=0.67188 acc_top1_avg=0.67353 acc_top5_avg=0.91394 lr=0.01000 gn=9.53340 time=50.05it/s +epoch=35 global_step=14000 loss=2.98349 loss_avg=3.41582 acc=0.72656 acc_top1_avg=0.67292 acc_top5_avg=0.91406 lr=0.01000 gn=9.62045 time=48.96it/s +epoch=35 global_step=14050 loss=3.44319 loss_avg=3.40768 acc=0.68750 acc_top1_avg=0.67363 acc_top5_avg=0.91428 lr=0.01000 gn=9.98121 time=55.33it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.51717 test_loss_avg=1.13331 acc=0.85938 test_acc_avg=0.71875 test_acc_top5_avg=0.98304 time=248.27it/s +epoch=35 global_step=14076 loss=1.49176 test_loss_avg=1.23921 acc=0.50000 test_acc_avg=0.70550 test_acc_top5_avg=0.96559 time=872.54it/s +curr_acc 0.7055 +BEST_ACC 0.8057 +curr_acc_top5 0.9656 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=3.38681 loss_avg=3.35714 acc=0.68750 acc_top1_avg=0.67904 acc_top5_avg=0.92155 lr=0.01000 gn=10.50448 time=56.08it/s +epoch=36 global_step=14150 loss=3.70355 loss_avg=3.31725 acc=0.64844 acc_top1_avg=0.68285 acc_top5_avg=0.91997 lr=0.01000 gn=9.24354 time=56.14it/s +epoch=36 global_step=14200 loss=2.96086 loss_avg=3.32805 acc=0.71094 acc_top1_avg=0.68177 acc_top5_avg=0.91854 lr=0.01000 gn=8.47078 time=56.63it/s +epoch=36 global_step=14250 loss=2.92389 loss_avg=3.36659 acc=0.74219 acc_top1_avg=0.67758 acc_top5_avg=0.91774 lr=0.01000 gn=10.45864 time=53.30it/s +epoch=36 global_step=14300 loss=3.76542 loss_avg=3.38180 acc=0.61719 acc_top1_avg=0.67637 acc_top5_avg=0.91640 lr=0.01000 gn=10.15370 time=59.92it/s +epoch=36 global_step=14350 loss=3.37699 loss_avg=3.39166 acc=0.69531 acc_top1_avg=0.67524 acc_top5_avg=0.91597 lr=0.01000 gn=8.67008 time=61.53it/s +epoch=36 global_step=14400 loss=3.96270 loss_avg=3.41180 acc=0.61719 acc_top1_avg=0.67286 acc_top5_avg=0.91435 lr=0.01000 gn=9.05328 time=55.97it/s +epoch=36 global_step=14450 loss=3.31631 loss_avg=3.41441 acc=0.67969 acc_top1_avg=0.67277 acc_top5_avg=0.91433 lr=0.01000 gn=9.56760 time=56.24it/s +====================Eval==================== +epoch=36 global_step=14467 loss=3.32725 test_loss_avg=2.81514 acc=0.30469 test_acc_avg=0.38281 test_acc_top5_avg=0.88021 time=239.89it/s +epoch=36 global_step=14467 loss=1.77758 test_loss_avg=1.30827 acc=0.64062 test_acc_avg=0.70299 test_acc_top5_avg=0.97559 time=245.60it/s +epoch=36 global_step=14467 loss=0.86998 test_loss_avg=1.28384 acc=0.81250 test_acc_avg=0.71025 test_acc_top5_avg=0.97013 time=853.19it/s +curr_acc 0.7102 +BEST_ACC 0.8057 +curr_acc_top5 0.9701 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=3.20983 loss_avg=3.37661 acc=0.67188 acc_top1_avg=0.68182 acc_top5_avg=0.91170 lr=0.01000 gn=9.08051 time=52.41it/s +epoch=37 global_step=14550 loss=3.37509 loss_avg=3.41147 acc=0.67188 acc_top1_avg=0.67658 acc_top5_avg=0.91199 lr=0.01000 gn=7.84854 time=58.79it/s +epoch=37 global_step=14600 loss=3.26169 loss_avg=3.42932 acc=0.70312 acc_top1_avg=0.67293 acc_top5_avg=0.91418 lr=0.01000 gn=9.90365 time=60.12it/s +epoch=37 global_step=14650 loss=3.70719 loss_avg=3.41591 acc=0.64062 acc_top1_avg=0.67435 acc_top5_avg=0.91372 lr=0.01000 gn=10.10061 time=63.48it/s +epoch=37 global_step=14700 loss=3.70574 loss_avg=3.41545 acc=0.66406 acc_top1_avg=0.67389 acc_top5_avg=0.91353 lr=0.01000 gn=7.93394 time=59.14it/s +epoch=37 global_step=14750 loss=3.36480 loss_avg=3.41139 acc=0.64844 acc_top1_avg=0.67411 acc_top5_avg=0.91412 lr=0.01000 gn=10.23347 time=57.56it/s +epoch=37 global_step=14800 loss=2.85042 loss_avg=3.40239 acc=0.71875 acc_top1_avg=0.67530 acc_top5_avg=0.91411 lr=0.01000 gn=9.14902 time=58.84it/s +epoch=37 global_step=14850 loss=3.15222 loss_avg=3.40758 acc=0.70312 acc_top1_avg=0.67424 acc_top5_avg=0.91329 lr=0.01000 gn=10.02171 time=56.97it/s +====================Eval==================== +epoch=37 global_step=14858 loss=1.31699 test_loss_avg=1.16290 acc=0.69531 test_acc_avg=0.71499 test_acc_top5_avg=0.98003 time=243.84it/s +epoch=37 global_step=14858 loss=0.55194 test_loss_avg=0.91861 acc=0.86719 test_acc_avg=0.77455 test_acc_top5_avg=0.98204 time=247.38it/s +epoch=37 global_step=14858 loss=0.07229 test_loss_avg=0.90113 acc=0.93750 test_acc_avg=0.77809 test_acc_top5_avg=0.98240 time=857.73it/s +curr_acc 0.7781 +BEST_ACC 0.8057 +curr_acc_top5 0.9824 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=3.49762 loss_avg=3.29522 acc=0.67969 acc_top1_avg=0.68285 acc_top5_avg=0.91481 lr=0.01000 gn=10.32139 time=61.62it/s +epoch=38 global_step=14950 loss=3.14120 loss_avg=3.31207 acc=0.71094 acc_top1_avg=0.68291 acc_top5_avg=0.91636 lr=0.01000 gn=9.55058 time=58.34it/s +epoch=38 global_step=15000 loss=3.54398 loss_avg=3.36694 acc=0.66406 acc_top1_avg=0.67815 acc_top5_avg=0.91555 lr=0.01000 gn=9.72549 time=62.37it/s +epoch=38 global_step=15050 loss=3.35994 loss_avg=3.38215 acc=0.67969 acc_top1_avg=0.67639 acc_top5_avg=0.91496 lr=0.01000 gn=11.53373 time=60.83it/s +epoch=38 global_step=15100 loss=3.22591 loss_avg=3.37707 acc=0.68750 acc_top1_avg=0.67723 acc_top5_avg=0.91568 lr=0.01000 gn=9.52057 time=53.07it/s +epoch=38 global_step=15150 loss=3.45412 loss_avg=3.39266 acc=0.68750 acc_top1_avg=0.67592 acc_top5_avg=0.91564 lr=0.01000 gn=9.73926 time=54.57it/s +epoch=38 global_step=15200 loss=3.46198 loss_avg=3.39481 acc=0.66406 acc_top1_avg=0.67551 acc_top5_avg=0.91550 lr=0.01000 gn=9.90025 time=52.89it/s +====================Eval==================== +epoch=38 global_step=15249 loss=0.31894 test_loss_avg=0.94181 acc=0.92188 test_acc_avg=0.75960 test_acc_top5_avg=0.97982 time=236.75it/s +epoch=38 global_step=15249 loss=0.36943 test_loss_avg=0.78413 acc=0.93750 test_acc_avg=0.79549 test_acc_top5_avg=0.98220 time=872.72it/s +curr_acc 0.7955 +BEST_ACC 0.8057 +curr_acc_top5 0.9822 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=3.07886 loss_avg=3.07886 acc=0.70312 acc_top1_avg=0.70312 acc_top5_avg=0.89844 lr=0.01000 gn=7.46308 time=33.77it/s +epoch=39 global_step=15300 loss=2.92646 loss_avg=3.36080 acc=0.72656 acc_top1_avg=0.67754 acc_top5_avg=0.91253 lr=0.01000 gn=9.22938 time=59.83it/s +epoch=39 global_step=15350 loss=3.45672 loss_avg=3.41726 acc=0.66406 acc_top1_avg=0.67157 acc_top5_avg=0.91515 lr=0.01000 gn=8.11540 time=57.10it/s +epoch=39 global_step=15400 loss=3.63525 loss_avg=3.40962 acc=0.65625 acc_top1_avg=0.67255 acc_top5_avg=0.91613 lr=0.01000 gn=12.03025 time=49.57it/s +epoch=39 global_step=15450 loss=4.20221 loss_avg=3.40279 acc=0.60156 acc_top1_avg=0.67370 acc_top5_avg=0.91566 lr=0.01000 gn=10.19192 time=63.40it/s +epoch=39 global_step=15500 loss=3.79026 loss_avg=3.40708 acc=0.64062 acc_top1_avg=0.67318 acc_top5_avg=0.91549 lr=0.01000 gn=8.78294 time=54.11it/s +epoch=39 global_step=15550 loss=3.52529 loss_avg=3.40924 acc=0.67188 acc_top1_avg=0.67268 acc_top5_avg=0.91518 lr=0.01000 gn=10.79178 time=59.49it/s +epoch=39 global_step=15600 loss=2.98396 loss_avg=3.40113 acc=0.71875 acc_top1_avg=0.67379 acc_top5_avg=0.91582 lr=0.01000 gn=10.64491 time=57.91it/s +====================Eval==================== +epoch=39 global_step=15640 loss=1.93869 test_loss_avg=1.46058 acc=0.57031 test_acc_avg=0.67023 test_acc_top5_avg=0.94285 time=232.75it/s +epoch=39 global_step=15640 loss=0.33691 test_loss_avg=1.26213 acc=0.89844 test_acc_avg=0.71954 test_acc_top5_avg=0.97339 time=245.48it/s +epoch=39 global_step=15640 loss=0.24416 test_loss_avg=1.15879 acc=0.93750 test_acc_avg=0.73922 test_acc_top5_avg=0.97567 time=858.43it/s +curr_acc 0.7392 +BEST_ACC 0.8057 +curr_acc_top5 0.9757 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=3.21866 loss_avg=3.22734 acc=0.71094 acc_top1_avg=0.69375 acc_top5_avg=0.92500 lr=0.00100 gn=10.06988 time=54.08it/s +epoch=40 global_step=15700 loss=3.14372 loss_avg=3.15527 acc=0.71094 acc_top1_avg=0.70156 acc_top5_avg=0.91745 lr=0.00100 gn=9.19659 time=56.49it/s +epoch=40 global_step=15750 loss=3.52911 loss_avg=3.08252 acc=0.64062 acc_top1_avg=0.70803 acc_top5_avg=0.91989 lr=0.00100 gn=9.03779 time=55.54it/s +epoch=40 global_step=15800 loss=2.52530 loss_avg=3.06575 acc=0.75781 acc_top1_avg=0.70947 acc_top5_avg=0.92021 lr=0.00100 gn=9.80477 time=58.11it/s +epoch=40 global_step=15850 loss=2.54070 loss_avg=3.05637 acc=0.76562 acc_top1_avg=0.71012 acc_top5_avg=0.92035 lr=0.00100 gn=6.72739 time=53.41it/s +epoch=40 global_step=15900 loss=2.65631 loss_avg=3.03939 acc=0.75000 acc_top1_avg=0.71187 acc_top5_avg=0.92043 lr=0.00100 gn=9.74919 time=53.21it/s +epoch=40 global_step=15950 loss=3.00679 loss_avg=3.02897 acc=0.72656 acc_top1_avg=0.71268 acc_top5_avg=0.92059 lr=0.00100 gn=9.87468 time=52.23it/s +epoch=40 global_step=16000 loss=2.61641 loss_avg=3.02245 acc=0.75781 acc_top1_avg=0.71324 acc_top5_avg=0.92138 lr=0.00100 gn=8.45518 time=59.59it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.79467 test_loss_avg=0.57623 acc=0.78125 test_acc_avg=0.83848 test_acc_top5_avg=0.99121 time=239.44it/s +epoch=40 global_step=16031 loss=0.33475 test_loss_avg=0.45090 acc=0.93750 test_acc_avg=0.87371 test_acc_top5_avg=0.99239 time=604.45it/s +curr_acc 0.8737 +BEST_ACC 0.8057 +curr_acc_top5 0.9924 +BEST_ACC_top5 0.9868 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=2.61835 loss_avg=2.89929 acc=0.76562 acc_top1_avg=0.72533 acc_top5_avg=0.92352 lr=0.00100 gn=7.32423 time=61.07it/s +epoch=41 global_step=16100 loss=2.51059 loss_avg=2.84840 acc=0.75781 acc_top1_avg=0.73041 acc_top5_avg=0.92414 lr=0.00100 gn=7.28934 time=63.69it/s +epoch=41 global_step=16150 loss=3.44315 loss_avg=2.85171 acc=0.66406 acc_top1_avg=0.72939 acc_top5_avg=0.92535 lr=0.00100 gn=8.47421 time=55.61it/s +epoch=41 global_step=16200 loss=2.70015 loss_avg=2.86491 acc=0.74219 acc_top1_avg=0.72772 acc_top5_avg=0.92474 lr=0.00100 gn=8.98733 time=62.40it/s +epoch=41 global_step=16250 loss=2.28477 loss_avg=2.85757 acc=0.79688 acc_top1_avg=0.72845 acc_top5_avg=0.92544 lr=0.00100 gn=7.89354 time=59.71it/s +epoch=41 global_step=16300 loss=2.91462 loss_avg=2.85035 acc=0.72656 acc_top1_avg=0.72906 acc_top5_avg=0.92472 lr=0.00100 gn=9.30819 time=60.74it/s +epoch=41 global_step=16350 loss=3.96282 loss_avg=2.85289 acc=0.61719 acc_top1_avg=0.72891 acc_top5_avg=0.92469 lr=0.00100 gn=11.03894 time=57.00it/s +epoch=41 global_step=16400 loss=3.22287 loss_avg=2.85718 acc=0.68750 acc_top1_avg=0.72866 acc_top5_avg=0.92562 lr=0.00100 gn=10.02186 time=57.93it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.21762 test_loss_avg=0.41317 acc=0.92969 test_acc_avg=0.87287 test_acc_top5_avg=0.99645 time=242.33it/s +epoch=41 global_step=16422 loss=0.16733 test_loss_avg=0.48105 acc=0.96094 test_acc_avg=0.86309 test_acc_top5_avg=0.99219 time=255.02it/s +epoch=41 global_step=16422 loss=0.12671 test_loss_avg=0.42076 acc=0.93750 test_acc_avg=0.87925 test_acc_top5_avg=0.99318 time=544.36it/s +curr_acc 0.8793 +BEST_ACC 0.8737 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9924 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=2.41391 loss_avg=2.79871 acc=0.78906 acc_top1_avg=0.73577 acc_top5_avg=0.92690 lr=0.00100 gn=7.72281 time=54.35it/s +epoch=42 global_step=16500 loss=2.96169 loss_avg=2.75198 acc=0.71875 acc_top1_avg=0.73908 acc_top5_avg=0.92758 lr=0.00100 gn=7.90034 time=58.30it/s +epoch=42 global_step=16550 loss=2.49653 loss_avg=2.79427 acc=0.76562 acc_top1_avg=0.73560 acc_top5_avg=0.92505 lr=0.00100 gn=8.45313 time=60.78it/s +epoch=42 global_step=16600 loss=3.07365 loss_avg=2.80125 acc=0.69531 acc_top1_avg=0.73490 acc_top5_avg=0.92578 lr=0.00100 gn=9.11207 time=52.36it/s +epoch=42 global_step=16650 loss=3.30495 loss_avg=2.79695 acc=0.67969 acc_top1_avg=0.73520 acc_top5_avg=0.92599 lr=0.00100 gn=10.35133 time=54.67it/s +epoch=42 global_step=16700 loss=3.31839 loss_avg=2.79373 acc=0.66406 acc_top1_avg=0.73502 acc_top5_avg=0.92595 lr=0.00100 gn=9.10195 time=55.94it/s +epoch=42 global_step=16750 loss=2.22123 loss_avg=2.78533 acc=0.78906 acc_top1_avg=0.73566 acc_top5_avg=0.92631 lr=0.00100 gn=7.07794 time=60.19it/s +epoch=42 global_step=16800 loss=2.65355 loss_avg=2.77268 acc=0.75781 acc_top1_avg=0.73710 acc_top5_avg=0.92723 lr=0.00100 gn=9.19876 time=55.84it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.50824 test_loss_avg=0.51257 acc=0.86719 test_acc_avg=0.85645 test_acc_top5_avg=0.99316 time=231.55it/s +epoch=42 global_step=16813 loss=0.27485 test_loss_avg=0.42215 acc=0.93750 test_acc_avg=0.88113 test_acc_top5_avg=0.99387 time=550.29it/s +curr_acc 0.8811 +BEST_ACC 0.8793 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9932 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=3.23622 loss_avg=2.73959 acc=0.69531 acc_top1_avg=0.74345 acc_top5_avg=0.92610 lr=0.00100 gn=9.13186 time=53.93it/s +epoch=43 global_step=16900 loss=2.83869 loss_avg=2.69994 acc=0.73438 acc_top1_avg=0.74569 acc_top5_avg=0.92924 lr=0.00100 gn=10.79316 time=60.97it/s +epoch=43 global_step=16950 loss=2.80595 loss_avg=2.70519 acc=0.73438 acc_top1_avg=0.74430 acc_top5_avg=0.92889 lr=0.00100 gn=6.00165 time=52.05it/s +epoch=43 global_step=17000 loss=2.40486 loss_avg=2.70501 acc=0.77344 acc_top1_avg=0.74423 acc_top5_avg=0.92827 lr=0.00100 gn=8.41333 time=53.43it/s +epoch=43 global_step=17050 loss=3.25968 loss_avg=2.71025 acc=0.69531 acc_top1_avg=0.74318 acc_top5_avg=0.92725 lr=0.00100 gn=11.96556 time=53.71it/s +epoch=43 global_step=17100 loss=2.51265 loss_avg=2.72164 acc=0.75781 acc_top1_avg=0.74219 acc_top5_avg=0.92743 lr=0.00100 gn=10.03238 time=52.88it/s +epoch=43 global_step=17150 loss=2.61650 loss_avg=2.72206 acc=0.75000 acc_top1_avg=0.74207 acc_top5_avg=0.92728 lr=0.00100 gn=9.77514 time=54.19it/s +epoch=43 global_step=17200 loss=2.58954 loss_avg=2.72382 acc=0.75781 acc_top1_avg=0.74168 acc_top5_avg=0.92785 lr=0.00100 gn=8.60972 time=58.57it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.32352 test_loss_avg=0.49517 acc=0.89062 test_acc_avg=0.84896 test_acc_top5_avg=0.99479 time=239.95it/s +epoch=43 global_step=17204 loss=0.15048 test_loss_avg=0.48210 acc=0.96094 test_acc_avg=0.86424 test_acc_top5_avg=0.99175 time=249.96it/s +epoch=43 global_step=17204 loss=0.18625 test_loss_avg=0.38920 acc=0.93750 test_acc_avg=0.88924 test_acc_top5_avg=0.99318 time=555.24it/s +curr_acc 0.8892 +BEST_ACC 0.8811 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=2.55577 loss_avg=2.68527 acc=0.75000 acc_top1_avg=0.74558 acc_top5_avg=0.93122 lr=0.00100 gn=10.11144 time=49.12it/s +epoch=44 global_step=17300 loss=2.45709 loss_avg=2.68257 acc=0.75781 acc_top1_avg=0.74438 acc_top5_avg=0.92961 lr=0.00100 gn=12.57378 time=58.74it/s +epoch=44 global_step=17350 loss=2.49448 loss_avg=2.67742 acc=0.77344 acc_top1_avg=0.74588 acc_top5_avg=0.92942 lr=0.00100 gn=8.44906 time=57.12it/s +epoch=44 global_step=17400 loss=2.53030 loss_avg=2.67642 acc=0.75781 acc_top1_avg=0.74574 acc_top5_avg=0.92881 lr=0.00100 gn=9.56148 time=59.32it/s +epoch=44 global_step=17450 loss=3.03424 loss_avg=2.68657 acc=0.71094 acc_top1_avg=0.74473 acc_top5_avg=0.92858 lr=0.00100 gn=8.37711 time=52.94it/s +epoch=44 global_step=17500 loss=3.03128 loss_avg=2.69274 acc=0.71094 acc_top1_avg=0.74440 acc_top5_avg=0.92858 lr=0.00100 gn=8.68558 time=60.20it/s +epoch=44 global_step=17550 loss=2.82814 loss_avg=2.70144 acc=0.73438 acc_top1_avg=0.74343 acc_top5_avg=0.92921 lr=0.00100 gn=8.41638 time=61.16it/s +====================Eval==================== +epoch=44 global_step=17595 loss=0.77132 test_loss_avg=0.46086 acc=0.77344 test_acc_avg=0.86784 test_acc_top5_avg=0.99089 time=228.87it/s +epoch=44 global_step=17595 loss=0.22412 test_loss_avg=0.40331 acc=0.92188 test_acc_avg=0.88587 test_acc_top5_avg=0.99335 time=238.71it/s +epoch=44 global_step=17595 loss=0.21359 test_loss_avg=0.39369 acc=0.93750 test_acc_avg=0.88815 test_acc_top5_avg=0.99367 time=559.09it/s +curr_acc 0.8882 +BEST_ACC 0.8892 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=2.49521 loss_avg=2.78537 acc=0.78125 acc_top1_avg=0.73594 acc_top5_avg=0.91875 lr=0.00100 gn=11.37111 time=59.76it/s +epoch=45 global_step=17650 loss=2.63745 loss_avg=2.67268 acc=0.74219 acc_top1_avg=0.74517 acc_top5_avg=0.93026 lr=0.00100 gn=9.70359 time=60.98it/s +epoch=45 global_step=17700 loss=2.33764 loss_avg=2.69762 acc=0.77344 acc_top1_avg=0.74315 acc_top5_avg=0.92879 lr=0.00100 gn=7.67651 time=51.12it/s +epoch=45 global_step=17750 loss=2.59833 loss_avg=2.66077 acc=0.76562 acc_top1_avg=0.74798 acc_top5_avg=0.92969 lr=0.00100 gn=11.73283 time=54.78it/s +epoch=45 global_step=17800 loss=3.19312 loss_avg=2.65657 acc=0.70312 acc_top1_avg=0.74878 acc_top5_avg=0.93007 lr=0.00100 gn=11.79508 time=53.01it/s +epoch=45 global_step=17850 loss=2.80450 loss_avg=2.67316 acc=0.71875 acc_top1_avg=0.74691 acc_top5_avg=0.92920 lr=0.00100 gn=8.00219 time=56.45it/s +epoch=45 global_step=17900 loss=2.40413 loss_avg=2.65827 acc=0.77344 acc_top1_avg=0.74828 acc_top5_avg=0.92966 lr=0.00100 gn=9.62221 time=60.23it/s +epoch=45 global_step=17950 loss=2.36217 loss_avg=2.65425 acc=0.78125 acc_top1_avg=0.74853 acc_top5_avg=0.93004 lr=0.00100 gn=10.26073 time=50.76it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.43871 test_loss_avg=0.49766 acc=0.88281 test_acc_avg=0.85920 test_acc_top5_avg=0.99097 time=239.58it/s +epoch=45 global_step=17986 loss=0.19678 test_loss_avg=0.38269 acc=0.93750 test_acc_avg=0.88954 test_acc_top5_avg=0.99318 time=544.93it/s +curr_acc 0.8895 +BEST_ACC 0.8892 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=2.06916 loss_avg=2.52185 acc=0.80469 acc_top1_avg=0.76116 acc_top5_avg=0.93583 lr=0.00100 gn=7.15069 time=55.72it/s +epoch=46 global_step=18050 loss=1.91286 loss_avg=2.59064 acc=0.82812 acc_top1_avg=0.75378 acc_top5_avg=0.93384 lr=0.00100 gn=9.44675 time=58.88it/s +epoch=46 global_step=18100 loss=3.26062 loss_avg=2.60498 acc=0.69531 acc_top1_avg=0.75281 acc_top5_avg=0.93209 lr=0.00100 gn=9.62516 time=61.50it/s +epoch=46 global_step=18150 loss=2.63806 loss_avg=2.61680 acc=0.75781 acc_top1_avg=0.75233 acc_top5_avg=0.93345 lr=0.00100 gn=12.56008 time=57.94it/s +epoch=46 global_step=18200 loss=2.95029 loss_avg=2.60355 acc=0.72656 acc_top1_avg=0.75442 acc_top5_avg=0.93286 lr=0.00100 gn=10.89727 time=53.34it/s +epoch=46 global_step=18250 loss=2.49860 loss_avg=2.61000 acc=0.75000 acc_top1_avg=0.75331 acc_top5_avg=0.93265 lr=0.00100 gn=9.03213 time=62.72it/s +epoch=46 global_step=18300 loss=2.45676 loss_avg=2.62632 acc=0.77344 acc_top1_avg=0.75174 acc_top5_avg=0.93148 lr=0.00100 gn=10.18546 time=60.86it/s +epoch=46 global_step=18350 loss=3.01443 loss_avg=2.62965 acc=0.69531 acc_top1_avg=0.75135 acc_top5_avg=0.93104 lr=0.00100 gn=10.33330 time=55.95it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.60506 test_loss_avg=0.30335 acc=0.85156 test_acc_avg=0.91455 test_acc_top5_avg=0.99756 time=233.80it/s +epoch=46 global_step=18377 loss=0.13577 test_loss_avg=0.41773 acc=0.94531 test_acc_avg=0.88565 test_acc_top5_avg=0.99290 time=224.65it/s +epoch=46 global_step=18377 loss=0.19961 test_loss_avg=0.37803 acc=0.93750 test_acc_avg=0.89527 test_acc_top5_avg=0.99377 time=883.57it/s +curr_acc 0.8953 +BEST_ACC 0.8895 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=2.90213 loss_avg=2.58784 acc=0.73438 acc_top1_avg=0.75747 acc_top5_avg=0.93716 lr=0.00100 gn=11.09713 time=51.58it/s +epoch=47 global_step=18450 loss=2.84037 loss_avg=2.56495 acc=0.74219 acc_top1_avg=0.75856 acc_top5_avg=0.93086 lr=0.00100 gn=11.17039 time=51.68it/s +epoch=47 global_step=18500 loss=2.83334 loss_avg=2.57839 acc=0.73438 acc_top1_avg=0.75629 acc_top5_avg=0.93108 lr=0.00100 gn=12.76363 time=59.54it/s +epoch=47 global_step=18550 loss=2.89897 loss_avg=2.58970 acc=0.73438 acc_top1_avg=0.75510 acc_top5_avg=0.93118 lr=0.00100 gn=11.24969 time=56.08it/s +epoch=47 global_step=18600 loss=2.94033 loss_avg=2.59476 acc=0.71875 acc_top1_avg=0.75511 acc_top5_avg=0.93158 lr=0.00100 gn=10.57567 time=48.87it/s +epoch=47 global_step=18650 loss=2.92319 loss_avg=2.58943 acc=0.74219 acc_top1_avg=0.75569 acc_top5_avg=0.93072 lr=0.00100 gn=14.80875 time=54.22it/s +epoch=47 global_step=18700 loss=3.01020 loss_avg=2.59196 acc=0.70312 acc_top1_avg=0.75513 acc_top5_avg=0.93049 lr=0.00100 gn=11.80136 time=60.33it/s +epoch=47 global_step=18750 loss=2.70398 loss_avg=2.59648 acc=0.74219 acc_top1_avg=0.75478 acc_top5_avg=0.93061 lr=0.00100 gn=13.84101 time=57.67it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.17720 test_loss_avg=0.44108 acc=0.93750 test_acc_avg=0.87373 test_acc_top5_avg=0.99324 time=248.33it/s +epoch=47 global_step=18768 loss=0.37595 test_loss_avg=0.38859 acc=0.93750 test_acc_avg=0.89023 test_acc_top5_avg=0.99318 time=851.98it/s +curr_acc 0.8902 +BEST_ACC 0.8953 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=2.83434 loss_avg=2.62397 acc=0.73438 acc_top1_avg=0.75146 acc_top5_avg=0.92920 lr=0.00100 gn=11.27000 time=61.13it/s +epoch=48 global_step=18850 loss=2.02658 loss_avg=2.57273 acc=0.82031 acc_top1_avg=0.75772 acc_top5_avg=0.93007 lr=0.00100 gn=11.57290 time=59.16it/s +epoch=48 global_step=18900 loss=2.55793 loss_avg=2.57993 acc=0.75000 acc_top1_avg=0.75692 acc_top5_avg=0.92785 lr=0.00100 gn=8.20074 time=60.90it/s +epoch=48 global_step=18950 loss=3.01430 loss_avg=2.56892 acc=0.71094 acc_top1_avg=0.75764 acc_top5_avg=0.92836 lr=0.00100 gn=10.51068 time=59.26it/s +epoch=48 global_step=19000 loss=2.71599 loss_avg=2.56308 acc=0.74219 acc_top1_avg=0.75852 acc_top5_avg=0.92871 lr=0.00100 gn=12.94163 time=52.89it/s +epoch=48 global_step=19050 loss=3.12702 loss_avg=2.56266 acc=0.70312 acc_top1_avg=0.75834 acc_top5_avg=0.93013 lr=0.00100 gn=10.77640 time=55.96it/s +epoch=48 global_step=19100 loss=1.91519 loss_avg=2.57907 acc=0.85156 acc_top1_avg=0.75678 acc_top5_avg=0.93021 lr=0.00100 gn=12.57677 time=60.77it/s +epoch=48 global_step=19150 loss=2.32441 loss_avg=2.57562 acc=0.77344 acc_top1_avg=0.75724 acc_top5_avg=0.93067 lr=0.00100 gn=9.48937 time=60.02it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.23614 test_loss_avg=0.48667 acc=0.89844 test_acc_avg=0.84668 test_acc_top5_avg=0.98730 time=241.23it/s +epoch=48 global_step=19159 loss=0.17601 test_loss_avg=0.42947 acc=0.95312 test_acc_avg=0.87850 test_acc_top5_avg=0.99327 time=233.15it/s +epoch=48 global_step=19159 loss=0.39000 test_loss_avg=0.38525 acc=0.93750 test_acc_avg=0.89023 test_acc_top5_avg=0.99426 time=543.23it/s +curr_acc 0.8902 +BEST_ACC 0.8953 +curr_acc_top5 0.9943 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=2.66082 loss_avg=2.53443 acc=0.74219 acc_top1_avg=0.76105 acc_top5_avg=0.93140 lr=0.00100 gn=13.36380 time=52.99it/s +epoch=49 global_step=19250 loss=1.76463 loss_avg=2.52313 acc=0.85156 acc_top1_avg=0.76408 acc_top5_avg=0.93312 lr=0.00100 gn=12.99742 time=61.66it/s +epoch=49 global_step=19300 loss=2.52814 loss_avg=2.53642 acc=0.76562 acc_top1_avg=0.76191 acc_top5_avg=0.93135 lr=0.00100 gn=11.69604 time=54.05it/s +epoch=49 global_step=19350 loss=2.28132 loss_avg=2.54850 acc=0.78906 acc_top1_avg=0.76096 acc_top5_avg=0.93149 lr=0.00100 gn=11.29234 time=56.17it/s +epoch=49 global_step=19400 loss=3.21298 loss_avg=2.55966 acc=0.67969 acc_top1_avg=0.75943 acc_top5_avg=0.93131 lr=0.00100 gn=8.48669 time=59.58it/s +epoch=49 global_step=19450 loss=1.96786 loss_avg=2.55285 acc=0.82031 acc_top1_avg=0.76012 acc_top5_avg=0.93208 lr=0.00100 gn=9.98550 time=54.33it/s +epoch=49 global_step=19500 loss=2.82290 loss_avg=2.55483 acc=0.71875 acc_top1_avg=0.75944 acc_top5_avg=0.93196 lr=0.00100 gn=12.26673 time=60.70it/s +epoch=49 global_step=19550 loss=2.54993 loss_avg=2.55528 acc=0.76250 acc_top1_avg=0.75944 acc_top5_avg=0.93162 lr=0.00100 gn=16.83487 time=71.92it/s +====================Eval==================== +epoch=49 global_step=19550 loss=0.51334 test_loss_avg=0.45646 acc=0.85156 test_acc_avg=0.87662 test_acc_top5_avg=0.99111 time=76.57it/s +epoch=49 global_step=19550 loss=0.24854 test_loss_avg=0.37509 acc=0.93750 test_acc_avg=0.89765 test_acc_top5_avg=0.99466 time=550.72it/s +epoch=49 global_step=19550 loss=0.24854 test_loss_avg=0.37509 acc=0.93750 test_acc_avg=0.89765 test_acc_top5_avg=0.99466 time=550.72it/s +curr_acc 0.8976 +BEST_ACC 0.8953 +curr_acc_top5 0.9947 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=2.16499 loss_avg=2.47472 acc=0.79688 acc_top1_avg=0.76906 acc_top5_avg=0.93484 lr=0.00100 gn=10.03039 time=53.31it/s +epoch=50 global_step=19650 loss=2.17695 loss_avg=2.50244 acc=0.79688 acc_top1_avg=0.76578 acc_top5_avg=0.93258 lr=0.00100 gn=11.66552 time=60.18it/s +epoch=50 global_step=19700 loss=2.38215 loss_avg=2.51382 acc=0.79688 acc_top1_avg=0.76469 acc_top5_avg=0.93193 lr=0.00100 gn=14.39836 time=57.56it/s +epoch=50 global_step=19750 loss=2.40568 loss_avg=2.51484 acc=0.76562 acc_top1_avg=0.76355 acc_top5_avg=0.93332 lr=0.00100 gn=9.51894 time=60.73it/s +epoch=50 global_step=19800 loss=2.60025 loss_avg=2.51317 acc=0.76562 acc_top1_avg=0.76366 acc_top5_avg=0.93356 lr=0.00100 gn=12.93401 time=54.36it/s +epoch=50 global_step=19850 loss=2.36021 loss_avg=2.52541 acc=0.77344 acc_top1_avg=0.76234 acc_top5_avg=0.93299 lr=0.00100 gn=11.08055 time=59.76it/s +epoch=50 global_step=19900 loss=2.64997 loss_avg=2.52752 acc=0.75781 acc_top1_avg=0.76210 acc_top5_avg=0.93339 lr=0.00100 gn=14.59172 time=55.41it/s +====================Eval==================== +epoch=50 global_step=19941 loss=0.13551 test_loss_avg=0.45803 acc=0.96094 test_acc_avg=0.87141 test_acc_top5_avg=0.99234 time=224.25it/s +epoch=50 global_step=19941 loss=0.26085 test_loss_avg=0.36444 acc=0.93750 test_acc_avg=0.89527 test_acc_top5_avg=0.99397 time=878.94it/s +curr_acc 0.8953 +BEST_ACC 0.8976 +curr_acc_top5 0.9940 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=2.72421 loss_avg=2.42812 acc=0.74219 acc_top1_avg=0.77170 acc_top5_avg=0.93576 lr=0.00100 gn=13.50916 time=56.99it/s +epoch=51 global_step=20000 loss=2.97479 loss_avg=2.48331 acc=0.71094 acc_top1_avg=0.76602 acc_top5_avg=0.93114 lr=0.00100 gn=12.32633 time=54.50it/s +epoch=51 global_step=20050 loss=2.59994 loss_avg=2.52065 acc=0.75000 acc_top1_avg=0.76312 acc_top5_avg=0.93055 lr=0.00100 gn=12.84667 time=56.32it/s +epoch=51 global_step=20100 loss=2.32615 loss_avg=2.48687 acc=0.78125 acc_top1_avg=0.76685 acc_top5_avg=0.93259 lr=0.00100 gn=11.07521 time=39.75it/s +epoch=51 global_step=20150 loss=2.46460 loss_avg=2.48704 acc=0.75781 acc_top1_avg=0.76712 acc_top5_avg=0.93268 lr=0.00100 gn=12.34535 time=62.10it/s +epoch=51 global_step=20200 loss=2.31115 loss_avg=2.50376 acc=0.77344 acc_top1_avg=0.76538 acc_top5_avg=0.93195 lr=0.00100 gn=8.70036 time=54.36it/s +epoch=51 global_step=20250 loss=2.07681 loss_avg=2.50502 acc=0.79688 acc_top1_avg=0.76492 acc_top5_avg=0.93302 lr=0.00100 gn=13.60535 time=59.73it/s +epoch=51 global_step=20300 loss=2.32334 loss_avg=2.50018 acc=0.78125 acc_top1_avg=0.76519 acc_top5_avg=0.93293 lr=0.00100 gn=10.64699 time=51.88it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.57808 test_loss_avg=0.38219 acc=0.83594 test_acc_avg=0.88504 test_acc_top5_avg=0.99479 time=243.74it/s +epoch=51 global_step=20332 loss=0.25487 test_loss_avg=0.39681 acc=0.92969 test_acc_avg=0.88710 test_acc_top5_avg=0.99263 time=228.88it/s +epoch=51 global_step=20332 loss=0.29389 test_loss_avg=0.38304 acc=0.93750 test_acc_avg=0.89142 test_acc_top5_avg=0.99298 time=807.68it/s +curr_acc 0.8914 +BEST_ACC 0.8976 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=2.53629 loss_avg=2.45766 acc=0.76562 acc_top1_avg=0.77040 acc_top5_avg=0.92578 lr=0.00100 gn=12.05405 time=48.82it/s +epoch=52 global_step=20400 loss=2.95628 loss_avg=2.43631 acc=0.71875 acc_top1_avg=0.77148 acc_top5_avg=0.93279 lr=0.00100 gn=12.24555 time=53.75it/s +epoch=52 global_step=20450 loss=2.27332 loss_avg=2.44458 acc=0.78906 acc_top1_avg=0.76973 acc_top5_avg=0.93498 lr=0.00100 gn=8.82679 time=57.05it/s +epoch=52 global_step=20500 loss=2.33376 loss_avg=2.46695 acc=0.77344 acc_top1_avg=0.76688 acc_top5_avg=0.93462 lr=0.00100 gn=11.94707 time=61.72it/s +epoch=52 global_step=20550 loss=2.51714 loss_avg=2.48077 acc=0.76562 acc_top1_avg=0.76555 acc_top5_avg=0.93417 lr=0.00100 gn=11.05256 time=52.12it/s +epoch=52 global_step=20600 loss=2.61936 loss_avg=2.49043 acc=0.76562 acc_top1_avg=0.76472 acc_top5_avg=0.93362 lr=0.00100 gn=10.37387 time=55.74it/s +epoch=52 global_step=20650 loss=1.98815 loss_avg=2.48273 acc=0.81250 acc_top1_avg=0.76597 acc_top5_avg=0.93413 lr=0.00100 gn=10.01793 time=57.14it/s +epoch=52 global_step=20700 loss=2.41982 loss_avg=2.49245 acc=0.77344 acc_top1_avg=0.76516 acc_top5_avg=0.93444 lr=0.00100 gn=13.12885 time=58.42it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.63962 test_loss_avg=0.46074 acc=0.80469 test_acc_avg=0.86868 test_acc_top5_avg=0.99033 time=248.51it/s +epoch=52 global_step=20723 loss=0.13655 test_loss_avg=0.37392 acc=0.93750 test_acc_avg=0.89260 test_acc_top5_avg=0.99278 time=728.43it/s +curr_acc 0.8926 +BEST_ACC 0.8976 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=2.34713 loss_avg=2.51586 acc=0.79688 acc_top1_avg=0.76302 acc_top5_avg=0.93837 lr=0.00100 gn=14.88880 time=58.70it/s +epoch=53 global_step=20800 loss=2.63118 loss_avg=2.46391 acc=0.75000 acc_top1_avg=0.76826 acc_top5_avg=0.93831 lr=0.00100 gn=12.40528 time=61.16it/s +epoch=53 global_step=20850 loss=2.94523 loss_avg=2.47433 acc=0.71875 acc_top1_avg=0.76710 acc_top5_avg=0.93670 lr=0.00100 gn=13.28477 time=50.85it/s +epoch=53 global_step=20900 loss=2.35050 loss_avg=2.46272 acc=0.78125 acc_top1_avg=0.76889 acc_top5_avg=0.93547 lr=0.00100 gn=13.97008 time=47.94it/s +epoch=53 global_step=20950 loss=2.41300 loss_avg=2.47353 acc=0.76562 acc_top1_avg=0.76741 acc_top5_avg=0.93323 lr=0.00100 gn=13.29623 time=53.06it/s +epoch=53 global_step=21000 loss=1.85376 loss_avg=2.47431 acc=0.81250 acc_top1_avg=0.76766 acc_top5_avg=0.93237 lr=0.00100 gn=13.99950 time=53.11it/s +epoch=53 global_step=21050 loss=2.23414 loss_avg=2.47979 acc=0.80469 acc_top1_avg=0.76708 acc_top5_avg=0.93203 lr=0.00100 gn=13.09398 time=55.32it/s +epoch=53 global_step=21100 loss=2.53335 loss_avg=2.47910 acc=0.77344 acc_top1_avg=0.76710 acc_top5_avg=0.93215 lr=0.00100 gn=15.79647 time=58.77it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.06298 test_loss_avg=0.37626 acc=0.96875 test_acc_avg=0.88522 test_acc_top5_avg=0.99159 time=242.89it/s +epoch=53 global_step=21114 loss=0.17240 test_loss_avg=0.45610 acc=0.93750 test_acc_avg=0.87351 test_acc_top5_avg=0.99318 time=235.45it/s +epoch=53 global_step=21114 loss=0.14277 test_loss_avg=0.40042 acc=0.93750 test_acc_avg=0.88776 test_acc_top5_avg=0.99387 time=582.14it/s +curr_acc 0.8878 +BEST_ACC 0.8976 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=2.73061 loss_avg=2.40748 acc=0.72656 acc_top1_avg=0.77474 acc_top5_avg=0.93186 lr=0.00100 gn=10.16174 time=48.41it/s +epoch=54 global_step=21200 loss=2.00709 loss_avg=2.45484 acc=0.82031 acc_top1_avg=0.77062 acc_top5_avg=0.93232 lr=0.00100 gn=12.94914 time=58.41it/s +epoch=54 global_step=21250 loss=2.20129 loss_avg=2.46400 acc=0.78906 acc_top1_avg=0.76855 acc_top5_avg=0.93176 lr=0.00100 gn=9.66175 time=49.57it/s +epoch=54 global_step=21300 loss=2.13863 loss_avg=2.46021 acc=0.78906 acc_top1_avg=0.76857 acc_top5_avg=0.93217 lr=0.00100 gn=13.70249 time=56.34it/s +epoch=54 global_step=21350 loss=2.68869 loss_avg=2.44014 acc=0.75000 acc_top1_avg=0.77066 acc_top5_avg=0.93343 lr=0.00100 gn=13.57876 time=55.36it/s +epoch=54 global_step=21400 loss=2.54084 loss_avg=2.46028 acc=0.76562 acc_top1_avg=0.76898 acc_top5_avg=0.93275 lr=0.00100 gn=11.62250 time=60.43it/s +epoch=54 global_step=21450 loss=2.61016 loss_avg=2.46494 acc=0.74219 acc_top1_avg=0.76837 acc_top5_avg=0.93264 lr=0.00100 gn=15.23705 time=55.51it/s +epoch=54 global_step=21500 loss=2.46889 loss_avg=2.46489 acc=0.77344 acc_top1_avg=0.76852 acc_top5_avg=0.93299 lr=0.00100 gn=11.66717 time=57.92it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.29415 test_loss_avg=0.47948 acc=0.91406 test_acc_avg=0.86213 test_acc_top5_avg=0.99012 time=237.07it/s +epoch=54 global_step=21505 loss=0.21661 test_loss_avg=0.37862 acc=0.93750 test_acc_avg=0.89339 test_acc_top5_avg=0.99357 time=868.75it/s +curr_acc 0.8934 +BEST_ACC 0.8976 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=2.67558 loss_avg=2.35184 acc=0.75000 acc_top1_avg=0.78125 acc_top5_avg=0.93559 lr=0.00100 gn=14.20714 time=53.35it/s +epoch=55 global_step=21600 loss=1.88585 loss_avg=2.38602 acc=0.84375 acc_top1_avg=0.77664 acc_top5_avg=0.93372 lr=0.00100 gn=17.58836 time=45.38it/s +epoch=55 global_step=21650 loss=2.82789 loss_avg=2.39813 acc=0.71875 acc_top1_avg=0.77522 acc_top5_avg=0.93421 lr=0.00100 gn=10.78920 time=55.75it/s +epoch=55 global_step=21700 loss=2.47663 loss_avg=2.40913 acc=0.76562 acc_top1_avg=0.77372 acc_top5_avg=0.93381 lr=0.00100 gn=13.62872 time=52.02it/s +epoch=55 global_step=21750 loss=2.21909 loss_avg=2.41419 acc=0.79688 acc_top1_avg=0.77372 acc_top5_avg=0.93291 lr=0.00100 gn=16.09180 time=59.51it/s +epoch=55 global_step=21800 loss=2.23367 loss_avg=2.41579 acc=0.78906 acc_top1_avg=0.77405 acc_top5_avg=0.93355 lr=0.00100 gn=13.93539 time=57.32it/s +epoch=55 global_step=21850 loss=2.42439 loss_avg=2.42022 acc=0.78125 acc_top1_avg=0.77375 acc_top5_avg=0.93440 lr=0.00100 gn=17.54867 time=53.65it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.25052 test_loss_avg=0.40493 acc=0.91406 test_acc_avg=0.87969 test_acc_top5_avg=0.99531 time=246.81it/s +epoch=55 global_step=21896 loss=0.39112 test_loss_avg=0.46182 acc=0.89844 test_acc_avg=0.87173 test_acc_top5_avg=0.98935 time=236.69it/s +epoch=55 global_step=21896 loss=0.11523 test_loss_avg=0.37903 acc=0.93750 test_acc_avg=0.89231 test_acc_top5_avg=0.99150 time=856.50it/s +curr_acc 0.8923 +BEST_ACC 0.8976 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=2.06832 loss_avg=2.18423 acc=0.82031 acc_top1_avg=0.80273 acc_top5_avg=0.94727 lr=0.00100 gn=16.35531 time=53.98it/s +epoch=56 global_step=21950 loss=2.91113 loss_avg=2.39489 acc=0.71094 acc_top1_avg=0.77575 acc_top5_avg=0.93533 lr=0.00100 gn=18.13442 time=54.32it/s +epoch=56 global_step=22000 loss=2.62803 loss_avg=2.41479 acc=0.75000 acc_top1_avg=0.77366 acc_top5_avg=0.93502 lr=0.00100 gn=12.05213 time=60.75it/s +epoch=56 global_step=22050 loss=2.70600 loss_avg=2.43072 acc=0.74219 acc_top1_avg=0.77232 acc_top5_avg=0.93415 lr=0.00100 gn=16.60218 time=46.07it/s +epoch=56 global_step=22100 loss=2.78079 loss_avg=2.43855 acc=0.74219 acc_top1_avg=0.77171 acc_top5_avg=0.93371 lr=0.00100 gn=13.11219 time=57.02it/s +epoch=56 global_step=22150 loss=2.50185 loss_avg=2.44135 acc=0.77344 acc_top1_avg=0.77153 acc_top5_avg=0.93421 lr=0.00100 gn=14.65131 time=56.02it/s +epoch=56 global_step=22200 loss=2.99212 loss_avg=2.43254 acc=0.71094 acc_top1_avg=0.77254 acc_top5_avg=0.93460 lr=0.00100 gn=14.45988 time=60.76it/s +epoch=56 global_step=22250 loss=2.37728 loss_avg=2.43313 acc=0.77344 acc_top1_avg=0.77269 acc_top5_avg=0.93445 lr=0.00100 gn=9.85928 time=59.27it/s +====================Eval==================== +epoch=56 global_step=22287 loss=0.58002 test_loss_avg=0.44406 acc=0.83594 test_acc_avg=0.87560 test_acc_top5_avg=0.99159 time=229.59it/s +epoch=56 global_step=22287 loss=0.18157 test_loss_avg=0.37209 acc=0.92969 test_acc_avg=0.89319 test_acc_top5_avg=0.99394 time=236.49it/s +epoch=56 global_step=22287 loss=0.24223 test_loss_avg=0.36747 acc=0.93750 test_acc_avg=0.89409 test_acc_top5_avg=0.99417 time=547.70it/s +curr_acc 0.8941 +BEST_ACC 0.8976 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=2.34900 loss_avg=2.34937 acc=0.77344 acc_top1_avg=0.77945 acc_top5_avg=0.93630 lr=0.00100 gn=14.12326 time=59.08it/s +epoch=57 global_step=22350 loss=2.67206 loss_avg=2.32697 acc=0.75000 acc_top1_avg=0.78348 acc_top5_avg=0.93614 lr=0.00100 gn=17.47614 time=56.58it/s +epoch=57 global_step=22400 loss=2.18054 loss_avg=2.37724 acc=0.78906 acc_top1_avg=0.77862 acc_top5_avg=0.93639 lr=0.00100 gn=9.93425 time=41.63it/s +epoch=57 global_step=22450 loss=3.17503 loss_avg=2.40367 acc=0.68750 acc_top1_avg=0.77488 acc_top5_avg=0.93563 lr=0.00100 gn=15.63353 time=59.65it/s +epoch=57 global_step=22500 loss=2.76244 loss_avg=2.40022 acc=0.74219 acc_top1_avg=0.77545 acc_top5_avg=0.93530 lr=0.00100 gn=13.21930 time=56.13it/s +epoch=57 global_step=22550 loss=2.81274 loss_avg=2.39535 acc=0.73438 acc_top1_avg=0.77614 acc_top5_avg=0.93590 lr=0.00100 gn=15.27127 time=58.67it/s +epoch=57 global_step=22600 loss=2.78521 loss_avg=2.41390 acc=0.74219 acc_top1_avg=0.77434 acc_top5_avg=0.93498 lr=0.00100 gn=16.12866 time=59.66it/s +epoch=57 global_step=22650 loss=2.14075 loss_avg=2.41070 acc=0.78906 acc_top1_avg=0.77462 acc_top5_avg=0.93533 lr=0.00100 gn=10.81697 time=45.41it/s +====================Eval==================== +epoch=57 global_step=22678 loss=0.76426 test_loss_avg=0.51886 acc=0.74219 test_acc_avg=0.85273 test_acc_top5_avg=0.99186 time=245.84it/s +epoch=57 global_step=22678 loss=0.13844 test_loss_avg=0.38286 acc=0.93750 test_acc_avg=0.89043 test_acc_top5_avg=0.99436 time=869.29it/s +curr_acc 0.8904 +BEST_ACC 0.8976 +curr_acc_top5 0.9944 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=2.14683 loss_avg=2.28844 acc=0.80469 acc_top1_avg=0.78977 acc_top5_avg=0.94141 lr=0.00100 gn=18.61018 time=51.97it/s +epoch=58 global_step=22750 loss=1.77309 loss_avg=2.31917 acc=0.84375 acc_top1_avg=0.78483 acc_top5_avg=0.93772 lr=0.00100 gn=18.33169 time=60.29it/s +epoch=58 global_step=22800 loss=2.81414 loss_avg=2.35366 acc=0.74219 acc_top1_avg=0.78112 acc_top5_avg=0.93584 lr=0.00100 gn=18.38693 time=59.74it/s +epoch=58 global_step=22850 loss=2.57764 loss_avg=2.36904 acc=0.75000 acc_top1_avg=0.77930 acc_top5_avg=0.93477 lr=0.00100 gn=11.98594 time=50.91it/s +epoch=58 global_step=22900 loss=2.33895 loss_avg=2.36386 acc=0.78906 acc_top1_avg=0.78005 acc_top5_avg=0.93563 lr=0.00100 gn=15.93298 time=55.36it/s +epoch=58 global_step=22950 loss=2.35907 loss_avg=2.38255 acc=0.78125 acc_top1_avg=0.77826 acc_top5_avg=0.93448 lr=0.00100 gn=15.02047 time=46.90it/s +epoch=58 global_step=23000 loss=2.38359 loss_avg=2.39348 acc=0.78906 acc_top1_avg=0.77725 acc_top5_avg=0.93452 lr=0.00100 gn=16.93476 time=55.95it/s +epoch=58 global_step=23050 loss=2.44692 loss_avg=2.39615 acc=0.78125 acc_top1_avg=0.77678 acc_top5_avg=0.93481 lr=0.00100 gn=16.34493 time=59.13it/s +====================Eval==================== +epoch=58 global_step=23069 loss=0.36437 test_loss_avg=0.33826 acc=0.89844 test_acc_avg=0.89887 test_acc_top5_avg=0.99219 time=232.64it/s +epoch=58 global_step=23069 loss=0.21363 test_loss_avg=0.38731 acc=0.94531 test_acc_avg=0.88925 test_acc_top5_avg=0.99334 time=212.91it/s +epoch=58 global_step=23069 loss=0.38342 test_loss_avg=0.38222 acc=0.93750 test_acc_avg=0.89102 test_acc_top5_avg=0.99387 time=870.91it/s +curr_acc 0.8910 +BEST_ACC 0.8976 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=3.06949 loss_avg=2.43552 acc=0.71094 acc_top1_avg=0.76991 acc_top5_avg=0.92616 lr=0.00100 gn=11.86005 time=46.59it/s +epoch=59 global_step=23150 loss=1.72258 loss_avg=2.35988 acc=0.84375 acc_top1_avg=0.77816 acc_top5_avg=0.93229 lr=0.00100 gn=12.00970 time=55.90it/s +epoch=59 global_step=23200 loss=2.10173 loss_avg=2.36298 acc=0.81250 acc_top1_avg=0.77886 acc_top5_avg=0.93589 lr=0.00100 gn=13.24135 time=57.28it/s +epoch=59 global_step=23250 loss=2.68813 loss_avg=2.38488 acc=0.75000 acc_top1_avg=0.77637 acc_top5_avg=0.93474 lr=0.00100 gn=18.45471 time=58.23it/s +epoch=59 global_step=23300 loss=2.94237 loss_avg=2.38035 acc=0.71094 acc_top1_avg=0.77689 acc_top5_avg=0.93611 lr=0.00100 gn=15.50950 time=49.35it/s +epoch=59 global_step=23350 loss=2.53957 loss_avg=2.37050 acc=0.75781 acc_top1_avg=0.77844 acc_top5_avg=0.93636 lr=0.00100 gn=13.37905 time=58.70it/s +epoch=59 global_step=23400 loss=2.19186 loss_avg=2.37348 acc=0.78906 acc_top1_avg=0.77821 acc_top5_avg=0.93653 lr=0.00100 gn=14.22671 time=51.28it/s +epoch=59 global_step=23450 loss=2.59617 loss_avg=2.38136 acc=0.75000 acc_top1_avg=0.77744 acc_top5_avg=0.93578 lr=0.00100 gn=12.28649 time=61.70it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.26389 test_loss_avg=0.45214 acc=0.92188 test_acc_avg=0.86999 test_acc_top5_avg=0.99279 time=247.58it/s +epoch=59 global_step=23460 loss=0.21385 test_loss_avg=0.36553 acc=0.93750 test_acc_avg=0.89537 test_acc_top5_avg=0.99417 time=829.90it/s +curr_acc 0.8954 +BEST_ACC 0.8976 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=2.09103 loss_avg=2.33777 acc=0.81250 acc_top1_avg=0.78203 acc_top5_avg=0.93555 lr=0.00100 gn=19.57376 time=63.08it/s +epoch=60 global_step=23550 loss=2.11269 loss_avg=2.36047 acc=0.80469 acc_top1_avg=0.77917 acc_top5_avg=0.93628 lr=0.00100 gn=12.73687 time=60.56it/s +epoch=60 global_step=23600 loss=1.96982 loss_avg=2.33468 acc=0.82031 acc_top1_avg=0.78248 acc_top5_avg=0.93689 lr=0.00100 gn=12.09866 time=59.37it/s +epoch=60 global_step=23650 loss=2.58195 loss_avg=2.33075 acc=0.77344 acc_top1_avg=0.78298 acc_top5_avg=0.93627 lr=0.00100 gn=18.66555 time=60.06it/s +epoch=60 global_step=23700 loss=2.31494 loss_avg=2.33827 acc=0.79688 acc_top1_avg=0.78226 acc_top5_avg=0.93665 lr=0.00100 gn=19.57981 time=58.40it/s +epoch=60 global_step=23750 loss=2.29704 loss_avg=2.35239 acc=0.78906 acc_top1_avg=0.78052 acc_top5_avg=0.93580 lr=0.00100 gn=22.53273 time=58.10it/s +epoch=60 global_step=23800 loss=2.00683 loss_avg=2.36466 acc=0.82031 acc_top1_avg=0.77927 acc_top5_avg=0.93529 lr=0.00100 gn=19.02559 time=61.91it/s +epoch=60 global_step=23850 loss=2.96925 loss_avg=2.37630 acc=0.69531 acc_top1_avg=0.77812 acc_top5_avg=0.93512 lr=0.00100 gn=15.57895 time=53.41it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.09857 test_loss_avg=0.40134 acc=0.98438 test_acc_avg=0.87422 test_acc_top5_avg=0.99531 time=233.98it/s +epoch=60 global_step=23851 loss=0.17305 test_loss_avg=0.41269 acc=0.94531 test_acc_avg=0.88203 test_acc_top5_avg=0.99271 time=230.30it/s +epoch=60 global_step=23851 loss=0.16873 test_loss_avg=0.36597 acc=0.93750 test_acc_avg=0.89458 test_acc_top5_avg=0.99367 time=697.19it/s +curr_acc 0.8946 +BEST_ACC 0.8976 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=2.52684 loss_avg=2.35984 acc=0.76562 acc_top1_avg=0.78109 acc_top5_avg=0.93862 lr=0.00100 gn=18.77491 time=48.19it/s +epoch=61 global_step=23950 loss=1.66774 loss_avg=2.35892 acc=0.86719 acc_top1_avg=0.78070 acc_top5_avg=0.93711 lr=0.00100 gn=15.67131 time=51.33it/s +epoch=61 global_step=24000 loss=2.53407 loss_avg=2.36053 acc=0.76562 acc_top1_avg=0.78052 acc_top5_avg=0.93687 lr=0.00100 gn=16.92740 time=52.25it/s +epoch=61 global_step=24050 loss=2.12545 loss_avg=2.36651 acc=0.80469 acc_top1_avg=0.78015 acc_top5_avg=0.93703 lr=0.00100 gn=15.90769 time=58.86it/s +epoch=61 global_step=24100 loss=2.30234 loss_avg=2.35141 acc=0.79688 acc_top1_avg=0.78128 acc_top5_avg=0.93747 lr=0.00100 gn=20.14676 time=52.70it/s +epoch=61 global_step=24150 loss=2.66757 loss_avg=2.34913 acc=0.75781 acc_top1_avg=0.78156 acc_top5_avg=0.93716 lr=0.00100 gn=16.40236 time=58.99it/s +epoch=61 global_step=24200 loss=2.75984 loss_avg=2.35428 acc=0.74219 acc_top1_avg=0.78116 acc_top5_avg=0.93699 lr=0.00100 gn=18.67057 time=63.53it/s +====================Eval==================== +epoch=61 global_step=24242 loss=0.98948 test_loss_avg=0.61961 acc=0.68750 test_acc_avg=0.82863 test_acc_top5_avg=0.98715 time=216.88it/s +epoch=61 global_step=24242 loss=0.33302 test_loss_avg=0.39231 acc=0.93750 test_acc_avg=0.88944 test_acc_top5_avg=0.99308 time=703.39it/s +curr_acc 0.8894 +BEST_ACC 0.8976 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=2.70679 loss_avg=2.54443 acc=0.73438 acc_top1_avg=0.76074 acc_top5_avg=0.93359 lr=0.00100 gn=18.30934 time=58.22it/s +epoch=62 global_step=24300 loss=2.10586 loss_avg=2.42194 acc=0.80469 acc_top1_avg=0.77344 acc_top5_avg=0.93225 lr=0.00100 gn=15.66933 time=53.37it/s +epoch=62 global_step=24350 loss=2.24862 loss_avg=2.34754 acc=0.78906 acc_top1_avg=0.78096 acc_top5_avg=0.93591 lr=0.00100 gn=18.70810 time=58.48it/s +epoch=62 global_step=24400 loss=2.10621 loss_avg=2.34368 acc=0.80469 acc_top1_avg=0.78174 acc_top5_avg=0.93527 lr=0.00100 gn=15.83527 time=53.55it/s +epoch=62 global_step=24450 loss=1.87216 loss_avg=2.32855 acc=0.83594 acc_top1_avg=0.78354 acc_top5_avg=0.93585 lr=0.00100 gn=14.28134 time=57.86it/s +epoch=62 global_step=24500 loss=1.86408 loss_avg=2.34056 acc=0.84375 acc_top1_avg=0.78267 acc_top5_avg=0.93468 lr=0.00100 gn=16.19056 time=53.11it/s +epoch=62 global_step=24550 loss=2.47151 loss_avg=2.34775 acc=0.76562 acc_top1_avg=0.78186 acc_top5_avg=0.93372 lr=0.00100 gn=16.82473 time=59.60it/s +epoch=62 global_step=24600 loss=2.73585 loss_avg=2.35494 acc=0.74219 acc_top1_avg=0.78125 acc_top5_avg=0.93405 lr=0.00100 gn=22.15053 time=56.28it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.54178 test_loss_avg=0.54104 acc=0.82812 test_acc_avg=0.83594 test_acc_top5_avg=0.99219 time=217.34it/s +epoch=62 global_step=24633 loss=0.10453 test_loss_avg=0.42322 acc=0.95312 test_acc_avg=0.88296 test_acc_top5_avg=0.99264 time=243.13it/s +epoch=62 global_step=24633 loss=0.44835 test_loss_avg=0.37782 acc=0.93750 test_acc_avg=0.89300 test_acc_top5_avg=0.99377 time=884.50it/s +curr_acc 0.8930 +BEST_ACC 0.8976 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=2.50828 loss_avg=2.34618 acc=0.78125 acc_top1_avg=0.78125 acc_top5_avg=0.93474 lr=0.00100 gn=22.45998 time=61.47it/s +epoch=63 global_step=24700 loss=2.71323 loss_avg=2.34681 acc=0.74219 acc_top1_avg=0.78183 acc_top5_avg=0.93657 lr=0.00100 gn=16.66783 time=52.31it/s +epoch=63 global_step=24750 loss=2.12751 loss_avg=2.34023 acc=0.79688 acc_top1_avg=0.78279 acc_top5_avg=0.93476 lr=0.00100 gn=13.28138 time=62.20it/s +epoch=63 global_step=24800 loss=2.58734 loss_avg=2.31850 acc=0.77344 acc_top1_avg=0.78574 acc_top5_avg=0.93483 lr=0.00100 gn=16.97298 time=50.10it/s +epoch=63 global_step=24850 loss=1.82607 loss_avg=2.32841 acc=0.83594 acc_top1_avg=0.78427 acc_top5_avg=0.93473 lr=0.00100 gn=11.20928 time=55.25it/s +epoch=63 global_step=24900 loss=2.31817 loss_avg=2.33412 acc=0.78906 acc_top1_avg=0.78362 acc_top5_avg=0.93446 lr=0.00100 gn=17.01339 time=60.60it/s +epoch=63 global_step=24950 loss=1.48249 loss_avg=2.32643 acc=0.87500 acc_top1_avg=0.78445 acc_top5_avg=0.93508 lr=0.00100 gn=14.21930 time=58.74it/s +epoch=63 global_step=25000 loss=2.57122 loss_avg=2.32836 acc=0.75781 acc_top1_avg=0.78438 acc_top5_avg=0.93501 lr=0.00100 gn=15.06577 time=53.40it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.90293 test_loss_avg=0.51926 acc=0.73438 test_acc_avg=0.84918 test_acc_top5_avg=0.99049 time=242.45it/s +epoch=63 global_step=25024 loss=0.26137 test_loss_avg=0.39664 acc=0.91406 test_acc_avg=0.88560 test_acc_top5_avg=0.99251 time=247.10it/s +epoch=63 global_step=25024 loss=0.33368 test_loss_avg=0.38403 acc=0.93750 test_acc_avg=0.88944 test_acc_top5_avg=0.99288 time=868.75it/s +curr_acc 0.8894 +BEST_ACC 0.8976 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=1.87862 loss_avg=2.23954 acc=0.84375 acc_top1_avg=0.79177 acc_top5_avg=0.93780 lr=0.00100 gn=21.90140 time=47.71it/s +epoch=64 global_step=25100 loss=2.14677 loss_avg=2.31334 acc=0.80469 acc_top1_avg=0.78505 acc_top5_avg=0.93596 lr=0.00100 gn=18.01454 time=60.50it/s +epoch=64 global_step=25150 loss=2.03404 loss_avg=2.29513 acc=0.82031 acc_top1_avg=0.78664 acc_top5_avg=0.93601 lr=0.00100 gn=15.54815 time=60.16it/s +epoch=64 global_step=25200 loss=2.60637 loss_avg=2.33310 acc=0.75781 acc_top1_avg=0.78254 acc_top5_avg=0.93581 lr=0.00100 gn=17.53920 time=59.04it/s +epoch=64 global_step=25250 loss=2.39432 loss_avg=2.31921 acc=0.78125 acc_top1_avg=0.78422 acc_top5_avg=0.93750 lr=0.00100 gn=18.40198 time=50.96it/s +epoch=64 global_step=25300 loss=2.49839 loss_avg=2.33404 acc=0.76562 acc_top1_avg=0.78250 acc_top5_avg=0.93696 lr=0.00100 gn=15.40542 time=56.34it/s +epoch=64 global_step=25350 loss=2.18902 loss_avg=2.32699 acc=0.79688 acc_top1_avg=0.78298 acc_top5_avg=0.93760 lr=0.00100 gn=16.01384 time=55.19it/s +epoch=64 global_step=25400 loss=2.36818 loss_avg=2.33193 acc=0.78906 acc_top1_avg=0.78264 acc_top5_avg=0.93727 lr=0.00100 gn=11.73146 time=55.66it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.73554 test_loss_avg=0.45763 acc=0.84375 test_acc_avg=0.87340 test_acc_top5_avg=0.99006 time=249.28it/s +epoch=64 global_step=25415 loss=0.44559 test_loss_avg=0.37521 acc=0.87500 test_acc_avg=0.89458 test_acc_top5_avg=0.99288 time=555.68it/s +curr_acc 0.8946 +BEST_ACC 0.8976 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=2.89662 loss_avg=2.36879 acc=0.72656 acc_top1_avg=0.78080 acc_top5_avg=0.93750 lr=0.00100 gn=12.86425 time=54.75it/s +epoch=65 global_step=25500 loss=2.79877 loss_avg=2.34472 acc=0.72656 acc_top1_avg=0.78125 acc_top5_avg=0.93493 lr=0.00100 gn=15.58404 time=58.92it/s +epoch=65 global_step=25550 loss=2.13930 loss_avg=2.31975 acc=0.81250 acc_top1_avg=0.78414 acc_top5_avg=0.93646 lr=0.00100 gn=20.27082 time=61.98it/s +epoch=65 global_step=25600 loss=2.38657 loss_avg=2.31685 acc=0.78125 acc_top1_avg=0.78438 acc_top5_avg=0.93636 lr=0.00100 gn=19.71940 time=59.44it/s +epoch=65 global_step=25650 loss=2.67982 loss_avg=2.32366 acc=0.73438 acc_top1_avg=0.78394 acc_top5_avg=0.93604 lr=0.00100 gn=14.53543 time=54.05it/s +epoch=65 global_step=25700 loss=1.97716 loss_avg=2.32354 acc=0.82812 acc_top1_avg=0.78446 acc_top5_avg=0.93544 lr=0.00100 gn=19.74163 time=56.29it/s +epoch=65 global_step=25750 loss=2.25256 loss_avg=2.32077 acc=0.78906 acc_top1_avg=0.78496 acc_top5_avg=0.93589 lr=0.00100 gn=20.58582 time=58.72it/s +epoch=65 global_step=25800 loss=2.94206 loss_avg=2.31286 acc=0.71875 acc_top1_avg=0.78582 acc_top5_avg=0.93592 lr=0.00100 gn=18.74233 time=57.03it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.32995 test_loss_avg=0.40622 acc=0.92188 test_acc_avg=0.88177 test_acc_top5_avg=0.99479 time=215.31it/s +epoch=65 global_step=25806 loss=0.14984 test_loss_avg=0.44843 acc=0.95312 test_acc_avg=0.87356 test_acc_top5_avg=0.99171 time=231.59it/s +epoch=65 global_step=25806 loss=0.06291 test_loss_avg=0.39450 acc=0.93750 test_acc_avg=0.88825 test_acc_top5_avg=0.99288 time=568.10it/s +curr_acc 0.8883 +BEST_ACC 0.8976 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=2.18290 loss_avg=2.24710 acc=0.79688 acc_top1_avg=0.79190 acc_top5_avg=0.94229 lr=0.00100 gn=23.80722 time=56.20it/s +epoch=66 global_step=25900 loss=2.99777 loss_avg=2.23039 acc=0.72656 acc_top1_avg=0.79446 acc_top5_avg=0.94141 lr=0.00100 gn=23.15604 time=57.28it/s +epoch=66 global_step=25950 loss=2.06158 loss_avg=2.27637 acc=0.81250 acc_top1_avg=0.78966 acc_top5_avg=0.93793 lr=0.00100 gn=14.62596 time=53.31it/s +epoch=66 global_step=26000 loss=2.94267 loss_avg=2.28135 acc=0.72656 acc_top1_avg=0.78898 acc_top5_avg=0.93730 lr=0.00100 gn=17.38104 time=53.20it/s +epoch=66 global_step=26050 loss=2.04952 loss_avg=2.29785 acc=0.81250 acc_top1_avg=0.78746 acc_top5_avg=0.93670 lr=0.00100 gn=15.79052 time=57.75it/s +epoch=66 global_step=26100 loss=2.23388 loss_avg=2.30558 acc=0.80469 acc_top1_avg=0.78667 acc_top5_avg=0.93692 lr=0.00100 gn=22.48694 time=53.44it/s +epoch=66 global_step=26150 loss=2.24838 loss_avg=2.30653 acc=0.78125 acc_top1_avg=0.78661 acc_top5_avg=0.93618 lr=0.00100 gn=20.21715 time=60.01it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.28869 test_loss_avg=0.48560 acc=0.92188 test_acc_avg=0.86458 test_acc_top5_avg=0.98741 time=241.40it/s +epoch=66 global_step=26197 loss=0.26633 test_loss_avg=0.37895 acc=0.93750 test_acc_avg=0.89072 test_acc_top5_avg=0.99120 time=776.72it/s +curr_acc 0.8907 +BEST_ACC 0.8976 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=2.12084 loss_avg=2.02033 acc=0.78906 acc_top1_avg=0.81250 acc_top5_avg=0.94271 lr=0.00100 gn=16.55757 time=57.12it/s +epoch=67 global_step=26250 loss=2.16853 loss_avg=2.27891 acc=0.80469 acc_top1_avg=0.79024 acc_top5_avg=0.93721 lr=0.00100 gn=16.23017 time=59.31it/s +epoch=67 global_step=26300 loss=2.06437 loss_avg=2.22653 acc=0.81250 acc_top1_avg=0.79536 acc_top5_avg=0.93887 lr=0.00100 gn=22.85110 time=60.13it/s +epoch=67 global_step=26350 loss=2.34948 loss_avg=2.25694 acc=0.79688 acc_top1_avg=0.79208 acc_top5_avg=0.93724 lr=0.00100 gn=20.03631 time=60.41it/s +epoch=67 global_step=26400 loss=2.18627 loss_avg=2.25536 acc=0.79688 acc_top1_avg=0.79268 acc_top5_avg=0.93854 lr=0.00100 gn=15.73449 time=57.92it/s +epoch=67 global_step=26450 loss=2.41100 loss_avg=2.27678 acc=0.77344 acc_top1_avg=0.78968 acc_top5_avg=0.93701 lr=0.00100 gn=11.37799 time=60.27it/s +epoch=67 global_step=26500 loss=2.51088 loss_avg=2.28942 acc=0.75781 acc_top1_avg=0.78829 acc_top5_avg=0.93716 lr=0.00100 gn=21.48949 time=61.00it/s +epoch=67 global_step=26550 loss=2.53812 loss_avg=2.30265 acc=0.76562 acc_top1_avg=0.78718 acc_top5_avg=0.93562 lr=0.00100 gn=16.27729 time=57.34it/s +====================Eval==================== +epoch=67 global_step=26588 loss=0.48806 test_loss_avg=0.58865 acc=0.83594 test_acc_avg=0.83036 test_acc_top5_avg=0.99107 time=243.12it/s +epoch=67 global_step=26588 loss=0.20780 test_loss_avg=0.45947 acc=0.93750 test_acc_avg=0.87130 test_acc_top5_avg=0.99109 time=220.93it/s +epoch=67 global_step=26588 loss=0.23667 test_loss_avg=0.39171 acc=0.93750 test_acc_avg=0.88914 test_acc_top5_avg=0.99288 time=886.56it/s +curr_acc 0.8891 +BEST_ACC 0.8976 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=2.40909 loss_avg=2.16982 acc=0.76562 acc_top1_avg=0.80078 acc_top5_avg=0.95247 lr=0.00100 gn=11.62932 time=53.21it/s +epoch=68 global_step=26650 loss=2.67307 loss_avg=2.24095 acc=0.74219 acc_top1_avg=0.79385 acc_top5_avg=0.93687 lr=0.00100 gn=14.01815 time=52.10it/s +epoch=68 global_step=26700 loss=1.83804 loss_avg=2.22739 acc=0.83594 acc_top1_avg=0.79632 acc_top5_avg=0.93862 lr=0.00100 gn=19.27129 time=58.67it/s +epoch=68 global_step=26750 loss=2.22058 loss_avg=2.25284 acc=0.79688 acc_top1_avg=0.79393 acc_top5_avg=0.93813 lr=0.00100 gn=23.66440 time=54.47it/s +epoch=68 global_step=26800 loss=2.24519 loss_avg=2.25748 acc=0.78906 acc_top1_avg=0.79323 acc_top5_avg=0.93761 lr=0.00100 gn=18.78945 time=64.45it/s +epoch=68 global_step=26850 loss=2.87142 loss_avg=2.27359 acc=0.73438 acc_top1_avg=0.79184 acc_top5_avg=0.93559 lr=0.00100 gn=21.83971 time=53.45it/s +epoch=68 global_step=26900 loss=2.25151 loss_avg=2.27361 acc=0.78125 acc_top1_avg=0.79142 acc_top5_avg=0.93532 lr=0.00100 gn=19.47201 time=60.23it/s +epoch=68 global_step=26950 loss=2.00218 loss_avg=2.27719 acc=0.81250 acc_top1_avg=0.79122 acc_top5_avg=0.93515 lr=0.00100 gn=18.27727 time=55.81it/s +====================Eval==================== +epoch=68 global_step=26979 loss=0.62964 test_loss_avg=0.58235 acc=0.82031 test_acc_avg=0.83984 test_acc_top5_avg=0.98661 time=226.90it/s +epoch=68 global_step=26979 loss=0.27920 test_loss_avg=0.40747 acc=0.92188 test_acc_avg=0.88532 test_acc_top5_avg=0.99219 time=246.52it/s +epoch=68 global_step=26979 loss=0.20702 test_loss_avg=0.40493 acc=0.93750 test_acc_avg=0.88598 test_acc_top5_avg=0.99229 time=842.06it/s +curr_acc 0.8860 +BEST_ACC 0.8976 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=2.65588 loss_avg=2.21297 acc=0.75781 acc_top1_avg=0.79539 acc_top5_avg=0.93415 lr=0.00100 gn=16.28438 time=58.72it/s +epoch=69 global_step=27050 loss=2.12100 loss_avg=2.20907 acc=0.80469 acc_top1_avg=0.79588 acc_top5_avg=0.93739 lr=0.00100 gn=21.90095 time=57.63it/s +epoch=69 global_step=27100 loss=2.46587 loss_avg=2.21859 acc=0.78125 acc_top1_avg=0.79558 acc_top5_avg=0.93705 lr=0.00100 gn=22.63415 time=59.21it/s +epoch=69 global_step=27150 loss=2.21107 loss_avg=2.23694 acc=0.80469 acc_top1_avg=0.79381 acc_top5_avg=0.93709 lr=0.00100 gn=17.70413 time=59.96it/s +epoch=69 global_step=27200 loss=2.27292 loss_avg=2.24406 acc=0.79688 acc_top1_avg=0.79341 acc_top5_avg=0.93704 lr=0.00100 gn=21.58432 time=53.85it/s +epoch=69 global_step=27250 loss=1.79696 loss_avg=2.24674 acc=0.83594 acc_top1_avg=0.79327 acc_top5_avg=0.93687 lr=0.00100 gn=20.27098 time=59.39it/s +epoch=69 global_step=27300 loss=2.85869 loss_avg=2.26178 acc=0.72656 acc_top1_avg=0.79176 acc_top5_avg=0.93633 lr=0.00100 gn=14.31528 time=60.48it/s +epoch=69 global_step=27350 loss=1.93337 loss_avg=2.27232 acc=0.82812 acc_top1_avg=0.79075 acc_top5_avg=0.93609 lr=0.00100 gn=16.97520 time=56.95it/s +====================Eval==================== +epoch=69 global_step=27370 loss=0.19400 test_loss_avg=0.48522 acc=0.93750 test_acc_avg=0.86304 test_acc_top5_avg=0.99043 time=237.68it/s +epoch=69 global_step=27370 loss=0.24960 test_loss_avg=0.37366 acc=0.93750 test_acc_avg=0.89369 test_acc_top5_avg=0.99337 time=546.77it/s +curr_acc 0.8937 +BEST_ACC 0.8976 +curr_acc_top5 0.9934 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=1.84775 loss_avg=2.29209 acc=0.83594 acc_top1_avg=0.79089 acc_top5_avg=0.93568 lr=0.00100 gn=17.54926 time=59.55it/s +epoch=70 global_step=27450 loss=2.61774 loss_avg=2.28607 acc=0.75000 acc_top1_avg=0.78994 acc_top5_avg=0.93594 lr=0.00100 gn=13.90848 time=58.88it/s +epoch=70 global_step=27500 loss=2.93215 loss_avg=2.24954 acc=0.73438 acc_top1_avg=0.79309 acc_top5_avg=0.93486 lr=0.00100 gn=24.12569 time=61.76it/s +epoch=70 global_step=27550 loss=2.45132 loss_avg=2.24500 acc=0.78125 acc_top1_avg=0.79366 acc_top5_avg=0.93498 lr=0.00100 gn=19.85975 time=56.98it/s +epoch=70 global_step=27600 loss=2.26805 loss_avg=2.26396 acc=0.79688 acc_top1_avg=0.79168 acc_top5_avg=0.93448 lr=0.00100 gn=19.47209 time=53.09it/s +epoch=70 global_step=27650 loss=2.56634 loss_avg=2.25960 acc=0.75781 acc_top1_avg=0.79208 acc_top5_avg=0.93404 lr=0.00100 gn=22.20391 time=54.98it/s +epoch=70 global_step=27700 loss=2.35004 loss_avg=2.25952 acc=0.77344 acc_top1_avg=0.79190 acc_top5_avg=0.93487 lr=0.00100 gn=14.99455 time=54.71it/s +epoch=70 global_step=27750 loss=1.82454 loss_avg=2.27530 acc=0.85156 acc_top1_avg=0.79019 acc_top5_avg=0.93491 lr=0.00100 gn=24.41039 time=60.79it/s +====================Eval==================== +epoch=70 global_step=27761 loss=0.64766 test_loss_avg=0.50650 acc=0.76562 test_acc_avg=0.85625 test_acc_top5_avg=0.98906 time=244.57it/s +epoch=70 global_step=27761 loss=0.13304 test_loss_avg=0.42276 acc=0.96875 test_acc_avg=0.88225 test_acc_top5_avg=0.99208 time=249.20it/s +epoch=70 global_step=27761 loss=0.26272 test_loss_avg=0.40485 acc=0.93750 test_acc_avg=0.88716 test_acc_top5_avg=0.99229 time=560.29it/s +curr_acc 0.8872 +BEST_ACC 0.8976 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=1.77267 loss_avg=2.25016 acc=0.85156 acc_top1_avg=0.79367 acc_top5_avg=0.93850 lr=0.00100 gn=23.45569 time=60.47it/s +epoch=71 global_step=27850 loss=2.39098 loss_avg=2.27145 acc=0.79688 acc_top1_avg=0.79055 acc_top5_avg=0.93425 lr=0.00100 gn=21.93606 time=51.96it/s +epoch=71 global_step=27900 loss=2.23274 loss_avg=2.25289 acc=0.78906 acc_top1_avg=0.79249 acc_top5_avg=0.93508 lr=0.00100 gn=24.21058 time=54.61it/s +epoch=71 global_step=27950 loss=2.82052 loss_avg=2.26513 acc=0.73438 acc_top1_avg=0.79105 acc_top5_avg=0.93432 lr=0.00100 gn=22.19276 time=54.92it/s +epoch=71 global_step=28000 loss=2.35994 loss_avg=2.26782 acc=0.78906 acc_top1_avg=0.79096 acc_top5_avg=0.93446 lr=0.00100 gn=20.47226 time=54.73it/s +epoch=71 global_step=28050 loss=1.92196 loss_avg=2.26795 acc=0.81250 acc_top1_avg=0.79090 acc_top5_avg=0.93531 lr=0.00100 gn=19.02151 time=53.11it/s +epoch=71 global_step=28100 loss=1.97975 loss_avg=2.27197 acc=0.82031 acc_top1_avg=0.79051 acc_top5_avg=0.93598 lr=0.00100 gn=16.98434 time=63.45it/s +epoch=71 global_step=28150 loss=2.39166 loss_avg=2.25844 acc=0.79688 acc_top1_avg=0.79205 acc_top5_avg=0.93629 lr=0.00100 gn=28.28102 time=61.02it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.52712 test_loss_avg=0.52573 acc=0.87500 test_acc_avg=0.85175 test_acc_top5_avg=0.98952 time=232.18it/s +epoch=71 global_step=28152 loss=0.24272 test_loss_avg=0.39358 acc=0.93750 test_acc_avg=0.88805 test_acc_top5_avg=0.99248 time=871.09it/s +curr_acc 0.8881 +BEST_ACC 0.8976 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=2.14016 loss_avg=2.31383 acc=0.81250 acc_top1_avg=0.78743 acc_top5_avg=0.93490 lr=0.00100 gn=26.02622 time=59.16it/s +epoch=72 global_step=28250 loss=2.32392 loss_avg=2.27808 acc=0.80469 acc_top1_avg=0.79082 acc_top5_avg=0.93535 lr=0.00100 gn=23.76976 time=57.61it/s +epoch=72 global_step=28300 loss=2.30058 loss_avg=2.24237 acc=0.78906 acc_top1_avg=0.79445 acc_top5_avg=0.93586 lr=0.00100 gn=17.88025 time=54.02it/s +epoch=72 global_step=28350 loss=2.38480 loss_avg=2.27178 acc=0.78125 acc_top1_avg=0.79131 acc_top5_avg=0.93541 lr=0.00100 gn=23.31921 time=53.14it/s +epoch=72 global_step=28400 loss=1.80152 loss_avg=2.25215 acc=0.85156 acc_top1_avg=0.79354 acc_top5_avg=0.93759 lr=0.00100 gn=19.08284 time=62.74it/s +epoch=72 global_step=28450 loss=2.53352 loss_avg=2.26296 acc=0.78125 acc_top1_avg=0.79279 acc_top5_avg=0.93763 lr=0.00100 gn=22.94767 time=56.25it/s +epoch=72 global_step=28500 loss=2.01832 loss_avg=2.25707 acc=0.81250 acc_top1_avg=0.79326 acc_top5_avg=0.93833 lr=0.00100 gn=24.50830 time=50.08it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.07830 test_loss_avg=0.44370 acc=0.98438 test_acc_avg=0.87109 test_acc_top5_avg=0.98893 time=235.89it/s +epoch=72 global_step=28543 loss=0.49019 test_loss_avg=0.43884 acc=0.83594 test_acc_avg=0.87613 test_acc_top5_avg=0.99168 time=232.40it/s +epoch=72 global_step=28543 loss=0.30621 test_loss_avg=0.40204 acc=0.93750 test_acc_avg=0.88558 test_acc_top5_avg=0.99278 time=544.08it/s +curr_acc 0.8856 +BEST_ACC 0.8976 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=2.30597 loss_avg=2.05156 acc=0.78906 acc_top1_avg=0.81138 acc_top5_avg=0.93862 lr=0.00100 gn=19.23693 time=47.79it/s +epoch=73 global_step=28600 loss=2.38359 loss_avg=2.23707 acc=0.78125 acc_top1_avg=0.79372 acc_top5_avg=0.93531 lr=0.00100 gn=19.47690 time=51.55it/s +epoch=73 global_step=28650 loss=1.76078 loss_avg=2.23624 acc=0.84375 acc_top1_avg=0.79381 acc_top5_avg=0.93801 lr=0.00100 gn=16.75041 time=58.31it/s +epoch=73 global_step=28700 loss=2.15659 loss_avg=2.25006 acc=0.78906 acc_top1_avg=0.79235 acc_top5_avg=0.93735 lr=0.00100 gn=17.34285 time=59.28it/s +epoch=73 global_step=28750 loss=2.51634 loss_avg=2.24953 acc=0.75781 acc_top1_avg=0.79204 acc_top5_avg=0.93776 lr=0.00100 gn=19.19341 time=54.31it/s +epoch=73 global_step=28800 loss=2.39242 loss_avg=2.24274 acc=0.78125 acc_top1_avg=0.79268 acc_top5_avg=0.93811 lr=0.00100 gn=21.31980 time=56.70it/s +epoch=73 global_step=28850 loss=2.48128 loss_avg=2.24679 acc=0.75000 acc_top1_avg=0.79275 acc_top5_avg=0.93712 lr=0.00100 gn=13.97535 time=59.55it/s +epoch=73 global_step=28900 loss=2.00965 loss_avg=2.24840 acc=0.81250 acc_top1_avg=0.79265 acc_top5_avg=0.93638 lr=0.00100 gn=20.36316 time=48.14it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.36290 test_loss_avg=0.58402 acc=0.89844 test_acc_avg=0.83570 test_acc_top5_avg=0.98532 time=229.36it/s +epoch=73 global_step=28934 loss=0.16186 test_loss_avg=0.39939 acc=0.93750 test_acc_avg=0.88647 test_acc_top5_avg=0.99090 time=772.15it/s +curr_acc 0.8865 +BEST_ACC 0.8976 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=1.92717 loss_avg=2.20050 acc=0.82812 acc_top1_avg=0.80078 acc_top5_avg=0.93457 lr=0.00100 gn=20.16460 time=52.67it/s +epoch=74 global_step=29000 loss=2.21619 loss_avg=2.19715 acc=0.79688 acc_top1_avg=0.79782 acc_top5_avg=0.94058 lr=0.00100 gn=26.24733 time=46.02it/s +epoch=74 global_step=29050 loss=2.05078 loss_avg=2.16452 acc=0.82031 acc_top1_avg=0.80179 acc_top5_avg=0.93918 lr=0.00100 gn=25.25120 time=61.27it/s +epoch=74 global_step=29100 loss=2.02942 loss_avg=2.18069 acc=0.81250 acc_top1_avg=0.80031 acc_top5_avg=0.93882 lr=0.00100 gn=17.29449 time=57.77it/s +epoch=74 global_step=29150 loss=1.56603 loss_avg=2.18858 acc=0.85938 acc_top1_avg=0.79962 acc_top5_avg=0.93783 lr=0.00100 gn=15.86622 time=59.88it/s +epoch=74 global_step=29200 loss=1.74100 loss_avg=2.22468 acc=0.83594 acc_top1_avg=0.79573 acc_top5_avg=0.93630 lr=0.00100 gn=24.21940 time=56.85it/s +epoch=74 global_step=29250 loss=1.84107 loss_avg=2.23505 acc=0.82031 acc_top1_avg=0.79465 acc_top5_avg=0.93602 lr=0.00100 gn=14.64622 time=46.87it/s +epoch=74 global_step=29300 loss=2.44428 loss_avg=2.24049 acc=0.75781 acc_top1_avg=0.79453 acc_top5_avg=0.93650 lr=0.00100 gn=25.81644 time=51.55it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.47984 test_loss_avg=0.58884 acc=0.85938 test_acc_avg=0.84180 test_acc_top5_avg=0.98242 time=245.90it/s +epoch=74 global_step=29325 loss=0.19598 test_loss_avg=0.44710 acc=0.94531 test_acc_avg=0.87486 test_acc_top5_avg=0.98872 time=232.77it/s +epoch=74 global_step=29325 loss=0.48018 test_loss_avg=0.39097 acc=0.87500 test_acc_avg=0.88904 test_acc_top5_avg=0.99130 time=867.67it/s +curr_acc 0.8890 +BEST_ACC 0.8976 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=2.22769 loss_avg=2.17241 acc=0.79688 acc_top1_avg=0.80437 acc_top5_avg=0.93812 lr=0.00100 gn=27.80674 time=62.39it/s +epoch=75 global_step=29400 loss=2.63614 loss_avg=2.18235 acc=0.75781 acc_top1_avg=0.80135 acc_top5_avg=0.93958 lr=0.00100 gn=19.25288 time=51.71it/s +epoch=75 global_step=29450 loss=2.00578 loss_avg=2.24097 acc=0.82031 acc_top1_avg=0.79481 acc_top5_avg=0.93856 lr=0.00100 gn=16.21866 time=58.15it/s +epoch=75 global_step=29500 loss=2.05133 loss_avg=2.22754 acc=0.79688 acc_top1_avg=0.79647 acc_top5_avg=0.93786 lr=0.00100 gn=27.95420 time=57.80it/s +epoch=75 global_step=29550 loss=3.08366 loss_avg=2.23141 acc=0.70312 acc_top1_avg=0.79642 acc_top5_avg=0.93830 lr=0.00100 gn=24.38924 time=59.26it/s +epoch=75 global_step=29600 loss=1.58966 loss_avg=2.24208 acc=0.85938 acc_top1_avg=0.79531 acc_top5_avg=0.93750 lr=0.00100 gn=19.85802 time=56.55it/s +epoch=75 global_step=29650 loss=1.73286 loss_avg=2.24399 acc=0.86719 acc_top1_avg=0.79486 acc_top5_avg=0.93707 lr=0.00100 gn=28.71323 time=51.69it/s +epoch=75 global_step=29700 loss=2.35154 loss_avg=2.23628 acc=0.78906 acc_top1_avg=0.79565 acc_top5_avg=0.93719 lr=0.00100 gn=24.42395 time=57.48it/s +====================Eval==================== +epoch=75 global_step=29716 loss=0.41897 test_loss_avg=0.44004 acc=0.88281 test_acc_avg=0.87531 test_acc_top5_avg=0.99219 time=243.88it/s +epoch=75 global_step=29716 loss=0.26026 test_loss_avg=0.42425 acc=0.92188 test_acc_avg=0.88010 test_acc_top5_avg=0.99229 time=241.48it/s +epoch=75 global_step=29716 loss=0.16077 test_loss_avg=0.41425 acc=0.93750 test_acc_avg=0.88232 test_acc_top5_avg=0.99268 time=876.37it/s +curr_acc 0.8823 +BEST_ACC 0.8976 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=2.39732 loss_avg=2.35653 acc=0.76562 acc_top1_avg=0.78148 acc_top5_avg=0.93153 lr=0.00100 gn=15.20196 time=54.85it/s +epoch=76 global_step=29800 loss=1.74628 loss_avg=2.30464 acc=0.84375 acc_top1_avg=0.78785 acc_top5_avg=0.93499 lr=0.00100 gn=19.70363 time=56.89it/s +epoch=76 global_step=29850 loss=2.18234 loss_avg=2.27592 acc=0.80469 acc_top1_avg=0.79128 acc_top5_avg=0.93686 lr=0.00100 gn=14.99101 time=56.39it/s +epoch=76 global_step=29900 loss=1.54691 loss_avg=2.25709 acc=0.86719 acc_top1_avg=0.79254 acc_top5_avg=0.93618 lr=0.00100 gn=15.43833 time=59.77it/s +epoch=76 global_step=29950 loss=2.55163 loss_avg=2.26675 acc=0.75781 acc_top1_avg=0.79207 acc_top5_avg=0.93533 lr=0.00100 gn=13.18046 time=53.90it/s +epoch=76 global_step=30000 loss=2.36780 loss_avg=2.25511 acc=0.78906 acc_top1_avg=0.79346 acc_top5_avg=0.93678 lr=0.00100 gn=21.90498 time=55.71it/s +epoch=76 global_step=30050 loss=2.08326 loss_avg=2.24647 acc=0.80469 acc_top1_avg=0.79397 acc_top5_avg=0.93734 lr=0.00100 gn=26.48047 time=51.44it/s +epoch=76 global_step=30100 loss=1.91699 loss_avg=2.24060 acc=0.82812 acc_top1_avg=0.79474 acc_top5_avg=0.93705 lr=0.00100 gn=23.49074 time=62.67it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.61574 test_loss_avg=0.58486 acc=0.84375 test_acc_avg=0.83916 test_acc_top5_avg=0.98641 time=232.14it/s +epoch=76 global_step=30107 loss=0.26318 test_loss_avg=0.41001 acc=0.93750 test_acc_avg=0.88538 test_acc_top5_avg=0.99150 time=860.55it/s +curr_acc 0.8854 +BEST_ACC 0.8976 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=2.70340 loss_avg=2.19937 acc=0.75781 acc_top1_avg=0.79833 acc_top5_avg=0.93368 lr=0.00100 gn=22.87884 time=58.09it/s +epoch=77 global_step=30200 loss=2.75383 loss_avg=2.25908 acc=0.75000 acc_top1_avg=0.79368 acc_top5_avg=0.93305 lr=0.00100 gn=27.39451 time=59.19it/s +epoch=77 global_step=30250 loss=2.72232 loss_avg=2.23375 acc=0.74219 acc_top1_avg=0.79638 acc_top5_avg=0.93357 lr=0.00100 gn=21.84720 time=50.41it/s +epoch=77 global_step=30300 loss=2.30082 loss_avg=2.21550 acc=0.79688 acc_top1_avg=0.79845 acc_top5_avg=0.93479 lr=0.00100 gn=25.15304 time=53.76it/s +epoch=77 global_step=30350 loss=1.90488 loss_avg=2.23319 acc=0.83594 acc_top1_avg=0.79633 acc_top5_avg=0.93451 lr=0.00100 gn=21.22097 time=53.00it/s +epoch=77 global_step=30400 loss=2.14107 loss_avg=2.21517 acc=0.79688 acc_top1_avg=0.79767 acc_top5_avg=0.93499 lr=0.00100 gn=27.44105 time=56.49it/s +epoch=77 global_step=30450 loss=2.16273 loss_avg=2.22148 acc=0.79688 acc_top1_avg=0.79703 acc_top5_avg=0.93552 lr=0.00100 gn=27.91052 time=48.96it/s +====================Eval==================== +epoch=77 global_step=30498 loss=1.29957 test_loss_avg=0.49833 acc=0.63281 test_acc_avg=0.85478 test_acc_top5_avg=0.98989 time=113.18it/s +epoch=77 global_step=30498 loss=0.17568 test_loss_avg=0.46508 acc=0.94531 test_acc_avg=0.87022 test_acc_top5_avg=0.99137 time=237.81it/s +epoch=77 global_step=30498 loss=0.27065 test_loss_avg=0.42983 acc=0.93750 test_acc_avg=0.87915 test_acc_top5_avg=0.99229 time=549.35it/s +curr_acc 0.8792 +BEST_ACC 0.8976 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=2.10214 loss_avg=1.78544 acc=0.81250 acc_top1_avg=0.84766 acc_top5_avg=0.95703 lr=0.00100 gn=19.64328 time=28.81it/s +epoch=78 global_step=30550 loss=2.55596 loss_avg=2.21615 acc=0.76562 acc_top1_avg=0.79748 acc_top5_avg=0.93404 lr=0.00100 gn=21.45373 time=54.79it/s +epoch=78 global_step=30600 loss=2.35561 loss_avg=2.23085 acc=0.78125 acc_top1_avg=0.79688 acc_top5_avg=0.93627 lr=0.00100 gn=26.75653 time=58.54it/s +epoch=78 global_step=30650 loss=2.73759 loss_avg=2.22074 acc=0.74219 acc_top1_avg=0.79765 acc_top5_avg=0.93796 lr=0.00100 gn=22.43807 time=57.07it/s +epoch=78 global_step=30700 loss=2.36854 loss_avg=2.21946 acc=0.78906 acc_top1_avg=0.79711 acc_top5_avg=0.93858 lr=0.00100 gn=26.72564 time=51.00it/s +epoch=78 global_step=30750 loss=2.35052 loss_avg=2.21929 acc=0.80469 acc_top1_avg=0.79765 acc_top5_avg=0.93862 lr=0.00100 gn=24.37061 time=57.11it/s +epoch=78 global_step=30800 loss=2.47669 loss_avg=2.22566 acc=0.78906 acc_top1_avg=0.79688 acc_top5_avg=0.93809 lr=0.00100 gn=27.12757 time=63.38it/s +epoch=78 global_step=30850 loss=2.28198 loss_avg=2.22588 acc=0.79688 acc_top1_avg=0.79679 acc_top5_avg=0.93808 lr=0.00100 gn=25.31789 time=54.49it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.25323 test_loss_avg=0.49130 acc=0.90625 test_acc_avg=0.85773 test_acc_top5_avg=0.98993 time=235.45it/s +epoch=78 global_step=30889 loss=0.33097 test_loss_avg=0.41745 acc=0.87500 test_acc_avg=0.88054 test_acc_top5_avg=0.99090 time=735.07it/s +curr_acc 0.8805 +BEST_ACC 0.8976 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=2.34223 loss_avg=2.11445 acc=0.78906 acc_top1_avg=0.80611 acc_top5_avg=0.94247 lr=0.00100 gn=25.09409 time=56.29it/s +epoch=79 global_step=30950 loss=1.71490 loss_avg=2.12609 acc=0.84375 acc_top1_avg=0.80610 acc_top5_avg=0.93981 lr=0.00100 gn=18.57885 time=56.45it/s +epoch=79 global_step=31000 loss=1.86808 loss_avg=2.15467 acc=0.82031 acc_top1_avg=0.80279 acc_top5_avg=0.93856 lr=0.00100 gn=23.40876 time=54.30it/s +epoch=79 global_step=31050 loss=2.46155 loss_avg=2.15885 acc=0.78125 acc_top1_avg=0.80231 acc_top5_avg=0.93823 lr=0.00100 gn=17.41003 time=61.03it/s +epoch=79 global_step=31100 loss=2.50834 loss_avg=2.16596 acc=0.78125 acc_top1_avg=0.80202 acc_top5_avg=0.93906 lr=0.00100 gn=26.32506 time=52.91it/s +epoch=79 global_step=31150 loss=2.34005 loss_avg=2.19396 acc=0.78125 acc_top1_avg=0.79930 acc_top5_avg=0.93783 lr=0.00100 gn=23.30837 time=56.70it/s +epoch=79 global_step=31200 loss=2.30971 loss_avg=2.19986 acc=0.77344 acc_top1_avg=0.79871 acc_top5_avg=0.93765 lr=0.00100 gn=21.93806 time=57.79it/s +epoch=79 global_step=31250 loss=2.56209 loss_avg=2.20152 acc=0.75781 acc_top1_avg=0.79858 acc_top5_avg=0.93808 lr=0.00100 gn=24.39646 time=58.85it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.17997 test_loss_avg=0.49912 acc=0.94531 test_acc_avg=0.83941 test_acc_top5_avg=0.99306 time=248.37it/s +epoch=79 global_step=31280 loss=0.15891 test_loss_avg=0.50543 acc=0.94531 test_acc_avg=0.85487 test_acc_top5_avg=0.98914 time=244.47it/s +epoch=79 global_step=31280 loss=0.20402 test_loss_avg=0.43341 acc=0.93750 test_acc_avg=0.87431 test_acc_top5_avg=0.99110 time=873.09it/s +curr_acc 0.8743 +BEST_ACC 0.8976 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=2.02513 loss_avg=2.13408 acc=0.83594 acc_top1_avg=0.80703 acc_top5_avg=0.93437 lr=0.00010 gn=26.14632 time=63.53it/s +epoch=80 global_step=31350 loss=1.49262 loss_avg=2.20014 acc=0.87500 acc_top1_avg=0.79710 acc_top5_avg=0.93348 lr=0.00010 gn=24.69552 time=56.39it/s +epoch=80 global_step=31400 loss=1.60207 loss_avg=2.16403 acc=0.85938 acc_top1_avg=0.80085 acc_top5_avg=0.93587 lr=0.00010 gn=17.63342 time=60.84it/s +epoch=80 global_step=31450 loss=1.73437 loss_avg=2.12669 acc=0.85156 acc_top1_avg=0.80464 acc_top5_avg=0.93824 lr=0.00010 gn=19.97881 time=59.56it/s +epoch=80 global_step=31500 loss=1.96127 loss_avg=2.10988 acc=0.81250 acc_top1_avg=0.80629 acc_top5_avg=0.93761 lr=0.00010 gn=16.77948 time=51.97it/s +epoch=80 global_step=31550 loss=1.76934 loss_avg=2.12260 acc=0.83594 acc_top1_avg=0.80518 acc_top5_avg=0.93756 lr=0.00010 gn=21.77144 time=54.45it/s +epoch=80 global_step=31600 loss=2.30749 loss_avg=2.11541 acc=0.77344 acc_top1_avg=0.80579 acc_top5_avg=0.93811 lr=0.00010 gn=15.85800 time=60.56it/s +epoch=80 global_step=31650 loss=2.65631 loss_avg=2.11193 acc=0.73438 acc_top1_avg=0.80598 acc_top5_avg=0.93826 lr=0.00010 gn=24.88035 time=48.42it/s +====================Eval==================== +epoch=80 global_step=31671 loss=0.59720 test_loss_avg=0.51481 acc=0.81250 test_acc_avg=0.85104 test_acc_top5_avg=0.98802 time=242.84it/s +epoch=80 global_step=31671 loss=0.23289 test_loss_avg=0.36828 acc=0.93750 test_acc_avg=0.89399 test_acc_top5_avg=0.99268 time=878.57it/s +curr_acc 0.8940 +BEST_ACC 0.8976 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=2.38899 loss_avg=2.10127 acc=0.76562 acc_top1_avg=0.80577 acc_top5_avg=0.93992 lr=0.00010 gn=22.71358 time=52.66it/s +epoch=81 global_step=31750 loss=1.64351 loss_avg=2.09532 acc=0.85156 acc_top1_avg=0.80746 acc_top5_avg=0.93879 lr=0.00010 gn=22.20182 time=61.52it/s +epoch=81 global_step=31800 loss=1.78248 loss_avg=2.06940 acc=0.85156 acc_top1_avg=0.81014 acc_top5_avg=0.93853 lr=0.00010 gn=24.28497 time=61.28it/s +epoch=81 global_step=31850 loss=2.04842 loss_avg=2.05796 acc=0.81250 acc_top1_avg=0.81071 acc_top5_avg=0.93911 lr=0.00010 gn=22.56709 time=53.33it/s +epoch=81 global_step=31900 loss=1.72966 loss_avg=2.07971 acc=0.83594 acc_top1_avg=0.80851 acc_top5_avg=0.93798 lr=0.00010 gn=13.57372 time=57.98it/s +epoch=81 global_step=31950 loss=1.89087 loss_avg=2.08885 acc=0.82812 acc_top1_avg=0.80777 acc_top5_avg=0.93814 lr=0.00010 gn=20.13952 time=60.84it/s +epoch=81 global_step=32000 loss=1.90846 loss_avg=2.08303 acc=0.82031 acc_top1_avg=0.80839 acc_top5_avg=0.93805 lr=0.00010 gn=21.11572 time=61.80it/s +epoch=81 global_step=32050 loss=2.06047 loss_avg=2.07637 acc=0.80469 acc_top1_avg=0.80900 acc_top5_avg=0.93878 lr=0.00010 gn=14.52084 time=53.31it/s +====================Eval==================== +epoch=81 global_step=32062 loss=0.72943 test_loss_avg=0.72943 acc=0.78125 test_acc_avg=0.78125 test_acc_top5_avg=0.98438 time=220.05it/s +epoch=81 global_step=32062 loss=0.13098 test_loss_avg=0.45603 acc=0.96875 test_acc_avg=0.87117 test_acc_top5_avg=0.99050 time=236.69it/s +epoch=81 global_step=32062 loss=0.21060 test_loss_avg=0.36896 acc=0.93750 test_acc_avg=0.89438 test_acc_top5_avg=0.99268 time=575.27it/s +curr_acc 0.8944 +BEST_ACC 0.8976 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=2.72742 loss_avg=2.02061 acc=0.75781 acc_top1_avg=0.81312 acc_top5_avg=0.93565 lr=0.00010 gn=23.19081 time=55.24it/s +epoch=82 global_step=32150 loss=2.01344 loss_avg=2.04344 acc=0.80469 acc_top1_avg=0.81179 acc_top5_avg=0.93652 lr=0.00010 gn=18.34650 time=60.54it/s +epoch=82 global_step=32200 loss=1.37490 loss_avg=2.01018 acc=0.88281 acc_top1_avg=0.81567 acc_top5_avg=0.93959 lr=0.00010 gn=33.05283 time=52.31it/s +epoch=82 global_step=32250 loss=1.66837 loss_avg=2.01963 acc=0.85156 acc_top1_avg=0.81449 acc_top5_avg=0.93983 lr=0.00010 gn=17.21784 time=58.08it/s +epoch=82 global_step=32300 loss=2.03617 loss_avg=2.04648 acc=0.80469 acc_top1_avg=0.81155 acc_top5_avg=0.93839 lr=0.00010 gn=19.65339 time=54.81it/s +epoch=82 global_step=32350 loss=1.93883 loss_avg=2.05599 acc=0.82812 acc_top1_avg=0.81055 acc_top5_avg=0.93888 lr=0.00010 gn=21.27724 time=54.71it/s +epoch=82 global_step=32400 loss=2.52267 loss_avg=2.05266 acc=0.77344 acc_top1_avg=0.81107 acc_top5_avg=0.93963 lr=0.00010 gn=29.62488 time=55.48it/s +epoch=82 global_step=32450 loss=1.74785 loss_avg=2.06382 acc=0.85156 acc_top1_avg=0.81010 acc_top5_avg=0.93847 lr=0.00010 gn=24.79595 time=59.88it/s +====================Eval==================== +epoch=82 global_step=32453 loss=0.86818 test_loss_avg=0.49514 acc=0.79688 test_acc_avg=0.86044 test_acc_top5_avg=0.98899 time=212.83it/s +epoch=82 global_step=32453 loss=0.38072 test_loss_avg=0.38374 acc=0.87500 test_acc_avg=0.89128 test_acc_top5_avg=0.99251 time=246.68it/s +epoch=82 global_step=32453 loss=0.32467 test_loss_avg=0.37873 acc=0.87500 test_acc_avg=0.89142 test_acc_top5_avg=0.99268 time=610.79it/s +curr_acc 0.8914 +BEST_ACC 0.8976 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=1.91547 loss_avg=2.11695 acc=0.82031 acc_top1_avg=0.80319 acc_top5_avg=0.93567 lr=0.00010 gn=21.83773 time=57.23it/s +epoch=83 global_step=32550 loss=1.99006 loss_avg=2.06129 acc=0.83594 acc_top1_avg=0.80944 acc_top5_avg=0.93629 lr=0.00010 gn=23.46397 time=59.75it/s +epoch=83 global_step=32600 loss=2.32357 loss_avg=2.03305 acc=0.78906 acc_top1_avg=0.81229 acc_top5_avg=0.93803 lr=0.00010 gn=27.39937 time=60.01it/s +epoch=83 global_step=32650 loss=2.79536 loss_avg=2.05137 acc=0.72656 acc_top1_avg=0.81111 acc_top5_avg=0.93798 lr=0.00010 gn=26.54881 time=55.68it/s +epoch=83 global_step=32700 loss=1.62775 loss_avg=2.06015 acc=0.85156 acc_top1_avg=0.81010 acc_top5_avg=0.93703 lr=0.00010 gn=20.54065 time=61.01it/s +epoch=83 global_step=32750 loss=2.53013 loss_avg=2.06262 acc=0.75781 acc_top1_avg=0.80971 acc_top5_avg=0.93703 lr=0.00010 gn=15.45994 time=53.95it/s +epoch=83 global_step=32800 loss=2.48759 loss_avg=2.06586 acc=0.78125 acc_top1_avg=0.80933 acc_top5_avg=0.93743 lr=0.00010 gn=25.54260 time=53.12it/s +====================Eval==================== +epoch=83 global_step=32844 loss=0.58810 test_loss_avg=0.47771 acc=0.83594 test_acc_avg=0.86156 test_acc_top5_avg=0.99073 time=247.92it/s +epoch=83 global_step=32844 loss=0.23709 test_loss_avg=0.36736 acc=0.93750 test_acc_avg=0.89399 test_acc_top5_avg=0.99337 time=878.20it/s +curr_acc 0.8940 +BEST_ACC 0.8976 +curr_acc_top5 0.9934 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=1.80080 loss_avg=2.13722 acc=0.82812 acc_top1_avg=0.79818 acc_top5_avg=0.93750 lr=0.00010 gn=18.87432 time=54.85it/s +epoch=84 global_step=32900 loss=1.98162 loss_avg=2.02868 acc=0.82031 acc_top1_avg=0.81208 acc_top5_avg=0.93736 lr=0.00010 gn=18.28582 time=49.78it/s +epoch=84 global_step=32950 loss=1.48880 loss_avg=2.02413 acc=0.87500 acc_top1_avg=0.81353 acc_top5_avg=0.93787 lr=0.00010 gn=20.85909 time=57.68it/s +epoch=84 global_step=33000 loss=1.94900 loss_avg=2.04367 acc=0.82031 acc_top1_avg=0.81170 acc_top5_avg=0.93915 lr=0.00010 gn=25.66724 time=61.15it/s +epoch=84 global_step=33050 loss=1.92412 loss_avg=2.05989 acc=0.82812 acc_top1_avg=0.81007 acc_top5_avg=0.93868 lr=0.00010 gn=24.17338 time=55.55it/s +epoch=84 global_step=33100 loss=2.66625 loss_avg=2.06366 acc=0.75000 acc_top1_avg=0.81009 acc_top5_avg=0.93854 lr=0.00010 gn=27.94222 time=58.48it/s +epoch=84 global_step=33150 loss=1.58144 loss_avg=2.05361 acc=0.85156 acc_top1_avg=0.81135 acc_top5_avg=0.93921 lr=0.00010 gn=15.31692 time=53.23it/s +epoch=84 global_step=33200 loss=1.89217 loss_avg=2.05070 acc=0.83594 acc_top1_avg=0.81160 acc_top5_avg=0.93926 lr=0.00010 gn=20.33830 time=53.80it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.09458 test_loss_avg=0.41832 acc=0.97656 test_acc_avg=0.87612 test_acc_top5_avg=0.99107 time=249.97it/s +epoch=84 global_step=33235 loss=0.19135 test_loss_avg=0.41367 acc=0.93750 test_acc_avg=0.88269 test_acc_top5_avg=0.99194 time=248.67it/s +epoch=84 global_step=33235 loss=0.28436 test_loss_avg=0.38012 acc=0.87500 test_acc_avg=0.89092 test_acc_top5_avg=0.99298 time=870.55it/s +curr_acc 0.8909 +BEST_ACC 0.8976 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=1.99149 loss_avg=1.98158 acc=0.81250 acc_top1_avg=0.82083 acc_top5_avg=0.93802 lr=0.00010 gn=20.04368 time=56.38it/s +epoch=85 global_step=33300 loss=2.13290 loss_avg=2.00154 acc=0.80469 acc_top1_avg=0.81719 acc_top5_avg=0.94099 lr=0.00010 gn=21.08631 time=58.89it/s +epoch=85 global_step=33350 loss=1.90073 loss_avg=2.00743 acc=0.83594 acc_top1_avg=0.81603 acc_top5_avg=0.94015 lr=0.00010 gn=20.86242 time=61.40it/s +epoch=85 global_step=33400 loss=1.58249 loss_avg=2.00367 acc=0.86719 acc_top1_avg=0.81563 acc_top5_avg=0.93991 lr=0.00010 gn=19.44583 time=58.55it/s +epoch=85 global_step=33450 loss=1.03644 loss_avg=1.99976 acc=0.91406 acc_top1_avg=0.81588 acc_top5_avg=0.93910 lr=0.00010 gn=11.89186 time=63.22it/s +epoch=85 global_step=33500 loss=2.30372 loss_avg=2.00927 acc=0.78125 acc_top1_avg=0.81459 acc_top5_avg=0.93948 lr=0.00010 gn=19.21364 time=59.91it/s +epoch=85 global_step=33550 loss=2.55979 loss_avg=2.02430 acc=0.76562 acc_top1_avg=0.81307 acc_top5_avg=0.93976 lr=0.00010 gn=27.60083 time=51.67it/s +epoch=85 global_step=33600 loss=2.18227 loss_avg=2.03386 acc=0.78906 acc_top1_avg=0.81211 acc_top5_avg=0.93891 lr=0.00010 gn=19.56467 time=59.79it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.17037 test_loss_avg=0.51761 acc=0.93750 test_acc_avg=0.84710 test_acc_top5_avg=0.98951 time=244.87it/s +epoch=85 global_step=33626 loss=0.30838 test_loss_avg=0.38332 acc=0.87500 test_acc_avg=0.88825 test_acc_top5_avg=0.99298 time=880.79it/s +curr_acc 0.8883 +BEST_ACC 0.8976 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=1.99952 loss_avg=2.05108 acc=0.81250 acc_top1_avg=0.81185 acc_top5_avg=0.93848 lr=0.00010 gn=20.23560 time=50.86it/s +epoch=86 global_step=33700 loss=2.15068 loss_avg=2.02600 acc=0.79688 acc_top1_avg=0.81377 acc_top5_avg=0.93792 lr=0.00010 gn=19.78069 time=53.58it/s +epoch=86 global_step=33750 loss=2.16978 loss_avg=2.01278 acc=0.80469 acc_top1_avg=0.81552 acc_top5_avg=0.93832 lr=0.00010 gn=26.83459 time=53.59it/s +epoch=86 global_step=33800 loss=1.93395 loss_avg=2.04006 acc=0.82031 acc_top1_avg=0.81241 acc_top5_avg=0.93660 lr=0.00010 gn=23.31818 time=59.40it/s +epoch=86 global_step=33850 loss=2.09594 loss_avg=2.03463 acc=0.80469 acc_top1_avg=0.81278 acc_top5_avg=0.93816 lr=0.00010 gn=17.25407 time=61.09it/s +epoch=86 global_step=33900 loss=1.80042 loss_avg=2.02666 acc=0.82812 acc_top1_avg=0.81355 acc_top5_avg=0.93824 lr=0.00010 gn=21.64261 time=48.92it/s +epoch=86 global_step=33950 loss=1.60192 loss_avg=2.02622 acc=0.85938 acc_top1_avg=0.81349 acc_top5_avg=0.93830 lr=0.00010 gn=22.00966 time=54.82it/s +epoch=86 global_step=34000 loss=1.85071 loss_avg=2.03423 acc=0.84375 acc_top1_avg=0.81279 acc_top5_avg=0.93861 lr=0.00010 gn=23.05799 time=52.20it/s +====================Eval==================== +epoch=86 global_step=34017 loss=0.67827 test_loss_avg=0.62890 acc=0.81250 test_acc_avg=0.80469 test_acc_top5_avg=0.98828 time=237.37it/s +epoch=86 global_step=34017 loss=0.23417 test_loss_avg=0.44492 acc=0.90625 test_acc_avg=0.87430 test_acc_top5_avg=0.99079 time=244.11it/s +epoch=86 global_step=34017 loss=0.28071 test_loss_avg=0.38118 acc=0.87500 test_acc_avg=0.89003 test_acc_top5_avg=0.99268 time=853.02it/s +curr_acc 0.8900 +BEST_ACC 0.8976 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=1.98725 loss_avg=2.04227 acc=0.81250 acc_top1_avg=0.81203 acc_top5_avg=0.94129 lr=0.00010 gn=21.74995 time=52.53it/s +epoch=87 global_step=34100 loss=1.90017 loss_avg=1.96667 acc=0.82812 acc_top1_avg=0.81899 acc_top5_avg=0.94230 lr=0.00010 gn=14.17191 time=53.81it/s +epoch=87 global_step=34150 loss=2.17327 loss_avg=1.99706 acc=0.78906 acc_top1_avg=0.81573 acc_top5_avg=0.93997 lr=0.00010 gn=21.95033 time=55.53it/s +epoch=87 global_step=34200 loss=1.54641 loss_avg=2.00776 acc=0.87500 acc_top1_avg=0.81451 acc_top5_avg=0.94027 lr=0.00010 gn=23.09286 time=59.91it/s +epoch=87 global_step=34250 loss=2.23266 loss_avg=2.00830 acc=0.80469 acc_top1_avg=0.81441 acc_top5_avg=0.94001 lr=0.00010 gn=24.55220 time=59.64it/s +epoch=87 global_step=34300 loss=2.07901 loss_avg=2.03032 acc=0.80469 acc_top1_avg=0.81242 acc_top5_avg=0.93913 lr=0.00010 gn=15.46929 time=56.46it/s +epoch=87 global_step=34350 loss=1.95861 loss_avg=2.02213 acc=0.82031 acc_top1_avg=0.81320 acc_top5_avg=0.93954 lr=0.00010 gn=23.25273 time=51.27it/s +epoch=87 global_step=34400 loss=1.96620 loss_avg=2.02341 acc=0.82031 acc_top1_avg=0.81309 acc_top5_avg=0.93909 lr=0.00010 gn=20.27827 time=52.12it/s +====================Eval==================== +epoch=87 global_step=34408 loss=0.60082 test_loss_avg=0.54505 acc=0.84375 test_acc_avg=0.84172 test_acc_top5_avg=0.98814 time=241.82it/s +epoch=87 global_step=34408 loss=0.44439 test_loss_avg=0.39356 acc=0.86719 test_acc_avg=0.88677 test_acc_top5_avg=0.99269 time=246.27it/s +epoch=87 global_step=34408 loss=0.28262 test_loss_avg=0.39017 acc=0.87500 test_acc_avg=0.88706 test_acc_top5_avg=0.99288 time=552.83it/s +curr_acc 0.8871 +BEST_ACC 0.8976 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=2.05610 loss_avg=2.06468 acc=0.80469 acc_top1_avg=0.80952 acc_top5_avg=0.93824 lr=0.00010 gn=19.45679 time=57.60it/s +epoch=88 global_step=34500 loss=1.74683 loss_avg=2.04931 acc=0.84375 acc_top1_avg=0.81072 acc_top5_avg=0.93708 lr=0.00010 gn=23.76432 time=51.79it/s +epoch=88 global_step=34550 loss=1.74513 loss_avg=2.02731 acc=0.84375 acc_top1_avg=0.81316 acc_top5_avg=0.93744 lr=0.00010 gn=19.33109 time=53.97it/s +epoch=88 global_step=34600 loss=2.04087 loss_avg=2.04027 acc=0.82031 acc_top1_avg=0.81177 acc_top5_avg=0.93673 lr=0.00010 gn=19.79773 time=53.47it/s +epoch=88 global_step=34650 loss=1.95024 loss_avg=2.03604 acc=0.82031 acc_top1_avg=0.81218 acc_top5_avg=0.93750 lr=0.00010 gn=25.30346 time=59.58it/s +epoch=88 global_step=34700 loss=1.97567 loss_avg=2.02935 acc=0.81250 acc_top1_avg=0.81263 acc_top5_avg=0.93782 lr=0.00010 gn=19.78593 time=61.35it/s +epoch=88 global_step=34750 loss=2.31210 loss_avg=2.01887 acc=0.78125 acc_top1_avg=0.81403 acc_top5_avg=0.93830 lr=0.00010 gn=16.53166 time=58.78it/s +====================Eval==================== +epoch=88 global_step=34799 loss=0.27538 test_loss_avg=0.50246 acc=0.92969 test_acc_avg=0.86035 test_acc_top5_avg=0.98893 time=235.93it/s +epoch=88 global_step=34799 loss=0.25584 test_loss_avg=0.38091 acc=0.93750 test_acc_avg=0.89250 test_acc_top5_avg=0.99229 time=545.28it/s +curr_acc 0.8925 +BEST_ACC 0.8976 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=1.91603 loss_avg=1.91603 acc=0.82031 acc_top1_avg=0.82031 acc_top5_avg=0.96094 lr=0.00010 gn=19.75311 time=49.58it/s +epoch=89 global_step=34850 loss=2.16501 loss_avg=2.00753 acc=0.78906 acc_top1_avg=0.81587 acc_top5_avg=0.93781 lr=0.00010 gn=17.99740 time=55.99it/s +epoch=89 global_step=34900 loss=2.24735 loss_avg=2.00988 acc=0.80469 acc_top1_avg=0.81498 acc_top5_avg=0.93928 lr=0.00010 gn=30.03063 time=52.09it/s +epoch=89 global_step=34950 loss=1.90167 loss_avg=1.99842 acc=0.82031 acc_top1_avg=0.81602 acc_top5_avg=0.93905 lr=0.00010 gn=20.89013 time=56.50it/s +epoch=89 global_step=35000 loss=2.40414 loss_avg=2.00182 acc=0.78125 acc_top1_avg=0.81557 acc_top5_avg=0.93820 lr=0.00010 gn=25.40974 time=53.39it/s +epoch=89 global_step=35050 loss=2.08647 loss_avg=2.01832 acc=0.81250 acc_top1_avg=0.81412 acc_top5_avg=0.93865 lr=0.00010 gn=22.78429 time=59.44it/s +epoch=89 global_step=35100 loss=2.45956 loss_avg=2.01185 acc=0.75781 acc_top1_avg=0.81463 acc_top5_avg=0.93880 lr=0.00010 gn=22.06124 time=47.99it/s +epoch=89 global_step=35150 loss=1.97650 loss_avg=2.01507 acc=0.81250 acc_top1_avg=0.81433 acc_top5_avg=0.93946 lr=0.00010 gn=19.80591 time=54.06it/s +====================Eval==================== +epoch=89 global_step=35190 loss=0.71293 test_loss_avg=0.44423 acc=0.81250 test_acc_avg=0.87582 test_acc_top5_avg=0.99013 time=236.93it/s +epoch=89 global_step=35190 loss=0.11996 test_loss_avg=0.39659 acc=0.94531 test_acc_avg=0.89074 test_acc_top5_avg=0.99196 time=230.23it/s +epoch=89 global_step=35190 loss=0.27271 test_loss_avg=0.38384 acc=0.87500 test_acc_avg=0.89221 test_acc_top5_avg=0.99239 time=881.53it/s +curr_acc 0.8922 +BEST_ACC 0.8976 +curr_acc_top5 0.9924 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=1.69711 loss_avg=1.95585 acc=0.84375 acc_top1_avg=0.82188 acc_top5_avg=0.94063 lr=0.00010 gn=20.51787 time=60.43it/s +epoch=90 global_step=35250 loss=2.38138 loss_avg=2.02361 acc=0.77344 acc_top1_avg=0.81172 acc_top5_avg=0.94180 lr=0.00010 gn=22.21180 time=61.98it/s +epoch=90 global_step=35300 loss=2.97805 loss_avg=2.01451 acc=0.71094 acc_top1_avg=0.81335 acc_top5_avg=0.94091 lr=0.00010 gn=20.91526 time=55.75it/s +epoch=90 global_step=35350 loss=1.87430 loss_avg=2.04033 acc=0.84375 acc_top1_avg=0.81113 acc_top5_avg=0.93906 lr=0.00010 gn=25.89823 time=56.35it/s +epoch=90 global_step=35400 loss=1.58906 loss_avg=2.03778 acc=0.86719 acc_top1_avg=0.81172 acc_top5_avg=0.93914 lr=0.00010 gn=24.41790 time=54.76it/s +epoch=90 global_step=35450 loss=1.90197 loss_avg=2.02023 acc=0.82031 acc_top1_avg=0.81349 acc_top5_avg=0.93867 lr=0.00010 gn=19.17438 time=56.60it/s +epoch=90 global_step=35500 loss=1.71365 loss_avg=2.01848 acc=0.84375 acc_top1_avg=0.81363 acc_top5_avg=0.93866 lr=0.00010 gn=15.08817 time=58.24it/s +epoch=90 global_step=35550 loss=1.97528 loss_avg=2.01890 acc=0.82031 acc_top1_avg=0.81359 acc_top5_avg=0.93924 lr=0.00010 gn=20.64912 time=50.83it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.63723 test_loss_avg=0.48162 acc=0.84375 test_acc_avg=0.86426 test_acc_top5_avg=0.98926 time=232.50it/s +epoch=90 global_step=35581 loss=0.32659 test_loss_avg=0.38354 acc=0.87500 test_acc_avg=0.89062 test_acc_top5_avg=0.99229 time=536.22it/s +curr_acc 0.8906 +BEST_ACC 0.8976 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=1.50550 loss_avg=1.90973 acc=0.86719 acc_top1_avg=0.82360 acc_top5_avg=0.93791 lr=0.00010 gn=18.69248 time=58.98it/s +epoch=91 global_step=35650 loss=1.97928 loss_avg=2.00237 acc=0.81250 acc_top1_avg=0.81499 acc_top5_avg=0.93682 lr=0.00010 gn=24.95722 time=53.70it/s +epoch=91 global_step=35700 loss=1.80624 loss_avg=2.03991 acc=0.83594 acc_top1_avg=0.81145 acc_top5_avg=0.93501 lr=0.00010 gn=21.45346 time=61.53it/s +epoch=91 global_step=35750 loss=1.71297 loss_avg=2.01760 acc=0.85156 acc_top1_avg=0.81412 acc_top5_avg=0.93658 lr=0.00010 gn=26.75372 time=53.10it/s +epoch=91 global_step=35800 loss=1.88928 loss_avg=2.01374 acc=0.82031 acc_top1_avg=0.81418 acc_top5_avg=0.93746 lr=0.00010 gn=18.18674 time=53.91it/s +epoch=91 global_step=35850 loss=1.46610 loss_avg=2.00913 acc=0.86719 acc_top1_avg=0.81482 acc_top5_avg=0.93834 lr=0.00010 gn=15.25457 time=57.92it/s +epoch=91 global_step=35900 loss=2.01608 loss_avg=2.00675 acc=0.81250 acc_top1_avg=0.81507 acc_top5_avg=0.93819 lr=0.00010 gn=18.41163 time=62.56it/s +epoch=91 global_step=35950 loss=2.37704 loss_avg=2.01107 acc=0.76562 acc_top1_avg=0.81466 acc_top5_avg=0.93839 lr=0.00010 gn=19.86857 time=56.94it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.13932 test_loss_avg=0.47163 acc=0.96875 test_acc_avg=0.85795 test_acc_top5_avg=0.98935 time=242.54it/s +epoch=91 global_step=35972 loss=0.12717 test_loss_avg=0.42772 acc=0.97656 test_acc_avg=0.87859 test_acc_top5_avg=0.99065 time=237.87it/s +epoch=91 global_step=35972 loss=0.31750 test_loss_avg=0.38782 acc=0.87500 test_acc_avg=0.88855 test_acc_top5_avg=0.99169 time=875.82it/s +curr_acc 0.8885 +BEST_ACC 0.8976 +curr_acc_top5 0.9917 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=1.95684 loss_avg=1.95125 acc=0.81250 acc_top1_avg=0.82003 acc_top5_avg=0.93834 lr=0.00010 gn=22.48585 time=52.28it/s +epoch=92 global_step=36050 loss=1.77488 loss_avg=1.96274 acc=0.83594 acc_top1_avg=0.81961 acc_top5_avg=0.93910 lr=0.00010 gn=17.37022 time=56.45it/s +epoch=92 global_step=36100 loss=1.62263 loss_avg=1.98547 acc=0.85156 acc_top1_avg=0.81744 acc_top5_avg=0.93951 lr=0.00010 gn=17.30008 time=51.51it/s +epoch=92 global_step=36150 loss=2.31916 loss_avg=1.99835 acc=0.78906 acc_top1_avg=0.81597 acc_top5_avg=0.93847 lr=0.00010 gn=27.56848 time=57.39it/s +epoch=92 global_step=36200 loss=2.10114 loss_avg=1.99397 acc=0.80469 acc_top1_avg=0.81695 acc_top5_avg=0.93976 lr=0.00010 gn=19.41857 time=57.63it/s +epoch=92 global_step=36250 loss=1.84781 loss_avg=1.99474 acc=0.84375 acc_top1_avg=0.81677 acc_top5_avg=0.93944 lr=0.00010 gn=23.50499 time=50.62it/s +epoch=92 global_step=36300 loss=2.02520 loss_avg=1.99977 acc=0.80469 acc_top1_avg=0.81617 acc_top5_avg=0.93931 lr=0.00010 gn=24.51371 time=57.59it/s +epoch=92 global_step=36350 loss=1.61972 loss_avg=1.99936 acc=0.85156 acc_top1_avg=0.81618 acc_top5_avg=0.93899 lr=0.00010 gn=26.72458 time=57.77it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.23423 test_loss_avg=0.53353 acc=0.89844 test_acc_avg=0.84741 test_acc_top5_avg=0.98682 time=246.19it/s +epoch=92 global_step=36363 loss=0.30221 test_loss_avg=0.38319 acc=0.87500 test_acc_avg=0.89152 test_acc_top5_avg=0.99209 time=845.28it/s +curr_acc 0.8915 +BEST_ACC 0.8976 +curr_acc_top5 0.9921 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=2.20225 loss_avg=2.11449 acc=0.78906 acc_top1_avg=0.80089 acc_top5_avg=0.94341 lr=0.00010 gn=22.24036 time=45.13it/s +epoch=93 global_step=36450 loss=1.60177 loss_avg=2.04291 acc=0.85156 acc_top1_avg=0.80963 acc_top5_avg=0.93894 lr=0.00010 gn=18.40445 time=53.97it/s +epoch=93 global_step=36500 loss=2.24393 loss_avg=2.05516 acc=0.78906 acc_top1_avg=0.80874 acc_top5_avg=0.93910 lr=0.00010 gn=21.79837 time=57.97it/s +epoch=93 global_step=36550 loss=1.53078 loss_avg=2.05167 acc=0.87500 acc_top1_avg=0.80995 acc_top5_avg=0.94005 lr=0.00010 gn=26.37144 time=53.15it/s +epoch=93 global_step=36600 loss=1.41915 loss_avg=2.02867 acc=0.88281 acc_top1_avg=0.81257 acc_top5_avg=0.93997 lr=0.00010 gn=22.27380 time=50.90it/s +epoch=93 global_step=36650 loss=1.46097 loss_avg=2.02055 acc=0.86719 acc_top1_avg=0.81332 acc_top5_avg=0.93970 lr=0.00010 gn=19.48669 time=56.53it/s +epoch=93 global_step=36700 loss=1.84824 loss_avg=2.01942 acc=0.82812 acc_top1_avg=0.81359 acc_top5_avg=0.93908 lr=0.00010 gn=25.25954 time=52.41it/s +epoch=93 global_step=36750 loss=1.88569 loss_avg=1.99980 acc=0.82812 acc_top1_avg=0.81575 acc_top5_avg=0.93966 lr=0.00010 gn=17.12795 time=57.56it/s +====================Eval==================== +epoch=93 global_step=36754 loss=0.56473 test_loss_avg=0.69790 acc=0.82812 test_acc_avg=0.78385 test_acc_top5_avg=0.98958 time=259.32it/s +epoch=93 global_step=36754 loss=0.09179 test_loss_avg=0.45935 acc=0.97656 test_acc_avg=0.87426 test_acc_top5_avg=0.98968 time=246.77it/s +epoch=93 global_step=36754 loss=0.26290 test_loss_avg=0.38337 acc=0.87500 test_acc_avg=0.89201 test_acc_top5_avg=0.99219 time=869.29it/s +curr_acc 0.8920 +BEST_ACC 0.8976 +curr_acc_top5 0.9922 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=1.75433 loss_avg=1.99797 acc=0.83594 acc_top1_avg=0.81793 acc_top5_avg=0.93682 lr=0.00010 gn=15.68661 time=57.02it/s +epoch=94 global_step=36850 loss=1.59920 loss_avg=1.96538 acc=0.85938 acc_top1_avg=0.82015 acc_top5_avg=0.93848 lr=0.00010 gn=27.22843 time=60.59it/s +epoch=94 global_step=36900 loss=1.91069 loss_avg=1.95987 acc=0.82812 acc_top1_avg=0.82122 acc_top5_avg=0.94012 lr=0.00010 gn=20.06100 time=59.98it/s +epoch=94 global_step=36950 loss=2.07501 loss_avg=1.97263 acc=0.80469 acc_top1_avg=0.81956 acc_top5_avg=0.93937 lr=0.00010 gn=17.67037 time=57.73it/s +epoch=94 global_step=37000 loss=2.33088 loss_avg=1.99838 acc=0.78906 acc_top1_avg=0.81685 acc_top5_avg=0.93826 lr=0.00010 gn=23.89406 time=56.11it/s +epoch=94 global_step=37050 loss=2.01239 loss_avg=1.99890 acc=0.82812 acc_top1_avg=0.81646 acc_top5_avg=0.93858 lr=0.00010 gn=25.89371 time=55.14it/s +epoch=94 global_step=37100 loss=1.26193 loss_avg=1.99964 acc=0.89844 acc_top1_avg=0.81643 acc_top5_avg=0.93870 lr=0.00010 gn=21.00789 time=52.99it/s +====================Eval==================== +epoch=94 global_step=37145 loss=0.83679 test_loss_avg=0.54959 acc=0.77344 test_acc_avg=0.84245 test_acc_top5_avg=0.98828 time=247.12it/s +epoch=94 global_step=37145 loss=0.30105 test_loss_avg=0.39752 acc=0.91406 test_acc_avg=0.88841 test_acc_top5_avg=0.99050 time=243.08it/s +epoch=94 global_step=37145 loss=0.35509 test_loss_avg=0.39480 acc=0.87500 test_acc_avg=0.88805 test_acc_top5_avg=0.99090 time=886.93it/s +curr_acc 0.8881 +BEST_ACC 0.8976 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=2.64815 loss_avg=2.10309 acc=0.74219 acc_top1_avg=0.80156 acc_top5_avg=0.92344 lr=0.00010 gn=17.25182 time=59.50it/s +epoch=95 global_step=37200 loss=1.77338 loss_avg=1.98429 acc=0.85156 acc_top1_avg=0.81648 acc_top5_avg=0.94020 lr=0.00010 gn=28.93540 time=53.00it/s +epoch=95 global_step=37250 loss=1.72412 loss_avg=2.01110 acc=0.85938 acc_top1_avg=0.81324 acc_top5_avg=0.93787 lr=0.00010 gn=29.47531 time=60.93it/s +epoch=95 global_step=37300 loss=2.42951 loss_avg=2.00315 acc=0.75000 acc_top1_avg=0.81467 acc_top5_avg=0.93866 lr=0.00010 gn=14.25708 time=57.40it/s +epoch=95 global_step=37350 loss=1.86928 loss_avg=1.99484 acc=0.83594 acc_top1_avg=0.81593 acc_top5_avg=0.93963 lr=0.00010 gn=20.87119 time=52.37it/s +epoch=95 global_step=37400 loss=2.20158 loss_avg=1.98978 acc=0.79688 acc_top1_avg=0.81679 acc_top5_avg=0.93912 lr=0.00010 gn=19.61335 time=59.81it/s +epoch=95 global_step=37450 loss=2.11432 loss_avg=1.99164 acc=0.80469 acc_top1_avg=0.81668 acc_top5_avg=0.93914 lr=0.00010 gn=27.21261 time=56.61it/s +epoch=95 global_step=37500 loss=1.86405 loss_avg=1.98308 acc=0.83594 acc_top1_avg=0.81767 acc_top5_avg=0.93952 lr=0.00010 gn=21.53210 time=50.60it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.47683 test_loss_avg=0.50022 acc=0.88281 test_acc_avg=0.85920 test_acc_top5_avg=0.98837 time=235.07it/s +epoch=95 global_step=37536 loss=0.33874 test_loss_avg=0.39001 acc=0.87500 test_acc_avg=0.88766 test_acc_top5_avg=0.99209 time=542.88it/s +curr_acc 0.8877 +BEST_ACC 0.8976 +curr_acc_top5 0.9921 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=2.04184 loss_avg=1.88761 acc=0.81250 acc_top1_avg=0.82589 acc_top5_avg=0.94364 lr=0.00010 gn=21.17427 time=52.35it/s +epoch=96 global_step=37600 loss=2.44693 loss_avg=1.93043 acc=0.76562 acc_top1_avg=0.82361 acc_top5_avg=0.94397 lr=0.00010 gn=20.48612 time=55.39it/s +epoch=96 global_step=37650 loss=2.13885 loss_avg=1.98476 acc=0.80469 acc_top1_avg=0.81737 acc_top5_avg=0.93997 lr=0.00010 gn=25.04396 time=54.66it/s +epoch=96 global_step=37700 loss=2.36956 loss_avg=1.96101 acc=0.76562 acc_top1_avg=0.81950 acc_top5_avg=0.94203 lr=0.00010 gn=21.58258 time=50.56it/s +epoch=96 global_step=37750 loss=2.22213 loss_avg=1.96925 acc=0.79688 acc_top1_avg=0.81856 acc_top5_avg=0.94100 lr=0.00010 gn=23.24159 time=54.83it/s +epoch=96 global_step=37800 loss=2.36374 loss_avg=1.96747 acc=0.77344 acc_top1_avg=0.81860 acc_top5_avg=0.94040 lr=0.00010 gn=15.24882 time=57.17it/s +epoch=96 global_step=37850 loss=1.77712 loss_avg=1.97570 acc=0.84375 acc_top1_avg=0.81812 acc_top5_avg=0.94024 lr=0.00010 gn=23.25887 time=55.56it/s +epoch=96 global_step=37900 loss=1.72933 loss_avg=1.98207 acc=0.84375 acc_top1_avg=0.81754 acc_top5_avg=0.93980 lr=0.00010 gn=19.51782 time=59.42it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.63225 test_loss_avg=0.40720 acc=0.82812 test_acc_avg=0.88086 test_acc_top5_avg=0.99121 time=235.00it/s +epoch=96 global_step=37927 loss=0.10445 test_loss_avg=0.41671 acc=0.95312 test_acc_avg=0.88281 test_acc_top5_avg=0.99077 time=239.11it/s +epoch=96 global_step=37927 loss=0.35826 test_loss_avg=0.39748 acc=0.87500 test_acc_avg=0.88687 test_acc_top5_avg=0.99150 time=543.23it/s +curr_acc 0.8869 +BEST_ACC 0.8976 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=2.32352 loss_avg=1.93970 acc=0.78125 acc_top1_avg=0.82235 acc_top5_avg=0.94192 lr=0.00010 gn=19.20870 time=62.39it/s +epoch=97 global_step=38000 loss=2.73697 loss_avg=1.99310 acc=0.72656 acc_top1_avg=0.81699 acc_top5_avg=0.93836 lr=0.00010 gn=26.85466 time=60.52it/s +epoch=97 global_step=38050 loss=1.81635 loss_avg=1.99687 acc=0.82812 acc_top1_avg=0.81593 acc_top5_avg=0.93788 lr=0.00010 gn=16.03671 time=52.28it/s +epoch=97 global_step=38100 loss=1.46688 loss_avg=1.98639 acc=0.88281 acc_top1_avg=0.81778 acc_top5_avg=0.93885 lr=0.00010 gn=22.37400 time=54.03it/s +epoch=97 global_step=38150 loss=2.02947 loss_avg=1.97628 acc=0.81250 acc_top1_avg=0.81863 acc_top5_avg=0.93946 lr=0.00010 gn=28.63967 time=59.70it/s +epoch=97 global_step=38200 loss=1.69576 loss_avg=1.96767 acc=0.84375 acc_top1_avg=0.81940 acc_top5_avg=0.93956 lr=0.00010 gn=22.21989 time=37.51it/s +epoch=97 global_step=38250 loss=2.07209 loss_avg=1.97754 acc=0.80469 acc_top1_avg=0.81821 acc_top5_avg=0.93963 lr=0.00010 gn=21.20249 time=60.73it/s +epoch=97 global_step=38300 loss=1.91957 loss_avg=1.98422 acc=0.82812 acc_top1_avg=0.81769 acc_top5_avg=0.93974 lr=0.00010 gn=19.22582 time=49.90it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.21143 test_loss_avg=0.50842 acc=0.93750 test_acc_avg=0.85262 test_acc_top5_avg=0.98796 time=244.25it/s +epoch=97 global_step=38318 loss=0.31649 test_loss_avg=0.39828 acc=0.87500 test_acc_avg=0.88647 test_acc_top5_avg=0.99140 time=843.92it/s +curr_acc 0.8865 +BEST_ACC 0.8976 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=2.23981 loss_avg=1.97792 acc=0.77344 acc_top1_avg=0.81885 acc_top5_avg=0.94214 lr=0.00010 gn=27.55748 time=53.26it/s +epoch=98 global_step=38400 loss=2.57943 loss_avg=2.02694 acc=0.75781 acc_top1_avg=0.81374 acc_top5_avg=0.94122 lr=0.00010 gn=22.42660 time=52.66it/s +epoch=98 global_step=38450 loss=1.89576 loss_avg=2.00903 acc=0.82812 acc_top1_avg=0.81469 acc_top5_avg=0.94040 lr=0.00010 gn=29.63164 time=54.24it/s +epoch=98 global_step=38500 loss=2.36181 loss_avg=1.99589 acc=0.77344 acc_top1_avg=0.81645 acc_top5_avg=0.94020 lr=0.00010 gn=19.07843 time=52.51it/s +epoch=98 global_step=38550 loss=2.08859 loss_avg=1.98704 acc=0.80469 acc_top1_avg=0.81735 acc_top5_avg=0.93982 lr=0.00010 gn=22.98799 time=58.42it/s +epoch=98 global_step=38600 loss=1.89541 loss_avg=1.98692 acc=0.82812 acc_top1_avg=0.81735 acc_top5_avg=0.93902 lr=0.00010 gn=21.13497 time=59.97it/s +epoch=98 global_step=38650 loss=2.15684 loss_avg=1.98143 acc=0.78906 acc_top1_avg=0.81768 acc_top5_avg=0.93945 lr=0.00010 gn=17.21430 time=54.26it/s +epoch=98 global_step=38700 loss=2.24157 loss_avg=1.97785 acc=0.78125 acc_top1_avg=0.81812 acc_top5_avg=0.93952 lr=0.00010 gn=18.57890 time=61.48it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.54215 test_loss_avg=0.68603 acc=0.84375 test_acc_avg=0.79004 test_acc_top5_avg=0.98730 time=230.51it/s +epoch=98 global_step=38709 loss=0.15635 test_loss_avg=0.45336 acc=0.95312 test_acc_avg=0.87392 test_acc_top5_avg=0.98922 time=232.75it/s +epoch=98 global_step=38709 loss=0.34845 test_loss_avg=0.39834 acc=0.87500 test_acc_avg=0.88697 test_acc_top5_avg=0.99130 time=863.91it/s +curr_acc 0.8870 +BEST_ACC 0.8976 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=2.17566 loss_avg=2.05000 acc=0.79688 acc_top1_avg=0.81021 acc_top5_avg=0.93693 lr=0.00010 gn=20.64950 time=61.12it/s +epoch=99 global_step=38800 loss=1.94253 loss_avg=1.98891 acc=0.82031 acc_top1_avg=0.81671 acc_top5_avg=0.93982 lr=0.00010 gn=18.73271 time=60.50it/s +epoch=99 global_step=38850 loss=2.54037 loss_avg=1.96237 acc=0.75781 acc_top1_avg=0.81976 acc_top5_avg=0.93955 lr=0.00010 gn=31.50675 time=55.05it/s +epoch=99 global_step=38900 loss=2.38228 loss_avg=1.98143 acc=0.78125 acc_top1_avg=0.81798 acc_top5_avg=0.93905 lr=0.00010 gn=23.76695 time=60.52it/s +epoch=99 global_step=38950 loss=1.99178 loss_avg=1.98132 acc=0.82031 acc_top1_avg=0.81830 acc_top5_avg=0.93990 lr=0.00010 gn=22.14813 time=47.78it/s +epoch=99 global_step=39000 loss=2.45011 loss_avg=1.98470 acc=0.78125 acc_top1_avg=0.81779 acc_top5_avg=0.93959 lr=0.00010 gn=25.67905 time=51.28it/s +epoch=99 global_step=39050 loss=1.97358 loss_avg=1.98026 acc=0.81250 acc_top1_avg=0.81816 acc_top5_avg=0.93961 lr=0.00010 gn=24.06510 time=50.28it/s +epoch=99 global_step=39100 loss=2.80178 loss_avg=1.97585 acc=0.72500 acc_top1_avg=0.81871 acc_top5_avg=0.94032 lr=0.00010 gn=14.60460 time=71.93it/s +====================Eval==================== +epoch=99 global_step=39100 loss=0.57901 test_loss_avg=0.58124 acc=0.84375 test_acc_avg=0.83405 test_acc_top5_avg=0.98653 time=245.96it/s +epoch=99 global_step=39100 loss=0.38378 test_loss_avg=0.41268 acc=0.87500 test_acc_avg=0.88202 test_acc_top5_avg=0.99140 time=860.72it/s +epoch=99 global_step=39100 loss=0.38378 test_loss_avg=0.41268 acc=0.87500 test_acc_avg=0.88202 test_acc_top5_avg=0.99140 time=860.72it/s +curr_acc 0.8820 +BEST_ACC 0.8976 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=1.70550 loss_avg=1.89443 acc=0.85156 acc_top1_avg=0.82812 acc_top5_avg=0.94516 lr=0.00010 gn=28.27412 time=60.48it/s +epoch=100 global_step=39200 loss=1.66088 loss_avg=1.91742 acc=0.84375 acc_top1_avg=0.82492 acc_top5_avg=0.94141 lr=0.00010 gn=20.49788 time=52.61it/s +epoch=100 global_step=39250 loss=1.87832 loss_avg=1.94135 acc=0.82812 acc_top1_avg=0.82276 acc_top5_avg=0.94057 lr=0.00010 gn=22.27778 time=52.48it/s +epoch=100 global_step=39300 loss=1.86326 loss_avg=1.93916 acc=0.82812 acc_top1_avg=0.82250 acc_top5_avg=0.94184 lr=0.00010 gn=20.22369 time=59.71it/s +epoch=100 global_step=39350 loss=2.35701 loss_avg=1.94440 acc=0.76562 acc_top1_avg=0.82203 acc_top5_avg=0.94169 lr=0.00010 gn=19.31809 time=54.12it/s +epoch=100 global_step=39400 loss=2.15283 loss_avg=1.96462 acc=0.80469 acc_top1_avg=0.81982 acc_top5_avg=0.94089 lr=0.00010 gn=31.94956 time=51.41it/s +epoch=100 global_step=39450 loss=2.14910 loss_avg=1.96426 acc=0.80469 acc_top1_avg=0.81978 acc_top5_avg=0.94092 lr=0.00010 gn=22.88789 time=54.50it/s +====================Eval==================== +epoch=100 global_step=39491 loss=0.14960 test_loss_avg=0.49720 acc=0.96875 test_acc_avg=0.85859 test_acc_top5_avg=0.98813 time=223.80it/s +epoch=100 global_step=39491 loss=0.40475 test_loss_avg=0.40312 acc=0.87500 test_acc_avg=0.88370 test_acc_top5_avg=0.99150 time=885.43it/s +curr_acc 0.8837 +BEST_ACC 0.8976 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=2.11049 loss_avg=2.02082 acc=0.80469 acc_top1_avg=0.81510 acc_top5_avg=0.94184 lr=0.00010 gn=27.78585 time=59.82it/s +epoch=101 global_step=39550 loss=2.05546 loss_avg=1.99693 acc=0.80469 acc_top1_avg=0.81687 acc_top5_avg=0.93909 lr=0.00010 gn=27.43879 time=59.86it/s +epoch=101 global_step=39600 loss=2.35828 loss_avg=1.98154 acc=0.78125 acc_top1_avg=0.81859 acc_top5_avg=0.94044 lr=0.00010 gn=20.95974 time=61.59it/s +epoch=101 global_step=39650 loss=1.76777 loss_avg=1.96842 acc=0.82812 acc_top1_avg=0.82012 acc_top5_avg=0.94015 lr=0.00010 gn=10.96460 time=59.69it/s +epoch=101 global_step=39700 loss=1.64637 loss_avg=1.95956 acc=0.84375 acc_top1_avg=0.82046 acc_top5_avg=0.93959 lr=0.00010 gn=25.44882 time=54.36it/s +epoch=101 global_step=39750 loss=2.06114 loss_avg=1.97338 acc=0.80469 acc_top1_avg=0.81902 acc_top5_avg=0.93937 lr=0.00010 gn=21.81891 time=52.48it/s +epoch=101 global_step=39800 loss=1.77070 loss_avg=1.97239 acc=0.85156 acc_top1_avg=0.81900 acc_top5_avg=0.93975 lr=0.00010 gn=23.34950 time=57.29it/s +epoch=101 global_step=39850 loss=1.61511 loss_avg=1.96695 acc=0.85938 acc_top1_avg=0.81972 acc_top5_avg=0.93992 lr=0.00010 gn=25.96259 time=53.53it/s +====================Eval==================== +epoch=101 global_step=39882 loss=1.02786 test_loss_avg=0.55371 acc=0.73438 test_acc_avg=0.83780 test_acc_top5_avg=0.98698 time=239.35it/s +epoch=101 global_step=39882 loss=0.34144 test_loss_avg=0.41181 acc=0.86719 test_acc_avg=0.88094 test_acc_top5_avg=0.99054 time=242.11it/s +epoch=101 global_step=39882 loss=0.35705 test_loss_avg=0.40853 acc=0.87500 test_acc_avg=0.88074 test_acc_top5_avg=0.99080 time=884.31it/s +curr_acc 0.8807 +BEST_ACC 0.8976 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=1.61828 loss_avg=2.03689 acc=0.85156 acc_top1_avg=0.81163 acc_top5_avg=0.94488 lr=0.00010 gn=16.15124 time=55.08it/s +epoch=102 global_step=39950 loss=2.71356 loss_avg=2.00376 acc=0.75000 acc_top1_avg=0.81572 acc_top5_avg=0.94267 lr=0.00010 gn=22.75569 time=57.83it/s +epoch=102 global_step=40000 loss=2.42675 loss_avg=1.99599 acc=0.77344 acc_top1_avg=0.81647 acc_top5_avg=0.94180 lr=0.00010 gn=24.23189 time=53.16it/s +epoch=102 global_step=40050 loss=1.62733 loss_avg=1.98238 acc=0.84375 acc_top1_avg=0.81794 acc_top5_avg=0.94076 lr=0.00010 gn=16.18549 time=53.55it/s +epoch=102 global_step=40100 loss=1.56268 loss_avg=1.96750 acc=0.85938 acc_top1_avg=0.81913 acc_top5_avg=0.94004 lr=0.00010 gn=18.62383 time=49.55it/s +epoch=102 global_step=40150 loss=2.23579 loss_avg=1.95916 acc=0.79688 acc_top1_avg=0.82037 acc_top5_avg=0.94030 lr=0.00010 gn=28.38035 time=54.91it/s +epoch=102 global_step=40200 loss=2.43533 loss_avg=1.96140 acc=0.77344 acc_top1_avg=0.81989 acc_top5_avg=0.94018 lr=0.00010 gn=21.32028 time=55.43it/s +epoch=102 global_step=40250 loss=1.78463 loss_avg=1.96486 acc=0.85156 acc_top1_avg=0.81972 acc_top5_avg=0.94013 lr=0.00010 gn=25.04363 time=54.29it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.55076 test_loss_avg=0.51932 acc=0.85938 test_acc_avg=0.85156 test_acc_top5_avg=0.98717 time=234.91it/s +epoch=102 global_step=40273 loss=0.36404 test_loss_avg=0.40360 acc=0.87500 test_acc_avg=0.88449 test_acc_top5_avg=0.99061 time=550.14it/s +curr_acc 0.8845 +BEST_ACC 0.8976 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=1.57431 loss_avg=1.99384 acc=0.85938 acc_top1_avg=0.81453 acc_top5_avg=0.93750 lr=0.00010 gn=23.29034 time=57.44it/s +epoch=103 global_step=40350 loss=2.05231 loss_avg=1.98165 acc=0.81250 acc_top1_avg=0.81849 acc_top5_avg=0.93811 lr=0.00010 gn=21.70317 time=59.38it/s +epoch=103 global_step=40400 loss=2.57791 loss_avg=1.99343 acc=0.75781 acc_top1_avg=0.81736 acc_top5_avg=0.93787 lr=0.00010 gn=20.28434 time=61.72it/s +epoch=103 global_step=40450 loss=2.49214 loss_avg=1.98318 acc=0.76562 acc_top1_avg=0.81815 acc_top5_avg=0.93816 lr=0.00010 gn=25.88751 time=51.19it/s +epoch=103 global_step=40500 loss=1.97005 loss_avg=1.96471 acc=0.82031 acc_top1_avg=0.82024 acc_top5_avg=0.93843 lr=0.00010 gn=28.57157 time=60.73it/s +epoch=103 global_step=40550 loss=2.00776 loss_avg=1.95895 acc=0.80469 acc_top1_avg=0.82071 acc_top5_avg=0.93897 lr=0.00010 gn=21.93782 time=54.31it/s +epoch=103 global_step=40600 loss=1.37366 loss_avg=1.96149 acc=0.88281 acc_top1_avg=0.82053 acc_top5_avg=0.93979 lr=0.00010 gn=22.34042 time=54.37it/s +epoch=103 global_step=40650 loss=2.26392 loss_avg=1.96251 acc=0.80469 acc_top1_avg=0.82031 acc_top5_avg=0.93947 lr=0.00010 gn=26.58220 time=52.04it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.07047 test_loss_avg=0.55026 acc=0.98438 test_acc_avg=0.83233 test_acc_top5_avg=0.98618 time=234.71it/s +epoch=103 global_step=40664 loss=0.09830 test_loss_avg=0.45629 acc=0.96875 test_acc_avg=0.86855 test_acc_top5_avg=0.98834 time=236.74it/s +epoch=103 global_step=40664 loss=0.30068 test_loss_avg=0.41647 acc=0.93750 test_acc_avg=0.87975 test_acc_top5_avg=0.98962 time=874.00it/s +curr_acc 0.8797 +BEST_ACC 0.8976 +curr_acc_top5 0.9896 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=2.34748 loss_avg=2.00174 acc=0.77344 acc_top1_avg=0.81597 acc_top5_avg=0.93273 lr=0.00010 gn=16.23459 time=53.94it/s +epoch=104 global_step=40750 loss=2.05241 loss_avg=1.97888 acc=0.80469 acc_top1_avg=0.81913 acc_top5_avg=0.93596 lr=0.00010 gn=23.39833 time=55.11it/s +epoch=104 global_step=40800 loss=1.83892 loss_avg=1.97389 acc=0.82812 acc_top1_avg=0.81974 acc_top5_avg=0.93894 lr=0.00010 gn=18.50038 time=58.46it/s +epoch=104 global_step=40850 loss=2.35315 loss_avg=1.96468 acc=0.77344 acc_top1_avg=0.81998 acc_top5_avg=0.94002 lr=0.00010 gn=19.35083 time=60.44it/s +epoch=104 global_step=40900 loss=1.81769 loss_avg=1.96749 acc=0.83594 acc_top1_avg=0.81948 acc_top5_avg=0.93972 lr=0.00010 gn=26.87197 time=59.35it/s +epoch=104 global_step=40950 loss=1.96923 loss_avg=1.94603 acc=0.82031 acc_top1_avg=0.82171 acc_top5_avg=0.94004 lr=0.00010 gn=19.76378 time=57.30it/s +epoch=104 global_step=41000 loss=2.23651 loss_avg=1.94914 acc=0.78906 acc_top1_avg=0.82150 acc_top5_avg=0.94071 lr=0.00010 gn=18.73825 time=60.35it/s +epoch=104 global_step=41050 loss=1.75310 loss_avg=1.95948 acc=0.84375 acc_top1_avg=0.82035 acc_top5_avg=0.94019 lr=0.00010 gn=25.86560 time=47.87it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.30548 test_loss_avg=0.55363 acc=0.90625 test_acc_avg=0.83938 test_acc_top5_avg=0.98690 time=237.64it/s +epoch=104 global_step=41055 loss=0.31521 test_loss_avg=0.40515 acc=0.87500 test_acc_avg=0.88252 test_acc_top5_avg=0.99090 time=891.08it/s +curr_acc 0.8825 +BEST_ACC 0.8976 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=1.61581 loss_avg=1.96885 acc=0.85938 acc_top1_avg=0.81910 acc_top5_avg=0.93576 lr=0.00010 gn=20.50507 time=51.24it/s +epoch=105 global_step=41150 loss=1.46764 loss_avg=1.96154 acc=0.88281 acc_top1_avg=0.82015 acc_top5_avg=0.93857 lr=0.00010 gn=26.60748 time=51.58it/s +epoch=105 global_step=41200 loss=1.62347 loss_avg=1.96591 acc=0.84375 acc_top1_avg=0.81956 acc_top5_avg=0.93852 lr=0.00010 gn=17.70293 time=54.08it/s +epoch=105 global_step=41250 loss=1.98419 loss_avg=1.96248 acc=0.81250 acc_top1_avg=0.81991 acc_top5_avg=0.93854 lr=0.00010 gn=21.93132 time=55.14it/s +epoch=105 global_step=41300 loss=1.91897 loss_avg=1.95555 acc=0.82812 acc_top1_avg=0.82054 acc_top5_avg=0.93897 lr=0.00010 gn=23.17679 time=58.79it/s +epoch=105 global_step=41350 loss=1.97050 loss_avg=1.96036 acc=0.82031 acc_top1_avg=0.81989 acc_top5_avg=0.93869 lr=0.00010 gn=22.45555 time=54.29it/s +epoch=105 global_step=41400 loss=1.92604 loss_avg=1.95076 acc=0.82031 acc_top1_avg=0.82074 acc_top5_avg=0.93947 lr=0.00010 gn=18.80384 time=60.54it/s +====================Eval==================== +epoch=105 global_step=41446 loss=0.60330 test_loss_avg=0.79353 acc=0.77344 test_acc_avg=0.75313 test_acc_top5_avg=0.98281 time=262.16it/s +epoch=105 global_step=41446 loss=0.32892 test_loss_avg=0.48576 acc=0.91406 test_acc_avg=0.86506 test_acc_top5_avg=0.98849 time=238.91it/s +epoch=105 global_step=41446 loss=0.30112 test_loss_avg=0.40886 acc=0.93750 test_acc_avg=0.88479 test_acc_top5_avg=0.99130 time=819.84it/s +curr_acc 0.8848 +BEST_ACC 0.8976 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=1.63408 loss_avg=1.93386 acc=0.84375 acc_top1_avg=0.82227 acc_top5_avg=0.94922 lr=0.00010 gn=17.66467 time=54.21it/s +epoch=106 global_step=41500 loss=1.84108 loss_avg=1.93922 acc=0.83594 acc_top1_avg=0.82133 acc_top5_avg=0.93764 lr=0.00010 gn=25.10238 time=56.82it/s +epoch=106 global_step=41550 loss=2.24292 loss_avg=1.92922 acc=0.78906 acc_top1_avg=0.82309 acc_top5_avg=0.94020 lr=0.00010 gn=24.63314 time=54.14it/s +epoch=106 global_step=41600 loss=1.55597 loss_avg=1.93564 acc=0.86719 acc_top1_avg=0.82315 acc_top5_avg=0.94014 lr=0.00010 gn=34.14705 time=53.43it/s +epoch=106 global_step=41650 loss=2.17353 loss_avg=1.96106 acc=0.79688 acc_top1_avg=0.81989 acc_top5_avg=0.93842 lr=0.00010 gn=27.87653 time=60.72it/s +epoch=106 global_step=41700 loss=1.93091 loss_avg=1.96302 acc=0.83594 acc_top1_avg=0.81979 acc_top5_avg=0.94011 lr=0.00010 gn=30.28851 time=60.95it/s +epoch=106 global_step=41750 loss=1.95717 loss_avg=1.95514 acc=0.82031 acc_top1_avg=0.82049 acc_top5_avg=0.94010 lr=0.00010 gn=20.86491 time=61.30it/s +epoch=106 global_step=41800 loss=1.69559 loss_avg=1.95094 acc=0.84375 acc_top1_avg=0.82075 acc_top5_avg=0.94010 lr=0.00010 gn=20.74559 time=53.37it/s +====================Eval==================== +epoch=106 global_step=41837 loss=0.61689 test_loss_avg=0.59285 acc=0.83594 test_acc_avg=0.82812 test_acc_top5_avg=0.98618 time=221.56it/s +epoch=106 global_step=41837 loss=0.30256 test_loss_avg=0.41284 acc=0.88281 test_acc_avg=0.88127 test_acc_top5_avg=0.99023 time=241.93it/s +epoch=106 global_step=41837 loss=0.34119 test_loss_avg=0.40985 acc=0.87500 test_acc_avg=0.88143 test_acc_top5_avg=0.99061 time=624.99it/s +curr_acc 0.8814 +BEST_ACC 0.8976 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=1.68425 loss_avg=1.88101 acc=0.85156 acc_top1_avg=0.82752 acc_top5_avg=0.94291 lr=0.00010 gn=20.75909 time=51.68it/s +epoch=107 global_step=41900 loss=1.54808 loss_avg=1.93079 acc=0.85938 acc_top1_avg=0.82192 acc_top5_avg=0.93924 lr=0.00010 gn=17.11170 time=54.65it/s +epoch=107 global_step=41950 loss=2.05629 loss_avg=1.90707 acc=0.81250 acc_top1_avg=0.82439 acc_top5_avg=0.94116 lr=0.00010 gn=23.61158 time=57.20it/s +epoch=107 global_step=42000 loss=2.38861 loss_avg=1.92427 acc=0.77344 acc_top1_avg=0.82295 acc_top5_avg=0.94081 lr=0.00010 gn=18.33438 time=60.10it/s +epoch=107 global_step=42050 loss=1.93346 loss_avg=1.93459 acc=0.82812 acc_top1_avg=0.82229 acc_top5_avg=0.93955 lr=0.00010 gn=24.13120 time=53.27it/s +epoch=107 global_step=42100 loss=1.54071 loss_avg=1.93611 acc=0.86719 acc_top1_avg=0.82218 acc_top5_avg=0.93937 lr=0.00010 gn=21.67620 time=57.04it/s +epoch=107 global_step=42150 loss=1.75745 loss_avg=1.94045 acc=0.83594 acc_top1_avg=0.82181 acc_top5_avg=0.93942 lr=0.00010 gn=19.31212 time=61.02it/s +epoch=107 global_step=42200 loss=2.29387 loss_avg=1.94209 acc=0.78125 acc_top1_avg=0.82143 acc_top5_avg=0.93985 lr=0.00010 gn=21.24031 time=53.60it/s +====================Eval==================== +epoch=107 global_step=42228 loss=0.47018 test_loss_avg=0.56217 acc=0.85156 test_acc_avg=0.84026 test_acc_top5_avg=0.98570 time=239.18it/s +epoch=107 global_step=42228 loss=0.37449 test_loss_avg=0.42489 acc=0.87500 test_acc_avg=0.87777 test_acc_top5_avg=0.99031 time=500.81it/s +curr_acc 0.8778 +BEST_ACC 0.8976 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=1.96650 loss_avg=1.77673 acc=0.81250 acc_top1_avg=0.84055 acc_top5_avg=0.95312 lr=0.00010 gn=20.13821 time=54.72it/s +epoch=108 global_step=42300 loss=2.54730 loss_avg=1.95231 acc=0.77344 acc_top1_avg=0.82053 acc_top5_avg=0.94173 lr=0.00010 gn=27.48726 time=53.93it/s +epoch=108 global_step=42350 loss=2.22078 loss_avg=1.95901 acc=0.81250 acc_top1_avg=0.82012 acc_top5_avg=0.94077 lr=0.00010 gn=27.36505 time=54.69it/s +epoch=108 global_step=42400 loss=2.21531 loss_avg=1.95509 acc=0.79688 acc_top1_avg=0.82040 acc_top5_avg=0.94086 lr=0.00010 gn=19.58498 time=51.74it/s +epoch=108 global_step=42450 loss=2.47501 loss_avg=1.94734 acc=0.77344 acc_top1_avg=0.82119 acc_top5_avg=0.94049 lr=0.00010 gn=32.55631 time=61.13it/s +epoch=108 global_step=42500 loss=1.77408 loss_avg=1.94497 acc=0.83594 acc_top1_avg=0.82109 acc_top5_avg=0.93974 lr=0.00010 gn=17.87251 time=51.01it/s +epoch=108 global_step=42550 loss=2.00358 loss_avg=1.94326 acc=0.81250 acc_top1_avg=0.82126 acc_top5_avg=0.93932 lr=0.00010 gn=24.31399 time=55.74it/s +epoch=108 global_step=42600 loss=1.84954 loss_avg=1.94271 acc=0.83594 acc_top1_avg=0.82117 acc_top5_avg=0.93945 lr=0.00010 gn=26.24757 time=52.66it/s +====================Eval==================== +epoch=108 global_step=42619 loss=0.88569 test_loss_avg=0.48257 acc=0.77344 test_acc_avg=0.86198 test_acc_top5_avg=0.98785 time=241.37it/s +epoch=108 global_step=42619 loss=0.24321 test_loss_avg=0.41854 acc=0.95312 test_acc_avg=0.88109 test_acc_top5_avg=0.99023 time=237.84it/s +epoch=108 global_step=42619 loss=0.34345 test_loss_avg=0.40124 acc=0.87500 test_acc_avg=0.88469 test_acc_top5_avg=0.99110 time=790.93it/s +curr_acc 0.8847 +BEST_ACC 0.8976 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=2.07444 loss_avg=1.99963 acc=0.81250 acc_top1_avg=0.81678 acc_top5_avg=0.94002 lr=0.00010 gn=27.49532 time=56.67it/s +epoch=109 global_step=42700 loss=1.86671 loss_avg=1.96655 acc=0.82812 acc_top1_avg=0.81916 acc_top5_avg=0.94145 lr=0.00010 gn=17.30794 time=60.43it/s +epoch=109 global_step=42750 loss=1.58483 loss_avg=1.92754 acc=0.85156 acc_top1_avg=0.82389 acc_top5_avg=0.94281 lr=0.00010 gn=19.77822 time=54.06it/s +epoch=109 global_step=42800 loss=1.67250 loss_avg=1.92514 acc=0.85156 acc_top1_avg=0.82394 acc_top5_avg=0.94251 lr=0.00010 gn=24.19773 time=57.93it/s +epoch=109 global_step=42850 loss=2.09398 loss_avg=1.93972 acc=0.80469 acc_top1_avg=0.82217 acc_top5_avg=0.94183 lr=0.00010 gn=25.69405 time=56.16it/s +epoch=109 global_step=42900 loss=1.92269 loss_avg=1.94642 acc=0.82812 acc_top1_avg=0.82145 acc_top5_avg=0.94089 lr=0.00010 gn=24.10113 time=58.67it/s +epoch=109 global_step=42950 loss=2.08437 loss_avg=1.94329 acc=0.80469 acc_top1_avg=0.82192 acc_top5_avg=0.94005 lr=0.00010 gn=21.66030 time=55.46it/s +epoch=109 global_step=43000 loss=2.86801 loss_avg=1.94162 acc=0.72656 acc_top1_avg=0.82214 acc_top5_avg=0.94008 lr=0.00010 gn=23.52646 time=56.65it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.19529 test_loss_avg=0.56044 acc=0.96094 test_acc_avg=0.84275 test_acc_top5_avg=0.98738 time=224.20it/s +epoch=109 global_step=43010 loss=0.33833 test_loss_avg=0.42597 acc=0.93750 test_acc_avg=0.87975 test_acc_top5_avg=0.99090 time=398.81it/s +curr_acc 0.8797 +BEST_ACC 0.8976 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=1.70135 loss_avg=1.87555 acc=0.85938 acc_top1_avg=0.83066 acc_top5_avg=0.93594 lr=0.00010 gn=26.69485 time=40.69it/s +epoch=110 global_step=43100 loss=1.31935 loss_avg=1.94432 acc=0.88281 acc_top1_avg=0.82144 acc_top5_avg=0.93646 lr=0.00010 gn=21.48058 time=60.78it/s +epoch=110 global_step=43150 loss=1.89395 loss_avg=1.93362 acc=0.82031 acc_top1_avg=0.82249 acc_top5_avg=0.93772 lr=0.00010 gn=22.07440 time=56.65it/s +epoch=110 global_step=43200 loss=1.91391 loss_avg=1.93886 acc=0.82812 acc_top1_avg=0.82196 acc_top5_avg=0.93865 lr=0.00010 gn=28.33300 time=54.63it/s +epoch=110 global_step=43250 loss=2.41229 loss_avg=1.92630 acc=0.77344 acc_top1_avg=0.82327 acc_top5_avg=0.93893 lr=0.00010 gn=23.11890 time=60.85it/s +epoch=110 global_step=43300 loss=2.12206 loss_avg=1.93807 acc=0.81250 acc_top1_avg=0.82204 acc_top5_avg=0.93879 lr=0.00010 gn=32.14098 time=54.72it/s +epoch=110 global_step=43350 loss=1.84619 loss_avg=1.93712 acc=0.82812 acc_top1_avg=0.82231 acc_top5_avg=0.93892 lr=0.00010 gn=26.59588 time=51.43it/s +epoch=110 global_step=43400 loss=1.68116 loss_avg=1.93569 acc=0.85938 acc_top1_avg=0.82270 acc_top5_avg=0.93950 lr=0.00010 gn=28.20021 time=57.94it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.10567 test_loss_avg=0.68894 acc=0.96094 test_acc_avg=0.79375 test_acc_top5_avg=0.98203 time=238.79it/s +epoch=110 global_step=43401 loss=0.19520 test_loss_avg=0.48003 acc=0.91406 test_acc_avg=0.86406 test_acc_top5_avg=0.98880 time=216.01it/s +epoch=110 global_step=43401 loss=0.40314 test_loss_avg=0.43195 acc=0.87500 test_acc_avg=0.87658 test_acc_top5_avg=0.99061 time=832.86it/s +curr_acc 0.8766 +BEST_ACC 0.8976 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=2.18447 loss_avg=1.96511 acc=0.78906 acc_top1_avg=0.82047 acc_top5_avg=0.93654 lr=0.00010 gn=18.68688 time=55.40it/s +epoch=111 global_step=43500 loss=1.94081 loss_avg=1.96564 acc=0.82031 acc_top1_avg=0.81992 acc_top5_avg=0.93679 lr=0.00010 gn=21.76460 time=51.97it/s +epoch=111 global_step=43550 loss=2.22507 loss_avg=1.95460 acc=0.78906 acc_top1_avg=0.82115 acc_top5_avg=0.93918 lr=0.00010 gn=17.10170 time=55.84it/s +epoch=111 global_step=43600 loss=2.14527 loss_avg=1.93831 acc=0.80469 acc_top1_avg=0.82267 acc_top5_avg=0.93931 lr=0.00010 gn=25.14128 time=55.05it/s +epoch=111 global_step=43650 loss=1.59914 loss_avg=1.94130 acc=0.85938 acc_top1_avg=0.82245 acc_top5_avg=0.93954 lr=0.00010 gn=21.92062 time=52.69it/s +epoch=111 global_step=43700 loss=1.69908 loss_avg=1.94039 acc=0.84375 acc_top1_avg=0.82256 acc_top5_avg=0.93975 lr=0.00010 gn=25.33306 time=59.97it/s +epoch=111 global_step=43750 loss=1.99363 loss_avg=1.93562 acc=0.82031 acc_top1_avg=0.82286 acc_top5_avg=0.94032 lr=0.00010 gn=23.56335 time=53.58it/s +====================Eval==================== +epoch=111 global_step=43792 loss=0.76989 test_loss_avg=0.58575 acc=0.78125 test_acc_avg=0.83342 test_acc_top5_avg=0.98438 time=240.22it/s +epoch=111 global_step=43792 loss=0.34373 test_loss_avg=0.41131 acc=0.93750 test_acc_avg=0.88192 test_acc_top5_avg=0.99031 time=858.43it/s +curr_acc 0.8819 +BEST_ACC 0.8976 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=1.56662 loss_avg=1.86041 acc=0.85156 acc_top1_avg=0.82617 acc_top5_avg=0.95312 lr=0.00010 gn=20.87183 time=62.07it/s +epoch=112 global_step=43850 loss=1.31726 loss_avg=1.91326 acc=0.88281 acc_top1_avg=0.82381 acc_top5_avg=0.94343 lr=0.00010 gn=16.26051 time=59.27it/s +epoch=112 global_step=43900 loss=1.89383 loss_avg=1.89424 acc=0.82812 acc_top1_avg=0.82668 acc_top5_avg=0.94314 lr=0.00010 gn=16.56502 time=53.33it/s +epoch=112 global_step=43950 loss=1.44512 loss_avg=1.92673 acc=0.87500 acc_top1_avg=0.82358 acc_top5_avg=0.94066 lr=0.00010 gn=22.13390 time=51.79it/s +epoch=112 global_step=44000 loss=1.99951 loss_avg=1.92890 acc=0.82031 acc_top1_avg=0.82339 acc_top5_avg=0.94114 lr=0.00010 gn=29.62367 time=57.14it/s +epoch=112 global_step=44050 loss=2.09950 loss_avg=1.93423 acc=0.80469 acc_top1_avg=0.82307 acc_top5_avg=0.94171 lr=0.00010 gn=27.89898 time=55.84it/s +epoch=112 global_step=44100 loss=1.90048 loss_avg=1.92906 acc=0.82812 acc_top1_avg=0.82343 acc_top5_avg=0.94141 lr=0.00010 gn=19.96536 time=53.12it/s +epoch=112 global_step=44150 loss=2.19111 loss_avg=1.92503 acc=0.78906 acc_top1_avg=0.82378 acc_top5_avg=0.94141 lr=0.00010 gn=23.75496 time=55.50it/s +====================Eval==================== +epoch=112 global_step=44183 loss=0.96858 test_loss_avg=0.94008 acc=0.71094 test_acc_avg=0.72266 test_acc_top5_avg=0.98438 time=246.09it/s +epoch=112 global_step=44183 loss=0.13511 test_loss_avg=0.51056 acc=0.96094 test_acc_avg=0.85637 test_acc_top5_avg=0.98708 time=232.99it/s +epoch=112 global_step=44183 loss=0.37364 test_loss_avg=0.42097 acc=0.87500 test_acc_avg=0.87935 test_acc_top5_avg=0.99051 time=865.52it/s +curr_acc 0.8794 +BEST_ACC 0.8976 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=1.79223 loss_avg=1.91846 acc=0.84375 acc_top1_avg=0.82353 acc_top5_avg=0.94531 lr=0.00010 gn=25.88586 time=59.67it/s +epoch=113 global_step=44250 loss=1.67354 loss_avg=1.93180 acc=0.84375 acc_top1_avg=0.82229 acc_top5_avg=0.94333 lr=0.00010 gn=22.99762 time=59.62it/s +epoch=113 global_step=44300 loss=1.85383 loss_avg=1.90424 acc=0.83594 acc_top1_avg=0.82525 acc_top5_avg=0.94191 lr=0.00010 gn=23.71186 time=54.15it/s +epoch=113 global_step=44350 loss=1.94876 loss_avg=1.92663 acc=0.82031 acc_top1_avg=0.82303 acc_top5_avg=0.94134 lr=0.00010 gn=25.51765 time=59.92it/s +epoch=113 global_step=44400 loss=1.61727 loss_avg=1.91169 acc=0.85938 acc_top1_avg=0.82445 acc_top5_avg=0.94268 lr=0.00010 gn=23.14631 time=53.66it/s +epoch=113 global_step=44450 loss=1.40763 loss_avg=1.91396 acc=0.87500 acc_top1_avg=0.82435 acc_top5_avg=0.94130 lr=0.00010 gn=20.38376 time=55.04it/s +epoch=113 global_step=44500 loss=1.51707 loss_avg=1.92216 acc=0.86719 acc_top1_avg=0.82361 acc_top5_avg=0.94046 lr=0.00010 gn=22.49056 time=60.41it/s +epoch=113 global_step=44550 loss=1.85402 loss_avg=1.93386 acc=0.82031 acc_top1_avg=0.82255 acc_top5_avg=0.93999 lr=0.00010 gn=21.31937 time=58.62it/s +====================Eval==================== +epoch=113 global_step=44574 loss=1.04527 test_loss_avg=0.59960 acc=0.67969 test_acc_avg=0.82711 test_acc_top5_avg=0.98573 time=231.77it/s +epoch=113 global_step=44574 loss=0.45520 test_loss_avg=0.42293 acc=0.85938 test_acc_avg=0.88035 test_acc_top5_avg=0.99080 time=230.98it/s +epoch=113 global_step=44574 loss=0.42084 test_loss_avg=0.42290 acc=0.87500 test_acc_avg=0.87955 test_acc_top5_avg=0.99120 time=542.88it/s +curr_acc 0.8795 +BEST_ACC 0.8976 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=1.78743 loss_avg=1.81923 acc=0.84375 acc_top1_avg=0.83654 acc_top5_avg=0.94681 lr=0.00010 gn=26.31831 time=53.81it/s +epoch=114 global_step=44650 loss=2.35151 loss_avg=1.88693 acc=0.78125 acc_top1_avg=0.82854 acc_top5_avg=0.94048 lr=0.00010 gn=24.99433 time=54.95it/s +epoch=114 global_step=44700 loss=1.64321 loss_avg=1.88432 acc=0.85156 acc_top1_avg=0.82893 acc_top5_avg=0.94110 lr=0.00010 gn=28.69881 time=58.33it/s +epoch=114 global_step=44750 loss=1.32050 loss_avg=1.90362 acc=0.89062 acc_top1_avg=0.82666 acc_top5_avg=0.94141 lr=0.00010 gn=21.62469 time=59.94it/s +epoch=114 global_step=44800 loss=1.56203 loss_avg=1.92141 acc=0.86719 acc_top1_avg=0.82436 acc_top5_avg=0.94134 lr=0.00010 gn=28.57141 time=54.10it/s +epoch=114 global_step=44850 loss=1.89995 loss_avg=1.93036 acc=0.82812 acc_top1_avg=0.82331 acc_top5_avg=0.94095 lr=0.00010 gn=23.50457 time=50.91it/s +epoch=114 global_step=44900 loss=1.54155 loss_avg=1.93180 acc=0.86719 acc_top1_avg=0.82331 acc_top5_avg=0.94064 lr=0.00010 gn=25.90442 time=58.26it/s +epoch=114 global_step=44950 loss=2.45078 loss_avg=1.92840 acc=0.76562 acc_top1_avg=0.82353 acc_top5_avg=0.94062 lr=0.00010 gn=32.95404 time=59.40it/s +====================Eval==================== +epoch=114 global_step=44965 loss=0.74253 test_loss_avg=0.54241 acc=0.82812 test_acc_avg=0.84819 test_acc_top5_avg=0.98597 time=238.52it/s +epoch=114 global_step=44965 loss=0.34996 test_loss_avg=0.41976 acc=0.87500 test_acc_avg=0.87925 test_acc_top5_avg=0.99001 time=895.84it/s +curr_acc 0.8793 +BEST_ACC 0.8976 +curr_acc_top5 0.9900 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=2.04969 loss_avg=1.90506 acc=0.81250 acc_top1_avg=0.82656 acc_top5_avg=0.94330 lr=0.00010 gn=24.00774 time=60.40it/s +epoch=115 global_step=45050 loss=2.30004 loss_avg=1.85330 acc=0.78125 acc_top1_avg=0.83208 acc_top5_avg=0.94219 lr=0.00010 gn=22.26251 time=52.14it/s +epoch=115 global_step=45100 loss=2.07183 loss_avg=1.88347 acc=0.80469 acc_top1_avg=0.82922 acc_top5_avg=0.94248 lr=0.00010 gn=20.84371 time=58.34it/s +epoch=115 global_step=45150 loss=1.86760 loss_avg=1.90403 acc=0.82812 acc_top1_avg=0.82686 acc_top5_avg=0.94113 lr=0.00010 gn=26.72068 time=61.65it/s +epoch=115 global_step=45200 loss=1.97630 loss_avg=1.89995 acc=0.82031 acc_top1_avg=0.82686 acc_top5_avg=0.94159 lr=0.00010 gn=26.75087 time=53.54it/s +epoch=115 global_step=45250 loss=1.47198 loss_avg=1.91851 acc=0.87500 acc_top1_avg=0.82511 acc_top5_avg=0.94076 lr=0.00010 gn=20.85926 time=61.42it/s +epoch=115 global_step=45300 loss=1.84211 loss_avg=1.92299 acc=0.83594 acc_top1_avg=0.82449 acc_top5_avg=0.94039 lr=0.00010 gn=21.95477 time=60.67it/s +epoch=115 global_step=45350 loss=1.47902 loss_avg=1.92586 acc=0.85938 acc_top1_avg=0.82419 acc_top5_avg=0.93996 lr=0.00010 gn=17.45930 time=58.78it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.19913 test_loss_avg=0.43420 acc=0.96094 test_acc_avg=0.87344 test_acc_top5_avg=0.99167 time=168.64it/s +epoch=115 global_step=45356 loss=0.19985 test_loss_avg=0.44016 acc=0.92969 test_acc_avg=0.87380 test_acc_top5_avg=0.99014 time=234.73it/s +epoch=115 global_step=45356 loss=0.33495 test_loss_avg=0.41821 acc=0.93750 test_acc_avg=0.87994 test_acc_top5_avg=0.99120 time=307.75it/s +curr_acc 0.8799 +BEST_ACC 0.8976 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=1.61660 loss_avg=1.82986 acc=0.85156 acc_top1_avg=0.83256 acc_top5_avg=0.94212 lr=0.00010 gn=21.23069 time=51.48it/s +epoch=116 global_step=45450 loss=1.50658 loss_avg=1.85650 acc=0.86719 acc_top1_avg=0.83087 acc_top5_avg=0.94149 lr=0.00010 gn=22.87589 time=51.69it/s +epoch=116 global_step=45500 loss=2.23407 loss_avg=1.89738 acc=0.79688 acc_top1_avg=0.82715 acc_top5_avg=0.94108 lr=0.00010 gn=27.35554 time=55.53it/s +epoch=116 global_step=45550 loss=2.07185 loss_avg=1.90318 acc=0.81250 acc_top1_avg=0.82688 acc_top5_avg=0.94060 lr=0.00010 gn=30.70736 time=57.99it/s +epoch=116 global_step=45600 loss=1.81015 loss_avg=1.90363 acc=0.83594 acc_top1_avg=0.82665 acc_top5_avg=0.94105 lr=0.00010 gn=19.88517 time=57.53it/s +epoch=116 global_step=45650 loss=1.75764 loss_avg=1.89830 acc=0.84375 acc_top1_avg=0.82722 acc_top5_avg=0.94141 lr=0.00010 gn=23.48635 time=61.80it/s +epoch=116 global_step=45700 loss=1.21330 loss_avg=1.92067 acc=0.89062 acc_top1_avg=0.82495 acc_top5_avg=0.94109 lr=0.00010 gn=15.95699 time=58.97it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.29840 test_loss_avg=0.57752 acc=0.91406 test_acc_avg=0.83811 test_acc_top5_avg=0.98546 time=206.26it/s +epoch=116 global_step=45747 loss=0.36445 test_loss_avg=0.42645 acc=0.87500 test_acc_avg=0.87826 test_acc_top5_avg=0.99061 time=893.36it/s +curr_acc 0.8783 +BEST_ACC 0.8976 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=1.70896 loss_avg=1.82523 acc=0.85938 acc_top1_avg=0.83854 acc_top5_avg=0.94531 lr=0.00010 gn=25.44437 time=56.28it/s +epoch=117 global_step=45800 loss=2.05363 loss_avg=1.94081 acc=0.81250 acc_top1_avg=0.82252 acc_top5_avg=0.93956 lr=0.00010 gn=20.85655 time=51.16it/s +epoch=117 global_step=45850 loss=1.43293 loss_avg=1.92507 acc=0.88281 acc_top1_avg=0.82441 acc_top5_avg=0.94266 lr=0.00010 gn=21.38698 time=55.69it/s +epoch=117 global_step=45900 loss=1.70795 loss_avg=1.92662 acc=0.85156 acc_top1_avg=0.82368 acc_top5_avg=0.94107 lr=0.00010 gn=24.78983 time=54.43it/s +epoch=117 global_step=45950 loss=1.96585 loss_avg=1.91855 acc=0.82031 acc_top1_avg=0.82466 acc_top5_avg=0.94081 lr=0.00010 gn=24.70065 time=56.65it/s +epoch=117 global_step=46000 loss=1.52473 loss_avg=1.92032 acc=0.85156 acc_top1_avg=0.82470 acc_top5_avg=0.94028 lr=0.00010 gn=19.38617 time=63.38it/s +epoch=117 global_step=46050 loss=2.51129 loss_avg=1.92109 acc=0.76562 acc_top1_avg=0.82457 acc_top5_avg=0.93967 lr=0.00010 gn=25.03548 time=60.36it/s +epoch=117 global_step=46100 loss=2.46116 loss_avg=1.91838 acc=0.76562 acc_top1_avg=0.82476 acc_top5_avg=0.94000 lr=0.00010 gn=20.58931 time=61.49it/s +====================Eval==================== +epoch=117 global_step=46138 loss=0.56167 test_loss_avg=0.75861 acc=0.82031 test_acc_avg=0.76897 test_acc_top5_avg=0.98103 time=251.40it/s +epoch=117 global_step=46138 loss=0.23254 test_loss_avg=0.48155 acc=0.90625 test_acc_avg=0.86212 test_acc_top5_avg=0.98780 time=245.41it/s +epoch=117 global_step=46138 loss=0.34571 test_loss_avg=0.42095 acc=0.87500 test_acc_avg=0.87866 test_acc_top5_avg=0.99051 time=871.45it/s +curr_acc 0.8787 +BEST_ACC 0.8976 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=1.65360 loss_avg=1.89249 acc=0.85156 acc_top1_avg=0.82682 acc_top5_avg=0.94010 lr=0.00010 gn=24.26549 time=52.93it/s +epoch=118 global_step=46200 loss=2.24971 loss_avg=1.93773 acc=0.78906 acc_top1_avg=0.82334 acc_top5_avg=0.94267 lr=0.00010 gn=26.54258 time=63.26it/s +epoch=118 global_step=46250 loss=2.00280 loss_avg=1.94947 acc=0.82031 acc_top1_avg=0.82129 acc_top5_avg=0.94301 lr=0.00010 gn=30.11576 time=59.29it/s +epoch=118 global_step=46300 loss=2.00729 loss_avg=1.92128 acc=0.82031 acc_top1_avg=0.82432 acc_top5_avg=0.94237 lr=0.00010 gn=27.87958 time=51.69it/s +epoch=118 global_step=46350 loss=1.90276 loss_avg=1.91197 acc=0.82031 acc_top1_avg=0.82525 acc_top5_avg=0.94236 lr=0.00010 gn=23.52468 time=49.31it/s +epoch=118 global_step=46400 loss=2.32099 loss_avg=1.92317 acc=0.77344 acc_top1_avg=0.82380 acc_top5_avg=0.94212 lr=0.00010 gn=26.29296 time=61.65it/s +epoch=118 global_step=46450 loss=2.09885 loss_avg=1.91990 acc=0.80469 acc_top1_avg=0.82422 acc_top5_avg=0.94121 lr=0.00010 gn=28.49948 time=59.59it/s +epoch=118 global_step=46500 loss=1.17228 loss_avg=1.91872 acc=0.91406 acc_top1_avg=0.82437 acc_top5_avg=0.94074 lr=0.00010 gn=34.43777 time=61.73it/s +====================Eval==================== +epoch=118 global_step=46529 loss=0.39760 test_loss_avg=0.64179 acc=0.86719 test_acc_avg=0.81892 test_acc_top5_avg=0.98438 time=229.22it/s +epoch=118 global_step=46529 loss=0.24146 test_loss_avg=0.42983 acc=0.92969 test_acc_avg=0.87770 test_acc_top5_avg=0.99018 time=248.68it/s +epoch=118 global_step=46529 loss=0.29470 test_loss_avg=0.42812 acc=0.93750 test_acc_avg=0.87846 test_acc_top5_avg=0.99031 time=876.37it/s +curr_acc 0.8785 +BEST_ACC 0.8976 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9947 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=2.25857 loss_avg=1.97525 acc=0.79688 acc_top1_avg=0.81510 acc_top5_avg=0.93341 lr=0.00010 gn=26.24622 time=60.71it/s +epoch=119 global_step=46600 loss=2.31167 loss_avg=1.97652 acc=0.79688 acc_top1_avg=0.81789 acc_top5_avg=0.93772 lr=0.00010 gn=31.99470 time=53.53it/s +epoch=119 global_step=46650 loss=1.39428 loss_avg=1.95644 acc=0.87500 acc_top1_avg=0.82012 acc_top5_avg=0.93931 lr=0.00010 gn=18.08836 time=54.08it/s +epoch=119 global_step=46700 loss=1.59118 loss_avg=1.93137 acc=0.86719 acc_top1_avg=0.82310 acc_top5_avg=0.93960 lr=0.00010 gn=27.43207 time=55.97it/s +epoch=119 global_step=46750 loss=2.49137 loss_avg=1.92492 acc=0.76562 acc_top1_avg=0.82406 acc_top5_avg=0.93980 lr=0.00010 gn=26.12596 time=62.67it/s +epoch=119 global_step=46800 loss=2.38820 loss_avg=1.91904 acc=0.77344 acc_top1_avg=0.82472 acc_top5_avg=0.94033 lr=0.00010 gn=28.20338 time=59.72it/s +epoch=119 global_step=46850 loss=1.73390 loss_avg=1.91147 acc=0.84375 acc_top1_avg=0.82555 acc_top5_avg=0.94032 lr=0.00010 gn=20.02258 time=61.28it/s +epoch=119 global_step=46900 loss=1.82072 loss_avg=1.91545 acc=0.82812 acc_top1_avg=0.82513 acc_top5_avg=0.94005 lr=0.00010 gn=21.98675 time=54.44it/s +====================Eval==================== +epoch=119 global_step=46920 loss=0.13712 test_loss_avg=0.54311 acc=0.95312 test_acc_avg=0.84423 test_acc_top5_avg=0.98533 time=207.00it/s +epoch=119 global_step=46920 loss=0.31311 test_loss_avg=0.42931 acc=0.93750 test_acc_avg=0.87530 test_acc_top5_avg=0.98981 time=895.84it/s +curr_acc 0.8753 +BEST_ACC 0.8976 +curr_acc_top5 0.9898 +BEST_ACC_top5 0.9947 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_2_4.log b/other_methods/sceloss/sceloss_results/out_2_4.log new file mode 100644 index 0000000..fe3a8eb --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_2_4.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.2__noise_amount__0.4.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=6.99048 loss_avg=7.58690 acc=0.30469 acc_top1_avg=0.22922 acc_top5_avg=0.67500 lr=0.01000 gn=8.50150 time=64.51it/s +epoch=0 global_step=100 loss=6.70740 loss_avg=7.34497 acc=0.31250 acc_top1_avg=0.25547 acc_top5_avg=0.69664 lr=0.01000 gn=6.10027 time=58.71it/s +epoch=0 global_step=150 loss=7.12669 loss_avg=7.19469 acc=0.27344 acc_top1_avg=0.27083 acc_top5_avg=0.71234 lr=0.01000 gn=5.87814 time=51.73it/s +epoch=0 global_step=200 loss=7.07940 loss_avg=7.10471 acc=0.26562 acc_top1_avg=0.28027 acc_top5_avg=0.72223 lr=0.01000 gn=5.07691 time=57.53it/s +epoch=0 global_step=250 loss=6.50473 loss_avg=7.02360 acc=0.35938 acc_top1_avg=0.28906 acc_top5_avg=0.73209 lr=0.01000 gn=5.54414 time=60.82it/s +epoch=0 global_step=300 loss=6.63131 loss_avg=6.94453 acc=0.31250 acc_top1_avg=0.29724 acc_top5_avg=0.73852 lr=0.01000 gn=4.59535 time=63.85it/s +epoch=0 global_step=350 loss=6.36039 loss_avg=6.86921 acc=0.36719 acc_top1_avg=0.30538 acc_top5_avg=0.74310 lr=0.01000 gn=5.89286 time=60.82it/s +====================Eval==================== +epoch=0 global_step=391 loss=4.37721 test_loss_avg=2.53587 acc=0.00000 test_acc_avg=0.34469 test_acc_top5_avg=0.88313 time=243.71it/s +epoch=0 global_step=391 loss=2.83900 test_loss_avg=2.26528 acc=0.56250 test_acc_avg=0.42267 test_acc_top5_avg=0.89349 time=32.16it/s +curr_acc 0.4227 +BEST_ACC 0.0000 +curr_acc_top5 0.8935 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.36107 loss_avg=6.21804 acc=0.35938 acc_top1_avg=0.37587 acc_top5_avg=0.78646 lr=0.01000 gn=5.23775 time=56.23it/s +epoch=1 global_step=450 loss=6.36047 loss_avg=6.29424 acc=0.36719 acc_top1_avg=0.36719 acc_top5_avg=0.78284 lr=0.01000 gn=5.96011 time=57.73it/s +epoch=1 global_step=500 loss=6.19881 loss_avg=6.32728 acc=0.35156 acc_top1_avg=0.36224 acc_top5_avg=0.78025 lr=0.01000 gn=3.90340 time=57.29it/s +epoch=1 global_step=550 loss=5.87366 loss_avg=6.26363 acc=0.40625 acc_top1_avg=0.36915 acc_top5_avg=0.78538 lr=0.01000 gn=3.93535 time=60.29it/s +epoch=1 global_step=600 loss=5.78763 loss_avg=6.24695 acc=0.39844 acc_top1_avg=0.37089 acc_top5_avg=0.78585 lr=0.01000 gn=3.82835 time=62.68it/s +epoch=1 global_step=650 loss=6.03959 loss_avg=6.21084 acc=0.39062 acc_top1_avg=0.37524 acc_top5_avg=0.78864 lr=0.01000 gn=5.41746 time=55.84it/s +epoch=1 global_step=700 loss=6.49177 loss_avg=6.19537 acc=0.35156 acc_top1_avg=0.37649 acc_top5_avg=0.79088 lr=0.01000 gn=4.76443 time=57.20it/s +epoch=1 global_step=750 loss=5.68440 loss_avg=6.16123 acc=0.44531 acc_top1_avg=0.38005 acc_top5_avg=0.79187 lr=0.01000 gn=4.92524 time=65.73it/s +====================Eval==================== +epoch=1 global_step=782 loss=1.88108 test_loss_avg=1.61610 acc=0.43750 test_acc_avg=0.56920 test_acc_top5_avg=0.92783 time=259.45it/s +epoch=1 global_step=782 loss=0.67422 test_loss_avg=1.68082 acc=0.81250 test_acc_avg=0.52707 test_acc_top5_avg=0.92925 time=257.16it/s +epoch=1 global_step=782 loss=0.79004 test_loss_avg=1.58803 acc=0.81250 test_acc_avg=0.55479 test_acc_top5_avg=0.93364 time=739.48it/s +curr_acc 0.5548 +BEST_ACC 0.4227 +curr_acc_top5 0.9336 +BEST_ACC_top5 0.8935 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=6.21936 loss_avg=5.75406 acc=0.35938 acc_top1_avg=0.42969 acc_top5_avg=0.81424 lr=0.01000 gn=5.13497 time=54.90it/s +epoch=2 global_step=850 loss=5.68853 loss_avg=5.83889 acc=0.46875 acc_top1_avg=0.41900 acc_top5_avg=0.80825 lr=0.01000 gn=5.74973 time=54.21it/s +epoch=2 global_step=900 loss=5.52194 loss_avg=5.84292 acc=0.47656 acc_top1_avg=0.41777 acc_top5_avg=0.80886 lr=0.01000 gn=4.99117 time=57.45it/s +epoch=2 global_step=950 loss=6.26396 loss_avg=5.86136 acc=0.36719 acc_top1_avg=0.41476 acc_top5_avg=0.80906 lr=0.01000 gn=4.68409 time=57.58it/s +epoch=2 global_step=1000 loss=5.72323 loss_avg=5.85979 acc=0.42188 acc_top1_avg=0.41499 acc_top5_avg=0.80867 lr=0.01000 gn=4.34730 time=53.61it/s +epoch=2 global_step=1050 loss=6.16400 loss_avg=5.84317 acc=0.38281 acc_top1_avg=0.41610 acc_top5_avg=0.80956 lr=0.01000 gn=5.13867 time=53.58it/s +epoch=2 global_step=1100 loss=5.98862 loss_avg=5.82532 acc=0.37500 acc_top1_avg=0.41713 acc_top5_avg=0.81112 lr=0.01000 gn=4.32030 time=55.56it/s +epoch=2 global_step=1150 loss=6.20338 loss_avg=5.79829 acc=0.37500 acc_top1_avg=0.42001 acc_top5_avg=0.81335 lr=0.01000 gn=4.28921 time=55.67it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.54222 test_loss_avg=1.90816 acc=0.60938 test_acc_avg=0.50186 test_acc_top5_avg=0.93657 time=233.97it/s +epoch=2 global_step=1173 loss=0.56802 test_loss_avg=1.62523 acc=0.87500 test_acc_avg=0.57031 test_acc_top5_avg=0.94027 time=882.08it/s +curr_acc 0.5703 +BEST_ACC 0.5548 +curr_acc_top5 0.9403 +BEST_ACC_top5 0.9336 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=5.00626 loss_avg=5.62028 acc=0.46094 acc_top1_avg=0.43576 acc_top5_avg=0.83420 lr=0.01000 gn=4.47667 time=60.51it/s +epoch=3 global_step=1250 loss=4.95247 loss_avg=5.60003 acc=0.48438 acc_top1_avg=0.43994 acc_top5_avg=0.82762 lr=0.01000 gn=4.68366 time=56.54it/s +epoch=3 global_step=1300 loss=5.97596 loss_avg=5.61792 acc=0.40625 acc_top1_avg=0.43842 acc_top5_avg=0.82277 lr=0.01000 gn=4.53383 time=56.35it/s +epoch=3 global_step=1350 loss=5.82195 loss_avg=5.61677 acc=0.39844 acc_top1_avg=0.43900 acc_top5_avg=0.82287 lr=0.01000 gn=4.17693 time=57.12it/s +epoch=3 global_step=1400 loss=5.48484 loss_avg=5.60034 acc=0.47656 acc_top1_avg=0.44101 acc_top5_avg=0.82486 lr=0.01000 gn=4.81654 time=61.90it/s +epoch=3 global_step=1450 loss=5.75228 loss_avg=5.58663 acc=0.42188 acc_top1_avg=0.44215 acc_top5_avg=0.82463 lr=0.01000 gn=4.14271 time=58.63it/s +epoch=3 global_step=1500 loss=5.33133 loss_avg=5.58782 acc=0.49219 acc_top1_avg=0.44237 acc_top5_avg=0.82406 lr=0.01000 gn=4.92337 time=60.93it/s +epoch=3 global_step=1550 loss=5.45189 loss_avg=5.58286 acc=0.50781 acc_top1_avg=0.44341 acc_top5_avg=0.82471 lr=0.01000 gn=5.16730 time=64.23it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.13272 test_loss_avg=0.75973 acc=0.94531 test_acc_avg=0.79327 test_acc_top5_avg=0.97536 time=233.30it/s +epoch=3 global_step=1564 loss=0.67411 test_loss_avg=1.50003 acc=0.82812 test_acc_avg=0.60218 test_acc_top5_avg=0.92374 time=134.43it/s +epoch=3 global_step=1564 loss=0.94981 test_loss_avg=1.37091 acc=0.62500 test_acc_avg=0.63657 test_acc_top5_avg=0.93631 time=626.20it/s +curr_acc 0.6366 +BEST_ACC 0.5703 +curr_acc_top5 0.9363 +BEST_ACC_top5 0.9403 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.43675 loss_avg=5.46786 acc=0.59375 acc_top1_avg=0.45399 acc_top5_avg=0.81641 lr=0.01000 gn=5.14482 time=61.43it/s +epoch=4 global_step=1650 loss=5.88341 loss_avg=5.54506 acc=0.40625 acc_top1_avg=0.44767 acc_top5_avg=0.82149 lr=0.01000 gn=4.85181 time=52.42it/s +epoch=4 global_step=1700 loss=5.11741 loss_avg=5.52701 acc=0.47656 acc_top1_avg=0.44813 acc_top5_avg=0.82663 lr=0.01000 gn=4.61854 time=53.80it/s +epoch=4 global_step=1750 loss=5.97976 loss_avg=5.48298 acc=0.41406 acc_top1_avg=0.45300 acc_top5_avg=0.82867 lr=0.01000 gn=4.67820 time=59.15it/s +epoch=4 global_step=1800 loss=4.68191 loss_avg=5.48147 acc=0.55469 acc_top1_avg=0.45312 acc_top5_avg=0.82875 lr=0.01000 gn=6.28442 time=50.44it/s +epoch=4 global_step=1850 loss=5.42594 loss_avg=5.46924 acc=0.44531 acc_top1_avg=0.45449 acc_top5_avg=0.83009 lr=0.01000 gn=3.98665 time=60.19it/s +epoch=4 global_step=1900 loss=5.46058 loss_avg=5.47571 acc=0.44531 acc_top1_avg=0.45410 acc_top5_avg=0.83054 lr=0.01000 gn=5.59659 time=53.71it/s +epoch=4 global_step=1950 loss=6.12252 loss_avg=5.48503 acc=0.37500 acc_top1_avg=0.45304 acc_top5_avg=0.83055 lr=0.01000 gn=3.84339 time=63.55it/s +====================Eval==================== +epoch=4 global_step=1955 loss=0.93751 test_loss_avg=1.60168 acc=0.73438 test_acc_avg=0.56687 test_acc_top5_avg=0.92739 time=211.03it/s +epoch=4 global_step=1955 loss=0.16810 test_loss_avg=1.18472 acc=0.93750 test_acc_avg=0.66297 test_acc_top5_avg=0.94818 time=486.47it/s +curr_acc 0.6630 +BEST_ACC 0.6366 +curr_acc_top5 0.9482 +BEST_ACC_top5 0.9403 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=5.70336 loss_avg=5.47971 acc=0.41406 acc_top1_avg=0.45747 acc_top5_avg=0.83177 lr=0.01000 gn=4.02155 time=49.91it/s +epoch=5 global_step=2050 loss=5.22473 loss_avg=5.41886 acc=0.48438 acc_top1_avg=0.46349 acc_top5_avg=0.83133 lr=0.01000 gn=5.97317 time=57.40it/s +epoch=5 global_step=2100 loss=4.94439 loss_avg=5.37198 acc=0.51562 acc_top1_avg=0.46756 acc_top5_avg=0.83572 lr=0.01000 gn=4.39427 time=56.59it/s +epoch=5 global_step=2150 loss=4.95420 loss_avg=5.37254 acc=0.47656 acc_top1_avg=0.46647 acc_top5_avg=0.83702 lr=0.01000 gn=5.19781 time=53.10it/s +epoch=5 global_step=2200 loss=5.18408 loss_avg=5.37247 acc=0.50000 acc_top1_avg=0.46610 acc_top5_avg=0.83760 lr=0.01000 gn=6.65007 time=54.73it/s +epoch=5 global_step=2250 loss=5.57521 loss_avg=5.37710 acc=0.44531 acc_top1_avg=0.46557 acc_top5_avg=0.83684 lr=0.01000 gn=5.22704 time=55.10it/s +epoch=5 global_step=2300 loss=5.45000 loss_avg=5.37305 acc=0.44531 acc_top1_avg=0.46560 acc_top5_avg=0.83705 lr=0.01000 gn=5.43604 time=56.13it/s +====================Eval==================== +epoch=5 global_step=2346 loss=1.42340 test_loss_avg=1.66486 acc=0.59375 test_acc_avg=0.56406 test_acc_top5_avg=0.96719 time=238.62it/s +epoch=5 global_step=2346 loss=1.85823 test_loss_avg=1.76231 acc=0.39062 test_acc_avg=0.51420 test_acc_top5_avg=0.90682 time=221.29it/s +epoch=5 global_step=2346 loss=0.54523 test_loss_avg=1.39409 acc=0.87500 test_acc_avg=0.61412 test_acc_top5_avg=0.93157 time=725.41it/s +curr_acc 0.6141 +BEST_ACC 0.6630 +curr_acc_top5 0.9316 +BEST_ACC_top5 0.9482 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=5.47884 loss_avg=5.25262 acc=0.45312 acc_top1_avg=0.48242 acc_top5_avg=0.82031 lr=0.01000 gn=5.30170 time=54.07it/s +epoch=6 global_step=2400 loss=5.22345 loss_avg=5.28061 acc=0.48438 acc_top1_avg=0.47801 acc_top5_avg=0.83883 lr=0.01000 gn=6.01323 time=60.07it/s +epoch=6 global_step=2450 loss=4.93295 loss_avg=5.31679 acc=0.51562 acc_top1_avg=0.47160 acc_top5_avg=0.83894 lr=0.01000 gn=5.20367 time=62.92it/s +epoch=6 global_step=2500 loss=5.48292 loss_avg=5.29205 acc=0.41406 acc_top1_avg=0.47342 acc_top5_avg=0.83949 lr=0.01000 gn=6.96406 time=53.49it/s +epoch=6 global_step=2550 loss=4.65142 loss_avg=5.29878 acc=0.54688 acc_top1_avg=0.47277 acc_top5_avg=0.84004 lr=0.01000 gn=5.70165 time=59.13it/s +epoch=6 global_step=2600 loss=5.12056 loss_avg=5.30140 acc=0.50000 acc_top1_avg=0.47256 acc_top5_avg=0.83904 lr=0.01000 gn=3.97819 time=56.87it/s +epoch=6 global_step=2650 loss=5.06824 loss_avg=5.31899 acc=0.49219 acc_top1_avg=0.47057 acc_top5_avg=0.83900 lr=0.01000 gn=3.60099 time=63.98it/s +epoch=6 global_step=2700 loss=4.86233 loss_avg=5.31962 acc=0.52344 acc_top1_avg=0.47065 acc_top5_avg=0.83887 lr=0.01000 gn=4.82536 time=58.19it/s +====================Eval==================== +epoch=6 global_step=2737 loss=2.59468 test_loss_avg=1.63536 acc=0.28906 test_acc_avg=0.60427 test_acc_top5_avg=0.89453 time=250.86it/s +epoch=6 global_step=2737 loss=0.17508 test_loss_avg=1.45757 acc=0.95312 test_acc_avg=0.61020 test_acc_top5_avg=0.92609 time=202.70it/s +epoch=6 global_step=2737 loss=0.17951 test_loss_avg=1.40997 acc=0.93750 test_acc_avg=0.62263 test_acc_top5_avg=0.92890 time=430.94it/s +curr_acc 0.6226 +BEST_ACC 0.6630 +curr_acc_top5 0.9289 +BEST_ACC_top5 0.9482 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=4.56565 loss_avg=5.17713 acc=0.55469 acc_top1_avg=0.48858 acc_top5_avg=0.84976 lr=0.01000 gn=5.50849 time=56.68it/s +epoch=7 global_step=2800 loss=5.51873 loss_avg=5.20587 acc=0.44531 acc_top1_avg=0.48313 acc_top5_avg=0.84288 lr=0.01000 gn=5.35545 time=55.84it/s +epoch=7 global_step=2850 loss=5.60504 loss_avg=5.25028 acc=0.44531 acc_top1_avg=0.47829 acc_top5_avg=0.84285 lr=0.01000 gn=6.23821 time=52.10it/s +epoch=7 global_step=2900 loss=5.05609 loss_avg=5.26550 acc=0.51562 acc_top1_avg=0.47714 acc_top5_avg=0.84202 lr=0.01000 gn=5.98313 time=57.18it/s +epoch=7 global_step=2950 loss=5.61421 loss_avg=5.24700 acc=0.43750 acc_top1_avg=0.47902 acc_top5_avg=0.84320 lr=0.01000 gn=5.73611 time=55.97it/s +epoch=7 global_step=3000 loss=5.58293 loss_avg=5.25626 acc=0.43750 acc_top1_avg=0.47754 acc_top5_avg=0.84212 lr=0.01000 gn=6.02856 time=53.93it/s +epoch=7 global_step=3050 loss=4.83555 loss_avg=5.27576 acc=0.50781 acc_top1_avg=0.47534 acc_top5_avg=0.84113 lr=0.01000 gn=5.30510 time=58.39it/s +epoch=7 global_step=3100 loss=4.98032 loss_avg=5.27762 acc=0.50000 acc_top1_avg=0.47544 acc_top5_avg=0.84164 lr=0.01000 gn=4.47252 time=60.64it/s +====================Eval==================== +epoch=7 global_step=3128 loss=2.11970 test_loss_avg=1.82973 acc=0.39062 test_acc_avg=0.56117 test_acc_top5_avg=0.91888 time=242.40it/s +epoch=7 global_step=3128 loss=0.00138 test_loss_avg=1.53275 acc=1.00000 test_acc_avg=0.59978 test_acc_top5_avg=0.94106 time=465.57it/s +curr_acc 0.5998 +BEST_ACC 0.6630 +curr_acc_top5 0.9411 +BEST_ACC_top5 0.9482 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=5.97796 loss_avg=5.22336 acc=0.39844 acc_top1_avg=0.47940 acc_top5_avg=0.85050 lr=0.01000 gn=3.98229 time=60.80it/s +epoch=8 global_step=3200 loss=5.60731 loss_avg=5.15588 acc=0.43750 acc_top1_avg=0.49002 acc_top5_avg=0.84928 lr=0.01000 gn=4.07262 time=57.19it/s +epoch=8 global_step=3250 loss=4.98829 loss_avg=5.21051 acc=0.50781 acc_top1_avg=0.48380 acc_top5_avg=0.84676 lr=0.01000 gn=5.97651 time=63.20it/s +epoch=8 global_step=3300 loss=5.18715 loss_avg=5.23187 acc=0.46875 acc_top1_avg=0.48101 acc_top5_avg=0.84552 lr=0.01000 gn=6.35920 time=56.55it/s +epoch=8 global_step=3350 loss=4.53649 loss_avg=5.24209 acc=0.57031 acc_top1_avg=0.47998 acc_top5_avg=0.84533 lr=0.01000 gn=4.76185 time=51.28it/s +epoch=8 global_step=3400 loss=4.60537 loss_avg=5.23684 acc=0.51562 acc_top1_avg=0.48090 acc_top5_avg=0.84504 lr=0.01000 gn=6.85893 time=55.60it/s +epoch=8 global_step=3450 loss=5.52498 loss_avg=5.23736 acc=0.45312 acc_top1_avg=0.48025 acc_top5_avg=0.84465 lr=0.01000 gn=5.81754 time=55.97it/s +epoch=8 global_step=3500 loss=5.56573 loss_avg=5.23083 acc=0.43750 acc_top1_avg=0.48089 acc_top5_avg=0.84488 lr=0.01000 gn=6.00843 time=55.83it/s +====================Eval==================== +epoch=8 global_step=3519 loss=3.45410 test_loss_avg=1.58573 acc=0.25781 test_acc_avg=0.62543 test_acc_top5_avg=0.94835 time=226.25it/s +epoch=8 global_step=3519 loss=0.71632 test_loss_avg=2.03390 acc=0.80469 test_acc_avg=0.51850 test_acc_top5_avg=0.89269 time=247.82it/s +epoch=8 global_step=3519 loss=0.00285 test_loss_avg=1.77317 acc=1.00000 test_acc_avg=0.57902 test_acc_top5_avg=0.90704 time=489.30it/s +curr_acc 0.5790 +BEST_ACC 0.6630 +curr_acc_top5 0.9070 +BEST_ACC_top5 0.9482 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=4.80514 loss_avg=5.16022 acc=0.54688 acc_top1_avg=0.49168 acc_top5_avg=0.84375 lr=0.01000 gn=5.89933 time=51.61it/s +epoch=9 global_step=3600 loss=4.90427 loss_avg=5.20734 acc=0.50781 acc_top1_avg=0.48380 acc_top5_avg=0.84346 lr=0.01000 gn=5.07896 time=54.74it/s +epoch=9 global_step=3650 loss=6.09451 loss_avg=5.19220 acc=0.35938 acc_top1_avg=0.48563 acc_top5_avg=0.84530 lr=0.01000 gn=6.24659 time=53.81it/s +epoch=9 global_step=3700 loss=4.75820 loss_avg=5.19022 acc=0.51562 acc_top1_avg=0.48489 acc_top5_avg=0.84647 lr=0.01000 gn=6.90951 time=53.65it/s +epoch=9 global_step=3750 loss=5.57442 loss_avg=5.19884 acc=0.42969 acc_top1_avg=0.48400 acc_top5_avg=0.84537 lr=0.01000 gn=6.28554 time=60.71it/s +epoch=9 global_step=3800 loss=5.15482 loss_avg=5.21275 acc=0.48438 acc_top1_avg=0.48268 acc_top5_avg=0.84350 lr=0.01000 gn=5.78248 time=47.53it/s +epoch=9 global_step=3850 loss=4.69181 loss_avg=5.22056 acc=0.53906 acc_top1_avg=0.48157 acc_top5_avg=0.84366 lr=0.01000 gn=6.08074 time=56.19it/s +epoch=9 global_step=3900 loss=5.38284 loss_avg=5.21476 acc=0.44531 acc_top1_avg=0.48196 acc_top5_avg=0.84426 lr=0.01000 gn=6.53477 time=52.20it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.38696 test_loss_avg=1.19449 acc=0.88281 test_acc_avg=0.67448 test_acc_top5_avg=0.96034 time=231.46it/s +epoch=9 global_step=3910 loss=0.14827 test_loss_avg=1.12176 acc=0.93750 test_acc_avg=0.68028 test_acc_top5_avg=0.95411 time=669.91it/s +curr_acc 0.6803 +BEST_ACC 0.6630 +curr_acc_top5 0.9541 +BEST_ACC_top5 0.9482 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=5.92428 loss_avg=5.18140 acc=0.39844 acc_top1_avg=0.48672 acc_top5_avg=0.83828 lr=0.01000 gn=6.19220 time=56.69it/s +epoch=10 global_step=4000 loss=5.46022 loss_avg=5.18305 acc=0.46875 acc_top1_avg=0.48568 acc_top5_avg=0.84080 lr=0.01000 gn=7.45538 time=56.29it/s +epoch=10 global_step=4050 loss=5.61342 loss_avg=5.17409 acc=0.44531 acc_top1_avg=0.48633 acc_top5_avg=0.84191 lr=0.01000 gn=4.79452 time=55.03it/s +epoch=10 global_step=4100 loss=5.18563 loss_avg=5.16227 acc=0.48438 acc_top1_avg=0.48725 acc_top5_avg=0.84404 lr=0.01000 gn=5.69446 time=54.49it/s +epoch=10 global_step=4150 loss=5.16354 loss_avg=5.18024 acc=0.48438 acc_top1_avg=0.48519 acc_top5_avg=0.84486 lr=0.01000 gn=5.49596 time=59.75it/s +epoch=10 global_step=4200 loss=5.68050 loss_avg=5.19327 acc=0.42969 acc_top1_avg=0.48349 acc_top5_avg=0.84413 lr=0.01000 gn=5.89684 time=52.40it/s +epoch=10 global_step=4250 loss=5.24083 loss_avg=5.20328 acc=0.46094 acc_top1_avg=0.48235 acc_top5_avg=0.84449 lr=0.01000 gn=6.72655 time=63.01it/s +epoch=10 global_step=4300 loss=5.12990 loss_avg=5.18869 acc=0.47656 acc_top1_avg=0.48369 acc_top5_avg=0.84551 lr=0.01000 gn=5.28221 time=56.02it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.78512 test_loss_avg=1.62965 acc=0.75781 test_acc_avg=0.60703 test_acc_top5_avg=0.98359 time=212.91it/s +epoch=10 global_step=4301 loss=1.08464 test_loss_avg=1.48185 acc=0.67969 test_acc_avg=0.61406 test_acc_top5_avg=0.92656 time=242.36it/s +epoch=10 global_step=4301 loss=0.14003 test_loss_avg=1.20858 acc=0.93750 test_acc_avg=0.68265 test_acc_top5_avg=0.94205 time=696.96it/s +curr_acc 0.6827 +BEST_ACC 0.6803 +curr_acc_top5 0.9420 +BEST_ACC_top5 0.9541 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=5.31776 loss_avg=5.18799 acc=0.44531 acc_top1_avg=0.48820 acc_top5_avg=0.85762 lr=0.01000 gn=7.16821 time=52.42it/s +epoch=11 global_step=4400 loss=5.51901 loss_avg=5.14721 acc=0.45312 acc_top1_avg=0.48879 acc_top5_avg=0.85472 lr=0.01000 gn=5.90253 time=58.48it/s +epoch=11 global_step=4450 loss=5.68983 loss_avg=5.18319 acc=0.43750 acc_top1_avg=0.48417 acc_top5_avg=0.85272 lr=0.01000 gn=5.91874 time=52.52it/s +epoch=11 global_step=4500 loss=5.47244 loss_avg=5.18174 acc=0.46875 acc_top1_avg=0.48422 acc_top5_avg=0.85223 lr=0.01000 gn=7.31127 time=54.61it/s +epoch=11 global_step=4550 loss=5.03108 loss_avg=5.17076 acc=0.49219 acc_top1_avg=0.48660 acc_top5_avg=0.85028 lr=0.01000 gn=5.80626 time=52.26it/s +epoch=11 global_step=4600 loss=5.34543 loss_avg=5.17058 acc=0.46094 acc_top1_avg=0.48644 acc_top5_avg=0.84937 lr=0.01000 gn=5.86273 time=54.58it/s +epoch=11 global_step=4650 loss=5.21297 loss_avg=5.17267 acc=0.46094 acc_top1_avg=0.48576 acc_top5_avg=0.84961 lr=0.01000 gn=5.51646 time=53.49it/s +====================Eval==================== +epoch=11 global_step=4692 loss=2.11037 test_loss_avg=1.68989 acc=0.46875 test_acc_avg=0.57434 test_acc_top5_avg=0.92918 time=223.13it/s +epoch=11 global_step=4692 loss=0.03723 test_loss_avg=1.41231 acc=1.00000 test_acc_avg=0.63064 test_acc_top5_avg=0.95243 time=456.80it/s +curr_acc 0.6306 +BEST_ACC 0.6827 +curr_acc_top5 0.9524 +BEST_ACC_top5 0.9541 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=5.55508 loss_avg=5.12798 acc=0.44531 acc_top1_avg=0.49414 acc_top5_avg=0.84961 lr=0.01000 gn=7.19924 time=47.21it/s +epoch=12 global_step=4750 loss=5.01093 loss_avg=5.13285 acc=0.50000 acc_top1_avg=0.48936 acc_top5_avg=0.84725 lr=0.01000 gn=7.20713 time=55.69it/s +epoch=12 global_step=4800 loss=5.06829 loss_avg=5.14248 acc=0.50000 acc_top1_avg=0.49016 acc_top5_avg=0.84585 lr=0.01000 gn=6.07374 time=63.36it/s +epoch=12 global_step=4850 loss=5.25716 loss_avg=5.17214 acc=0.46094 acc_top1_avg=0.48749 acc_top5_avg=0.84652 lr=0.01000 gn=5.30720 time=54.39it/s +epoch=12 global_step=4900 loss=4.93673 loss_avg=5.16813 acc=0.50000 acc_top1_avg=0.48719 acc_top5_avg=0.84724 lr=0.01000 gn=5.80897 time=57.34it/s +epoch=12 global_step=4950 loss=5.00684 loss_avg=5.15557 acc=0.49219 acc_top1_avg=0.48919 acc_top5_avg=0.84820 lr=0.01000 gn=5.31080 time=52.00it/s +epoch=12 global_step=5000 loss=5.24532 loss_avg=5.14695 acc=0.48438 acc_top1_avg=0.49021 acc_top5_avg=0.84865 lr=0.01000 gn=6.78048 time=54.93it/s +epoch=12 global_step=5050 loss=4.69045 loss_avg=5.14977 acc=0.53906 acc_top1_avg=0.48977 acc_top5_avg=0.84892 lr=0.01000 gn=6.30061 time=53.81it/s +====================Eval==================== +epoch=12 global_step=5083 loss=1.23803 test_loss_avg=1.23358 acc=0.62500 test_acc_avg=0.64062 test_acc_top5_avg=0.98047 time=229.30it/s +epoch=12 global_step=5083 loss=1.83825 test_loss_avg=1.25651 acc=0.42188 test_acc_avg=0.64213 test_acc_top5_avg=0.95087 time=244.59it/s +epoch=12 global_step=5083 loss=0.18144 test_loss_avg=1.05626 acc=0.93750 test_acc_avg=0.69897 test_acc_top5_avg=0.95985 time=472.65it/s +curr_acc 0.6990 +BEST_ACC 0.6827 +curr_acc_top5 0.9598 +BEST_ACC_top5 0.9541 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=4.59791 loss_avg=5.01555 acc=0.53906 acc_top1_avg=0.50322 acc_top5_avg=0.85202 lr=0.01000 gn=5.78819 time=63.25it/s +epoch=13 global_step=5150 loss=5.34506 loss_avg=5.06153 acc=0.44531 acc_top1_avg=0.49767 acc_top5_avg=0.85156 lr=0.01000 gn=7.02216 time=53.60it/s +epoch=13 global_step=5200 loss=5.28975 loss_avg=5.09964 acc=0.46094 acc_top1_avg=0.49439 acc_top5_avg=0.84882 lr=0.01000 gn=6.58801 time=55.47it/s +epoch=13 global_step=5250 loss=5.64150 loss_avg=5.07412 acc=0.42969 acc_top1_avg=0.49771 acc_top5_avg=0.85025 lr=0.01000 gn=5.86595 time=63.29it/s +epoch=13 global_step=5300 loss=4.86468 loss_avg=5.08786 acc=0.52344 acc_top1_avg=0.49622 acc_top5_avg=0.85174 lr=0.01000 gn=5.53459 time=62.14it/s +epoch=13 global_step=5350 loss=4.86285 loss_avg=5.09734 acc=0.51562 acc_top1_avg=0.49470 acc_top5_avg=0.85083 lr=0.01000 gn=6.68449 time=54.68it/s +epoch=13 global_step=5400 loss=5.03090 loss_avg=5.11454 acc=0.50000 acc_top1_avg=0.49290 acc_top5_avg=0.85085 lr=0.01000 gn=7.10111 time=56.28it/s +epoch=13 global_step=5450 loss=5.55689 loss_avg=5.10519 acc=0.43750 acc_top1_avg=0.49383 acc_top5_avg=0.85148 lr=0.01000 gn=5.63581 time=59.54it/s +====================Eval==================== +epoch=13 global_step=5474 loss=2.67761 test_loss_avg=1.83438 acc=0.42188 test_acc_avg=0.58662 test_acc_top5_avg=0.90082 time=235.40it/s +epoch=13 global_step=5474 loss=0.63867 test_loss_avg=1.66072 acc=0.80469 test_acc_avg=0.60113 test_acc_top5_avg=0.94328 time=240.10it/s +epoch=13 global_step=5474 loss=0.58255 test_loss_avg=1.58485 acc=0.87500 test_acc_avg=0.61867 test_acc_top5_avg=0.94640 time=533.56it/s +curr_acc 0.6187 +BEST_ACC 0.6990 +curr_acc_top5 0.9464 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=4.82351 loss_avg=5.06615 acc=0.54688 acc_top1_avg=0.49970 acc_top5_avg=0.86448 lr=0.01000 gn=7.36688 time=57.30it/s +epoch=14 global_step=5550 loss=5.09827 loss_avg=5.05046 acc=0.50000 acc_top1_avg=0.50154 acc_top5_avg=0.85722 lr=0.01000 gn=6.36528 time=55.31it/s +epoch=14 global_step=5600 loss=4.88056 loss_avg=5.10545 acc=0.50781 acc_top1_avg=0.49448 acc_top5_avg=0.85596 lr=0.01000 gn=5.08130 time=59.84it/s +epoch=14 global_step=5650 loss=5.70532 loss_avg=5.12627 acc=0.41406 acc_top1_avg=0.49214 acc_top5_avg=0.85050 lr=0.01000 gn=7.24614 time=55.95it/s +epoch=14 global_step=5700 loss=5.17145 loss_avg=5.12770 acc=0.50781 acc_top1_avg=0.49208 acc_top5_avg=0.84835 lr=0.01000 gn=7.16375 time=61.98it/s +epoch=14 global_step=5750 loss=5.69864 loss_avg=5.13455 acc=0.41406 acc_top1_avg=0.49165 acc_top5_avg=0.84825 lr=0.01000 gn=6.57850 time=53.50it/s +epoch=14 global_step=5800 loss=5.09215 loss_avg=5.12592 acc=0.50000 acc_top1_avg=0.49283 acc_top5_avg=0.84967 lr=0.01000 gn=6.23211 time=59.96it/s +epoch=14 global_step=5850 loss=5.12942 loss_avg=5.12536 acc=0.47656 acc_top1_avg=0.49231 acc_top5_avg=0.84963 lr=0.01000 gn=5.45242 time=53.81it/s +====================Eval==================== +epoch=14 global_step=5865 loss=2.96203 test_loss_avg=1.32266 acc=0.28125 test_acc_avg=0.64329 test_acc_top5_avg=0.95153 time=115.74it/s +epoch=14 global_step=5865 loss=0.48603 test_loss_avg=1.21503 acc=0.81250 test_acc_avg=0.64844 test_acc_top5_avg=0.95421 time=863.91it/s +curr_acc 0.6484 +BEST_ACC 0.6990 +curr_acc_top5 0.9542 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=4.73107 loss_avg=5.10459 acc=0.54688 acc_top1_avg=0.49799 acc_top5_avg=0.84978 lr=0.01000 gn=6.17148 time=59.93it/s +epoch=15 global_step=5950 loss=5.27829 loss_avg=5.12626 acc=0.45312 acc_top1_avg=0.49347 acc_top5_avg=0.85028 lr=0.01000 gn=8.34068 time=57.43it/s +epoch=15 global_step=6000 loss=4.95599 loss_avg=5.10205 acc=0.51562 acc_top1_avg=0.49693 acc_top5_avg=0.85122 lr=0.01000 gn=6.02544 time=59.37it/s +epoch=15 global_step=6050 loss=4.93183 loss_avg=5.08772 acc=0.50781 acc_top1_avg=0.49738 acc_top5_avg=0.85393 lr=0.01000 gn=5.97170 time=58.97it/s +epoch=15 global_step=6100 loss=5.24153 loss_avg=5.10081 acc=0.46875 acc_top1_avg=0.49611 acc_top5_avg=0.85322 lr=0.01000 gn=6.00012 time=61.70it/s +epoch=15 global_step=6150 loss=5.07378 loss_avg=5.11367 acc=0.52344 acc_top1_avg=0.49449 acc_top5_avg=0.85203 lr=0.01000 gn=8.13619 time=59.58it/s +epoch=15 global_step=6200 loss=4.92183 loss_avg=5.11809 acc=0.51562 acc_top1_avg=0.49375 acc_top5_avg=0.85163 lr=0.01000 gn=6.34674 time=48.14it/s +epoch=15 global_step=6250 loss=4.74629 loss_avg=5.12218 acc=0.53125 acc_top1_avg=0.49324 acc_top5_avg=0.85207 lr=0.01000 gn=7.12427 time=54.40it/s +====================Eval==================== +epoch=15 global_step=6256 loss=0.40072 test_loss_avg=1.10066 acc=0.90625 test_acc_avg=0.73750 test_acc_top5_avg=0.96667 time=243.27it/s +epoch=15 global_step=6256 loss=0.60187 test_loss_avg=1.50159 acc=0.85156 test_acc_avg=0.58906 test_acc_top5_avg=0.94748 time=239.88it/s +epoch=15 global_step=6256 loss=0.00239 test_loss_avg=1.29142 acc=1.00000 test_acc_avg=0.64666 test_acc_top5_avg=0.95540 time=853.54it/s +curr_acc 0.6467 +BEST_ACC 0.6990 +curr_acc_top5 0.9554 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=5.75988 loss_avg=5.13715 acc=0.42188 acc_top1_avg=0.49432 acc_top5_avg=0.84783 lr=0.01000 gn=7.83725 time=53.46it/s +epoch=16 global_step=6350 loss=4.83745 loss_avg=5.11456 acc=0.52344 acc_top1_avg=0.49576 acc_top5_avg=0.84899 lr=0.01000 gn=8.29526 time=51.09it/s +epoch=16 global_step=6400 loss=5.24234 loss_avg=5.11616 acc=0.50000 acc_top1_avg=0.49523 acc_top5_avg=0.84825 lr=0.01000 gn=6.72212 time=52.88it/s +epoch=16 global_step=6450 loss=4.84884 loss_avg=5.11093 acc=0.51562 acc_top1_avg=0.49601 acc_top5_avg=0.85108 lr=0.01000 gn=5.84108 time=54.99it/s +epoch=16 global_step=6500 loss=4.84395 loss_avg=5.11444 acc=0.53906 acc_top1_avg=0.49555 acc_top5_avg=0.85233 lr=0.01000 gn=6.36904 time=54.53it/s +epoch=16 global_step=6550 loss=4.45037 loss_avg=5.12131 acc=0.55469 acc_top1_avg=0.49455 acc_top5_avg=0.85340 lr=0.01000 gn=5.23488 time=57.46it/s +epoch=16 global_step=6600 loss=5.00663 loss_avg=5.11380 acc=0.53125 acc_top1_avg=0.49500 acc_top5_avg=0.85265 lr=0.01000 gn=6.96569 time=52.16it/s +====================Eval==================== +epoch=16 global_step=6647 loss=0.79940 test_loss_avg=1.71792 acc=0.82031 test_acc_avg=0.58919 test_acc_top5_avg=0.94792 time=237.89it/s +epoch=16 global_step=6647 loss=0.10396 test_loss_avg=1.50508 acc=0.93750 test_acc_avg=0.62777 test_acc_top5_avg=0.95016 time=530.52it/s +curr_acc 0.6278 +BEST_ACC 0.6990 +curr_acc_top5 0.9502 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=4.54638 loss_avg=5.17333 acc=0.55469 acc_top1_avg=0.49479 acc_top5_avg=0.88281 lr=0.01000 gn=6.57443 time=52.92it/s +epoch=17 global_step=6700 loss=4.79851 loss_avg=5.13975 acc=0.53906 acc_top1_avg=0.49130 acc_top5_avg=0.85318 lr=0.01000 gn=6.27192 time=51.37it/s +epoch=17 global_step=6750 loss=4.54131 loss_avg=5.09288 acc=0.55469 acc_top1_avg=0.49636 acc_top5_avg=0.85680 lr=0.01000 gn=8.01952 time=57.23it/s +epoch=17 global_step=6800 loss=4.97625 loss_avg=5.11701 acc=0.51562 acc_top1_avg=0.49403 acc_top5_avg=0.85544 lr=0.01000 gn=7.24017 time=52.79it/s +epoch=17 global_step=6850 loss=4.83330 loss_avg=5.09954 acc=0.51562 acc_top1_avg=0.49565 acc_top5_avg=0.85337 lr=0.01000 gn=5.88363 time=46.95it/s +epoch=17 global_step=6900 loss=4.74092 loss_avg=5.12042 acc=0.53125 acc_top1_avg=0.49336 acc_top5_avg=0.85416 lr=0.01000 gn=6.95100 time=53.55it/s +epoch=17 global_step=6950 loss=4.72522 loss_avg=5.10103 acc=0.53906 acc_top1_avg=0.49567 acc_top5_avg=0.85507 lr=0.01000 gn=7.07032 time=62.37it/s +epoch=17 global_step=7000 loss=4.95896 loss_avg=5.09883 acc=0.50781 acc_top1_avg=0.49535 acc_top5_avg=0.85475 lr=0.01000 gn=5.60439 time=60.31it/s +====================Eval==================== +epoch=17 global_step=7038 loss=1.49220 test_loss_avg=1.43632 acc=0.56250 test_acc_avg=0.57478 test_acc_top5_avg=0.96540 time=180.07it/s +epoch=17 global_step=7038 loss=0.41537 test_loss_avg=1.40552 acc=0.88281 test_acc_avg=0.59718 test_acc_top5_avg=0.94874 time=240.78it/s +epoch=17 global_step=7038 loss=0.26236 test_loss_avg=1.12753 acc=0.87500 test_acc_avg=0.67781 test_acc_top5_avg=0.96034 time=818.88it/s +curr_acc 0.6778 +BEST_ACC 0.6990 +curr_acc_top5 0.9603 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=5.21160 loss_avg=5.03527 acc=0.46875 acc_top1_avg=0.50846 acc_top5_avg=0.85742 lr=0.01000 gn=7.35580 time=55.93it/s +epoch=18 global_step=7100 loss=4.82055 loss_avg=5.08050 acc=0.50781 acc_top1_avg=0.49408 acc_top5_avg=0.85522 lr=0.01000 gn=8.21905 time=59.60it/s +epoch=18 global_step=7150 loss=5.33548 loss_avg=5.10645 acc=0.46094 acc_top1_avg=0.49226 acc_top5_avg=0.85317 lr=0.01000 gn=6.49098 time=55.78it/s +epoch=18 global_step=7200 loss=5.41837 loss_avg=5.11341 acc=0.45312 acc_top1_avg=0.49253 acc_top5_avg=0.85195 lr=0.01000 gn=8.55592 time=57.77it/s +epoch=18 global_step=7250 loss=5.31748 loss_avg=5.10432 acc=0.46875 acc_top1_avg=0.49403 acc_top5_avg=0.85204 lr=0.01000 gn=6.94247 time=56.44it/s +epoch=18 global_step=7300 loss=4.94909 loss_avg=5.08878 acc=0.50781 acc_top1_avg=0.49580 acc_top5_avg=0.85177 lr=0.01000 gn=8.77075 time=57.60it/s +epoch=18 global_step=7350 loss=4.86672 loss_avg=5.08973 acc=0.50781 acc_top1_avg=0.49589 acc_top5_avg=0.85191 lr=0.01000 gn=6.29488 time=63.77it/s +epoch=18 global_step=7400 loss=5.19402 loss_avg=5.10145 acc=0.47656 acc_top1_avg=0.49435 acc_top5_avg=0.85137 lr=0.01000 gn=6.36995 time=51.12it/s +====================Eval==================== +epoch=18 global_step=7429 loss=4.69939 test_loss_avg=1.50063 acc=0.08594 test_acc_avg=0.67132 test_acc_top5_avg=0.93443 time=241.18it/s +epoch=18 global_step=7429 loss=0.22361 test_loss_avg=1.71898 acc=0.93750 test_acc_avg=0.60717 test_acc_top5_avg=0.91336 time=256.25it/s +epoch=18 global_step=7429 loss=0.37334 test_loss_avg=1.70195 acc=0.87500 test_acc_avg=0.61056 test_acc_top5_avg=0.91446 time=869.29it/s +curr_acc 0.6106 +BEST_ACC 0.6990 +curr_acc_top5 0.9145 +BEST_ACC_top5 0.9603 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=5.28139 loss_avg=5.20110 acc=0.46875 acc_top1_avg=0.48289 acc_top5_avg=0.84301 lr=0.01000 gn=5.65286 time=60.41it/s +epoch=19 global_step=7500 loss=5.39627 loss_avg=5.07830 acc=0.46094 acc_top1_avg=0.49615 acc_top5_avg=0.84793 lr=0.01000 gn=7.46368 time=56.11it/s +epoch=19 global_step=7550 loss=5.28585 loss_avg=5.02827 acc=0.50000 acc_top1_avg=0.50168 acc_top5_avg=0.84911 lr=0.01000 gn=7.02710 time=55.75it/s +epoch=19 global_step=7600 loss=4.68929 loss_avg=5.04569 acc=0.54688 acc_top1_avg=0.50032 acc_top5_avg=0.84864 lr=0.01000 gn=7.67741 time=59.16it/s +epoch=19 global_step=7650 loss=4.71948 loss_avg=5.05190 acc=0.54688 acc_top1_avg=0.49996 acc_top5_avg=0.84979 lr=0.01000 gn=6.84773 time=50.06it/s +epoch=19 global_step=7700 loss=5.10915 loss_avg=5.06444 acc=0.51562 acc_top1_avg=0.49905 acc_top5_avg=0.85139 lr=0.01000 gn=6.75583 time=59.13it/s +epoch=19 global_step=7750 loss=5.05418 loss_avg=5.08008 acc=0.50000 acc_top1_avg=0.49698 acc_top5_avg=0.85120 lr=0.01000 gn=6.99075 time=57.35it/s +epoch=19 global_step=7800 loss=5.66955 loss_avg=5.09485 acc=0.43750 acc_top1_avg=0.49539 acc_top5_avg=0.85066 lr=0.01000 gn=7.15550 time=62.29it/s +====================Eval==================== +epoch=19 global_step=7820 loss=1.43863 test_loss_avg=1.02717 acc=0.50781 test_acc_avg=0.71572 test_acc_top5_avg=0.96126 time=234.74it/s +epoch=19 global_step=7820 loss=0.73813 test_loss_avg=0.96896 acc=0.81250 test_acc_avg=0.72488 test_acc_top5_avg=0.96964 time=851.29it/s +curr_acc 0.7249 +BEST_ACC 0.6990 +curr_acc_top5 0.9696 +BEST_ACC_top5 0.9603 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=4.40231 loss_avg=4.93209 acc=0.57812 acc_top1_avg=0.51354 acc_top5_avg=0.85990 lr=0.01000 gn=6.20373 time=58.01it/s +epoch=20 global_step=7900 loss=4.86543 loss_avg=5.03910 acc=0.50781 acc_top1_avg=0.50107 acc_top5_avg=0.85244 lr=0.01000 gn=5.42805 time=51.86it/s +epoch=20 global_step=7950 loss=5.30296 loss_avg=5.06438 acc=0.47656 acc_top1_avg=0.49820 acc_top5_avg=0.85144 lr=0.01000 gn=6.00610 time=53.41it/s +epoch=20 global_step=8000 loss=4.89723 loss_avg=5.06061 acc=0.51562 acc_top1_avg=0.49848 acc_top5_avg=0.85273 lr=0.01000 gn=5.68519 time=58.75it/s +epoch=20 global_step=8050 loss=4.63222 loss_avg=5.07296 acc=0.54688 acc_top1_avg=0.49755 acc_top5_avg=0.85319 lr=0.01000 gn=7.48375 time=56.11it/s +epoch=20 global_step=8100 loss=5.29595 loss_avg=5.08082 acc=0.48438 acc_top1_avg=0.49632 acc_top5_avg=0.85472 lr=0.01000 gn=7.48212 time=54.44it/s +epoch=20 global_step=8150 loss=5.40182 loss_avg=5.09340 acc=0.44531 acc_top1_avg=0.49474 acc_top5_avg=0.85447 lr=0.01000 gn=6.30580 time=55.62it/s +epoch=20 global_step=8200 loss=5.19762 loss_avg=5.08624 acc=0.47656 acc_top1_avg=0.49609 acc_top5_avg=0.85436 lr=0.01000 gn=8.23664 time=54.66it/s +====================Eval==================== +epoch=20 global_step=8211 loss=1.49504 test_loss_avg=0.68088 acc=0.61719 test_acc_avg=0.80430 test_acc_top5_avg=0.98008 time=233.90it/s +epoch=20 global_step=8211 loss=0.26940 test_loss_avg=1.10055 acc=0.92188 test_acc_avg=0.67589 test_acc_top5_avg=0.94442 time=236.87it/s +epoch=20 global_step=8211 loss=0.23599 test_loss_avg=1.02833 acc=0.87500 test_acc_avg=0.69610 test_acc_top5_avg=0.94877 time=505.83it/s +curr_acc 0.6961 +BEST_ACC 0.7249 +curr_acc_top5 0.9488 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=5.09066 loss_avg=4.96149 acc=0.48438 acc_top1_avg=0.51302 acc_top5_avg=0.86138 lr=0.01000 gn=7.82528 time=55.99it/s +epoch=21 global_step=8300 loss=5.90217 loss_avg=4.96068 acc=0.40625 acc_top1_avg=0.51334 acc_top5_avg=0.85736 lr=0.01000 gn=6.50526 time=52.49it/s +epoch=21 global_step=8350 loss=5.57857 loss_avg=5.01028 acc=0.42969 acc_top1_avg=0.50573 acc_top5_avg=0.85763 lr=0.01000 gn=6.82146 time=56.59it/s +epoch=21 global_step=8400 loss=5.25655 loss_avg=5.06213 acc=0.46875 acc_top1_avg=0.49996 acc_top5_avg=0.85441 lr=0.01000 gn=6.97236 time=58.31it/s +epoch=21 global_step=8450 loss=4.65049 loss_avg=5.08161 acc=0.52344 acc_top1_avg=0.49722 acc_top5_avg=0.85450 lr=0.01000 gn=6.94027 time=53.27it/s +epoch=21 global_step=8500 loss=4.73655 loss_avg=5.07552 acc=0.56250 acc_top1_avg=0.49838 acc_top5_avg=0.85516 lr=0.01000 gn=7.05847 time=56.83it/s +epoch=21 global_step=8550 loss=4.67321 loss_avg=5.06018 acc=0.54688 acc_top1_avg=0.49968 acc_top5_avg=0.85562 lr=0.01000 gn=7.54343 time=57.85it/s +epoch=21 global_step=8600 loss=4.71120 loss_avg=5.06436 acc=0.53906 acc_top1_avg=0.49904 acc_top5_avg=0.85560 lr=0.01000 gn=5.25212 time=63.76it/s +====================Eval==================== +epoch=21 global_step=8602 loss=2.83781 test_loss_avg=1.23759 acc=0.35156 test_acc_avg=0.67283 test_acc_top5_avg=0.94074 time=233.29it/s +epoch=21 global_step=8602 loss=0.01294 test_loss_avg=1.10043 acc=1.00000 test_acc_avg=0.70184 test_acc_top5_avg=0.94294 time=576.22it/s +curr_acc 0.7018 +BEST_ACC 0.7249 +curr_acc_top5 0.9429 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=4.52544 loss_avg=4.98150 acc=0.57812 acc_top1_avg=0.50749 acc_top5_avg=0.85693 lr=0.01000 gn=7.63557 time=54.26it/s +epoch=22 global_step=8700 loss=5.71311 loss_avg=5.03092 acc=0.42969 acc_top1_avg=0.50430 acc_top5_avg=0.85523 lr=0.01000 gn=8.61690 time=52.48it/s +epoch=22 global_step=8750 loss=4.74513 loss_avg=5.06108 acc=0.53906 acc_top1_avg=0.50153 acc_top5_avg=0.85320 lr=0.01000 gn=6.52587 time=53.92it/s +epoch=22 global_step=8800 loss=4.91415 loss_avg=5.05672 acc=0.51562 acc_top1_avg=0.50205 acc_top5_avg=0.85421 lr=0.01000 gn=6.64734 time=52.24it/s +epoch=22 global_step=8850 loss=4.99829 loss_avg=5.07388 acc=0.51562 acc_top1_avg=0.49997 acc_top5_avg=0.85490 lr=0.01000 gn=6.29966 time=54.01it/s +epoch=22 global_step=8900 loss=4.98679 loss_avg=5.07066 acc=0.51562 acc_top1_avg=0.49974 acc_top5_avg=0.85487 lr=0.01000 gn=8.90420 time=53.57it/s +epoch=22 global_step=8950 loss=5.07714 loss_avg=5.05642 acc=0.50000 acc_top1_avg=0.50094 acc_top5_avg=0.85587 lr=0.01000 gn=6.09090 time=60.58it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.73471 test_loss_avg=1.06460 acc=0.81250 test_acc_avg=0.73438 test_acc_top5_avg=0.97266 time=229.60it/s +epoch=22 global_step=8993 loss=0.72316 test_loss_avg=1.38571 acc=0.78906 test_acc_avg=0.60635 test_acc_top5_avg=0.95149 time=232.49it/s +epoch=22 global_step=8993 loss=0.00642 test_loss_avg=1.16259 acc=1.00000 test_acc_avg=0.67237 test_acc_top5_avg=0.95965 time=519.10it/s +curr_acc 0.6724 +BEST_ACC 0.7249 +curr_acc_top5 0.9597 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=4.88295 loss_avg=5.06042 acc=0.53125 acc_top1_avg=0.51004 acc_top5_avg=0.85045 lr=0.01000 gn=7.39905 time=54.53it/s +epoch=23 global_step=9050 loss=5.17848 loss_avg=5.03679 acc=0.48438 acc_top1_avg=0.50397 acc_top5_avg=0.85732 lr=0.01000 gn=7.69533 time=62.30it/s +epoch=23 global_step=9100 loss=4.75486 loss_avg=5.08421 acc=0.53125 acc_top1_avg=0.49679 acc_top5_avg=0.85558 lr=0.01000 gn=7.19112 time=54.94it/s +epoch=23 global_step=9150 loss=5.13021 loss_avg=5.05416 acc=0.48438 acc_top1_avg=0.49990 acc_top5_avg=0.85510 lr=0.01000 gn=6.62023 time=54.32it/s +epoch=23 global_step=9200 loss=4.63649 loss_avg=5.03678 acc=0.53906 acc_top1_avg=0.50204 acc_top5_avg=0.85522 lr=0.01000 gn=6.70484 time=49.47it/s +epoch=23 global_step=9250 loss=5.04710 loss_avg=5.03377 acc=0.50781 acc_top1_avg=0.50258 acc_top5_avg=0.85466 lr=0.01000 gn=6.73045 time=31.76it/s +epoch=23 global_step=9300 loss=4.59146 loss_avg=5.05446 acc=0.54688 acc_top1_avg=0.50031 acc_top5_avg=0.85451 lr=0.01000 gn=6.30097 time=55.35it/s +epoch=23 global_step=9350 loss=5.03088 loss_avg=5.05699 acc=0.50000 acc_top1_avg=0.49956 acc_top5_avg=0.85441 lr=0.01000 gn=7.29574 time=49.60it/s +====================Eval==================== +epoch=23 global_step=9384 loss=0.49855 test_loss_avg=1.51520 acc=0.85938 test_acc_avg=0.63352 test_acc_top5_avg=0.95975 time=118.60it/s +epoch=23 global_step=9384 loss=0.90245 test_loss_avg=1.43043 acc=0.81250 test_acc_avg=0.64399 test_acc_top5_avg=0.94136 time=861.08it/s +curr_acc 0.6440 +BEST_ACC 0.7249 +curr_acc_top5 0.9414 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=4.19954 loss_avg=5.09988 acc=0.60156 acc_top1_avg=0.49072 acc_top5_avg=0.84473 lr=0.01000 gn=6.55889 time=54.55it/s +epoch=24 global_step=9450 loss=4.62839 loss_avg=5.04196 acc=0.54688 acc_top1_avg=0.50201 acc_top5_avg=0.85535 lr=0.01000 gn=7.31934 time=53.66it/s +epoch=24 global_step=9500 loss=4.74824 loss_avg=5.06074 acc=0.53125 acc_top1_avg=0.50054 acc_top5_avg=0.85607 lr=0.01000 gn=6.43330 time=62.44it/s +epoch=24 global_step=9550 loss=4.68017 loss_avg=5.05671 acc=0.54688 acc_top1_avg=0.50099 acc_top5_avg=0.85693 lr=0.01000 gn=6.31343 time=54.31it/s +epoch=24 global_step=9600 loss=5.35016 loss_avg=5.06760 acc=0.46875 acc_top1_avg=0.49910 acc_top5_avg=0.85659 lr=0.01000 gn=7.52334 time=53.89it/s +epoch=24 global_step=9650 loss=6.12098 loss_avg=5.07916 acc=0.39062 acc_top1_avg=0.49715 acc_top5_avg=0.85597 lr=0.01000 gn=7.60601 time=56.91it/s +epoch=24 global_step=9700 loss=5.22199 loss_avg=5.07228 acc=0.48438 acc_top1_avg=0.49805 acc_top5_avg=0.85579 lr=0.01000 gn=6.27566 time=55.37it/s +epoch=24 global_step=9750 loss=5.28397 loss_avg=5.06677 acc=0.47656 acc_top1_avg=0.49866 acc_top5_avg=0.85577 lr=0.01000 gn=6.54514 time=60.62it/s +====================Eval==================== +epoch=24 global_step=9775 loss=0.87884 test_loss_avg=0.85631 acc=0.75000 test_acc_avg=0.76172 test_acc_top5_avg=0.96875 time=231.35it/s +epoch=24 global_step=9775 loss=3.05819 test_loss_avg=1.36914 acc=0.10938 test_acc_avg=0.62818 test_acc_top5_avg=0.95385 time=194.94it/s +epoch=24 global_step=9775 loss=0.22741 test_loss_avg=1.09533 acc=0.93750 test_acc_avg=0.69788 test_acc_top5_avg=0.96559 time=849.57it/s +curr_acc 0.6979 +BEST_ACC 0.7249 +curr_acc_top5 0.9656 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=5.48054 loss_avg=5.01055 acc=0.44531 acc_top1_avg=0.49969 acc_top5_avg=0.85281 lr=0.01000 gn=8.45976 time=63.51it/s +epoch=25 global_step=9850 loss=5.10368 loss_avg=4.99571 acc=0.50781 acc_top1_avg=0.50250 acc_top5_avg=0.85490 lr=0.01000 gn=6.75250 time=57.14it/s +epoch=25 global_step=9900 loss=5.19110 loss_avg=4.99993 acc=0.50781 acc_top1_avg=0.50406 acc_top5_avg=0.85613 lr=0.01000 gn=5.97503 time=56.17it/s +epoch=25 global_step=9950 loss=4.55592 loss_avg=4.96657 acc=0.53906 acc_top1_avg=0.50871 acc_top5_avg=0.85634 lr=0.01000 gn=6.52887 time=51.03it/s +epoch=25 global_step=10000 loss=4.86660 loss_avg=4.97366 acc=0.50781 acc_top1_avg=0.50837 acc_top5_avg=0.85667 lr=0.01000 gn=7.50095 time=59.94it/s +epoch=25 global_step=10050 loss=5.06475 loss_avg=5.00072 acc=0.50781 acc_top1_avg=0.50534 acc_top5_avg=0.85605 lr=0.01000 gn=6.95029 time=55.10it/s +epoch=25 global_step=10100 loss=4.98287 loss_avg=5.01609 acc=0.50781 acc_top1_avg=0.50404 acc_top5_avg=0.85582 lr=0.01000 gn=6.74174 time=50.49it/s +epoch=25 global_step=10150 loss=4.97514 loss_avg=5.03367 acc=0.51562 acc_top1_avg=0.50217 acc_top5_avg=0.85546 lr=0.01000 gn=6.93227 time=35.81it/s +====================Eval==================== +epoch=25 global_step=10166 loss=1.10428 test_loss_avg=1.11974 acc=0.65625 test_acc_avg=0.71281 test_acc_top5_avg=0.95344 time=236.95it/s +epoch=25 global_step=10166 loss=0.14185 test_loss_avg=1.24620 acc=0.95312 test_acc_avg=0.65615 test_acc_top5_avg=0.93917 time=252.91it/s +epoch=25 global_step=10166 loss=0.14424 test_loss_avg=1.19308 acc=0.93750 test_acc_avg=0.67039 test_acc_top5_avg=0.94215 time=851.46it/s +curr_acc 0.6704 +BEST_ACC 0.7249 +curr_acc_top5 0.9421 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=5.52420 loss_avg=5.07513 acc=0.47656 acc_top1_avg=0.49747 acc_top5_avg=0.86420 lr=0.01000 gn=6.47773 time=46.07it/s +epoch=26 global_step=10250 loss=5.90530 loss_avg=5.06726 acc=0.40625 acc_top1_avg=0.50065 acc_top5_avg=0.85928 lr=0.01000 gn=6.77622 time=59.49it/s +epoch=26 global_step=10300 loss=4.40249 loss_avg=5.04870 acc=0.58594 acc_top1_avg=0.50321 acc_top5_avg=0.85652 lr=0.01000 gn=6.79928 time=59.75it/s +epoch=26 global_step=10350 loss=5.60874 loss_avg=5.05756 acc=0.43750 acc_top1_avg=0.50119 acc_top5_avg=0.85555 lr=0.01000 gn=6.57286 time=51.75it/s +epoch=26 global_step=10400 loss=4.69021 loss_avg=5.05271 acc=0.54688 acc_top1_avg=0.50224 acc_top5_avg=0.85367 lr=0.01000 gn=7.89159 time=63.72it/s +epoch=26 global_step=10450 loss=5.58555 loss_avg=5.05249 acc=0.42969 acc_top1_avg=0.50206 acc_top5_avg=0.85442 lr=0.01000 gn=6.20043 time=58.04it/s +epoch=26 global_step=10500 loss=5.02824 loss_avg=5.05864 acc=0.48438 acc_top1_avg=0.50101 acc_top5_avg=0.85495 lr=0.01000 gn=7.30807 time=54.34it/s +epoch=26 global_step=10550 loss=4.72334 loss_avg=5.06070 acc=0.53906 acc_top1_avg=0.50051 acc_top5_avg=0.85457 lr=0.01000 gn=6.66043 time=51.38it/s +====================Eval==================== +epoch=26 global_step=10557 loss=1.15484 test_loss_avg=1.04587 acc=0.66406 test_acc_avg=0.70805 test_acc_top5_avg=0.94022 time=229.54it/s +epoch=26 global_step=10557 loss=0.81563 test_loss_avg=0.99250 acc=0.81250 test_acc_avg=0.70896 test_acc_top5_avg=0.95303 time=772.29it/s +curr_acc 0.7090 +BEST_ACC 0.7249 +curr_acc_top5 0.9530 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=5.31687 loss_avg=5.00737 acc=0.46875 acc_top1_avg=0.50600 acc_top5_avg=0.85392 lr=0.01000 gn=5.98812 time=56.21it/s +epoch=27 global_step=10650 loss=4.80888 loss_avg=5.04725 acc=0.55469 acc_top1_avg=0.50269 acc_top5_avg=0.84946 lr=0.01000 gn=7.80512 time=55.36it/s +epoch=27 global_step=10700 loss=4.71195 loss_avg=5.03545 acc=0.56250 acc_top1_avg=0.50404 acc_top5_avg=0.84981 lr=0.01000 gn=7.55046 time=63.20it/s +epoch=27 global_step=10750 loss=5.16207 loss_avg=5.07067 acc=0.47656 acc_top1_avg=0.50040 acc_top5_avg=0.84982 lr=0.01000 gn=6.39189 time=31.94it/s +epoch=27 global_step=10800 loss=5.03301 loss_avg=5.06835 acc=0.50000 acc_top1_avg=0.50035 acc_top5_avg=0.85204 lr=0.01000 gn=6.28458 time=58.94it/s +epoch=27 global_step=10850 loss=5.02316 loss_avg=5.06619 acc=0.51562 acc_top1_avg=0.49981 acc_top5_avg=0.85324 lr=0.01000 gn=6.64086 time=60.20it/s +epoch=27 global_step=10900 loss=4.79830 loss_avg=5.05613 acc=0.52344 acc_top1_avg=0.50096 acc_top5_avg=0.85304 lr=0.01000 gn=6.76242 time=52.89it/s +====================Eval==================== +epoch=27 global_step=10948 loss=2.64029 test_loss_avg=1.06538 acc=0.39844 test_acc_avg=0.73667 test_acc_top5_avg=0.96553 time=225.89it/s +epoch=27 global_step=10948 loss=0.39162 test_loss_avg=1.57865 acc=0.86719 test_acc_avg=0.58512 test_acc_top5_avg=0.91581 time=250.96it/s +epoch=27 global_step=10948 loss=0.82162 test_loss_avg=1.44612 acc=0.81250 test_acc_avg=0.62193 test_acc_top5_avg=0.92712 time=859.31it/s +curr_acc 0.6219 +BEST_ACC 0.7249 +curr_acc_top5 0.9271 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=4.49654 loss_avg=4.86511 acc=0.53906 acc_top1_avg=0.50781 acc_top5_avg=0.85938 lr=0.01000 gn=6.36438 time=22.70it/s +epoch=28 global_step=11000 loss=5.22560 loss_avg=4.98956 acc=0.46875 acc_top1_avg=0.50481 acc_top5_avg=0.85712 lr=0.01000 gn=6.27981 time=54.41it/s +epoch=28 global_step=11050 loss=5.31943 loss_avg=5.04676 acc=0.50781 acc_top1_avg=0.50046 acc_top5_avg=0.85263 lr=0.01000 gn=7.04111 time=55.02it/s +epoch=28 global_step=11100 loss=5.01014 loss_avg=5.05273 acc=0.47656 acc_top1_avg=0.49913 acc_top5_avg=0.85331 lr=0.01000 gn=5.06022 time=62.17it/s +epoch=28 global_step=11150 loss=4.84313 loss_avg=5.04984 acc=0.54688 acc_top1_avg=0.49992 acc_top5_avg=0.85446 lr=0.01000 gn=6.83757 time=52.10it/s +epoch=28 global_step=11200 loss=5.01242 loss_avg=5.04275 acc=0.49219 acc_top1_avg=0.50130 acc_top5_avg=0.85472 lr=0.01000 gn=6.25481 time=57.44it/s +epoch=28 global_step=11250 loss=4.88561 loss_avg=5.04583 acc=0.52344 acc_top1_avg=0.50044 acc_top5_avg=0.85511 lr=0.01000 gn=7.68996 time=51.40it/s +epoch=28 global_step=11300 loss=4.65179 loss_avg=5.04705 acc=0.53125 acc_top1_avg=0.50042 acc_top5_avg=0.85529 lr=0.01000 gn=6.59893 time=50.55it/s +====================Eval==================== +epoch=28 global_step=11339 loss=0.39096 test_loss_avg=2.06875 acc=0.89062 test_acc_avg=0.53824 test_acc_top5_avg=0.91242 time=204.06it/s +epoch=28 global_step=11339 loss=0.05061 test_loss_avg=1.55151 acc=1.00000 test_acc_avg=0.62747 test_acc_top5_avg=0.94907 time=502.49it/s +curr_acc 0.6275 +BEST_ACC 0.7249 +curr_acc_top5 0.9491 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=5.29963 loss_avg=5.13885 acc=0.46875 acc_top1_avg=0.48864 acc_top5_avg=0.84020 lr=0.01000 gn=7.35058 time=53.16it/s +epoch=29 global_step=11400 loss=5.54271 loss_avg=5.01539 acc=0.45312 acc_top1_avg=0.50423 acc_top5_avg=0.85643 lr=0.01000 gn=5.32915 time=59.34it/s +epoch=29 global_step=11450 loss=4.69078 loss_avg=5.01312 acc=0.53906 acc_top1_avg=0.50457 acc_top5_avg=0.85515 lr=0.01000 gn=6.55264 time=63.72it/s +epoch=29 global_step=11500 loss=4.70039 loss_avg=5.02610 acc=0.53125 acc_top1_avg=0.50340 acc_top5_avg=0.85389 lr=0.01000 gn=6.46694 time=50.72it/s +epoch=29 global_step=11550 loss=4.93516 loss_avg=5.03059 acc=0.51562 acc_top1_avg=0.50300 acc_top5_avg=0.85360 lr=0.01000 gn=8.85436 time=54.73it/s +epoch=29 global_step=11600 loss=4.56272 loss_avg=5.05003 acc=0.55469 acc_top1_avg=0.50036 acc_top5_avg=0.85429 lr=0.01000 gn=7.03473 time=57.92it/s +epoch=29 global_step=11650 loss=5.24888 loss_avg=5.04194 acc=0.49219 acc_top1_avg=0.50138 acc_top5_avg=0.85536 lr=0.01000 gn=8.17570 time=55.03it/s +epoch=29 global_step=11700 loss=5.37513 loss_avg=5.03922 acc=0.45312 acc_top1_avg=0.50171 acc_top5_avg=0.85485 lr=0.01000 gn=6.48205 time=46.76it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.30157 test_loss_avg=0.63513 acc=0.89062 test_acc_avg=0.82118 test_acc_top5_avg=0.97135 time=237.53it/s +epoch=29 global_step=11730 loss=0.86210 test_loss_avg=1.36778 acc=0.74219 test_acc_avg=0.62394 test_acc_top5_avg=0.94055 time=149.53it/s +epoch=29 global_step=11730 loss=0.87344 test_loss_avg=1.16770 acc=0.75000 test_acc_avg=0.67731 test_acc_top5_avg=0.95125 time=438.05it/s +curr_acc 0.6773 +BEST_ACC 0.7249 +curr_acc_top5 0.9512 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=4.80560 loss_avg=5.00973 acc=0.53125 acc_top1_avg=0.50742 acc_top5_avg=0.86992 lr=0.01000 gn=5.39268 time=56.61it/s +epoch=30 global_step=11800 loss=5.42225 loss_avg=5.05081 acc=0.45312 acc_top1_avg=0.50100 acc_top5_avg=0.86194 lr=0.01000 gn=7.24664 time=57.58it/s +epoch=30 global_step=11850 loss=5.48098 loss_avg=5.04531 acc=0.45312 acc_top1_avg=0.50117 acc_top5_avg=0.85853 lr=0.01000 gn=6.42684 time=57.68it/s +epoch=30 global_step=11900 loss=5.44077 loss_avg=5.06738 acc=0.47656 acc_top1_avg=0.49931 acc_top5_avg=0.85588 lr=0.01000 gn=6.23350 time=54.99it/s +epoch=30 global_step=11950 loss=4.95775 loss_avg=5.06401 acc=0.51562 acc_top1_avg=0.49961 acc_top5_avg=0.85604 lr=0.01000 gn=7.60215 time=40.91it/s +epoch=30 global_step=12000 loss=4.67882 loss_avg=5.06636 acc=0.53906 acc_top1_avg=0.49899 acc_top5_avg=0.85541 lr=0.01000 gn=7.01187 time=56.02it/s +epoch=30 global_step=12050 loss=4.36319 loss_avg=5.05564 acc=0.57031 acc_top1_avg=0.50027 acc_top5_avg=0.85588 lr=0.01000 gn=7.40465 time=54.96it/s +epoch=30 global_step=12100 loss=5.49086 loss_avg=5.04832 acc=0.45312 acc_top1_avg=0.50093 acc_top5_avg=0.85579 lr=0.01000 gn=8.26121 time=52.13it/s +====================Eval==================== +epoch=30 global_step=12121 loss=1.64013 test_loss_avg=1.11706 acc=0.53125 test_acc_avg=0.69401 test_acc_top5_avg=0.94115 time=225.19it/s +epoch=30 global_step=12121 loss=0.19581 test_loss_avg=1.07650 acc=0.93750 test_acc_avg=0.68799 test_acc_top5_avg=0.94887 time=511.13it/s +curr_acc 0.6880 +BEST_ACC 0.7249 +curr_acc_top5 0.9489 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=4.37332 loss_avg=5.01290 acc=0.56250 acc_top1_avg=0.50458 acc_top5_avg=0.85991 lr=0.01000 gn=7.09689 time=56.89it/s +epoch=31 global_step=12200 loss=4.74694 loss_avg=5.01640 acc=0.53906 acc_top1_avg=0.50326 acc_top5_avg=0.85720 lr=0.01000 gn=8.22812 time=53.32it/s +epoch=31 global_step=12250 loss=5.19165 loss_avg=5.01673 acc=0.48438 acc_top1_avg=0.50466 acc_top5_avg=0.85659 lr=0.01000 gn=6.42478 time=55.40it/s +epoch=31 global_step=12300 loss=4.77558 loss_avg=4.99590 acc=0.53125 acc_top1_avg=0.50690 acc_top5_avg=0.85711 lr=0.01000 gn=8.46371 time=51.46it/s +epoch=31 global_step=12350 loss=5.16471 loss_avg=5.00102 acc=0.48438 acc_top1_avg=0.50556 acc_top5_avg=0.85801 lr=0.01000 gn=5.73397 time=56.60it/s +epoch=31 global_step=12400 loss=5.22491 loss_avg=5.01271 acc=0.48438 acc_top1_avg=0.50462 acc_top5_avg=0.85758 lr=0.01000 gn=7.17796 time=59.02it/s +epoch=31 global_step=12450 loss=4.73587 loss_avg=5.03619 acc=0.54688 acc_top1_avg=0.50161 acc_top5_avg=0.85636 lr=0.01000 gn=7.06585 time=54.51it/s +epoch=31 global_step=12500 loss=5.02572 loss_avg=5.03774 acc=0.48438 acc_top1_avg=0.50117 acc_top5_avg=0.85637 lr=0.01000 gn=7.48265 time=50.72it/s +====================Eval==================== +epoch=31 global_step=12512 loss=2.08722 test_loss_avg=2.08722 acc=0.51562 test_acc_avg=0.51562 test_acc_top5_avg=0.89062 time=188.20it/s +epoch=31 global_step=12512 loss=1.22158 test_loss_avg=1.46631 acc=0.62500 test_acc_avg=0.63373 test_acc_top5_avg=0.91958 time=236.70it/s +epoch=31 global_step=12512 loss=0.14943 test_loss_avg=1.24281 acc=0.93750 test_acc_avg=0.67959 test_acc_top5_avg=0.93938 time=520.84it/s +curr_acc 0.6796 +BEST_ACC 0.7249 +curr_acc_top5 0.9394 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=5.05211 loss_avg=4.93044 acc=0.51562 acc_top1_avg=0.51254 acc_top5_avg=0.86102 lr=0.01000 gn=7.93419 time=54.34it/s +epoch=32 global_step=12600 loss=4.91123 loss_avg=4.98845 acc=0.50781 acc_top1_avg=0.50577 acc_top5_avg=0.85591 lr=0.01000 gn=7.79974 time=55.14it/s +epoch=32 global_step=12650 loss=5.28040 loss_avg=4.99807 acc=0.47656 acc_top1_avg=0.50459 acc_top5_avg=0.85615 lr=0.01000 gn=7.25715 time=62.00it/s +epoch=32 global_step=12700 loss=5.35721 loss_avg=5.00627 acc=0.44531 acc_top1_avg=0.50370 acc_top5_avg=0.85522 lr=0.01000 gn=6.70485 time=56.19it/s +epoch=32 global_step=12750 loss=4.72668 loss_avg=5.00899 acc=0.52344 acc_top1_avg=0.50391 acc_top5_avg=0.85471 lr=0.01000 gn=6.01665 time=57.04it/s +epoch=32 global_step=12800 loss=4.92498 loss_avg=5.02161 acc=0.51562 acc_top1_avg=0.50274 acc_top5_avg=0.85569 lr=0.01000 gn=5.41519 time=63.84it/s +epoch=32 global_step=12850 loss=4.82177 loss_avg=5.01659 acc=0.49219 acc_top1_avg=0.50294 acc_top5_avg=0.85660 lr=0.01000 gn=6.33015 time=54.81it/s +epoch=32 global_step=12900 loss=4.79251 loss_avg=5.02413 acc=0.52344 acc_top1_avg=0.50211 acc_top5_avg=0.85650 lr=0.01000 gn=7.91439 time=58.91it/s +====================Eval==================== +epoch=32 global_step=12903 loss=2.49425 test_loss_avg=1.15661 acc=0.42188 test_acc_avg=0.72195 test_acc_top5_avg=0.92720 time=239.61it/s +epoch=32 global_step=12903 loss=1.11810 test_loss_avg=1.35190 acc=0.71875 test_acc_avg=0.63780 test_acc_top5_avg=0.93728 time=233.87it/s +epoch=32 global_step=12903 loss=1.09954 test_loss_avg=1.32341 acc=0.75000 test_acc_avg=0.64775 test_acc_top5_avg=0.94195 time=853.89it/s +curr_acc 0.6477 +BEST_ACC 0.7249 +curr_acc_top5 0.9420 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=4.56705 loss_avg=4.88996 acc=0.55469 acc_top1_avg=0.51928 acc_top5_avg=0.86021 lr=0.01000 gn=7.30543 time=54.05it/s +epoch=33 global_step=13000 loss=5.15266 loss_avg=4.96523 acc=0.49219 acc_top1_avg=0.50966 acc_top5_avg=0.85954 lr=0.01000 gn=6.95506 time=62.58it/s +epoch=33 global_step=13050 loss=4.72591 loss_avg=4.96946 acc=0.54688 acc_top1_avg=0.50797 acc_top5_avg=0.86092 lr=0.01000 gn=6.66147 time=53.47it/s +epoch=33 global_step=13100 loss=5.41011 loss_avg=5.02640 acc=0.45312 acc_top1_avg=0.50107 acc_top5_avg=0.85989 lr=0.01000 gn=8.39818 time=56.56it/s +epoch=33 global_step=13150 loss=5.46546 loss_avg=5.02827 acc=0.44531 acc_top1_avg=0.50187 acc_top5_avg=0.86004 lr=0.01000 gn=8.43438 time=55.25it/s +epoch=33 global_step=13200 loss=4.96432 loss_avg=5.03259 acc=0.50781 acc_top1_avg=0.50189 acc_top5_avg=0.85906 lr=0.01000 gn=7.64465 time=50.86it/s +epoch=33 global_step=13250 loss=4.76250 loss_avg=5.03242 acc=0.53906 acc_top1_avg=0.50171 acc_top5_avg=0.85778 lr=0.01000 gn=6.82867 time=56.70it/s +====================Eval==================== +epoch=33 global_step=13294 loss=1.32923 test_loss_avg=1.18304 acc=0.61719 test_acc_avg=0.69059 test_acc_top5_avg=0.95912 time=161.39it/s +epoch=33 global_step=13294 loss=0.35183 test_loss_avg=1.02075 acc=0.93750 test_acc_avg=0.72073 test_acc_top5_avg=0.96766 time=561.26it/s +curr_acc 0.7207 +BEST_ACC 0.7249 +curr_acc_top5 0.9677 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=5.03957 loss_avg=4.78817 acc=0.50000 acc_top1_avg=0.53646 acc_top5_avg=0.86198 lr=0.01000 gn=7.11758 time=58.92it/s +epoch=34 global_step=13350 loss=4.54545 loss_avg=4.93461 acc=0.57812 acc_top1_avg=0.51549 acc_top5_avg=0.86077 lr=0.01000 gn=8.10825 time=53.78it/s +epoch=34 global_step=13400 loss=4.78344 loss_avg=5.02969 acc=0.53906 acc_top1_avg=0.50332 acc_top5_avg=0.85451 lr=0.01000 gn=6.71781 time=58.84it/s +epoch=34 global_step=13450 loss=4.93856 loss_avg=5.02437 acc=0.53125 acc_top1_avg=0.50346 acc_top5_avg=0.85607 lr=0.01000 gn=9.44813 time=53.43it/s +epoch=34 global_step=13500 loss=4.80461 loss_avg=5.02672 acc=0.50781 acc_top1_avg=0.50262 acc_top5_avg=0.85850 lr=0.01000 gn=7.93791 time=56.35it/s +epoch=34 global_step=13550 loss=5.14254 loss_avg=5.03628 acc=0.49219 acc_top1_avg=0.50168 acc_top5_avg=0.85739 lr=0.01000 gn=7.81952 time=58.38it/s +epoch=34 global_step=13600 loss=4.85911 loss_avg=5.03056 acc=0.50000 acc_top1_avg=0.50222 acc_top5_avg=0.85616 lr=0.01000 gn=8.43501 time=59.38it/s +epoch=34 global_step=13650 loss=4.90515 loss_avg=5.03138 acc=0.53125 acc_top1_avg=0.50237 acc_top5_avg=0.85580 lr=0.01000 gn=7.41818 time=53.64it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.20487 test_loss_avg=0.67375 acc=0.93750 test_acc_avg=0.81975 test_acc_top5_avg=0.99442 time=230.24it/s +epoch=34 global_step=13685 loss=0.27954 test_loss_avg=1.47316 acc=0.94531 test_acc_avg=0.59888 test_acc_top5_avg=0.93506 time=240.98it/s +epoch=34 global_step=13685 loss=0.38546 test_loss_avg=1.26504 acc=0.87500 test_acc_avg=0.65516 test_acc_top5_avg=0.94680 time=840.04it/s +curr_acc 0.6552 +BEST_ACC 0.7249 +curr_acc_top5 0.9468 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=5.57377 loss_avg=5.08476 acc=0.43750 acc_top1_avg=0.49167 acc_top5_avg=0.85625 lr=0.01000 gn=6.97326 time=51.13it/s +epoch=35 global_step=13750 loss=4.39852 loss_avg=4.95288 acc=0.56250 acc_top1_avg=0.50925 acc_top5_avg=0.86082 lr=0.01000 gn=7.22438 time=53.51it/s +epoch=35 global_step=13800 loss=5.43487 loss_avg=5.00703 acc=0.46094 acc_top1_avg=0.50503 acc_top5_avg=0.86107 lr=0.01000 gn=6.18601 time=53.37it/s +epoch=35 global_step=13850 loss=4.75731 loss_avg=5.02170 acc=0.55469 acc_top1_avg=0.50365 acc_top5_avg=0.86009 lr=0.01000 gn=7.40884 time=55.87it/s +epoch=35 global_step=13900 loss=4.28357 loss_avg=5.01304 acc=0.59375 acc_top1_avg=0.50407 acc_top5_avg=0.85977 lr=0.01000 gn=5.92533 time=59.77it/s +epoch=35 global_step=13950 loss=4.91070 loss_avg=5.02241 acc=0.53906 acc_top1_avg=0.50342 acc_top5_avg=0.85781 lr=0.01000 gn=6.73933 time=63.30it/s +epoch=35 global_step=14000 loss=4.80308 loss_avg=5.03115 acc=0.51562 acc_top1_avg=0.50258 acc_top5_avg=0.85754 lr=0.01000 gn=5.72610 time=59.84it/s +epoch=35 global_step=14050 loss=5.05001 loss_avg=5.02379 acc=0.51562 acc_top1_avg=0.50349 acc_top5_avg=0.85828 lr=0.01000 gn=7.53298 time=58.92it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.34709 test_loss_avg=1.24895 acc=0.92188 test_acc_avg=0.67121 test_acc_top5_avg=0.96205 time=240.96it/s +epoch=35 global_step=14076 loss=0.00357 test_loss_avg=1.07238 acc=1.00000 test_acc_avg=0.70550 test_acc_top5_avg=0.96232 time=494.26it/s +curr_acc 0.7055 +BEST_ACC 0.7249 +curr_acc_top5 0.9623 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=4.95786 loss_avg=5.02334 acc=0.48438 acc_top1_avg=0.50651 acc_top5_avg=0.86393 lr=0.01000 gn=7.22326 time=59.47it/s +epoch=36 global_step=14150 loss=5.10498 loss_avg=5.06782 acc=0.48438 acc_top1_avg=0.50032 acc_top5_avg=0.85938 lr=0.01000 gn=8.00205 time=54.93it/s +epoch=36 global_step=14200 loss=4.91079 loss_avg=5.03134 acc=0.50781 acc_top1_avg=0.50277 acc_top5_avg=0.86051 lr=0.01000 gn=9.21644 time=50.77it/s +epoch=36 global_step=14250 loss=4.83000 loss_avg=5.02800 acc=0.53125 acc_top1_avg=0.50242 acc_top5_avg=0.85978 lr=0.01000 gn=7.47857 time=59.42it/s +epoch=36 global_step=14300 loss=5.24117 loss_avg=5.00681 acc=0.49219 acc_top1_avg=0.50530 acc_top5_avg=0.86025 lr=0.01000 gn=7.15506 time=57.67it/s +epoch=36 global_step=14350 loss=5.39611 loss_avg=5.00390 acc=0.46875 acc_top1_avg=0.50536 acc_top5_avg=0.85920 lr=0.01000 gn=7.21696 time=55.15it/s +epoch=36 global_step=14400 loss=4.08098 loss_avg=5.02692 acc=0.61719 acc_top1_avg=0.50287 acc_top5_avg=0.85733 lr=0.01000 gn=7.13498 time=60.12it/s +epoch=36 global_step=14450 loss=4.86613 loss_avg=5.02753 acc=0.52344 acc_top1_avg=0.50284 acc_top5_avg=0.85676 lr=0.01000 gn=7.42855 time=55.75it/s +====================Eval==================== +epoch=36 global_step=14467 loss=1.31354 test_loss_avg=1.29613 acc=0.61719 test_acc_avg=0.63411 test_acc_top5_avg=0.96354 time=219.12it/s +epoch=36 global_step=14467 loss=0.82800 test_loss_avg=1.35379 acc=0.78125 test_acc_avg=0.62486 test_acc_top5_avg=0.95550 time=238.49it/s +epoch=36 global_step=14467 loss=0.32715 test_loss_avg=1.10651 acc=0.93750 test_acc_avg=0.69185 test_acc_top5_avg=0.96479 time=492.87it/s +curr_acc 0.6919 +BEST_ACC 0.7249 +curr_acc_top5 0.9648 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=5.01241 loss_avg=4.97522 acc=0.51562 acc_top1_avg=0.50781 acc_top5_avg=0.85393 lr=0.01000 gn=7.78733 time=56.02it/s +epoch=37 global_step=14550 loss=4.47933 loss_avg=5.08674 acc=0.57031 acc_top1_avg=0.49708 acc_top5_avg=0.85495 lr=0.01000 gn=7.66218 time=56.02it/s +epoch=37 global_step=14600 loss=5.26464 loss_avg=5.10558 acc=0.47656 acc_top1_avg=0.49436 acc_top5_avg=0.85520 lr=0.01000 gn=5.87035 time=52.48it/s +epoch=37 global_step=14650 loss=4.95980 loss_avg=5.07121 acc=0.50000 acc_top1_avg=0.49752 acc_top5_avg=0.85596 lr=0.01000 gn=6.61006 time=60.49it/s +epoch=37 global_step=14700 loss=5.21063 loss_avg=5.04786 acc=0.46875 acc_top1_avg=0.50064 acc_top5_avg=0.85595 lr=0.01000 gn=8.61966 time=55.52it/s +epoch=37 global_step=14750 loss=5.07268 loss_avg=5.05118 acc=0.50781 acc_top1_avg=0.50066 acc_top5_avg=0.85559 lr=0.01000 gn=6.85929 time=53.29it/s +epoch=37 global_step=14800 loss=4.83698 loss_avg=5.03670 acc=0.53906 acc_top1_avg=0.50223 acc_top5_avg=0.85604 lr=0.01000 gn=6.75845 time=54.16it/s +epoch=37 global_step=14850 loss=5.58137 loss_avg=5.04035 acc=0.45312 acc_top1_avg=0.50163 acc_top5_avg=0.85564 lr=0.01000 gn=5.91086 time=59.62it/s +====================Eval==================== +epoch=37 global_step=14858 loss=2.81915 test_loss_avg=1.25088 acc=0.25781 test_acc_avg=0.65914 test_acc_top5_avg=0.94387 time=232.37it/s +epoch=37 global_step=14858 loss=0.27165 test_loss_avg=1.07757 acc=0.92969 test_acc_avg=0.69724 test_acc_top5_avg=0.94450 time=241.43it/s +epoch=37 global_step=14858 loss=0.29365 test_loss_avg=1.05675 acc=0.87500 test_acc_avg=0.70253 test_acc_top5_avg=0.94591 time=509.57it/s +curr_acc 0.7025 +BEST_ACC 0.7249 +curr_acc_top5 0.9459 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=5.60003 loss_avg=5.02735 acc=0.43750 acc_top1_avg=0.50391 acc_top5_avg=0.85100 lr=0.01000 gn=8.09050 time=53.20it/s +epoch=38 global_step=14950 loss=4.57341 loss_avg=4.98381 acc=0.53906 acc_top1_avg=0.50790 acc_top5_avg=0.85521 lr=0.01000 gn=6.16543 time=63.26it/s +epoch=38 global_step=15000 loss=5.30937 loss_avg=4.99427 acc=0.44531 acc_top1_avg=0.50660 acc_top5_avg=0.85404 lr=0.01000 gn=7.78034 time=60.92it/s +epoch=38 global_step=15050 loss=5.50276 loss_avg=4.99712 acc=0.45312 acc_top1_avg=0.50675 acc_top5_avg=0.85433 lr=0.01000 gn=7.28663 time=53.22it/s +epoch=38 global_step=15100 loss=4.82588 loss_avg=4.99209 acc=0.55469 acc_top1_avg=0.50743 acc_top5_avg=0.85553 lr=0.01000 gn=8.43828 time=55.14it/s +epoch=38 global_step=15150 loss=5.35245 loss_avg=5.01423 acc=0.48438 acc_top1_avg=0.50511 acc_top5_avg=0.85571 lr=0.01000 gn=6.44354 time=56.78it/s +epoch=38 global_step=15200 loss=5.09984 loss_avg=5.02155 acc=0.50000 acc_top1_avg=0.50393 acc_top5_avg=0.85501 lr=0.01000 gn=8.89179 time=54.54it/s +====================Eval==================== +epoch=38 global_step=15249 loss=3.98935 test_loss_avg=1.69539 acc=0.03125 test_acc_avg=0.60531 test_acc_top5_avg=0.91911 time=228.72it/s +epoch=38 global_step=15249 loss=0.00040 test_loss_avg=1.60000 acc=1.00000 test_acc_avg=0.61630 test_acc_top5_avg=0.92939 time=502.85it/s +curr_acc 0.6163 +BEST_ACC 0.7249 +curr_acc_top5 0.9294 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=5.15447 loss_avg=5.15447 acc=0.50000 acc_top1_avg=0.50000 acc_top5_avg=0.82812 lr=0.01000 gn=6.96999 time=50.44it/s +epoch=39 global_step=15300 loss=5.56804 loss_avg=4.95423 acc=0.46094 acc_top1_avg=0.51180 acc_top5_avg=0.85983 lr=0.01000 gn=7.32118 time=54.86it/s +epoch=39 global_step=15350 loss=5.18734 loss_avg=5.00613 acc=0.48438 acc_top1_avg=0.50526 acc_top5_avg=0.85760 lr=0.01000 gn=7.70258 time=48.82it/s +epoch=39 global_step=15400 loss=4.54628 loss_avg=4.99738 acc=0.56250 acc_top1_avg=0.50647 acc_top5_avg=0.85720 lr=0.01000 gn=6.13094 time=63.59it/s +epoch=39 global_step=15450 loss=5.23978 loss_avg=4.99873 acc=0.46094 acc_top1_avg=0.50630 acc_top5_avg=0.85829 lr=0.01000 gn=6.18886 time=53.86it/s +epoch=39 global_step=15500 loss=5.63131 loss_avg=5.01159 acc=0.43750 acc_top1_avg=0.50526 acc_top5_avg=0.85807 lr=0.01000 gn=7.18980 time=55.80it/s +epoch=39 global_step=15550 loss=5.31381 loss_avg=5.03129 acc=0.46875 acc_top1_avg=0.50298 acc_top5_avg=0.85719 lr=0.01000 gn=6.36810 time=55.41it/s +epoch=39 global_step=15600 loss=5.62442 loss_avg=5.01748 acc=0.43750 acc_top1_avg=0.50434 acc_top5_avg=0.85737 lr=0.01000 gn=6.06482 time=43.83it/s +====================Eval==================== +epoch=39 global_step=15640 loss=0.97168 test_loss_avg=0.59012 acc=0.74219 test_acc_avg=0.84005 test_acc_top5_avg=0.98602 time=223.01it/s +epoch=39 global_step=15640 loss=0.34718 test_loss_avg=1.07202 acc=0.90625 test_acc_avg=0.69022 test_acc_top5_avg=0.95494 time=235.83it/s +epoch=39 global_step=15640 loss=0.45177 test_loss_avg=0.99422 acc=0.87500 test_acc_avg=0.71430 test_acc_top5_avg=0.95945 time=841.38it/s +curr_acc 0.7143 +BEST_ACC 0.7249 +curr_acc_top5 0.9595 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=4.97972 loss_avg=4.82029 acc=0.50781 acc_top1_avg=0.52109 acc_top5_avg=0.87187 lr=0.00100 gn=5.44098 time=55.28it/s +epoch=40 global_step=15700 loss=4.79098 loss_avg=4.83242 acc=0.54688 acc_top1_avg=0.52305 acc_top5_avg=0.86875 lr=0.00100 gn=6.67314 time=54.38it/s +epoch=40 global_step=15750 loss=5.13837 loss_avg=4.80926 acc=0.47656 acc_top1_avg=0.52464 acc_top5_avg=0.87074 lr=0.00100 gn=6.23035 time=50.39it/s +epoch=40 global_step=15800 loss=4.67822 loss_avg=4.78016 acc=0.53906 acc_top1_avg=0.52808 acc_top5_avg=0.87192 lr=0.00100 gn=6.58726 time=59.80it/s +epoch=40 global_step=15850 loss=4.67841 loss_avg=4.75295 acc=0.54688 acc_top1_avg=0.53132 acc_top5_avg=0.87225 lr=0.00100 gn=7.80913 time=55.11it/s +epoch=40 global_step=15900 loss=3.86474 loss_avg=4.72645 acc=0.59375 acc_top1_avg=0.53356 acc_top5_avg=0.87151 lr=0.00100 gn=4.12166 time=60.27it/s +epoch=40 global_step=15950 loss=3.78104 loss_avg=4.70503 acc=0.65625 acc_top1_avg=0.53576 acc_top5_avg=0.87127 lr=0.00100 gn=6.69693 time=53.45it/s +epoch=40 global_step=16000 loss=4.20601 loss_avg=4.69574 acc=0.60156 acc_top1_avg=0.53694 acc_top5_avg=0.87151 lr=0.00100 gn=6.05378 time=59.35it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.95383 test_loss_avg=0.73739 acc=0.74219 test_acc_avg=0.79707 test_acc_top5_avg=0.97246 time=236.46it/s +epoch=40 global_step=16031 loss=0.23992 test_loss_avg=0.69204 acc=0.93750 test_acc_avg=0.79203 test_acc_top5_avg=0.97973 time=527.45it/s +curr_acc 0.7920 +BEST_ACC 0.7249 +curr_acc_top5 0.9797 +BEST_ACC_top5 0.9696 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=4.63207 loss_avg=4.58085 acc=0.55469 acc_top1_avg=0.54153 acc_top5_avg=0.87418 lr=0.00100 gn=6.52656 time=56.27it/s +epoch=41 global_step=16100 loss=4.40960 loss_avg=4.55045 acc=0.54688 acc_top1_avg=0.54903 acc_top5_avg=0.87036 lr=0.00100 gn=4.99074 time=59.52it/s +epoch=41 global_step=16150 loss=4.55946 loss_avg=4.57297 acc=0.54688 acc_top1_avg=0.54793 acc_top5_avg=0.87520 lr=0.00100 gn=7.64872 time=46.36it/s +epoch=41 global_step=16200 loss=4.73167 loss_avg=4.58188 acc=0.53906 acc_top1_avg=0.54808 acc_top5_avg=0.87468 lr=0.00100 gn=7.10883 time=52.97it/s +epoch=41 global_step=16250 loss=4.27145 loss_avg=4.58029 acc=0.56250 acc_top1_avg=0.54841 acc_top5_avg=0.87461 lr=0.00100 gn=6.78347 time=54.31it/s +epoch=41 global_step=16300 loss=5.06560 loss_avg=4.56946 acc=0.50000 acc_top1_avg=0.54963 acc_top5_avg=0.87404 lr=0.00100 gn=6.49685 time=56.08it/s +epoch=41 global_step=16350 loss=5.02933 loss_avg=4.58073 acc=0.48438 acc_top1_avg=0.54886 acc_top5_avg=0.87397 lr=0.00100 gn=5.31870 time=56.67it/s +epoch=41 global_step=16400 loss=4.58102 loss_avg=4.58268 acc=0.55469 acc_top1_avg=0.54861 acc_top5_avg=0.87424 lr=0.00100 gn=7.00097 time=54.47it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.10695 test_loss_avg=0.48879 acc=0.96094 test_acc_avg=0.86435 test_acc_top5_avg=0.98864 time=227.44it/s +epoch=41 global_step=16422 loss=0.17321 test_loss_avg=0.76368 acc=0.94531 test_acc_avg=0.76806 test_acc_top5_avg=0.97695 time=127.77it/s +epoch=41 global_step=16422 loss=0.06940 test_loss_avg=0.63428 acc=0.93750 test_acc_avg=0.80627 test_acc_top5_avg=0.98111 time=480.12it/s +curr_acc 0.8063 +BEST_ACC 0.7920 +curr_acc_top5 0.9811 +BEST_ACC_top5 0.9797 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=4.51720 loss_avg=4.52759 acc=0.57031 acc_top1_avg=0.55608 acc_top5_avg=0.87751 lr=0.00100 gn=6.97450 time=54.99it/s +epoch=42 global_step=16500 loss=4.72231 loss_avg=4.49370 acc=0.50781 acc_top1_avg=0.55899 acc_top5_avg=0.87670 lr=0.00100 gn=6.14724 time=63.41it/s +epoch=42 global_step=16550 loss=5.34724 loss_avg=4.48447 acc=0.47656 acc_top1_avg=0.56012 acc_top5_avg=0.87494 lr=0.00100 gn=6.38870 time=56.36it/s +epoch=42 global_step=16600 loss=3.91892 loss_avg=4.51389 acc=0.64062 acc_top1_avg=0.55736 acc_top5_avg=0.87434 lr=0.00100 gn=6.75012 time=55.54it/s +epoch=42 global_step=16650 loss=4.49100 loss_avg=4.50660 acc=0.55469 acc_top1_avg=0.55794 acc_top5_avg=0.87538 lr=0.00100 gn=4.99473 time=56.45it/s +epoch=42 global_step=16700 loss=4.06002 loss_avg=4.52222 acc=0.60938 acc_top1_avg=0.55595 acc_top5_avg=0.87598 lr=0.00100 gn=7.84784 time=63.43it/s +epoch=42 global_step=16750 loss=4.07918 loss_avg=4.53040 acc=0.58594 acc_top1_avg=0.55459 acc_top5_avg=0.87552 lr=0.00100 gn=5.60692 time=55.45it/s +epoch=42 global_step=16800 loss=4.30839 loss_avg=4.52724 acc=0.58594 acc_top1_avg=0.55525 acc_top5_avg=0.87610 lr=0.00100 gn=7.47608 time=60.80it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.67372 test_loss_avg=0.70194 acc=0.78906 test_acc_avg=0.80396 test_acc_top5_avg=0.97754 time=237.33it/s +epoch=42 global_step=16813 loss=0.13536 test_loss_avg=0.62510 acc=0.93750 test_acc_avg=0.80953 test_acc_top5_avg=0.98200 time=513.63it/s +curr_acc 0.8095 +BEST_ACC 0.8063 +curr_acc_top5 0.9820 +BEST_ACC_top5 0.9811 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=4.74376 loss_avg=4.43994 acc=0.52344 acc_top1_avg=0.56018 acc_top5_avg=0.87648 lr=0.00100 gn=6.44381 time=54.87it/s +epoch=43 global_step=16900 loss=4.85485 loss_avg=4.44493 acc=0.53906 acc_top1_avg=0.56340 acc_top5_avg=0.87635 lr=0.00100 gn=7.90585 time=62.43it/s +epoch=43 global_step=16950 loss=4.43472 loss_avg=4.50107 acc=0.56250 acc_top1_avg=0.55811 acc_top5_avg=0.87460 lr=0.00100 gn=5.79377 time=43.04it/s +epoch=43 global_step=17000 loss=3.84416 loss_avg=4.45898 acc=0.62500 acc_top1_avg=0.56313 acc_top5_avg=0.87538 lr=0.00100 gn=6.63665 time=58.59it/s +epoch=43 global_step=17050 loss=4.70758 loss_avg=4.46496 acc=0.52344 acc_top1_avg=0.56250 acc_top5_avg=0.87437 lr=0.00100 gn=6.42098 time=62.82it/s +epoch=43 global_step=17100 loss=4.38080 loss_avg=4.46585 acc=0.55469 acc_top1_avg=0.56236 acc_top5_avg=0.87478 lr=0.00100 gn=6.44864 time=56.96it/s +epoch=43 global_step=17150 loss=3.94574 loss_avg=4.48938 acc=0.61719 acc_top1_avg=0.56016 acc_top5_avg=0.87514 lr=0.00100 gn=8.75643 time=52.42it/s +epoch=43 global_step=17200 loss=4.23489 loss_avg=4.48776 acc=0.56250 acc_top1_avg=0.56012 acc_top5_avg=0.87565 lr=0.00100 gn=7.99595 time=60.12it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.75252 test_loss_avg=0.84924 acc=0.79688 test_acc_avg=0.77865 test_acc_top5_avg=0.97396 time=239.81it/s +epoch=43 global_step=17204 loss=1.55644 test_loss_avg=0.83060 acc=0.50000 test_acc_avg=0.74912 test_acc_top5_avg=0.97671 time=224.26it/s +epoch=43 global_step=17204 loss=0.15329 test_loss_avg=0.64394 acc=0.93750 test_acc_avg=0.80301 test_acc_top5_avg=0.98309 time=626.86it/s +curr_acc 0.8030 +BEST_ACC 0.8095 +curr_acc_top5 0.9831 +BEST_ACC_top5 0.9820 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=4.40147 loss_avg=4.37860 acc=0.55469 acc_top1_avg=0.57150 acc_top5_avg=0.87993 lr=0.00100 gn=8.00020 time=63.69it/s +epoch=44 global_step=17300 loss=5.01254 loss_avg=4.36724 acc=0.51562 acc_top1_avg=0.57284 acc_top5_avg=0.87948 lr=0.00100 gn=7.35136 time=53.62it/s +epoch=44 global_step=17350 loss=4.84005 loss_avg=4.41808 acc=0.53125 acc_top1_avg=0.56609 acc_top5_avg=0.87725 lr=0.00100 gn=7.94552 time=58.71it/s +epoch=44 global_step=17400 loss=4.91741 loss_avg=4.46466 acc=0.50781 acc_top1_avg=0.56107 acc_top5_avg=0.87548 lr=0.00100 gn=8.42428 time=55.45it/s +epoch=44 global_step=17450 loss=4.36423 loss_avg=4.45645 acc=0.57812 acc_top1_avg=0.56215 acc_top5_avg=0.87637 lr=0.00100 gn=8.20248 time=56.04it/s +epoch=44 global_step=17500 loss=4.49407 loss_avg=4.45293 acc=0.56250 acc_top1_avg=0.56255 acc_top5_avg=0.87669 lr=0.00100 gn=5.96678 time=59.85it/s +epoch=44 global_step=17550 loss=4.03743 loss_avg=4.45583 acc=0.63281 acc_top1_avg=0.56225 acc_top5_avg=0.87791 lr=0.00100 gn=9.46691 time=54.95it/s +====================Eval==================== +epoch=44 global_step=17595 loss=1.20414 test_loss_avg=0.63059 acc=0.67969 test_acc_avg=0.82292 test_acc_top5_avg=0.97624 time=127.35it/s +epoch=44 global_step=17595 loss=0.17081 test_loss_avg=0.65865 acc=0.96094 test_acc_avg=0.79677 test_acc_top5_avg=0.98079 time=233.33it/s +epoch=44 global_step=17595 loss=0.17038 test_loss_avg=0.62666 acc=0.93750 test_acc_avg=0.80578 test_acc_top5_avg=0.98190 time=553.92it/s +curr_acc 0.8058 +BEST_ACC 0.8095 +curr_acc_top5 0.9819 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=4.05822 loss_avg=4.43726 acc=0.59375 acc_top1_avg=0.56563 acc_top5_avg=0.85781 lr=0.00100 gn=6.28244 time=56.39it/s +epoch=45 global_step=17650 loss=3.81264 loss_avg=4.32529 acc=0.63281 acc_top1_avg=0.57713 acc_top5_avg=0.87955 lr=0.00100 gn=10.14011 time=59.16it/s +epoch=45 global_step=17700 loss=4.47176 loss_avg=4.33675 acc=0.56250 acc_top1_avg=0.57716 acc_top5_avg=0.88177 lr=0.00100 gn=8.48860 time=39.06it/s +epoch=45 global_step=17750 loss=4.25369 loss_avg=4.35011 acc=0.57031 acc_top1_avg=0.57460 acc_top5_avg=0.88296 lr=0.00100 gn=10.00451 time=54.67it/s +epoch=45 global_step=17800 loss=4.28420 loss_avg=4.35900 acc=0.58594 acc_top1_avg=0.57428 acc_top5_avg=0.88201 lr=0.00100 gn=8.21015 time=53.60it/s +epoch=45 global_step=17850 loss=3.56648 loss_avg=4.37407 acc=0.66406 acc_top1_avg=0.57258 acc_top5_avg=0.88100 lr=0.00100 gn=8.11605 time=55.77it/s +epoch=45 global_step=17900 loss=4.96994 loss_avg=4.39809 acc=0.51562 acc_top1_avg=0.57031 acc_top5_avg=0.88030 lr=0.00100 gn=8.48909 time=62.21it/s +epoch=45 global_step=17950 loss=4.29486 loss_avg=4.41073 acc=0.58594 acc_top1_avg=0.56849 acc_top5_avg=0.87982 lr=0.00100 gn=7.63883 time=55.54it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.72755 test_loss_avg=0.68323 acc=0.79688 test_acc_avg=0.80833 test_acc_top5_avg=0.97691 time=241.25it/s +epoch=45 global_step=17986 loss=0.24146 test_loss_avg=0.62168 acc=0.93750 test_acc_avg=0.81112 test_acc_top5_avg=0.98299 time=845.28it/s +curr_acc 0.8111 +BEST_ACC 0.8095 +curr_acc_top5 0.9830 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=4.07947 loss_avg=4.36702 acc=0.61719 acc_top1_avg=0.57868 acc_top5_avg=0.87165 lr=0.00100 gn=9.05449 time=56.38it/s +epoch=46 global_step=18050 loss=4.51373 loss_avg=4.43501 acc=0.55469 acc_top1_avg=0.56848 acc_top5_avg=0.86780 lr=0.00100 gn=10.78262 time=57.26it/s +epoch=46 global_step=18100 loss=4.14015 loss_avg=4.38899 acc=0.60156 acc_top1_avg=0.57237 acc_top5_avg=0.87404 lr=0.00100 gn=8.56245 time=54.39it/s +epoch=46 global_step=18150 loss=4.12351 loss_avg=4.36806 acc=0.59375 acc_top1_avg=0.57403 acc_top5_avg=0.87576 lr=0.00100 gn=7.40261 time=51.24it/s +epoch=46 global_step=18200 loss=4.63809 loss_avg=4.36779 acc=0.55469 acc_top1_avg=0.57396 acc_top5_avg=0.87624 lr=0.00100 gn=10.24172 time=56.31it/s +epoch=46 global_step=18250 loss=4.41537 loss_avg=4.37956 acc=0.55469 acc_top1_avg=0.57221 acc_top5_avg=0.87692 lr=0.00100 gn=9.44788 time=57.37it/s +epoch=46 global_step=18300 loss=3.99768 loss_avg=4.39220 acc=0.60938 acc_top1_avg=0.57006 acc_top5_avg=0.87816 lr=0.00100 gn=9.49200 time=59.50it/s +epoch=46 global_step=18350 loss=3.76264 loss_avg=4.39596 acc=0.62500 acc_top1_avg=0.56943 acc_top5_avg=0.87788 lr=0.00100 gn=9.80162 time=55.44it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.71274 test_loss_avg=0.47222 acc=0.83594 test_acc_avg=0.86572 test_acc_top5_avg=0.99072 time=235.33it/s +epoch=46 global_step=18377 loss=0.16750 test_loss_avg=0.72734 acc=0.93750 test_acc_avg=0.77616 test_acc_top5_avg=0.97727 time=240.17it/s +epoch=46 global_step=18377 loss=0.24194 test_loss_avg=0.63249 acc=0.93750 test_acc_avg=0.80528 test_acc_top5_avg=0.98081 time=512.94it/s +curr_acc 0.8053 +BEST_ACC 0.8111 +curr_acc_top5 0.9808 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=4.57372 loss_avg=4.38773 acc=0.55469 acc_top1_avg=0.57065 acc_top5_avg=0.87874 lr=0.00100 gn=11.10393 time=58.81it/s +epoch=47 global_step=18450 loss=4.56361 loss_avg=4.35385 acc=0.57031 acc_top1_avg=0.57374 acc_top5_avg=0.88281 lr=0.00100 gn=9.93071 time=57.67it/s +epoch=47 global_step=18500 loss=3.78940 loss_avg=4.37611 acc=0.61719 acc_top1_avg=0.57095 acc_top5_avg=0.88459 lr=0.00100 gn=8.58205 time=53.94it/s +epoch=47 global_step=18550 loss=4.62637 loss_avg=4.37809 acc=0.54688 acc_top1_avg=0.57171 acc_top5_avg=0.88444 lr=0.00100 gn=6.87673 time=55.53it/s +epoch=47 global_step=18600 loss=4.45752 loss_avg=4.37571 acc=0.55469 acc_top1_avg=0.57182 acc_top5_avg=0.88302 lr=0.00100 gn=8.58972 time=51.86it/s +epoch=47 global_step=18650 loss=4.03061 loss_avg=4.36131 acc=0.62500 acc_top1_avg=0.57320 acc_top5_avg=0.88241 lr=0.00100 gn=11.11721 time=53.76it/s +epoch=47 global_step=18700 loss=4.35137 loss_avg=4.36886 acc=0.57812 acc_top1_avg=0.57266 acc_top5_avg=0.88112 lr=0.00100 gn=9.14032 time=54.54it/s +epoch=47 global_step=18750 loss=4.60150 loss_avg=4.36609 acc=0.54688 acc_top1_avg=0.57310 acc_top5_avg=0.88032 lr=0.00100 gn=8.95612 time=60.93it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.28020 test_loss_avg=0.70415 acc=0.91406 test_acc_avg=0.80511 test_acc_top5_avg=0.97297 time=231.35it/s +epoch=47 global_step=18768 loss=0.22528 test_loss_avg=0.63474 acc=0.93750 test_acc_avg=0.80696 test_acc_top5_avg=0.98111 time=854.59it/s +curr_acc 0.8070 +BEST_ACC 0.8111 +curr_acc_top5 0.9811 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=4.18950 loss_avg=4.36684 acc=0.58594 acc_top1_avg=0.56787 acc_top5_avg=0.87646 lr=0.00100 gn=9.07403 time=45.99it/s +epoch=48 global_step=18850 loss=4.91195 loss_avg=4.34325 acc=0.51562 acc_top1_avg=0.57260 acc_top5_avg=0.87786 lr=0.00100 gn=8.52662 time=51.94it/s +epoch=48 global_step=18900 loss=4.88529 loss_avg=4.39147 acc=0.52344 acc_top1_avg=0.56954 acc_top5_avg=0.87760 lr=0.00100 gn=8.72491 time=49.65it/s +epoch=48 global_step=18950 loss=4.81117 loss_avg=4.37919 acc=0.53125 acc_top1_avg=0.57160 acc_top5_avg=0.87908 lr=0.00100 gn=10.28727 time=53.74it/s +epoch=48 global_step=19000 loss=5.04248 loss_avg=4.37406 acc=0.50781 acc_top1_avg=0.57243 acc_top5_avg=0.87985 lr=0.00100 gn=11.78219 time=56.74it/s +epoch=48 global_step=19050 loss=4.14753 loss_avg=4.35923 acc=0.59375 acc_top1_avg=0.57425 acc_top5_avg=0.88090 lr=0.00100 gn=10.57279 time=57.85it/s +epoch=48 global_step=19100 loss=4.24116 loss_avg=4.34302 acc=0.59375 acc_top1_avg=0.57603 acc_top5_avg=0.88062 lr=0.00100 gn=10.34251 time=54.89it/s +epoch=48 global_step=19150 loss=3.88375 loss_avg=4.33915 acc=0.63281 acc_top1_avg=0.57622 acc_top5_avg=0.88077 lr=0.00100 gn=11.22765 time=56.41it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.38462 test_loss_avg=0.64337 acc=0.85938 test_acc_avg=0.81250 test_acc_top5_avg=0.98926 time=237.57it/s +epoch=48 global_step=19159 loss=0.25611 test_loss_avg=0.79299 acc=0.92188 test_acc_avg=0.75970 test_acc_top5_avg=0.97629 time=237.92it/s +epoch=48 global_step=19159 loss=0.24480 test_loss_avg=0.62846 acc=0.93750 test_acc_avg=0.80914 test_acc_top5_avg=0.98161 time=500.63it/s +curr_acc 0.8091 +BEST_ACC 0.8111 +curr_acc_top5 0.9816 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=4.62547 loss_avg=4.28337 acc=0.55469 acc_top1_avg=0.58613 acc_top5_avg=0.88205 lr=0.00100 gn=9.76257 time=63.26it/s +epoch=49 global_step=19250 loss=4.23728 loss_avg=4.31149 acc=0.59375 acc_top1_avg=0.58027 acc_top5_avg=0.87826 lr=0.00100 gn=11.33720 time=54.75it/s +epoch=49 global_step=19300 loss=4.48409 loss_avg=4.31824 acc=0.57031 acc_top1_avg=0.57912 acc_top5_avg=0.87888 lr=0.00100 gn=11.34977 time=62.09it/s +epoch=49 global_step=19350 loss=4.56344 loss_avg=4.30832 acc=0.54688 acc_top1_avg=0.57972 acc_top5_avg=0.87970 lr=0.00100 gn=12.68357 time=59.77it/s +epoch=49 global_step=19400 loss=4.58993 loss_avg=4.29674 acc=0.53125 acc_top1_avg=0.58072 acc_top5_avg=0.88022 lr=0.00100 gn=9.52212 time=54.85it/s +epoch=49 global_step=19450 loss=4.27988 loss_avg=4.29976 acc=0.57812 acc_top1_avg=0.58022 acc_top5_avg=0.88069 lr=0.00100 gn=11.62968 time=61.94it/s +epoch=49 global_step=19500 loss=4.35880 loss_avg=4.29058 acc=0.57031 acc_top1_avg=0.58142 acc_top5_avg=0.88190 lr=0.00100 gn=9.61910 time=52.42it/s +epoch=49 global_step=19550 loss=4.00229 loss_avg=4.29040 acc=0.62500 acc_top1_avg=0.58144 acc_top5_avg=0.88225 lr=0.00100 gn=13.46015 time=53.32it/s +====================Eval==================== +epoch=49 global_step=19550 loss=1.39514 test_loss_avg=0.83113 acc=0.61719 test_acc_avg=0.77128 test_acc_top5_avg=0.97360 time=186.90it/s +epoch=49 global_step=19550 loss=0.34133 test_loss_avg=0.68974 acc=0.93750 test_acc_avg=0.80083 test_acc_top5_avg=0.98161 time=493.97it/s +epoch=49 global_step=19550 loss=0.34133 test_loss_avg=0.68974 acc=0.93750 test_acc_avg=0.80083 test_acc_top5_avg=0.98161 time=493.97it/s +curr_acc 0.8008 +BEST_ACC 0.8111 +curr_acc_top5 0.9816 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=4.00864 loss_avg=4.28085 acc=0.62500 acc_top1_avg=0.58344 acc_top5_avg=0.88047 lr=0.00100 gn=7.80217 time=57.32it/s +epoch=50 global_step=19650 loss=3.56635 loss_avg=4.31948 acc=0.64844 acc_top1_avg=0.57898 acc_top5_avg=0.88094 lr=0.00100 gn=10.32413 time=55.19it/s +epoch=50 global_step=19700 loss=4.25127 loss_avg=4.31239 acc=0.59375 acc_top1_avg=0.57911 acc_top5_avg=0.88083 lr=0.00100 gn=10.64327 time=55.88it/s +epoch=50 global_step=19750 loss=4.28080 loss_avg=4.28764 acc=0.58594 acc_top1_avg=0.58113 acc_top5_avg=0.88152 lr=0.00100 gn=10.92847 time=54.38it/s +epoch=50 global_step=19800 loss=4.27757 loss_avg=4.28873 acc=0.59375 acc_top1_avg=0.58088 acc_top5_avg=0.88138 lr=0.00100 gn=12.33938 time=61.45it/s +epoch=50 global_step=19850 loss=4.11217 loss_avg=4.28796 acc=0.60938 acc_top1_avg=0.58073 acc_top5_avg=0.88112 lr=0.00100 gn=13.95512 time=63.37it/s +epoch=50 global_step=19900 loss=4.04860 loss_avg=4.28952 acc=0.60938 acc_top1_avg=0.58076 acc_top5_avg=0.88118 lr=0.00100 gn=11.17938 time=53.77it/s +====================Eval==================== +epoch=50 global_step=19941 loss=1.42121 test_loss_avg=0.77073 acc=0.47656 test_acc_avg=0.77594 test_acc_top5_avg=0.97609 time=251.02it/s +epoch=50 global_step=19941 loss=0.30270 test_loss_avg=0.63755 acc=0.93750 test_acc_avg=0.81121 test_acc_top5_avg=0.98210 time=508.03it/s +curr_acc 0.8112 +BEST_ACC 0.8111 +curr_acc_top5 0.9821 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=4.30505 loss_avg=4.15261 acc=0.57812 acc_top1_avg=0.59375 acc_top5_avg=0.87066 lr=0.00100 gn=10.96868 time=50.06it/s +epoch=51 global_step=20000 loss=4.51664 loss_avg=4.25731 acc=0.55469 acc_top1_avg=0.58528 acc_top5_avg=0.88083 lr=0.00100 gn=10.32916 time=62.42it/s +epoch=51 global_step=20050 loss=4.13972 loss_avg=4.22102 acc=0.60156 acc_top1_avg=0.58916 acc_top5_avg=0.87937 lr=0.00100 gn=9.53700 time=55.16it/s +epoch=51 global_step=20100 loss=4.57254 loss_avg=4.24036 acc=0.57031 acc_top1_avg=0.58707 acc_top5_avg=0.87888 lr=0.00100 gn=11.54351 time=55.74it/s +epoch=51 global_step=20150 loss=4.65532 loss_avg=4.24556 acc=0.54688 acc_top1_avg=0.58695 acc_top5_avg=0.88023 lr=0.00100 gn=14.48805 time=58.68it/s +epoch=51 global_step=20200 loss=4.26074 loss_avg=4.23579 acc=0.57812 acc_top1_avg=0.58775 acc_top5_avg=0.88124 lr=0.00100 gn=11.50760 time=56.40it/s +epoch=51 global_step=20250 loss=4.98023 loss_avg=4.24839 acc=0.51562 acc_top1_avg=0.58642 acc_top5_avg=0.88046 lr=0.00100 gn=13.85645 time=58.06it/s +epoch=51 global_step=20300 loss=4.91843 loss_avg=4.25683 acc=0.52344 acc_top1_avg=0.58555 acc_top5_avg=0.88096 lr=0.00100 gn=13.32428 time=54.29it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.82991 test_loss_avg=0.55633 acc=0.78906 test_acc_avg=0.84896 test_acc_top5_avg=0.98103 time=238.98it/s +epoch=51 global_step=20332 loss=0.16667 test_loss_avg=0.65889 acc=0.94531 test_acc_avg=0.80205 test_acc_top5_avg=0.98030 time=252.71it/s +epoch=51 global_step=20332 loss=0.46322 test_loss_avg=0.61425 acc=0.93750 test_acc_avg=0.81576 test_acc_top5_avg=0.98210 time=832.53it/s +curr_acc 0.8158 +BEST_ACC 0.8112 +curr_acc_top5 0.9821 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=3.63097 loss_avg=4.14195 acc=0.64844 acc_top1_avg=0.59505 acc_top5_avg=0.90061 lr=0.00100 gn=11.72958 time=55.68it/s +epoch=52 global_step=20400 loss=4.48095 loss_avg=4.21865 acc=0.56250 acc_top1_avg=0.58789 acc_top5_avg=0.88718 lr=0.00100 gn=10.02230 time=54.62it/s +epoch=52 global_step=20450 loss=3.53067 loss_avg=4.19690 acc=0.67188 acc_top1_avg=0.59097 acc_top5_avg=0.88427 lr=0.00100 gn=10.43056 time=52.78it/s +epoch=52 global_step=20500 loss=3.77258 loss_avg=4.21178 acc=0.64844 acc_top1_avg=0.59012 acc_top5_avg=0.88360 lr=0.00100 gn=12.67021 time=55.67it/s +epoch=52 global_step=20550 loss=4.05387 loss_avg=4.21591 acc=0.62500 acc_top1_avg=0.58956 acc_top5_avg=0.88278 lr=0.00100 gn=13.07471 time=53.51it/s +epoch=52 global_step=20600 loss=4.74881 loss_avg=4.22961 acc=0.50781 acc_top1_avg=0.58780 acc_top5_avg=0.88086 lr=0.00100 gn=12.75893 time=54.04it/s +epoch=52 global_step=20650 loss=4.65391 loss_avg=4.24432 acc=0.55469 acc_top1_avg=0.58616 acc_top5_avg=0.88156 lr=0.00100 gn=13.15618 time=50.59it/s +epoch=52 global_step=20700 loss=3.67070 loss_avg=4.23895 acc=0.64844 acc_top1_avg=0.58691 acc_top5_avg=0.88090 lr=0.00100 gn=11.39522 time=49.97it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.90176 test_loss_avg=0.72348 acc=0.71875 test_acc_avg=0.79204 test_acc_top5_avg=0.97433 time=186.90it/s +epoch=52 global_step=20723 loss=0.25134 test_loss_avg=0.74394 acc=0.93750 test_acc_avg=0.78600 test_acc_top5_avg=0.98101 time=509.82it/s +curr_acc 0.7860 +BEST_ACC 0.8158 +curr_acc_top5 0.9810 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=4.57649 loss_avg=4.21216 acc=0.56250 acc_top1_avg=0.59172 acc_top5_avg=0.88108 lr=0.00100 gn=13.45307 time=53.49it/s +epoch=53 global_step=20800 loss=3.99956 loss_avg=4.22156 acc=0.60938 acc_top1_avg=0.58989 acc_top5_avg=0.88464 lr=0.00100 gn=10.87594 time=57.89it/s +epoch=53 global_step=20850 loss=3.86891 loss_avg=4.19202 acc=0.60938 acc_top1_avg=0.59289 acc_top5_avg=0.88410 lr=0.00100 gn=9.44353 time=62.20it/s +epoch=53 global_step=20900 loss=3.61490 loss_avg=4.20442 acc=0.65625 acc_top1_avg=0.59168 acc_top5_avg=0.88233 lr=0.00100 gn=11.16529 time=51.59it/s +epoch=53 global_step=20950 loss=4.18083 loss_avg=4.21775 acc=0.59375 acc_top1_avg=0.59045 acc_top5_avg=0.88302 lr=0.00100 gn=15.11813 time=60.70it/s +epoch=53 global_step=21000 loss=4.02516 loss_avg=4.20672 acc=0.61719 acc_top1_avg=0.59158 acc_top5_avg=0.88324 lr=0.00100 gn=15.38382 time=50.81it/s +epoch=53 global_step=21050 loss=4.61766 loss_avg=4.21204 acc=0.53906 acc_top1_avg=0.59095 acc_top5_avg=0.88403 lr=0.00100 gn=12.53837 time=53.42it/s +epoch=53 global_step=21100 loss=3.88434 loss_avg=4.22114 acc=0.62500 acc_top1_avg=0.59019 acc_top5_avg=0.88263 lr=0.00100 gn=12.72662 time=59.63it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.14283 test_loss_avg=0.56128 acc=0.93750 test_acc_avg=0.82812 test_acc_top5_avg=0.98438 time=239.26it/s +epoch=53 global_step=21114 loss=0.09371 test_loss_avg=0.75986 acc=0.96875 test_acc_avg=0.76860 test_acc_top5_avg=0.97557 time=228.78it/s +epoch=53 global_step=21114 loss=0.28147 test_loss_avg=0.63788 acc=0.93750 test_acc_avg=0.80637 test_acc_top5_avg=0.98012 time=523.37it/s +curr_acc 0.8064 +BEST_ACC 0.8158 +curr_acc_top5 0.9801 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=4.37455 loss_avg=4.26035 acc=0.58594 acc_top1_avg=0.58377 acc_top5_avg=0.88325 lr=0.00100 gn=13.93572 time=56.37it/s +epoch=54 global_step=21200 loss=4.19347 loss_avg=4.17469 acc=0.57031 acc_top1_avg=0.59339 acc_top5_avg=0.88272 lr=0.00100 gn=14.35000 time=50.89it/s +epoch=54 global_step=21250 loss=4.20780 loss_avg=4.15790 acc=0.60938 acc_top1_avg=0.59628 acc_top5_avg=0.88528 lr=0.00100 gn=14.44424 time=59.98it/s +epoch=54 global_step=21300 loss=4.22405 loss_avg=4.16990 acc=0.59375 acc_top1_avg=0.59543 acc_top5_avg=0.88567 lr=0.00100 gn=13.23109 time=56.77it/s +epoch=54 global_step=21350 loss=4.56388 loss_avg=4.18072 acc=0.53906 acc_top1_avg=0.59428 acc_top5_avg=0.88414 lr=0.00100 gn=14.86706 time=60.42it/s +epoch=54 global_step=21400 loss=4.25583 loss_avg=4.18341 acc=0.59375 acc_top1_avg=0.59383 acc_top5_avg=0.88298 lr=0.00100 gn=15.31751 time=61.87it/s +epoch=54 global_step=21450 loss=4.10398 loss_avg=4.18741 acc=0.60156 acc_top1_avg=0.59370 acc_top5_avg=0.88323 lr=0.00100 gn=12.97213 time=54.29it/s +epoch=54 global_step=21500 loss=4.44515 loss_avg=4.20253 acc=0.57031 acc_top1_avg=0.59235 acc_top5_avg=0.88208 lr=0.00100 gn=14.19859 time=61.53it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.45450 test_loss_avg=0.77383 acc=0.87500 test_acc_avg=0.77436 test_acc_top5_avg=0.97633 time=242.43it/s +epoch=54 global_step=21505 loss=0.18129 test_loss_avg=0.64954 acc=0.93750 test_acc_avg=0.80538 test_acc_top5_avg=0.98141 time=517.18it/s +curr_acc 0.8054 +BEST_ACC 0.8158 +curr_acc_top5 0.9814 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=4.04090 loss_avg=4.22386 acc=0.59375 acc_top1_avg=0.58854 acc_top5_avg=0.88299 lr=0.00100 gn=14.11688 time=54.52it/s +epoch=55 global_step=21600 loss=4.75420 loss_avg=4.19748 acc=0.53906 acc_top1_avg=0.59128 acc_top5_avg=0.87911 lr=0.00100 gn=15.06755 time=51.79it/s +epoch=55 global_step=21650 loss=4.63012 loss_avg=4.18741 acc=0.53906 acc_top1_avg=0.59283 acc_top5_avg=0.88060 lr=0.00100 gn=13.55424 time=53.32it/s +epoch=55 global_step=21700 loss=4.14941 loss_avg=4.20298 acc=0.57812 acc_top1_avg=0.59151 acc_top5_avg=0.88017 lr=0.00100 gn=11.84676 time=53.73it/s +epoch=55 global_step=21750 loss=4.54751 loss_avg=4.18727 acc=0.53906 acc_top1_avg=0.59295 acc_top5_avg=0.88115 lr=0.00100 gn=15.23574 time=62.27it/s +epoch=55 global_step=21800 loss=3.50389 loss_avg=4.18177 acc=0.65625 acc_top1_avg=0.59378 acc_top5_avg=0.88146 lr=0.00100 gn=14.14701 time=62.00it/s +epoch=55 global_step=21850 loss=4.16541 loss_avg=4.16250 acc=0.58594 acc_top1_avg=0.59565 acc_top5_avg=0.88243 lr=0.00100 gn=16.73588 time=55.78it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.64088 test_loss_avg=0.93786 acc=0.78125 test_acc_avg=0.75000 test_acc_top5_avg=0.96719 time=236.71it/s +epoch=55 global_step=21896 loss=1.02044 test_loss_avg=0.88546 acc=0.67188 test_acc_avg=0.74134 test_acc_top5_avg=0.97202 time=219.76it/s +epoch=55 global_step=21896 loss=0.12921 test_loss_avg=0.66972 acc=0.93750 test_acc_avg=0.80291 test_acc_top5_avg=0.97953 time=460.56it/s +curr_acc 0.8029 +BEST_ACC 0.8158 +curr_acc_top5 0.9795 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=3.14157 loss_avg=3.97256 acc=0.71875 acc_top1_avg=0.62109 acc_top5_avg=0.87500 lr=0.00100 gn=16.18772 time=53.49it/s +epoch=56 global_step=21950 loss=3.92490 loss_avg=4.20344 acc=0.62500 acc_top1_avg=0.59158 acc_top5_avg=0.88209 lr=0.00100 gn=15.37731 time=60.15it/s +epoch=56 global_step=22000 loss=4.63276 loss_avg=4.17835 acc=0.55469 acc_top1_avg=0.59458 acc_top5_avg=0.88161 lr=0.00100 gn=13.61275 time=50.67it/s +epoch=56 global_step=22050 loss=4.26049 loss_avg=4.16159 acc=0.57031 acc_top1_avg=0.59578 acc_top5_avg=0.88220 lr=0.00100 gn=12.88349 time=52.80it/s +epoch=56 global_step=22100 loss=4.07584 loss_avg=4.15869 acc=0.60938 acc_top1_avg=0.59628 acc_top5_avg=0.88258 lr=0.00100 gn=17.34602 time=57.90it/s +epoch=56 global_step=22150 loss=4.05366 loss_avg=4.14946 acc=0.59375 acc_top1_avg=0.59756 acc_top5_avg=0.88355 lr=0.00100 gn=15.18971 time=51.05it/s +epoch=56 global_step=22200 loss=4.11557 loss_avg=4.14195 acc=0.59375 acc_top1_avg=0.59835 acc_top5_avg=0.88387 lr=0.00100 gn=14.82464 time=32.81it/s +epoch=56 global_step=22250 loss=4.46167 loss_avg=4.15332 acc=0.56250 acc_top1_avg=0.59706 acc_top5_avg=0.88367 lr=0.00100 gn=14.04719 time=56.23it/s +====================Eval==================== +epoch=56 global_step=22287 loss=1.26933 test_loss_avg=0.72915 acc=0.66406 test_acc_avg=0.79537 test_acc_top5_avg=0.97296 time=240.86it/s +epoch=56 global_step=22287 loss=0.12263 test_loss_avg=0.66753 acc=0.94531 test_acc_avg=0.80376 test_acc_top5_avg=0.98119 time=242.26it/s +epoch=56 global_step=22287 loss=0.13480 test_loss_avg=0.64570 acc=0.93750 test_acc_avg=0.81003 test_acc_top5_avg=0.98190 time=557.31it/s +curr_acc 0.8100 +BEST_ACC 0.8158 +curr_acc_top5 0.9819 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=3.65563 loss_avg=4.17148 acc=0.65625 acc_top1_avg=0.59976 acc_top5_avg=0.87680 lr=0.00100 gn=15.95138 time=55.62it/s +epoch=57 global_step=22350 loss=3.62976 loss_avg=4.14792 acc=0.65625 acc_top1_avg=0.59809 acc_top5_avg=0.87612 lr=0.00100 gn=11.95635 time=59.48it/s +epoch=57 global_step=22400 loss=4.69717 loss_avg=4.14740 acc=0.55469 acc_top1_avg=0.59859 acc_top5_avg=0.87949 lr=0.00100 gn=17.98426 time=59.78it/s +epoch=57 global_step=22450 loss=3.95874 loss_avg=4.15387 acc=0.60938 acc_top1_avg=0.59768 acc_top5_avg=0.88128 lr=0.00100 gn=18.14940 time=59.19it/s +epoch=57 global_step=22500 loss=5.13787 loss_avg=4.14256 acc=0.48438 acc_top1_avg=0.59885 acc_top5_avg=0.88256 lr=0.00100 gn=14.80470 time=63.02it/s +epoch=57 global_step=22550 loss=3.90843 loss_avg=4.13680 acc=0.62500 acc_top1_avg=0.59948 acc_top5_avg=0.88350 lr=0.00100 gn=20.63313 time=51.94it/s +epoch=57 global_step=22600 loss=4.83304 loss_avg=4.14559 acc=0.52344 acc_top1_avg=0.59867 acc_top5_avg=0.88309 lr=0.00100 gn=15.23443 time=53.82it/s +epoch=57 global_step=22650 loss=3.71275 loss_avg=4.13383 acc=0.63281 acc_top1_avg=0.60010 acc_top5_avg=0.88283 lr=0.00100 gn=13.88929 time=30.91it/s +====================Eval==================== +epoch=57 global_step=22678 loss=1.01472 test_loss_avg=0.70255 acc=0.67188 test_acc_avg=0.79405 test_acc_top5_avg=0.97756 time=213.05it/s +epoch=57 global_step=22678 loss=0.36583 test_loss_avg=0.63618 acc=0.93750 test_acc_avg=0.80825 test_acc_top5_avg=0.98329 time=541.34it/s +curr_acc 0.8082 +BEST_ACC 0.8158 +curr_acc_top5 0.9833 +BEST_ACC_top5 0.9831 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=3.05459 loss_avg=3.98965 acc=0.71875 acc_top1_avg=0.61222 acc_top5_avg=0.89666 lr=0.00100 gn=14.83246 time=54.67it/s +epoch=58 global_step=22750 loss=4.70796 loss_avg=4.08092 acc=0.53125 acc_top1_avg=0.60319 acc_top5_avg=0.89062 lr=0.00100 gn=19.14984 time=52.88it/s +epoch=58 global_step=22800 loss=4.62363 loss_avg=4.11946 acc=0.56250 acc_top1_avg=0.60079 acc_top5_avg=0.88973 lr=0.00100 gn=15.86907 time=56.32it/s +epoch=58 global_step=22850 loss=3.70546 loss_avg=4.10895 acc=0.64062 acc_top1_avg=0.60202 acc_top5_avg=0.88667 lr=0.00100 gn=12.70603 time=61.21it/s +epoch=58 global_step=22900 loss=4.36989 loss_avg=4.11862 acc=0.58594 acc_top1_avg=0.60135 acc_top5_avg=0.88559 lr=0.00100 gn=17.63757 time=62.66it/s +epoch=58 global_step=22950 loss=3.88502 loss_avg=4.11729 acc=0.61719 acc_top1_avg=0.60122 acc_top5_avg=0.88442 lr=0.00100 gn=14.87936 time=57.17it/s +epoch=58 global_step=23000 loss=3.97678 loss_avg=4.11111 acc=0.60938 acc_top1_avg=0.60229 acc_top5_avg=0.88466 lr=0.00100 gn=17.89461 time=61.96it/s +epoch=58 global_step=23050 loss=4.02006 loss_avg=4.10611 acc=0.60938 acc_top1_avg=0.60299 acc_top5_avg=0.88519 lr=0.00100 gn=18.39541 time=55.87it/s +====================Eval==================== +epoch=58 global_step=23069 loss=1.08260 test_loss_avg=0.60670 acc=0.70312 test_acc_avg=0.82769 test_acc_top5_avg=0.98481 time=226.92it/s +epoch=58 global_step=23069 loss=0.27857 test_loss_avg=0.83145 acc=0.90625 test_acc_avg=0.75551 test_acc_top5_avg=0.97369 time=236.95it/s +epoch=58 global_step=23069 loss=0.39983 test_loss_avg=0.73595 acc=0.93750 test_acc_avg=0.78412 test_acc_top5_avg=0.97716 time=842.91it/s +curr_acc 0.7841 +BEST_ACC 0.8158 +curr_acc_top5 0.9772 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=4.11994 loss_avg=4.11921 acc=0.58594 acc_top1_avg=0.60081 acc_top5_avg=0.87349 lr=0.00100 gn=16.16046 time=58.01it/s +epoch=59 global_step=23150 loss=3.95949 loss_avg=4.05662 acc=0.61719 acc_top1_avg=0.60629 acc_top5_avg=0.88166 lr=0.00100 gn=18.78082 time=53.25it/s +epoch=59 global_step=23200 loss=4.49621 loss_avg=4.07818 acc=0.59375 acc_top1_avg=0.60466 acc_top5_avg=0.88156 lr=0.00100 gn=19.53017 time=49.10it/s +epoch=59 global_step=23250 loss=4.20613 loss_avg=4.07935 acc=0.58594 acc_top1_avg=0.60545 acc_top5_avg=0.88238 lr=0.00100 gn=18.04874 time=57.85it/s +epoch=59 global_step=23300 loss=3.90556 loss_avg=4.06442 acc=0.62500 acc_top1_avg=0.60701 acc_top5_avg=0.88356 lr=0.00100 gn=19.66951 time=54.56it/s +epoch=59 global_step=23350 loss=4.11914 loss_avg=4.06589 acc=0.60938 acc_top1_avg=0.60715 acc_top5_avg=0.88337 lr=0.00100 gn=17.32070 time=52.82it/s +epoch=59 global_step=23400 loss=4.49310 loss_avg=4.07909 acc=0.55469 acc_top1_avg=0.60588 acc_top5_avg=0.88383 lr=0.00100 gn=17.49453 time=59.41it/s +epoch=59 global_step=23450 loss=3.92930 loss_avg=4.08808 acc=0.64844 acc_top1_avg=0.60509 acc_top5_avg=0.88408 lr=0.00100 gn=23.41916 time=54.37it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.29999 test_loss_avg=0.92279 acc=0.93750 test_acc_avg=0.73818 test_acc_top5_avg=0.96695 time=222.06it/s +epoch=59 global_step=23460 loss=0.25738 test_loss_avg=0.74636 acc=0.93750 test_acc_avg=0.78590 test_acc_top5_avg=0.97785 time=498.79it/s +curr_acc 0.7859 +BEST_ACC 0.8158 +curr_acc_top5 0.9778 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=4.37552 loss_avg=4.01849 acc=0.57812 acc_top1_avg=0.61738 acc_top5_avg=0.89180 lr=0.00100 gn=17.29481 time=53.88it/s +epoch=60 global_step=23550 loss=4.04315 loss_avg=4.03294 acc=0.61719 acc_top1_avg=0.61233 acc_top5_avg=0.89149 lr=0.00100 gn=15.77723 time=60.24it/s +epoch=60 global_step=23600 loss=3.58306 loss_avg=4.04556 acc=0.65625 acc_top1_avg=0.60982 acc_top5_avg=0.88968 lr=0.00100 gn=17.39421 time=54.71it/s +epoch=60 global_step=23650 loss=4.46073 loss_avg=4.07700 acc=0.56250 acc_top1_avg=0.60592 acc_top5_avg=0.88701 lr=0.00100 gn=16.79293 time=62.80it/s +epoch=60 global_step=23700 loss=3.84082 loss_avg=4.08545 acc=0.63281 acc_top1_avg=0.60531 acc_top5_avg=0.88561 lr=0.00100 gn=22.79928 time=54.79it/s +epoch=60 global_step=23750 loss=4.10489 loss_avg=4.08977 acc=0.58594 acc_top1_avg=0.60474 acc_top5_avg=0.88521 lr=0.00100 gn=17.68149 time=48.45it/s +epoch=60 global_step=23800 loss=4.21466 loss_avg=4.08586 acc=0.60938 acc_top1_avg=0.60508 acc_top5_avg=0.88536 lr=0.00100 gn=16.60766 time=55.79it/s +epoch=60 global_step=23850 loss=3.87171 loss_avg=4.08384 acc=0.63281 acc_top1_avg=0.60541 acc_top5_avg=0.88566 lr=0.00100 gn=19.55937 time=55.54it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.21228 test_loss_avg=0.74629 acc=0.95312 test_acc_avg=0.77422 test_acc_top5_avg=0.98047 time=235.40it/s +epoch=60 global_step=23851 loss=0.14472 test_loss_avg=0.85353 acc=0.96094 test_acc_avg=0.75182 test_acc_top5_avg=0.97604 time=234.67it/s +epoch=60 global_step=23851 loss=0.46688 test_loss_avg=0.69500 acc=0.93750 test_acc_avg=0.79846 test_acc_top5_avg=0.98111 time=501.17it/s +curr_acc 0.7985 +BEST_ACC 0.8158 +curr_acc_top5 0.9811 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=3.77345 loss_avg=3.93064 acc=0.64062 acc_top1_avg=0.62213 acc_top5_avg=0.88329 lr=0.00100 gn=20.61778 time=61.58it/s +epoch=61 global_step=23950 loss=3.89154 loss_avg=3.95907 acc=0.64062 acc_top1_avg=0.61908 acc_top5_avg=0.88455 lr=0.00100 gn=20.29259 time=63.65it/s +epoch=61 global_step=24000 loss=4.27393 loss_avg=3.99358 acc=0.58594 acc_top1_avg=0.61582 acc_top5_avg=0.88402 lr=0.00100 gn=19.85966 time=60.37it/s +epoch=61 global_step=24050 loss=4.45744 loss_avg=4.01922 acc=0.53906 acc_top1_avg=0.61287 acc_top5_avg=0.88250 lr=0.00100 gn=21.85406 time=60.66it/s +epoch=61 global_step=24100 loss=3.84911 loss_avg=4.02653 acc=0.63281 acc_top1_avg=0.61151 acc_top5_avg=0.88215 lr=0.00100 gn=19.08073 time=48.73it/s +epoch=61 global_step=24150 loss=3.97565 loss_avg=4.05149 acc=0.63281 acc_top1_avg=0.60922 acc_top5_avg=0.88114 lr=0.00100 gn=22.22437 time=52.02it/s +epoch=61 global_step=24200 loss=4.55450 loss_avg=4.05635 acc=0.55469 acc_top1_avg=0.60850 acc_top5_avg=0.88277 lr=0.00100 gn=25.14777 time=53.45it/s +====================Eval==================== +epoch=61 global_step=24242 loss=1.01542 test_loss_avg=0.81816 acc=0.66406 test_acc_avg=0.76361 test_acc_top5_avg=0.96648 time=228.72it/s +epoch=61 global_step=24242 loss=0.31623 test_loss_avg=0.66707 acc=0.93750 test_acc_avg=0.80073 test_acc_top5_avg=0.97963 time=552.03it/s +curr_acc 0.8007 +BEST_ACC 0.8158 +curr_acc_top5 0.9796 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=4.64877 loss_avg=3.94261 acc=0.55469 acc_top1_avg=0.62012 acc_top5_avg=0.87598 lr=0.00100 gn=21.88585 time=53.46it/s +epoch=62 global_step=24300 loss=4.72674 loss_avg=4.05972 acc=0.53906 acc_top1_avg=0.60843 acc_top5_avg=0.88012 lr=0.00100 gn=17.93221 time=54.75it/s +epoch=62 global_step=24350 loss=3.75597 loss_avg=4.02935 acc=0.63281 acc_top1_avg=0.61191 acc_top5_avg=0.88433 lr=0.00100 gn=18.87313 time=55.64it/s +epoch=62 global_step=24400 loss=4.51657 loss_avg=4.03614 acc=0.53125 acc_top1_avg=0.61214 acc_top5_avg=0.88489 lr=0.00100 gn=17.76973 time=63.69it/s +epoch=62 global_step=24450 loss=3.60660 loss_avg=4.02254 acc=0.65625 acc_top1_avg=0.61291 acc_top5_avg=0.88454 lr=0.00100 gn=19.44344 time=48.66it/s +epoch=62 global_step=24500 loss=4.07829 loss_avg=4.01414 acc=0.62500 acc_top1_avg=0.61358 acc_top5_avg=0.88557 lr=0.00100 gn=21.13111 time=55.83it/s +epoch=62 global_step=24550 loss=3.47098 loss_avg=4.02968 acc=0.67188 acc_top1_avg=0.61262 acc_top5_avg=0.88378 lr=0.00100 gn=18.86269 time=54.13it/s +epoch=62 global_step=24600 loss=4.10376 loss_avg=4.03663 acc=0.60938 acc_top1_avg=0.61202 acc_top5_avg=0.88395 lr=0.00100 gn=20.43323 time=53.87it/s +====================Eval==================== +epoch=62 global_step=24633 loss=1.22248 test_loss_avg=1.12449 acc=0.65625 test_acc_avg=0.67578 test_acc_top5_avg=0.97266 time=230.29it/s +epoch=62 global_step=24633 loss=3.05270 test_loss_avg=0.98129 acc=0.26562 test_acc_avg=0.73603 test_acc_top5_avg=0.96920 time=112.53it/s +epoch=62 global_step=24633 loss=0.37658 test_loss_avg=0.80792 acc=0.93750 test_acc_avg=0.78313 test_acc_top5_avg=0.97636 time=504.24it/s +curr_acc 0.7831 +BEST_ACC 0.8158 +curr_acc_top5 0.9764 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=3.91929 loss_avg=3.88425 acc=0.63281 acc_top1_avg=0.63097 acc_top5_avg=0.89017 lr=0.00100 gn=15.24995 time=54.94it/s +epoch=63 global_step=24700 loss=4.22861 loss_avg=3.93649 acc=0.60938 acc_top1_avg=0.62290 acc_top5_avg=0.88363 lr=0.00100 gn=20.13947 time=48.88it/s +epoch=63 global_step=24750 loss=3.90833 loss_avg=3.98893 acc=0.63281 acc_top1_avg=0.61752 acc_top5_avg=0.88268 lr=0.00100 gn=20.75831 time=60.04it/s +epoch=63 global_step=24800 loss=4.74813 loss_avg=4.01120 acc=0.52344 acc_top1_avg=0.61504 acc_top5_avg=0.88253 lr=0.00100 gn=19.72598 time=63.19it/s +epoch=63 global_step=24850 loss=3.80784 loss_avg=4.01247 acc=0.62500 acc_top1_avg=0.61492 acc_top5_avg=0.88299 lr=0.00100 gn=18.44279 time=54.42it/s +epoch=63 global_step=24900 loss=4.00071 loss_avg=4.00602 acc=0.60938 acc_top1_avg=0.61531 acc_top5_avg=0.88351 lr=0.00100 gn=17.69670 time=52.61it/s +epoch=63 global_step=24950 loss=3.95593 loss_avg=4.00482 acc=0.60156 acc_top1_avg=0.61571 acc_top5_avg=0.88451 lr=0.00100 gn=16.67972 time=56.61it/s +epoch=63 global_step=25000 loss=4.33444 loss_avg=4.01106 acc=0.57031 acc_top1_avg=0.61497 acc_top5_avg=0.88486 lr=0.00100 gn=19.64301 time=55.56it/s +====================Eval==================== +epoch=63 global_step=25024 loss=1.06529 test_loss_avg=0.87308 acc=0.68750 test_acc_avg=0.75408 test_acc_top5_avg=0.96671 time=239.52it/s +epoch=63 global_step=25024 loss=0.17900 test_loss_avg=0.80347 acc=0.95312 test_acc_avg=0.76274 test_acc_top5_avg=0.97688 time=257.02it/s +epoch=63 global_step=25024 loss=0.20100 test_loss_avg=0.75024 acc=0.93750 test_acc_avg=0.77809 test_acc_top5_avg=0.97854 time=854.24it/s +curr_acc 0.7781 +BEST_ACC 0.8158 +curr_acc_top5 0.9785 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=3.80759 loss_avg=3.92571 acc=0.62500 acc_top1_avg=0.62410 acc_top5_avg=0.88431 lr=0.00100 gn=21.14818 time=52.62it/s +epoch=64 global_step=25100 loss=3.59983 loss_avg=3.94914 acc=0.67188 acc_top1_avg=0.62284 acc_top5_avg=0.88189 lr=0.00100 gn=23.77949 time=54.31it/s +epoch=64 global_step=25150 loss=3.97720 loss_avg=3.94108 acc=0.61719 acc_top1_avg=0.62376 acc_top5_avg=0.88535 lr=0.00100 gn=17.67149 time=52.97it/s +epoch=64 global_step=25200 loss=4.25018 loss_avg=4.00106 acc=0.60156 acc_top1_avg=0.61648 acc_top5_avg=0.88588 lr=0.00100 gn=22.86172 time=62.10it/s +epoch=64 global_step=25250 loss=4.39604 loss_avg=4.03327 acc=0.58594 acc_top1_avg=0.61273 acc_top5_avg=0.88468 lr=0.00100 gn=27.81745 time=52.56it/s +epoch=64 global_step=25300 loss=3.77975 loss_avg=4.01354 acc=0.63281 acc_top1_avg=0.61487 acc_top5_avg=0.88587 lr=0.00100 gn=19.30945 time=48.32it/s +epoch=64 global_step=25350 loss=3.72857 loss_avg=4.02555 acc=0.66406 acc_top1_avg=0.61354 acc_top5_avg=0.88449 lr=0.00100 gn=25.76477 time=53.54it/s +epoch=64 global_step=25400 loss=3.81760 loss_avg=4.01333 acc=0.62500 acc_top1_avg=0.61463 acc_top5_avg=0.88514 lr=0.00100 gn=22.06564 time=51.66it/s +====================Eval==================== +epoch=64 global_step=25415 loss=2.23776 test_loss_avg=0.87463 acc=0.41406 test_acc_avg=0.74609 test_acc_top5_avg=0.96804 time=241.72it/s +epoch=64 global_step=25415 loss=0.42762 test_loss_avg=0.79783 acc=0.93750 test_acc_avg=0.76909 test_acc_top5_avg=0.97468 time=568.87it/s +curr_acc 0.7691 +BEST_ACC 0.8158 +curr_acc_top5 0.9747 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=3.64437 loss_avg=4.07890 acc=0.66406 acc_top1_avg=0.60915 acc_top5_avg=0.88259 lr=0.00100 gn=23.06721 time=60.84it/s +epoch=65 global_step=25500 loss=4.32950 loss_avg=3.98720 acc=0.57031 acc_top1_avg=0.61691 acc_top5_avg=0.88575 lr=0.00100 gn=17.96274 time=57.04it/s +epoch=65 global_step=25550 loss=4.03899 loss_avg=3.95206 acc=0.60938 acc_top1_avg=0.62078 acc_top5_avg=0.88559 lr=0.00100 gn=18.54419 time=53.58it/s +epoch=65 global_step=25600 loss=4.41338 loss_avg=3.97295 acc=0.57812 acc_top1_avg=0.61888 acc_top5_avg=0.88606 lr=0.00100 gn=23.26032 time=57.71it/s +epoch=65 global_step=25650 loss=4.50811 loss_avg=3.97782 acc=0.56250 acc_top1_avg=0.61802 acc_top5_avg=0.88534 lr=0.00100 gn=19.48171 time=51.66it/s +epoch=65 global_step=25700 loss=4.31701 loss_avg=3.97782 acc=0.58594 acc_top1_avg=0.61834 acc_top5_avg=0.88596 lr=0.00100 gn=20.47101 time=54.18it/s +epoch=65 global_step=25750 loss=3.23092 loss_avg=3.98393 acc=0.69531 acc_top1_avg=0.61716 acc_top5_avg=0.88645 lr=0.00100 gn=17.81980 time=57.25it/s +epoch=65 global_step=25800 loss=3.63855 loss_avg=3.99302 acc=0.66406 acc_top1_avg=0.61656 acc_top5_avg=0.88636 lr=0.00100 gn=22.80478 time=57.13it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.27253 test_loss_avg=0.65378 acc=0.93750 test_acc_avg=0.80521 test_acc_top5_avg=0.98177 time=104.45it/s +epoch=65 global_step=25806 loss=0.20079 test_loss_avg=0.90821 acc=0.95312 test_acc_avg=0.74159 test_acc_top5_avg=0.97188 time=238.69it/s +epoch=65 global_step=25806 loss=0.51198 test_loss_avg=0.77933 acc=0.93750 test_acc_avg=0.77937 test_acc_top5_avg=0.97627 time=517.88it/s +curr_acc 0.7794 +BEST_ACC 0.8158 +curr_acc_top5 0.9763 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=3.74582 loss_avg=3.97903 acc=0.64062 acc_top1_avg=0.61754 acc_top5_avg=0.88530 lr=0.00100 gn=22.99418 time=54.44it/s +epoch=66 global_step=25900 loss=4.07856 loss_avg=3.93545 acc=0.61719 acc_top1_avg=0.62317 acc_top5_avg=0.88597 lr=0.00100 gn=22.17400 time=54.94it/s +epoch=66 global_step=25950 loss=3.77815 loss_avg=3.94272 acc=0.64062 acc_top1_avg=0.62288 acc_top5_avg=0.88601 lr=0.00100 gn=24.65894 time=50.61it/s +epoch=66 global_step=26000 loss=4.60192 loss_avg=3.96884 acc=0.54688 acc_top1_avg=0.61964 acc_top5_avg=0.88475 lr=0.00100 gn=23.80637 time=58.14it/s +epoch=66 global_step=26050 loss=4.10986 loss_avg=3.98602 acc=0.60156 acc_top1_avg=0.61799 acc_top5_avg=0.88381 lr=0.00100 gn=18.96024 time=63.48it/s +epoch=66 global_step=26100 loss=3.42649 loss_avg=3.96428 acc=0.67188 acc_top1_avg=0.62046 acc_top5_avg=0.88571 lr=0.00100 gn=17.06495 time=62.38it/s +epoch=66 global_step=26150 loss=4.81020 loss_avg=3.96880 acc=0.52344 acc_top1_avg=0.62034 acc_top5_avg=0.88601 lr=0.00100 gn=18.93464 time=54.23it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.33318 test_loss_avg=0.94799 acc=0.90625 test_acc_avg=0.71072 test_acc_top5_avg=0.96984 time=229.23it/s +epoch=66 global_step=26197 loss=0.46573 test_loss_avg=0.81882 acc=0.93750 test_acc_avg=0.75633 test_acc_top5_avg=0.97597 time=834.52it/s +curr_acc 0.7563 +BEST_ACC 0.8158 +curr_acc_top5 0.9760 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=4.27276 loss_avg=4.33691 acc=0.57031 acc_top1_avg=0.57812 acc_top5_avg=0.86979 lr=0.00100 gn=24.47869 time=44.85it/s +epoch=67 global_step=26250 loss=4.09946 loss_avg=3.93113 acc=0.60938 acc_top1_avg=0.62588 acc_top5_avg=0.88606 lr=0.00100 gn=19.89550 time=62.50it/s +epoch=67 global_step=26300 loss=4.03902 loss_avg=3.94737 acc=0.59375 acc_top1_avg=0.62181 acc_top5_avg=0.88486 lr=0.00100 gn=17.08443 time=49.55it/s +epoch=67 global_step=26350 loss=3.79074 loss_avg=3.96818 acc=0.64062 acc_top1_avg=0.61918 acc_top5_avg=0.88424 lr=0.00100 gn=23.07167 time=53.49it/s +epoch=67 global_step=26400 loss=3.73935 loss_avg=3.95681 acc=0.64062 acc_top1_avg=0.62034 acc_top5_avg=0.88351 lr=0.00100 gn=26.56794 time=58.05it/s +epoch=67 global_step=26450 loss=3.79970 loss_avg=3.98016 acc=0.64062 acc_top1_avg=0.61787 acc_top5_avg=0.88460 lr=0.00100 gn=23.47897 time=54.34it/s +epoch=67 global_step=26500 loss=4.35416 loss_avg=3.97161 acc=0.57812 acc_top1_avg=0.61876 acc_top5_avg=0.88413 lr=0.00100 gn=29.98436 time=54.81it/s +epoch=67 global_step=26550 loss=3.87877 loss_avg=3.97075 acc=0.61719 acc_top1_avg=0.61896 acc_top5_avg=0.88432 lr=0.00100 gn=24.06716 time=53.91it/s +====================Eval==================== +epoch=67 global_step=26588 loss=1.09993 test_loss_avg=1.12441 acc=0.64062 test_acc_avg=0.64844 test_acc_top5_avg=0.95982 time=238.60it/s +epoch=67 global_step=26588 loss=0.33439 test_loss_avg=0.80545 acc=0.88281 test_acc_avg=0.75795 test_acc_top5_avg=0.97067 time=122.73it/s +epoch=67 global_step=26588 loss=0.42387 test_loss_avg=0.64151 acc=0.93750 test_acc_avg=0.80785 test_acc_top5_avg=0.97765 time=859.14it/s +curr_acc 0.8079 +BEST_ACC 0.8158 +curr_acc_top5 0.9777 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=4.19405 loss_avg=4.09123 acc=0.60938 acc_top1_avg=0.60026 acc_top5_avg=0.87891 lr=0.00100 gn=24.19132 time=50.35it/s +epoch=68 global_step=26650 loss=3.69028 loss_avg=3.89447 acc=0.64062 acc_top1_avg=0.62828 acc_top5_avg=0.88962 lr=0.00100 gn=20.77281 time=56.95it/s +epoch=68 global_step=26700 loss=3.40244 loss_avg=3.88613 acc=0.67188 acc_top1_avg=0.62919 acc_top5_avg=0.88993 lr=0.00100 gn=18.34065 time=62.15it/s +epoch=68 global_step=26750 loss=3.54686 loss_avg=3.89929 acc=0.65625 acc_top1_avg=0.62746 acc_top5_avg=0.88797 lr=0.00100 gn=25.09408 time=57.77it/s +epoch=68 global_step=26800 loss=3.58954 loss_avg=3.87574 acc=0.67969 acc_top1_avg=0.63016 acc_top5_avg=0.88786 lr=0.00100 gn=25.59358 time=57.66it/s +epoch=68 global_step=26850 loss=4.17649 loss_avg=3.89371 acc=0.60156 acc_top1_avg=0.62861 acc_top5_avg=0.88890 lr=0.00100 gn=21.96690 time=44.90it/s +epoch=68 global_step=26900 loss=3.29070 loss_avg=3.90678 acc=0.69531 acc_top1_avg=0.62678 acc_top5_avg=0.88817 lr=0.00100 gn=25.00136 time=52.90it/s +epoch=68 global_step=26950 loss=3.79189 loss_avg=3.92666 acc=0.62500 acc_top1_avg=0.62500 acc_top5_avg=0.88827 lr=0.00100 gn=22.52025 time=58.78it/s +====================Eval==================== +epoch=68 global_step=26979 loss=1.19763 test_loss_avg=0.89964 acc=0.64844 test_acc_avg=0.73465 test_acc_top5_avg=0.96959 time=181.05it/s +epoch=68 global_step=26979 loss=0.09335 test_loss_avg=0.76233 acc=0.97656 test_acc_avg=0.77183 test_acc_top5_avg=0.97917 time=255.07it/s +epoch=68 global_step=26979 loss=0.17649 test_loss_avg=0.75492 acc=0.93750 test_acc_avg=0.77393 test_acc_top5_avg=0.97943 time=802.89it/s +curr_acc 0.7739 +BEST_ACC 0.8158 +curr_acc_top5 0.9794 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=4.00979 loss_avg=3.85558 acc=0.60938 acc_top1_avg=0.63467 acc_top5_avg=0.88318 lr=0.00100 gn=22.60405 time=55.60it/s +epoch=69 global_step=27050 loss=3.64828 loss_avg=3.88003 acc=0.65625 acc_top1_avg=0.63116 acc_top5_avg=0.88413 lr=0.00100 gn=28.78623 time=52.18it/s +epoch=69 global_step=27100 loss=3.63816 loss_avg=3.88074 acc=0.65625 acc_top1_avg=0.63042 acc_top5_avg=0.88830 lr=0.00100 gn=23.62930 time=50.40it/s +epoch=69 global_step=27150 loss=3.88932 loss_avg=3.88188 acc=0.61719 acc_top1_avg=0.63003 acc_top5_avg=0.88857 lr=0.00100 gn=21.97659 time=55.94it/s +epoch=69 global_step=27200 loss=3.75164 loss_avg=3.90506 acc=0.64844 acc_top1_avg=0.62783 acc_top5_avg=0.88727 lr=0.00100 gn=25.00381 time=51.11it/s +epoch=69 global_step=27250 loss=4.41924 loss_avg=3.91962 acc=0.57812 acc_top1_avg=0.62656 acc_top5_avg=0.88682 lr=0.00100 gn=27.75337 time=56.50it/s +epoch=69 global_step=27300 loss=3.89309 loss_avg=3.92673 acc=0.63281 acc_top1_avg=0.62612 acc_top5_avg=0.88612 lr=0.00100 gn=22.66156 time=56.97it/s +epoch=69 global_step=27350 loss=3.96611 loss_avg=3.91650 acc=0.64062 acc_top1_avg=0.62730 acc_top5_avg=0.88677 lr=0.00100 gn=25.14942 time=55.58it/s +====================Eval==================== +epoch=69 global_step=27370 loss=3.51188 test_loss_avg=1.01248 acc=0.26562 test_acc_avg=0.72784 test_acc_top5_avg=0.96253 time=213.60it/s +epoch=69 global_step=27370 loss=0.32767 test_loss_avg=0.96524 acc=0.93750 test_acc_avg=0.75702 test_acc_top5_avg=0.97004 time=498.20it/s +curr_acc 0.7570 +BEST_ACC 0.8158 +curr_acc_top5 0.9700 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=3.88033 loss_avg=3.89660 acc=0.60156 acc_top1_avg=0.62813 acc_top5_avg=0.89349 lr=0.00100 gn=21.80764 time=62.13it/s +epoch=70 global_step=27450 loss=3.79388 loss_avg=3.85100 acc=0.64844 acc_top1_avg=0.63281 acc_top5_avg=0.89023 lr=0.00100 gn=28.72351 time=54.36it/s +epoch=70 global_step=27500 loss=3.30656 loss_avg=3.87122 acc=0.70312 acc_top1_avg=0.63095 acc_top5_avg=0.88792 lr=0.00100 gn=25.93264 time=35.90it/s +epoch=70 global_step=27550 loss=3.53464 loss_avg=3.87217 acc=0.64844 acc_top1_avg=0.63116 acc_top5_avg=0.88702 lr=0.00100 gn=20.50853 time=57.49it/s +epoch=70 global_step=27600 loss=3.87837 loss_avg=3.90874 acc=0.61719 acc_top1_avg=0.62782 acc_top5_avg=0.88689 lr=0.00100 gn=24.90532 time=56.50it/s +epoch=70 global_step=27650 loss=4.33895 loss_avg=3.92012 acc=0.57031 acc_top1_avg=0.62609 acc_top5_avg=0.88717 lr=0.00100 gn=23.63530 time=62.91it/s +epoch=70 global_step=27700 loss=3.98444 loss_avg=3.91401 acc=0.61719 acc_top1_avg=0.62644 acc_top5_avg=0.88705 lr=0.00100 gn=24.68997 time=55.16it/s +epoch=70 global_step=27750 loss=4.24417 loss_avg=3.91723 acc=0.58594 acc_top1_avg=0.62609 acc_top5_avg=0.88676 lr=0.00100 gn=21.00683 time=55.68it/s +====================Eval==================== +epoch=70 global_step=27761 loss=1.23358 test_loss_avg=0.81497 acc=0.64062 test_acc_avg=0.76328 test_acc_top5_avg=0.97266 time=227.31it/s +epoch=70 global_step=27761 loss=0.15911 test_loss_avg=0.85034 acc=0.94531 test_acc_avg=0.75290 test_acc_top5_avg=0.97221 time=231.46it/s +epoch=70 global_step=27761 loss=0.23341 test_loss_avg=0.77128 acc=0.93750 test_acc_avg=0.77551 test_acc_top5_avg=0.97518 time=775.43it/s +curr_acc 0.7755 +BEST_ACC 0.8158 +curr_acc_top5 0.9752 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=3.16264 loss_avg=3.79663 acc=0.69531 acc_top1_avg=0.64363 acc_top5_avg=0.89083 lr=0.00100 gn=22.51052 time=62.79it/s +epoch=71 global_step=27850 loss=3.80280 loss_avg=3.88939 acc=0.64844 acc_top1_avg=0.63097 acc_top5_avg=0.88896 lr=0.00100 gn=26.91809 time=62.18it/s +epoch=71 global_step=27900 loss=4.19662 loss_avg=3.89975 acc=0.60156 acc_top1_avg=0.62910 acc_top5_avg=0.88894 lr=0.00100 gn=28.27511 time=59.35it/s +epoch=71 global_step=27950 loss=3.81302 loss_avg=3.87100 acc=0.64062 acc_top1_avg=0.63194 acc_top5_avg=0.88922 lr=0.00100 gn=25.17985 time=53.20it/s +epoch=71 global_step=28000 loss=3.62692 loss_avg=3.87541 acc=0.66406 acc_top1_avg=0.63147 acc_top5_avg=0.88951 lr=0.00100 gn=29.01106 time=54.22it/s +epoch=71 global_step=28050 loss=3.97468 loss_avg=3.88801 acc=0.60938 acc_top1_avg=0.63060 acc_top5_avg=0.88784 lr=0.00100 gn=31.23370 time=59.69it/s +epoch=71 global_step=28100 loss=4.07064 loss_avg=3.89466 acc=0.60938 acc_top1_avg=0.62966 acc_top5_avg=0.88848 lr=0.00100 gn=27.53252 time=56.03it/s +epoch=71 global_step=28150 loss=4.42249 loss_avg=3.90594 acc=0.58594 acc_top1_avg=0.62809 acc_top5_avg=0.88733 lr=0.00100 gn=25.43941 time=61.69it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.98917 test_loss_avg=0.89722 acc=0.67188 test_acc_avg=0.73628 test_acc_top5_avg=0.95960 time=162.05it/s +epoch=71 global_step=28152 loss=0.08843 test_loss_avg=0.79209 acc=0.93750 test_acc_avg=0.76968 test_acc_top5_avg=0.96865 time=499.86it/s +curr_acc 0.7697 +BEST_ACC 0.8158 +curr_acc_top5 0.9687 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=3.83469 loss_avg=3.84563 acc=0.64062 acc_top1_avg=0.63298 acc_top5_avg=0.89274 lr=0.00100 gn=25.83850 time=52.59it/s +epoch=72 global_step=28250 loss=3.60432 loss_avg=3.83936 acc=0.67188 acc_top1_avg=0.63568 acc_top5_avg=0.88831 lr=0.00100 gn=25.36793 time=61.18it/s +epoch=72 global_step=28300 loss=3.82797 loss_avg=3.84882 acc=0.64844 acc_top1_avg=0.63524 acc_top5_avg=0.88619 lr=0.00100 gn=31.29765 time=56.29it/s +epoch=72 global_step=28350 loss=3.45248 loss_avg=3.85906 acc=0.68750 acc_top1_avg=0.63463 acc_top5_avg=0.88648 lr=0.00100 gn=34.02357 time=61.60it/s +epoch=72 global_step=28400 loss=4.62929 loss_avg=3.85017 acc=0.52344 acc_top1_avg=0.63536 acc_top5_avg=0.88647 lr=0.00100 gn=21.50289 time=54.65it/s +epoch=72 global_step=28450 loss=4.30615 loss_avg=3.86436 acc=0.58594 acc_top1_avg=0.63389 acc_top5_avg=0.88627 lr=0.00100 gn=29.41498 time=59.13it/s +epoch=72 global_step=28500 loss=4.70086 loss_avg=3.87933 acc=0.53906 acc_top1_avg=0.63194 acc_top5_avg=0.88553 lr=0.00100 gn=20.41355 time=58.08it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.21496 test_loss_avg=1.08089 acc=0.93750 test_acc_avg=0.67057 test_acc_top5_avg=0.97656 time=235.32it/s +epoch=72 global_step=28543 loss=0.42356 test_loss_avg=1.05196 acc=0.87500 test_acc_avg=0.69733 test_acc_top5_avg=0.96547 time=106.14it/s +epoch=72 global_step=28543 loss=0.19118 test_loss_avg=0.85037 acc=0.93750 test_acc_avg=0.75346 test_acc_top5_avg=0.97271 time=653.42it/s +curr_acc 0.7535 +BEST_ACC 0.8158 +curr_acc_top5 0.9727 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=4.12932 loss_avg=3.83435 acc=0.61719 acc_top1_avg=0.63728 acc_top5_avg=0.90737 lr=0.00100 gn=27.84700 time=62.78it/s +epoch=73 global_step=28600 loss=3.72566 loss_avg=3.73476 acc=0.65625 acc_top1_avg=0.64391 acc_top5_avg=0.89803 lr=0.00100 gn=28.59692 time=49.67it/s +epoch=73 global_step=28650 loss=4.34944 loss_avg=3.79377 acc=0.57031 acc_top1_avg=0.63953 acc_top5_avg=0.89413 lr=0.00100 gn=26.89914 time=55.48it/s +epoch=73 global_step=28700 loss=3.23436 loss_avg=3.85366 acc=0.70312 acc_top1_avg=0.63222 acc_top5_avg=0.88844 lr=0.00100 gn=28.15224 time=50.63it/s +epoch=73 global_step=28750 loss=3.86485 loss_avg=3.86941 acc=0.64062 acc_top1_avg=0.63070 acc_top5_avg=0.88776 lr=0.00100 gn=29.10092 time=57.02it/s +epoch=73 global_step=28800 loss=3.78808 loss_avg=3.86715 acc=0.65625 acc_top1_avg=0.63166 acc_top5_avg=0.88774 lr=0.00100 gn=23.63515 time=57.88it/s +epoch=73 global_step=28850 loss=4.48130 loss_avg=3.86830 acc=0.56250 acc_top1_avg=0.63106 acc_top5_avg=0.88795 lr=0.00100 gn=33.80774 time=57.51it/s +epoch=73 global_step=28900 loss=3.81802 loss_avg=3.86226 acc=0.64062 acc_top1_avg=0.63220 acc_top5_avg=0.88815 lr=0.00100 gn=29.44044 time=58.68it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.47444 test_loss_avg=1.05041 acc=0.85156 test_acc_avg=0.69413 test_acc_top5_avg=0.95857 time=235.12it/s +epoch=73 global_step=28934 loss=0.14588 test_loss_avg=0.88172 acc=0.93750 test_acc_avg=0.75603 test_acc_top5_avg=0.96875 time=500.63it/s +curr_acc 0.7560 +BEST_ACC 0.8158 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=3.80314 loss_avg=3.87505 acc=0.64844 acc_top1_avg=0.62988 acc_top5_avg=0.88867 lr=0.00100 gn=30.59849 time=57.81it/s +epoch=74 global_step=29000 loss=3.67918 loss_avg=3.81462 acc=0.67188 acc_top1_avg=0.63885 acc_top5_avg=0.89145 lr=0.00100 gn=35.60380 time=54.06it/s +epoch=74 global_step=29050 loss=3.31463 loss_avg=3.82395 acc=0.69531 acc_top1_avg=0.63786 acc_top5_avg=0.88665 lr=0.00100 gn=24.06127 time=40.72it/s +epoch=74 global_step=29100 loss=4.39198 loss_avg=3.84126 acc=0.57031 acc_top1_avg=0.63667 acc_top5_avg=0.88804 lr=0.00100 gn=22.97258 time=54.36it/s +epoch=74 global_step=29150 loss=3.61068 loss_avg=3.83445 acc=0.67188 acc_top1_avg=0.63686 acc_top5_avg=0.88849 lr=0.00100 gn=25.64876 time=58.13it/s +epoch=74 global_step=29200 loss=3.88697 loss_avg=3.84062 acc=0.61719 acc_top1_avg=0.63628 acc_top5_avg=0.88833 lr=0.00100 gn=32.74196 time=57.99it/s +epoch=74 global_step=29250 loss=3.00868 loss_avg=3.86005 acc=0.74219 acc_top1_avg=0.63452 acc_top5_avg=0.88739 lr=0.00100 gn=25.71141 time=58.13it/s +epoch=74 global_step=29300 loss=3.73800 loss_avg=3.86082 acc=0.63281 acc_top1_avg=0.63448 acc_top5_avg=0.88719 lr=0.00100 gn=25.75652 time=52.69it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.74317 test_loss_avg=0.94092 acc=0.73438 test_acc_avg=0.68945 test_acc_top5_avg=0.97266 time=229.33it/s +epoch=74 global_step=29325 loss=1.83435 test_loss_avg=1.04685 acc=0.44531 test_acc_avg=0.69777 test_acc_top5_avg=0.96672 time=237.97it/s +epoch=74 global_step=29325 loss=0.19533 test_loss_avg=0.79416 acc=0.93750 test_acc_avg=0.76958 test_acc_top5_avg=0.97577 time=489.87it/s +curr_acc 0.7696 +BEST_ACC 0.8158 +curr_acc_top5 0.9758 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=3.81512 loss_avg=3.68806 acc=0.63281 acc_top1_avg=0.64875 acc_top5_avg=0.89062 lr=0.00100 gn=27.02232 time=54.70it/s +epoch=75 global_step=29400 loss=3.63458 loss_avg=3.76999 acc=0.65625 acc_top1_avg=0.64240 acc_top5_avg=0.89104 lr=0.00100 gn=27.73799 time=61.77it/s +epoch=75 global_step=29450 loss=4.15730 loss_avg=3.80795 acc=0.60938 acc_top1_avg=0.63962 acc_top5_avg=0.88938 lr=0.00100 gn=29.58325 time=62.36it/s +epoch=75 global_step=29500 loss=3.20961 loss_avg=3.81310 acc=0.71875 acc_top1_avg=0.63964 acc_top5_avg=0.88991 lr=0.00100 gn=33.11939 time=57.47it/s +epoch=75 global_step=29550 loss=3.87203 loss_avg=3.82661 acc=0.61719 acc_top1_avg=0.63788 acc_top5_avg=0.88750 lr=0.00100 gn=24.94003 time=53.60it/s +epoch=75 global_step=29600 loss=3.14716 loss_avg=3.82402 acc=0.71875 acc_top1_avg=0.63815 acc_top5_avg=0.88889 lr=0.00100 gn=29.88225 time=54.49it/s +epoch=75 global_step=29650 loss=4.27596 loss_avg=3.84186 acc=0.57812 acc_top1_avg=0.63603 acc_top5_avg=0.88728 lr=0.00100 gn=33.83953 time=62.73it/s +epoch=75 global_step=29700 loss=3.80943 loss_avg=3.85710 acc=0.66406 acc_top1_avg=0.63498 acc_top5_avg=0.88633 lr=0.00100 gn=33.38141 time=53.63it/s +====================Eval==================== +epoch=75 global_step=29716 loss=1.55126 test_loss_avg=0.84809 acc=0.60156 test_acc_avg=0.75625 test_acc_top5_avg=0.96531 time=218.98it/s +epoch=75 global_step=29716 loss=0.14711 test_loss_avg=0.79610 acc=0.94531 test_acc_avg=0.76885 test_acc_top5_avg=0.97104 time=244.39it/s +epoch=75 global_step=29716 loss=0.28598 test_loss_avg=0.76388 acc=0.93750 test_acc_avg=0.77828 test_acc_top5_avg=0.97251 time=496.37it/s +curr_acc 0.7783 +BEST_ACC 0.8158 +curr_acc_top5 0.9725 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=4.25017 loss_avg=3.78540 acc=0.59375 acc_top1_avg=0.64361 acc_top5_avg=0.88764 lr=0.00100 gn=30.37684 time=60.95it/s +epoch=76 global_step=29800 loss=3.98765 loss_avg=3.74914 acc=0.61719 acc_top1_avg=0.64797 acc_top5_avg=0.89239 lr=0.00100 gn=25.88539 time=55.47it/s +epoch=76 global_step=29850 loss=4.05385 loss_avg=3.80291 acc=0.62500 acc_top1_avg=0.64173 acc_top5_avg=0.88765 lr=0.00100 gn=22.24609 time=55.12it/s +epoch=76 global_step=29900 loss=3.57511 loss_avg=3.82958 acc=0.65625 acc_top1_avg=0.63808 acc_top5_avg=0.88638 lr=0.00100 gn=29.85343 time=59.28it/s +epoch=76 global_step=29950 loss=3.44398 loss_avg=3.83765 acc=0.67188 acc_top1_avg=0.63789 acc_top5_avg=0.88722 lr=0.00100 gn=26.07810 time=54.80it/s +epoch=76 global_step=30000 loss=3.68931 loss_avg=3.82682 acc=0.67188 acc_top1_avg=0.63884 acc_top5_avg=0.88809 lr=0.00100 gn=34.62844 time=54.98it/s +epoch=76 global_step=30050 loss=3.42510 loss_avg=3.82666 acc=0.70312 acc_top1_avg=0.63852 acc_top5_avg=0.88810 lr=0.00100 gn=27.60596 time=62.58it/s +epoch=76 global_step=30100 loss=4.33103 loss_avg=3.83479 acc=0.55469 acc_top1_avg=0.63784 acc_top5_avg=0.88804 lr=0.00100 gn=28.62658 time=54.07it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.88254 test_loss_avg=0.85351 acc=0.74219 test_acc_avg=0.74338 test_acc_top5_avg=0.96824 time=243.47it/s +epoch=76 global_step=30107 loss=0.22994 test_loss_avg=0.80840 acc=0.93750 test_acc_avg=0.76622 test_acc_top5_avg=0.97251 time=488.90it/s +curr_acc 0.7662 +BEST_ACC 0.8158 +curr_acc_top5 0.9725 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=4.07948 loss_avg=3.77830 acc=0.61719 acc_top1_avg=0.64408 acc_top5_avg=0.88808 lr=0.00100 gn=30.23842 time=49.22it/s +epoch=77 global_step=30200 loss=4.06593 loss_avg=3.84031 acc=0.63281 acc_top1_avg=0.63768 acc_top5_avg=0.88659 lr=0.00100 gn=25.71472 time=55.43it/s +epoch=77 global_step=30250 loss=3.66518 loss_avg=3.82055 acc=0.62500 acc_top1_avg=0.64057 acc_top5_avg=0.88604 lr=0.00100 gn=26.95767 time=53.92it/s +epoch=77 global_step=30300 loss=3.44079 loss_avg=3.83206 acc=0.67969 acc_top1_avg=0.64014 acc_top5_avg=0.88589 lr=0.00100 gn=34.43152 time=62.45it/s +epoch=77 global_step=30350 loss=4.15047 loss_avg=3.82082 acc=0.62500 acc_top1_avg=0.64108 acc_top5_avg=0.88715 lr=0.00100 gn=28.46049 time=61.85it/s +epoch=77 global_step=30400 loss=3.51855 loss_avg=3.81516 acc=0.66406 acc_top1_avg=0.64084 acc_top5_avg=0.88775 lr=0.00100 gn=25.99227 time=61.13it/s +epoch=77 global_step=30450 loss=3.50342 loss_avg=3.81678 acc=0.67188 acc_top1_avg=0.64067 acc_top5_avg=0.88760 lr=0.00100 gn=27.72773 time=56.53it/s +====================Eval==================== +epoch=77 global_step=30498 loss=1.31730 test_loss_avg=0.82090 acc=0.60156 test_acc_avg=0.75276 test_acc_top5_avg=0.97932 time=234.79it/s +epoch=77 global_step=30498 loss=0.07722 test_loss_avg=0.96998 acc=0.96875 test_acc_avg=0.72318 test_acc_top5_avg=0.96514 time=230.58it/s +epoch=77 global_step=30498 loss=0.10752 test_loss_avg=0.84369 acc=0.93750 test_acc_avg=0.75742 test_acc_top5_avg=0.97023 time=515.27it/s +curr_acc 0.7574 +BEST_ACC 0.8158 +curr_acc_top5 0.9702 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=3.14134 loss_avg=3.46510 acc=0.71094 acc_top1_avg=0.67578 acc_top5_avg=0.90234 lr=0.00100 gn=31.58383 time=60.70it/s +epoch=78 global_step=30550 loss=3.39624 loss_avg=3.73983 acc=0.67969 acc_top1_avg=0.64874 acc_top5_avg=0.89017 lr=0.00100 gn=26.26777 time=48.96it/s +epoch=78 global_step=30600 loss=3.98414 loss_avg=3.73538 acc=0.61719 acc_top1_avg=0.64844 acc_top5_avg=0.88909 lr=0.00100 gn=29.79968 time=60.07it/s +epoch=78 global_step=30650 loss=3.74631 loss_avg=3.74907 acc=0.64844 acc_top1_avg=0.64690 acc_top5_avg=0.89109 lr=0.00100 gn=30.04986 time=42.50it/s +epoch=78 global_step=30700 loss=3.59371 loss_avg=3.77667 acc=0.66406 acc_top1_avg=0.64376 acc_top5_avg=0.89004 lr=0.00100 gn=32.13365 time=57.39it/s +epoch=78 global_step=30750 loss=4.17674 loss_avg=3.77949 acc=0.60156 acc_top1_avg=0.64413 acc_top5_avg=0.89087 lr=0.00100 gn=32.06376 time=57.04it/s +epoch=78 global_step=30800 loss=4.25310 loss_avg=3.80340 acc=0.60156 acc_top1_avg=0.64171 acc_top5_avg=0.88949 lr=0.00100 gn=34.11331 time=52.41it/s +epoch=78 global_step=30850 loss=4.00244 loss_avg=3.80359 acc=0.63281 acc_top1_avg=0.64211 acc_top5_avg=0.88952 lr=0.00100 gn=34.74261 time=55.10it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.59746 test_loss_avg=0.98181 acc=0.85156 test_acc_avg=0.71217 test_acc_top5_avg=0.96320 time=235.89it/s +epoch=78 global_step=30889 loss=0.03680 test_loss_avg=0.79383 acc=1.00000 test_acc_avg=0.76800 test_acc_top5_avg=0.97241 time=514.07it/s +curr_acc 0.7680 +BEST_ACC 0.8158 +curr_acc_top5 0.9724 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=3.90021 loss_avg=3.86077 acc=0.63281 acc_top1_avg=0.63068 acc_top5_avg=0.88281 lr=0.00100 gn=30.62096 time=57.33it/s +epoch=79 global_step=30950 loss=4.33568 loss_avg=3.78973 acc=0.58594 acc_top1_avg=0.64408 acc_top5_avg=0.88998 lr=0.00100 gn=28.33982 time=54.05it/s +epoch=79 global_step=31000 loss=3.31102 loss_avg=3.78345 acc=0.71094 acc_top1_avg=0.64478 acc_top5_avg=0.89013 lr=0.00100 gn=34.64497 time=62.36it/s +epoch=79 global_step=31050 loss=3.88667 loss_avg=3.77499 acc=0.62500 acc_top1_avg=0.64582 acc_top5_avg=0.88970 lr=0.00100 gn=30.94021 time=55.89it/s +epoch=79 global_step=31100 loss=3.98284 loss_avg=3.79313 acc=0.60156 acc_top1_avg=0.64336 acc_top5_avg=0.89018 lr=0.00100 gn=24.07964 time=50.20it/s +epoch=79 global_step=31150 loss=4.21196 loss_avg=3.78903 acc=0.60156 acc_top1_avg=0.64356 acc_top5_avg=0.88880 lr=0.00100 gn=35.74316 time=62.31it/s +epoch=79 global_step=31200 loss=3.66074 loss_avg=3.79810 acc=0.65625 acc_top1_avg=0.64216 acc_top5_avg=0.88899 lr=0.00100 gn=29.63246 time=57.55it/s +epoch=79 global_step=31250 loss=3.47140 loss_avg=3.80072 acc=0.68750 acc_top1_avg=0.64229 acc_top5_avg=0.88807 lr=0.00100 gn=27.83181 time=55.67it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.19757 test_loss_avg=0.75467 acc=0.94531 test_acc_avg=0.75434 test_acc_top5_avg=0.97743 time=229.40it/s +epoch=79 global_step=31280 loss=0.33340 test_loss_avg=0.91631 acc=0.88281 test_acc_avg=0.73636 test_acc_top5_avg=0.96901 time=215.08it/s +epoch=79 global_step=31280 loss=0.19567 test_loss_avg=0.73496 acc=0.93750 test_acc_avg=0.78738 test_acc_top5_avg=0.97587 time=493.62it/s +curr_acc 0.7874 +BEST_ACC 0.8158 +curr_acc_top5 0.9759 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=3.57095 loss_avg=3.68130 acc=0.66406 acc_top1_avg=0.65469 acc_top5_avg=0.88164 lr=0.00010 gn=32.38764 time=52.93it/s +epoch=80 global_step=31350 loss=3.56005 loss_avg=3.73590 acc=0.64844 acc_top1_avg=0.64766 acc_top5_avg=0.88616 lr=0.00010 gn=25.11635 time=56.37it/s +epoch=80 global_step=31400 loss=3.67435 loss_avg=3.73731 acc=0.67969 acc_top1_avg=0.64753 acc_top5_avg=0.88730 lr=0.00010 gn=29.53544 time=59.44it/s +epoch=80 global_step=31450 loss=3.35933 loss_avg=3.72181 acc=0.69531 acc_top1_avg=0.64903 acc_top5_avg=0.88805 lr=0.00010 gn=34.73542 time=57.00it/s +epoch=80 global_step=31500 loss=3.87421 loss_avg=3.69881 acc=0.64062 acc_top1_avg=0.65188 acc_top5_avg=0.88935 lr=0.00010 gn=29.48978 time=53.97it/s +epoch=80 global_step=31550 loss=3.47699 loss_avg=3.69417 acc=0.67188 acc_top1_avg=0.65179 acc_top5_avg=0.88932 lr=0.00010 gn=28.56607 time=60.11it/s +epoch=80 global_step=31600 loss=3.64795 loss_avg=3.68520 acc=0.66406 acc_top1_avg=0.65293 acc_top5_avg=0.88977 lr=0.00010 gn=26.78456 time=54.85it/s +epoch=80 global_step=31650 loss=3.81896 loss_avg=3.67445 acc=0.63281 acc_top1_avg=0.65414 acc_top5_avg=0.88970 lr=0.00010 gn=22.86441 time=54.52it/s +====================Eval==================== +epoch=80 global_step=31671 loss=1.32139 test_loss_avg=0.96629 acc=0.60156 test_acc_avg=0.71120 test_acc_top5_avg=0.96510 time=240.28it/s +epoch=80 global_step=31671 loss=0.16221 test_loss_avg=0.79091 acc=0.93750 test_acc_avg=0.77215 test_acc_top5_avg=0.97379 time=478.75it/s +curr_acc 0.7722 +BEST_ACC 0.8158 +curr_acc_top5 0.9738 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=4.57725 loss_avg=3.67263 acc=0.55469 acc_top1_avg=0.65302 acc_top5_avg=0.88847 lr=0.00010 gn=31.51464 time=54.00it/s +epoch=81 global_step=31750 loss=3.89103 loss_avg=3.64326 acc=0.62500 acc_top1_avg=0.65724 acc_top5_avg=0.88677 lr=0.00010 gn=33.87821 time=58.77it/s +epoch=81 global_step=31800 loss=4.08241 loss_avg=3.64680 acc=0.60156 acc_top1_avg=0.65625 acc_top5_avg=0.88748 lr=0.00010 gn=25.97409 time=56.48it/s +epoch=81 global_step=31850 loss=3.67772 loss_avg=3.63898 acc=0.66406 acc_top1_avg=0.65717 acc_top5_avg=0.88748 lr=0.00010 gn=28.40066 time=62.52it/s +epoch=81 global_step=31900 loss=3.17246 loss_avg=3.61978 acc=0.71875 acc_top1_avg=0.66007 acc_top5_avg=0.88895 lr=0.00010 gn=30.62298 time=60.76it/s +epoch=81 global_step=31950 loss=3.52145 loss_avg=3.63000 acc=0.67188 acc_top1_avg=0.65841 acc_top5_avg=0.88970 lr=0.00010 gn=29.52222 time=57.02it/s +epoch=81 global_step=32000 loss=3.51302 loss_avg=3.62014 acc=0.65625 acc_top1_avg=0.65962 acc_top5_avg=0.89034 lr=0.00010 gn=29.12487 time=60.10it/s +epoch=81 global_step=32050 loss=2.79041 loss_avg=3.60507 acc=0.74219 acc_top1_avg=0.66095 acc_top5_avg=0.89038 lr=0.00010 gn=29.19930 time=57.24it/s +====================Eval==================== +epoch=81 global_step=32062 loss=1.38179 test_loss_avg=1.38179 acc=0.54688 test_acc_avg=0.54688 test_acc_top5_avg=0.95312 time=208.94it/s +epoch=81 global_step=32062 loss=2.19448 test_loss_avg=0.97360 acc=0.46875 test_acc_avg=0.71584 test_acc_top5_avg=0.96661 time=186.48it/s +epoch=81 global_step=32062 loss=0.18088 test_loss_avg=0.78616 acc=0.93750 test_acc_avg=0.77047 test_acc_top5_avg=0.97439 time=829.57it/s +curr_acc 0.7705 +BEST_ACC 0.8158 +curr_acc_top5 0.9744 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=2.89953 loss_avg=3.61100 acc=0.75781 acc_top1_avg=0.66427 acc_top5_avg=0.88939 lr=0.00010 gn=34.36278 time=59.58it/s +epoch=82 global_step=32150 loss=3.64080 loss_avg=3.57943 acc=0.64844 acc_top1_avg=0.66442 acc_top5_avg=0.89560 lr=0.00010 gn=34.29107 time=52.88it/s +epoch=82 global_step=32200 loss=3.46100 loss_avg=3.57668 acc=0.65625 acc_top1_avg=0.66463 acc_top5_avg=0.89487 lr=0.00010 gn=29.21475 time=52.31it/s +epoch=82 global_step=32250 loss=3.37279 loss_avg=3.57637 acc=0.68750 acc_top1_avg=0.66435 acc_top5_avg=0.89374 lr=0.00010 gn=27.26176 time=53.91it/s +epoch=82 global_step=32300 loss=3.45604 loss_avg=3.58038 acc=0.67969 acc_top1_avg=0.66400 acc_top5_avg=0.89309 lr=0.00010 gn=30.14162 time=56.17it/s +epoch=82 global_step=32350 loss=3.38278 loss_avg=3.59198 acc=0.69531 acc_top1_avg=0.66252 acc_top5_avg=0.89231 lr=0.00010 gn=31.62833 time=53.86it/s +epoch=82 global_step=32400 loss=3.28210 loss_avg=3.58802 acc=0.70312 acc_top1_avg=0.66291 acc_top5_avg=0.89185 lr=0.00010 gn=32.70341 time=55.97it/s +epoch=82 global_step=32450 loss=3.68956 loss_avg=3.58518 acc=0.64062 acc_top1_avg=0.66279 acc_top5_avg=0.89175 lr=0.00010 gn=33.06615 time=55.05it/s +====================Eval==================== +epoch=82 global_step=32453 loss=1.20929 test_loss_avg=0.90813 acc=0.66406 test_acc_avg=0.72763 test_acc_top5_avg=0.96982 time=160.27it/s +epoch=82 global_step=32453 loss=0.14349 test_loss_avg=0.88954 acc=0.95312 test_acc_avg=0.74436 test_acc_top5_avg=0.96984 time=246.58it/s +epoch=82 global_step=32453 loss=0.13158 test_loss_avg=0.82149 acc=0.93750 test_acc_avg=0.76335 test_acc_top5_avg=0.97231 time=549.64it/s +curr_acc 0.7634 +BEST_ACC 0.8158 +curr_acc_top5 0.9723 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=3.91681 loss_avg=3.58339 acc=0.61719 acc_top1_avg=0.66140 acc_top5_avg=0.89279 lr=0.00010 gn=29.32626 time=54.43it/s +epoch=83 global_step=32550 loss=4.27636 loss_avg=3.52166 acc=0.59375 acc_top1_avg=0.66986 acc_top5_avg=0.89562 lr=0.00010 gn=33.57980 time=58.63it/s +epoch=83 global_step=32600 loss=3.68791 loss_avg=3.53501 acc=0.64844 acc_top1_avg=0.66858 acc_top5_avg=0.89605 lr=0.00010 gn=26.63441 time=63.53it/s +epoch=83 global_step=32650 loss=3.16377 loss_avg=3.53928 acc=0.70312 acc_top1_avg=0.66846 acc_top5_avg=0.89376 lr=0.00010 gn=29.94818 time=52.20it/s +epoch=83 global_step=32700 loss=3.82774 loss_avg=3.56592 acc=0.64062 acc_top1_avg=0.66615 acc_top5_avg=0.89138 lr=0.00010 gn=31.37514 time=56.89it/s +epoch=83 global_step=32750 loss=3.38925 loss_avg=3.57587 acc=0.67969 acc_top1_avg=0.66517 acc_top5_avg=0.89123 lr=0.00010 gn=31.52013 time=54.05it/s +epoch=83 global_step=32800 loss=3.54113 loss_avg=3.57236 acc=0.65625 acc_top1_avg=0.66539 acc_top5_avg=0.89056 lr=0.00010 gn=26.32892 time=50.83it/s +====================Eval==================== +epoch=83 global_step=32844 loss=1.12798 test_loss_avg=0.86848 acc=0.69531 test_acc_avg=0.74328 test_acc_top5_avg=0.96930 time=239.87it/s +epoch=83 global_step=32844 loss=0.15703 test_loss_avg=0.78734 acc=0.93750 test_acc_avg=0.77195 test_acc_top5_avg=0.97399 time=521.29it/s +curr_acc 0.7720 +BEST_ACC 0.8158 +curr_acc_top5 0.9740 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=3.40507 loss_avg=3.38470 acc=0.67969 acc_top1_avg=0.68750 acc_top5_avg=0.91276 lr=0.00010 gn=26.50885 time=56.75it/s +epoch=84 global_step=32900 loss=3.39452 loss_avg=3.52924 acc=0.69531 acc_top1_avg=0.66908 acc_top5_avg=0.89858 lr=0.00010 gn=29.66358 time=55.79it/s +epoch=84 global_step=32950 loss=3.30140 loss_avg=3.54917 acc=0.68750 acc_top1_avg=0.66745 acc_top5_avg=0.89586 lr=0.00010 gn=21.90733 time=53.96it/s +epoch=84 global_step=33000 loss=2.97847 loss_avg=3.55750 acc=0.71875 acc_top1_avg=0.66602 acc_top5_avg=0.89628 lr=0.00010 gn=27.02496 time=61.03it/s +epoch=84 global_step=33050 loss=3.20019 loss_avg=3.55036 acc=0.70312 acc_top1_avg=0.66615 acc_top5_avg=0.89415 lr=0.00010 gn=29.41538 time=54.89it/s +epoch=84 global_step=33100 loss=3.06664 loss_avg=3.56137 acc=0.71094 acc_top1_avg=0.66544 acc_top5_avg=0.89371 lr=0.00010 gn=32.84958 time=53.28it/s +epoch=84 global_step=33150 loss=2.98984 loss_avg=3.55563 acc=0.71875 acc_top1_avg=0.66628 acc_top5_avg=0.89295 lr=0.00010 gn=25.94815 time=60.66it/s +epoch=84 global_step=33200 loss=3.77947 loss_avg=3.56321 acc=0.63281 acc_top1_avg=0.66509 acc_top5_avg=0.89131 lr=0.00010 gn=27.99163 time=52.38it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.26081 test_loss_avg=0.78900 acc=0.92188 test_acc_avg=0.75949 test_acc_top5_avg=0.97600 time=217.68it/s +epoch=84 global_step=33235 loss=0.15500 test_loss_avg=0.95238 acc=0.95312 test_acc_avg=0.72546 test_acc_top5_avg=0.96667 time=220.87it/s +epoch=84 global_step=33235 loss=0.14629 test_loss_avg=0.80026 acc=0.93750 test_acc_avg=0.76859 test_acc_top5_avg=0.97280 time=508.83it/s +curr_acc 0.7686 +BEST_ACC 0.8158 +curr_acc_top5 0.9728 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=3.11820 loss_avg=3.40301 acc=0.71094 acc_top1_avg=0.68073 acc_top5_avg=0.89948 lr=0.00010 gn=35.15874 time=52.26it/s +epoch=85 global_step=33300 loss=3.32893 loss_avg=3.42071 acc=0.69531 acc_top1_avg=0.67981 acc_top5_avg=0.89880 lr=0.00010 gn=34.46257 time=63.19it/s +epoch=85 global_step=33350 loss=3.24542 loss_avg=3.47398 acc=0.70312 acc_top1_avg=0.67480 acc_top5_avg=0.89545 lr=0.00010 gn=30.84789 time=53.83it/s +epoch=85 global_step=33400 loss=3.34085 loss_avg=3.49720 acc=0.68750 acc_top1_avg=0.67206 acc_top5_avg=0.89498 lr=0.00010 gn=31.09201 time=58.15it/s +epoch=85 global_step=33450 loss=2.55981 loss_avg=3.50816 acc=0.76562 acc_top1_avg=0.67140 acc_top5_avg=0.89364 lr=0.00010 gn=28.46774 time=55.03it/s +epoch=85 global_step=33500 loss=3.28684 loss_avg=3.51654 acc=0.70312 acc_top1_avg=0.67043 acc_top5_avg=0.89407 lr=0.00010 gn=33.76033 time=54.26it/s +epoch=85 global_step=33550 loss=2.96597 loss_avg=3.53386 acc=0.73438 acc_top1_avg=0.66820 acc_top5_avg=0.89239 lr=0.00010 gn=35.20297 time=60.24it/s +epoch=85 global_step=33600 loss=3.75446 loss_avg=3.53718 acc=0.66406 acc_top1_avg=0.66789 acc_top5_avg=0.89131 lr=0.00010 gn=39.43789 time=52.25it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.34145 test_loss_avg=0.93887 acc=0.89844 test_acc_avg=0.71964 test_acc_top5_avg=0.96652 time=235.97it/s +epoch=85 global_step=33626 loss=0.15072 test_loss_avg=0.84517 acc=0.93750 test_acc_avg=0.76058 test_acc_top5_avg=0.96984 time=490.79it/s +curr_acc 0.7606 +BEST_ACC 0.8158 +curr_acc_top5 0.9698 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=3.00979 loss_avg=3.53313 acc=0.71875 acc_top1_avg=0.66797 acc_top5_avg=0.89746 lr=0.00010 gn=31.27813 time=46.93it/s +epoch=86 global_step=33700 loss=3.87401 loss_avg=3.53813 acc=0.62500 acc_top1_avg=0.66522 acc_top5_avg=0.89295 lr=0.00010 gn=34.76575 time=59.86it/s +epoch=86 global_step=33750 loss=3.69676 loss_avg=3.53937 acc=0.65625 acc_top1_avg=0.66652 acc_top5_avg=0.89239 lr=0.00010 gn=31.89193 time=58.51it/s +epoch=86 global_step=33800 loss=3.60903 loss_avg=3.56533 acc=0.67188 acc_top1_avg=0.66384 acc_top5_avg=0.89130 lr=0.00010 gn=29.67759 time=62.74it/s +epoch=86 global_step=33850 loss=3.90704 loss_avg=3.56221 acc=0.61719 acc_top1_avg=0.66445 acc_top5_avg=0.89111 lr=0.00010 gn=28.01414 time=59.49it/s +epoch=86 global_step=33900 loss=3.85470 loss_avg=3.55274 acc=0.60938 acc_top1_avg=0.66594 acc_top5_avg=0.89254 lr=0.00010 gn=29.57839 time=58.03it/s +epoch=86 global_step=33950 loss=3.54122 loss_avg=3.53962 acc=0.65625 acc_top1_avg=0.66792 acc_top5_avg=0.89335 lr=0.00010 gn=28.55471 time=55.30it/s +epoch=86 global_step=34000 loss=3.11140 loss_avg=3.53777 acc=0.70312 acc_top1_avg=0.66780 acc_top5_avg=0.89276 lr=0.00010 gn=29.74486 time=54.24it/s +====================Eval==================== +epoch=86 global_step=34017 loss=1.21109 test_loss_avg=1.27286 acc=0.65625 test_acc_avg=0.61198 test_acc_top5_avg=0.96484 time=213.82it/s +epoch=86 global_step=34017 loss=0.28936 test_loss_avg=1.04938 acc=0.90625 test_acc_avg=0.69782 test_acc_top5_avg=0.96345 time=236.38it/s +epoch=86 global_step=34017 loss=0.17765 test_loss_avg=0.80282 acc=0.93750 test_acc_avg=0.76810 test_acc_top5_avg=0.97330 time=544.15it/s +curr_acc 0.7681 +BEST_ACC 0.8158 +curr_acc_top5 0.9733 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=4.10685 loss_avg=3.52924 acc=0.60156 acc_top1_avg=0.67045 acc_top5_avg=0.88281 lr=0.00010 gn=26.90929 time=53.49it/s +epoch=87 global_step=34100 loss=3.66773 loss_avg=3.54418 acc=0.64844 acc_top1_avg=0.66802 acc_top5_avg=0.88686 lr=0.00010 gn=27.17389 time=55.02it/s +epoch=87 global_step=34150 loss=3.30802 loss_avg=3.53044 acc=0.67969 acc_top1_avg=0.66947 acc_top5_avg=0.88863 lr=0.00010 gn=30.16041 time=60.59it/s +epoch=87 global_step=34200 loss=3.83485 loss_avg=3.52752 acc=0.65625 acc_top1_avg=0.66995 acc_top5_avg=0.88930 lr=0.00010 gn=32.53479 time=52.13it/s +epoch=87 global_step=34250 loss=3.58925 loss_avg=3.53724 acc=0.66406 acc_top1_avg=0.66876 acc_top5_avg=0.89073 lr=0.00010 gn=25.99171 time=63.56it/s +epoch=87 global_step=34300 loss=3.94020 loss_avg=3.54116 acc=0.61719 acc_top1_avg=0.66862 acc_top5_avg=0.89167 lr=0.00010 gn=27.20068 time=52.68it/s +epoch=87 global_step=34350 loss=4.21589 loss_avg=3.54773 acc=0.60156 acc_top1_avg=0.66796 acc_top5_avg=0.89159 lr=0.00010 gn=32.31427 time=55.45it/s +epoch=87 global_step=34400 loss=4.00519 loss_avg=3.54761 acc=0.64062 acc_top1_avg=0.66802 acc_top5_avg=0.89111 lr=0.00010 gn=32.83195 time=61.11it/s +====================Eval==================== +epoch=87 global_step=34408 loss=1.47474 test_loss_avg=1.03627 acc=0.55469 test_acc_avg=0.69358 test_acc_top5_avg=0.96209 time=246.83it/s +epoch=87 global_step=34408 loss=0.12102 test_loss_avg=0.87643 acc=0.95312 test_acc_avg=0.75091 test_acc_top5_avg=0.96946 time=254.69it/s +epoch=87 global_step=34408 loss=0.19950 test_loss_avg=0.85787 acc=0.93750 test_acc_avg=0.75603 test_acc_top5_avg=0.97023 time=834.85it/s +curr_acc 0.7560 +BEST_ACC 0.8158 +curr_acc_top5 0.9702 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=3.12157 loss_avg=3.47670 acc=0.72656 acc_top1_avg=0.67299 acc_top5_avg=0.89360 lr=0.00010 gn=29.01313 time=56.19it/s +epoch=88 global_step=34500 loss=3.38248 loss_avg=3.50958 acc=0.67969 acc_top1_avg=0.67009 acc_top5_avg=0.89122 lr=0.00010 gn=24.56161 time=55.14it/s +epoch=88 global_step=34550 loss=3.45151 loss_avg=3.52013 acc=0.67188 acc_top1_avg=0.66923 acc_top5_avg=0.89134 lr=0.00010 gn=33.28416 time=60.91it/s +epoch=88 global_step=34600 loss=3.45363 loss_avg=3.50227 acc=0.69531 acc_top1_avg=0.67179 acc_top5_avg=0.89286 lr=0.00010 gn=30.00114 time=48.64it/s +epoch=88 global_step=34650 loss=3.46629 loss_avg=3.50747 acc=0.67188 acc_top1_avg=0.67158 acc_top5_avg=0.89327 lr=0.00010 gn=27.19713 time=55.61it/s +epoch=88 global_step=34700 loss=3.51859 loss_avg=3.51438 acc=0.67188 acc_top1_avg=0.67145 acc_top5_avg=0.89250 lr=0.00010 gn=27.68728 time=51.44it/s +epoch=88 global_step=34750 loss=3.49349 loss_avg=3.51622 acc=0.67188 acc_top1_avg=0.67096 acc_top5_avg=0.89264 lr=0.00010 gn=34.56206 time=58.24it/s +====================Eval==================== +epoch=88 global_step=34799 loss=2.40196 test_loss_avg=0.93895 acc=0.36719 test_acc_avg=0.72656 test_acc_top5_avg=0.96452 time=157.00it/s +epoch=88 global_step=34799 loss=0.18877 test_loss_avg=0.84720 acc=0.93750 test_acc_avg=0.76048 test_acc_top5_avg=0.96984 time=501.83it/s +curr_acc 0.7605 +BEST_ACC 0.8158 +curr_acc_top5 0.9698 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=3.16075 loss_avg=3.16075 acc=0.70312 acc_top1_avg=0.70312 acc_top5_avg=0.92969 lr=0.00010 gn=33.18639 time=31.53it/s +epoch=89 global_step=34850 loss=3.36595 loss_avg=3.45544 acc=0.70312 acc_top1_avg=0.67770 acc_top5_avg=0.89629 lr=0.00010 gn=36.04727 time=48.88it/s +epoch=89 global_step=34900 loss=4.10369 loss_avg=3.50281 acc=0.59375 acc_top1_avg=0.67126 acc_top5_avg=0.89078 lr=0.00010 gn=28.57819 time=60.38it/s +epoch=89 global_step=34950 loss=3.63266 loss_avg=3.51752 acc=0.63281 acc_top1_avg=0.66924 acc_top5_avg=0.89166 lr=0.00010 gn=29.81847 time=51.53it/s +epoch=89 global_step=35000 loss=3.30104 loss_avg=3.50789 acc=0.71094 acc_top1_avg=0.67067 acc_top5_avg=0.89257 lr=0.00010 gn=42.77079 time=52.27it/s +epoch=89 global_step=35050 loss=3.51615 loss_avg=3.50648 acc=0.67188 acc_top1_avg=0.67088 acc_top5_avg=0.89349 lr=0.00010 gn=30.66390 time=60.48it/s +epoch=89 global_step=35100 loss=3.85378 loss_avg=3.50624 acc=0.63281 acc_top1_avg=0.67091 acc_top5_avg=0.89361 lr=0.00010 gn=31.96412 time=53.74it/s +epoch=89 global_step=35150 loss=3.74844 loss_avg=3.51264 acc=0.64062 acc_top1_avg=0.67023 acc_top5_avg=0.89345 lr=0.00010 gn=32.01430 time=54.91it/s +====================Eval==================== +epoch=89 global_step=35190 loss=1.06587 test_loss_avg=0.84873 acc=0.68750 test_acc_avg=0.75123 test_acc_top5_avg=0.97122 time=237.79it/s +epoch=89 global_step=35190 loss=0.17373 test_loss_avg=0.93437 acc=0.96094 test_acc_avg=0.73585 test_acc_top5_avg=0.96739 time=249.93it/s +epoch=89 global_step=35190 loss=0.21832 test_loss_avg=0.83477 acc=0.93750 test_acc_avg=0.76335 test_acc_top5_avg=0.97132 time=499.68it/s +curr_acc 0.7634 +BEST_ACC 0.8158 +curr_acc_top5 0.9713 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=3.03381 loss_avg=3.56340 acc=0.74219 acc_top1_avg=0.67031 acc_top5_avg=0.88750 lr=0.00010 gn=38.07122 time=51.54it/s +epoch=90 global_step=35250 loss=2.91930 loss_avg=3.46031 acc=0.74219 acc_top1_avg=0.67852 acc_top5_avg=0.89323 lr=0.00010 gn=31.32530 time=51.15it/s +epoch=90 global_step=35300 loss=3.92695 loss_avg=3.47048 acc=0.61719 acc_top1_avg=0.67734 acc_top5_avg=0.89339 lr=0.00010 gn=30.99426 time=45.26it/s +epoch=90 global_step=35350 loss=3.56132 loss_avg=3.49552 acc=0.64844 acc_top1_avg=0.67422 acc_top5_avg=0.89351 lr=0.00010 gn=28.23068 time=51.71it/s +epoch=90 global_step=35400 loss=4.24750 loss_avg=3.49745 acc=0.58594 acc_top1_avg=0.67333 acc_top5_avg=0.89182 lr=0.00010 gn=29.53965 time=61.10it/s +epoch=90 global_step=35450 loss=3.20693 loss_avg=3.50543 acc=0.71094 acc_top1_avg=0.67251 acc_top5_avg=0.89273 lr=0.00010 gn=33.20080 time=60.25it/s +epoch=90 global_step=35500 loss=3.56258 loss_avg=3.50534 acc=0.66406 acc_top1_avg=0.67273 acc_top5_avg=0.89335 lr=0.00010 gn=34.21594 time=55.72it/s +epoch=90 global_step=35550 loss=3.42036 loss_avg=3.50334 acc=0.67969 acc_top1_avg=0.67305 acc_top5_avg=0.89275 lr=0.00010 gn=30.75795 time=52.07it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.89590 test_loss_avg=0.92838 acc=0.75000 test_acc_avg=0.73027 test_acc_top5_avg=0.96465 time=230.17it/s +epoch=90 global_step=35581 loss=0.19908 test_loss_avg=0.85301 acc=0.93750 test_acc_avg=0.76068 test_acc_top5_avg=0.96875 time=832.04it/s +curr_acc 0.7607 +BEST_ACC 0.8158 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=3.42330 loss_avg=3.51573 acc=0.67188 acc_top1_avg=0.66817 acc_top5_avg=0.89967 lr=0.00010 gn=28.23381 time=59.40it/s +epoch=91 global_step=35650 loss=3.56588 loss_avg=3.47370 acc=0.67969 acc_top1_avg=0.67323 acc_top5_avg=0.89368 lr=0.00010 gn=35.13748 time=52.73it/s +epoch=91 global_step=35700 loss=3.09581 loss_avg=3.48427 acc=0.72656 acc_top1_avg=0.67371 acc_top5_avg=0.89305 lr=0.00010 gn=33.65653 time=52.55it/s +epoch=91 global_step=35750 loss=3.73575 loss_avg=3.48438 acc=0.64062 acc_top1_avg=0.67359 acc_top5_avg=0.89243 lr=0.00010 gn=29.62382 time=51.45it/s +epoch=91 global_step=35800 loss=3.07696 loss_avg=3.46753 acc=0.71875 acc_top1_avg=0.67544 acc_top5_avg=0.89248 lr=0.00010 gn=31.26288 time=55.87it/s +epoch=91 global_step=35850 loss=3.06313 loss_avg=3.46314 acc=0.71875 acc_top1_avg=0.67609 acc_top5_avg=0.89498 lr=0.00010 gn=40.73857 time=54.59it/s +epoch=91 global_step=35900 loss=2.83684 loss_avg=3.47286 acc=0.73438 acc_top1_avg=0.67503 acc_top5_avg=0.89464 lr=0.00010 gn=29.12695 time=52.62it/s +epoch=91 global_step=35950 loss=3.28555 loss_avg=3.49233 acc=0.69531 acc_top1_avg=0.67295 acc_top5_avg=0.89401 lr=0.00010 gn=29.75031 time=57.15it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.20015 test_loss_avg=0.98485 acc=0.93750 test_acc_avg=0.70312 test_acc_top5_avg=0.96875 time=179.31it/s +epoch=91 global_step=35972 loss=0.27087 test_loss_avg=1.01770 acc=0.89844 test_acc_avg=0.70389 test_acc_top5_avg=0.96337 time=190.01it/s +epoch=91 global_step=35972 loss=0.28073 test_loss_avg=0.82624 acc=0.93750 test_acc_avg=0.75949 test_acc_top5_avg=0.97102 time=509.08it/s +curr_acc 0.7595 +BEST_ACC 0.8158 +curr_acc_top5 0.9710 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=2.97588 loss_avg=3.45375 acc=0.73438 acc_top1_avg=0.67746 acc_top5_avg=0.89927 lr=0.00010 gn=28.66805 time=59.70it/s +epoch=92 global_step=36050 loss=3.37621 loss_avg=3.50260 acc=0.68750 acc_top1_avg=0.67338 acc_top5_avg=0.89583 lr=0.00010 gn=26.20402 time=53.26it/s +epoch=92 global_step=36100 loss=3.64304 loss_avg=3.47025 acc=0.66406 acc_top1_avg=0.67651 acc_top5_avg=0.89624 lr=0.00010 gn=33.28598 time=60.08it/s +epoch=92 global_step=36150 loss=3.93562 loss_avg=3.49189 acc=0.64062 acc_top1_avg=0.67385 acc_top5_avg=0.89554 lr=0.00010 gn=37.79680 time=49.16it/s +epoch=92 global_step=36200 loss=3.99161 loss_avg=3.48784 acc=0.61719 acc_top1_avg=0.67438 acc_top5_avg=0.89587 lr=0.00010 gn=28.67514 time=25.54it/s +epoch=92 global_step=36250 loss=3.40647 loss_avg=3.49072 acc=0.69531 acc_top1_avg=0.67426 acc_top5_avg=0.89411 lr=0.00010 gn=37.79554 time=51.77it/s +epoch=92 global_step=36300 loss=3.21087 loss_avg=3.49530 acc=0.72656 acc_top1_avg=0.67369 acc_top5_avg=0.89336 lr=0.00010 gn=34.40757 time=56.44it/s +epoch=92 global_step=36350 loss=3.93142 loss_avg=3.49395 acc=0.63281 acc_top1_avg=0.67371 acc_top5_avg=0.89325 lr=0.00010 gn=33.70195 time=59.27it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.59962 test_loss_avg=1.02367 acc=0.80469 test_acc_avg=0.69434 test_acc_top5_avg=0.96069 time=200.10it/s +epoch=92 global_step=36363 loss=0.25974 test_loss_avg=0.84389 acc=0.93750 test_acc_avg=0.75949 test_acc_top5_avg=0.96865 time=858.61it/s +curr_acc 0.7595 +BEST_ACC 0.8158 +curr_acc_top5 0.9687 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=3.44219 loss_avg=3.38474 acc=0.67969 acc_top1_avg=0.68307 acc_top5_avg=0.90034 lr=0.00010 gn=30.65686 time=50.96it/s +epoch=93 global_step=36450 loss=3.49857 loss_avg=3.41274 acc=0.67969 acc_top1_avg=0.68274 acc_top5_avg=0.89511 lr=0.00010 gn=31.88268 time=52.92it/s +epoch=93 global_step=36500 loss=3.62676 loss_avg=3.44119 acc=0.66406 acc_top1_avg=0.67940 acc_top5_avg=0.89370 lr=0.00010 gn=36.38036 time=63.34it/s +epoch=93 global_step=36550 loss=4.20877 loss_avg=3.45542 acc=0.58594 acc_top1_avg=0.67752 acc_top5_avg=0.89313 lr=0.00010 gn=28.76429 time=48.97it/s +epoch=93 global_step=36600 loss=3.41616 loss_avg=3.44902 acc=0.67188 acc_top1_avg=0.67754 acc_top5_avg=0.89395 lr=0.00010 gn=29.82313 time=52.07it/s +epoch=93 global_step=36650 loss=3.22525 loss_avg=3.46626 acc=0.69531 acc_top1_avg=0.67577 acc_top5_avg=0.89318 lr=0.00010 gn=40.89604 time=59.78it/s +epoch=93 global_step=36700 loss=3.87092 loss_avg=3.47641 acc=0.63281 acc_top1_avg=0.67477 acc_top5_avg=0.89287 lr=0.00010 gn=32.14919 time=56.22it/s +epoch=93 global_step=36750 loss=3.86366 loss_avg=3.47865 acc=0.63281 acc_top1_avg=0.67442 acc_top5_avg=0.89283 lr=0.00010 gn=30.71865 time=60.70it/s +====================Eval==================== +epoch=93 global_step=36754 loss=1.53272 test_loss_avg=1.55249 acc=0.56250 test_acc_avg=0.54948 test_acc_top5_avg=0.94271 time=232.75it/s +epoch=93 global_step=36754 loss=2.21865 test_loss_avg=1.08897 acc=0.40625 test_acc_avg=0.68824 test_acc_top5_avg=0.95755 time=237.57it/s +epoch=93 global_step=36754 loss=0.26384 test_loss_avg=0.83871 acc=0.93750 test_acc_avg=0.75890 test_acc_top5_avg=0.96895 time=508.96it/s +curr_acc 0.7589 +BEST_ACC 0.8158 +curr_acc_top5 0.9689 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=3.54568 loss_avg=3.47422 acc=0.67188 acc_top1_avg=0.67221 acc_top5_avg=0.89317 lr=0.00010 gn=28.92259 time=53.22it/s +epoch=94 global_step=36850 loss=3.99986 loss_avg=3.48460 acc=0.63281 acc_top1_avg=0.67367 acc_top5_avg=0.89225 lr=0.00010 gn=34.70485 time=46.85it/s +epoch=94 global_step=36900 loss=3.47792 loss_avg=3.50728 acc=0.67188 acc_top1_avg=0.67080 acc_top5_avg=0.89089 lr=0.00010 gn=34.58052 time=39.96it/s +epoch=94 global_step=36950 loss=3.55507 loss_avg=3.48878 acc=0.68750 acc_top1_avg=0.67319 acc_top5_avg=0.89078 lr=0.00010 gn=35.34734 time=54.01it/s +epoch=94 global_step=37000 loss=3.81043 loss_avg=3.50147 acc=0.63281 acc_top1_avg=0.67210 acc_top5_avg=0.88958 lr=0.00010 gn=39.95297 time=54.20it/s +epoch=94 global_step=37050 loss=4.24361 loss_avg=3.49300 acc=0.57031 acc_top1_avg=0.67261 acc_top5_avg=0.89015 lr=0.00010 gn=30.98137 time=58.01it/s +epoch=94 global_step=37100 loss=3.60725 loss_avg=3.48968 acc=0.65625 acc_top1_avg=0.67314 acc_top5_avg=0.89126 lr=0.00010 gn=35.68173 time=59.48it/s +====================Eval==================== +epoch=94 global_step=37145 loss=1.50799 test_loss_avg=0.98663 acc=0.53906 test_acc_avg=0.71061 test_acc_top5_avg=0.96257 time=99.30it/s +epoch=94 global_step=37145 loss=0.15152 test_loss_avg=0.92668 acc=0.95312 test_acc_avg=0.73902 test_acc_top5_avg=0.96569 time=229.02it/s +epoch=94 global_step=37145 loss=0.27842 test_loss_avg=0.87833 acc=0.93750 test_acc_avg=0.75218 test_acc_top5_avg=0.96786 time=518.71it/s +curr_acc 0.7522 +BEST_ACC 0.8158 +curr_acc_top5 0.9679 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=3.20363 loss_avg=3.45241 acc=0.68750 acc_top1_avg=0.67031 acc_top5_avg=0.89375 lr=0.00010 gn=25.36218 time=54.03it/s +epoch=95 global_step=37200 loss=3.54591 loss_avg=3.51782 acc=0.68750 acc_top1_avg=0.67031 acc_top5_avg=0.89034 lr=0.00010 gn=35.15741 time=51.78it/s +epoch=95 global_step=37250 loss=2.93799 loss_avg=3.46296 acc=0.74219 acc_top1_avg=0.67500 acc_top5_avg=0.89516 lr=0.00010 gn=35.08507 time=55.49it/s +epoch=95 global_step=37300 loss=4.39358 loss_avg=3.48725 acc=0.58594 acc_top1_avg=0.67248 acc_top5_avg=0.89340 lr=0.00010 gn=34.50633 time=54.67it/s +epoch=95 global_step=37350 loss=3.98624 loss_avg=3.49357 acc=0.61719 acc_top1_avg=0.67241 acc_top5_avg=0.89428 lr=0.00010 gn=35.86482 time=51.75it/s +epoch=95 global_step=37400 loss=3.58025 loss_avg=3.48903 acc=0.66406 acc_top1_avg=0.67328 acc_top5_avg=0.89396 lr=0.00010 gn=35.58378 time=61.89it/s +epoch=95 global_step=37450 loss=3.33834 loss_avg=3.48390 acc=0.69531 acc_top1_avg=0.67413 acc_top5_avg=0.89457 lr=0.00010 gn=29.54812 time=61.77it/s +epoch=95 global_step=37500 loss=3.00578 loss_avg=3.47817 acc=0.72656 acc_top1_avg=0.67526 acc_top5_avg=0.89395 lr=0.00010 gn=26.30739 time=57.75it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.82167 test_loss_avg=0.89090 acc=0.78125 test_acc_avg=0.73438 test_acc_top5_avg=0.96667 time=235.12it/s +epoch=95 global_step=37536 loss=0.25771 test_loss_avg=0.82236 acc=0.93750 test_acc_avg=0.76315 test_acc_top5_avg=0.96964 time=536.01it/s +curr_acc 0.7632 +BEST_ACC 0.8158 +curr_acc_top5 0.9696 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=3.22858 loss_avg=3.40058 acc=0.71094 acc_top1_avg=0.68359 acc_top5_avg=0.90848 lr=0.00010 gn=33.72376 time=61.28it/s +epoch=96 global_step=37600 loss=3.82193 loss_avg=3.50584 acc=0.62500 acc_top1_avg=0.66956 acc_top5_avg=0.89551 lr=0.00010 gn=26.03924 time=55.72it/s +epoch=96 global_step=37650 loss=3.18249 loss_avg=3.50951 acc=0.71094 acc_top1_avg=0.67030 acc_top5_avg=0.89220 lr=0.00010 gn=37.47835 time=57.16it/s +epoch=96 global_step=37700 loss=3.60645 loss_avg=3.50747 acc=0.67188 acc_top1_avg=0.67068 acc_top5_avg=0.89205 lr=0.00010 gn=38.25257 time=59.64it/s +epoch=96 global_step=37750 loss=2.91882 loss_avg=3.49302 acc=0.73438 acc_top1_avg=0.67268 acc_top5_avg=0.89296 lr=0.00010 gn=35.13262 time=55.13it/s +epoch=96 global_step=37800 loss=2.95632 loss_avg=3.47547 acc=0.75000 acc_top1_avg=0.67480 acc_top5_avg=0.89264 lr=0.00010 gn=36.40831 time=53.15it/s +epoch=96 global_step=37850 loss=3.72641 loss_avg=3.48233 acc=0.64844 acc_top1_avg=0.67404 acc_top5_avg=0.89252 lr=0.00010 gn=34.07688 time=52.61it/s +epoch=96 global_step=37900 loss=3.69185 loss_avg=3.47043 acc=0.65625 acc_top1_avg=0.67520 acc_top5_avg=0.89264 lr=0.00010 gn=34.74231 time=55.17it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.66002 test_loss_avg=0.82523 acc=0.79688 test_acc_avg=0.74902 test_acc_top5_avg=0.97705 time=231.97it/s +epoch=96 global_step=37927 loss=0.13321 test_loss_avg=1.00409 acc=0.94531 test_acc_avg=0.71153 test_acc_top5_avg=0.96248 time=240.61it/s +epoch=96 global_step=37927 loss=0.27221 test_loss_avg=0.86537 acc=0.93750 test_acc_avg=0.75049 test_acc_top5_avg=0.96835 time=525.73it/s +curr_acc 0.7505 +BEST_ACC 0.8158 +curr_acc_top5 0.9684 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=4.08265 loss_avg=3.48190 acc=0.62500 acc_top1_avg=0.67188 acc_top5_avg=0.89504 lr=0.00010 gn=33.30924 time=53.27it/s +epoch=97 global_step=38000 loss=3.73800 loss_avg=3.49467 acc=0.64844 acc_top1_avg=0.67198 acc_top5_avg=0.89062 lr=0.00010 gn=30.95231 time=60.21it/s +epoch=97 global_step=38050 loss=3.48235 loss_avg=3.49477 acc=0.67969 acc_top1_avg=0.67359 acc_top5_avg=0.88694 lr=0.00010 gn=32.97328 time=59.51it/s +epoch=97 global_step=38100 loss=3.43108 loss_avg=3.49453 acc=0.67969 acc_top1_avg=0.67359 acc_top5_avg=0.89076 lr=0.00010 gn=29.35919 time=61.14it/s +epoch=97 global_step=38150 loss=3.43379 loss_avg=3.48632 acc=0.67969 acc_top1_avg=0.67454 acc_top5_avg=0.89073 lr=0.00010 gn=38.97560 time=50.13it/s +epoch=97 global_step=38200 loss=3.89619 loss_avg=3.48248 acc=0.61719 acc_top1_avg=0.67491 acc_top5_avg=0.89180 lr=0.00010 gn=26.49850 time=56.66it/s +epoch=97 global_step=38250 loss=4.12375 loss_avg=3.46891 acc=0.60156 acc_top1_avg=0.67647 acc_top5_avg=0.89258 lr=0.00010 gn=32.73828 time=54.74it/s +epoch=97 global_step=38300 loss=3.49391 loss_avg=3.47030 acc=0.66406 acc_top1_avg=0.67592 acc_top5_avg=0.89268 lr=0.00010 gn=26.85562 time=56.58it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.43524 test_loss_avg=0.95897 acc=0.88281 test_acc_avg=0.71917 test_acc_top5_avg=0.96347 time=237.93it/s +epoch=97 global_step=38318 loss=0.28367 test_loss_avg=0.86511 acc=0.93750 test_acc_avg=0.75603 test_acc_top5_avg=0.96885 time=504.55it/s +curr_acc 0.7560 +BEST_ACC 0.8158 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=3.83955 loss_avg=3.44513 acc=0.63281 acc_top1_avg=0.67969 acc_top5_avg=0.88477 lr=0.00010 gn=29.50188 time=54.79it/s +epoch=98 global_step=38400 loss=3.19891 loss_avg=3.45739 acc=0.71094 acc_top1_avg=0.67902 acc_top5_avg=0.89529 lr=0.00010 gn=26.39767 time=54.81it/s +epoch=98 global_step=38450 loss=3.45460 loss_avg=3.43718 acc=0.67188 acc_top1_avg=0.68016 acc_top5_avg=0.89441 lr=0.00010 gn=27.44142 time=55.58it/s +epoch=98 global_step=38500 loss=3.36415 loss_avg=3.44038 acc=0.67969 acc_top1_avg=0.67960 acc_top5_avg=0.89526 lr=0.00010 gn=31.66630 time=56.04it/s +epoch=98 global_step=38550 loss=3.48718 loss_avg=3.43619 acc=0.67188 acc_top1_avg=0.67996 acc_top5_avg=0.89403 lr=0.00010 gn=34.10712 time=58.04it/s +epoch=98 global_step=38600 loss=3.77454 loss_avg=3.45303 acc=0.64062 acc_top1_avg=0.67833 acc_top5_avg=0.89304 lr=0.00010 gn=32.01922 time=53.58it/s +epoch=98 global_step=38650 loss=3.19231 loss_avg=3.45322 acc=0.71875 acc_top1_avg=0.67846 acc_top5_avg=0.89345 lr=0.00010 gn=29.74353 time=61.54it/s +epoch=98 global_step=38700 loss=3.77757 loss_avg=3.45451 acc=0.63281 acc_top1_avg=0.67836 acc_top5_avg=0.89375 lr=0.00010 gn=29.31259 time=54.54it/s +====================Eval==================== +epoch=98 global_step=38709 loss=1.15089 test_loss_avg=1.45121 acc=0.64062 test_acc_avg=0.57715 test_acc_top5_avg=0.95020 time=239.89it/s +epoch=98 global_step=38709 loss=0.37013 test_loss_avg=1.10324 acc=0.87500 test_acc_avg=0.68750 test_acc_top5_avg=0.95797 time=84.67it/s +epoch=98 global_step=38709 loss=0.28231 test_loss_avg=0.86276 acc=0.93750 test_acc_avg=0.75475 test_acc_top5_avg=0.96826 time=810.96it/s +curr_acc 0.7547 +BEST_ACC 0.8158 +curr_acc_top5 0.9683 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=3.06645 loss_avg=3.35502 acc=0.73438 acc_top1_avg=0.68941 acc_top5_avg=0.90777 lr=0.00010 gn=35.85428 time=52.23it/s +epoch=99 global_step=38800 loss=3.59278 loss_avg=3.35582 acc=0.64062 acc_top1_avg=0.68853 acc_top5_avg=0.90050 lr=0.00010 gn=26.10203 time=55.08it/s +epoch=99 global_step=38850 loss=3.07830 loss_avg=3.40057 acc=0.72656 acc_top1_avg=0.68362 acc_top5_avg=0.89833 lr=0.00010 gn=35.77259 time=55.08it/s +epoch=99 global_step=38900 loss=3.26177 loss_avg=3.41542 acc=0.71094 acc_top1_avg=0.68267 acc_top5_avg=0.89570 lr=0.00010 gn=32.31228 time=52.77it/s +epoch=99 global_step=38950 loss=3.46766 loss_avg=3.42132 acc=0.67969 acc_top1_avg=0.68189 acc_top5_avg=0.89442 lr=0.00010 gn=26.94794 time=48.48it/s +epoch=99 global_step=39000 loss=3.36334 loss_avg=3.43619 acc=0.70312 acc_top1_avg=0.68012 acc_top5_avg=0.89463 lr=0.00010 gn=34.68427 time=55.92it/s +epoch=99 global_step=39050 loss=3.35275 loss_avg=3.43744 acc=0.71094 acc_top1_avg=0.67999 acc_top5_avg=0.89436 lr=0.00010 gn=33.81160 time=54.72it/s +epoch=99 global_step=39100 loss=2.88619 loss_avg=3.44559 acc=0.73750 acc_top1_avg=0.67862 acc_top5_avg=0.89449 lr=0.00010 gn=39.94355 time=71.53it/s +====================Eval==================== +epoch=99 global_step=39100 loss=1.52794 test_loss_avg=1.02864 acc=0.60156 test_acc_avg=0.70178 test_acc_top5_avg=0.95690 time=234.38it/s +epoch=99 global_step=39100 loss=0.30617 test_loss_avg=0.87747 acc=0.93750 test_acc_avg=0.75475 test_acc_top5_avg=0.96657 time=525.14it/s +epoch=99 global_step=39100 loss=0.30617 test_loss_avg=0.87747 acc=0.93750 test_acc_avg=0.75475 test_acc_top5_avg=0.96657 time=525.14it/s +curr_acc 0.7547 +BEST_ACC 0.8158 +curr_acc_top5 0.9666 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=3.58456 loss_avg=3.43504 acc=0.67188 acc_top1_avg=0.67891 acc_top5_avg=0.88953 lr=0.00010 gn=36.48743 time=54.02it/s +epoch=100 global_step=39200 loss=3.36110 loss_avg=3.42149 acc=0.67969 acc_top1_avg=0.68047 acc_top5_avg=0.89078 lr=0.00010 gn=23.15418 time=28.77it/s +epoch=100 global_step=39250 loss=3.21626 loss_avg=3.39914 acc=0.69531 acc_top1_avg=0.68401 acc_top5_avg=0.89276 lr=0.00010 gn=27.58720 time=50.64it/s +epoch=100 global_step=39300 loss=3.52372 loss_avg=3.41534 acc=0.67188 acc_top1_avg=0.68242 acc_top5_avg=0.89137 lr=0.00010 gn=34.07586 time=54.89it/s +epoch=100 global_step=39350 loss=3.21298 loss_avg=3.41445 acc=0.70312 acc_top1_avg=0.68191 acc_top5_avg=0.89269 lr=0.00010 gn=29.24469 time=52.05it/s +epoch=100 global_step=39400 loss=3.21509 loss_avg=3.42222 acc=0.70312 acc_top1_avg=0.68094 acc_top5_avg=0.89326 lr=0.00010 gn=28.80342 time=49.53it/s +epoch=100 global_step=39450 loss=3.15885 loss_avg=3.42687 acc=0.69531 acc_top1_avg=0.68063 acc_top5_avg=0.89388 lr=0.00010 gn=29.11258 time=55.83it/s +====================Eval==================== +epoch=100 global_step=39491 loss=2.17171 test_loss_avg=1.03453 acc=0.43750 test_acc_avg=0.70094 test_acc_top5_avg=0.96016 time=232.11it/s +epoch=100 global_step=39491 loss=0.25746 test_loss_avg=0.84803 acc=0.93750 test_acc_avg=0.75811 test_acc_top5_avg=0.96895 time=807.37it/s +curr_acc 0.7581 +BEST_ACC 0.8158 +curr_acc_top5 0.9689 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=3.78485 loss_avg=3.49958 acc=0.62500 acc_top1_avg=0.67622 acc_top5_avg=0.88194 lr=0.00010 gn=30.28345 time=54.44it/s +epoch=101 global_step=39550 loss=3.03772 loss_avg=3.44683 acc=0.73438 acc_top1_avg=0.67876 acc_top5_avg=0.88904 lr=0.00010 gn=40.08885 time=50.79it/s +epoch=101 global_step=39600 loss=3.16530 loss_avg=3.47164 acc=0.71875 acc_top1_avg=0.67610 acc_top5_avg=0.88955 lr=0.00010 gn=33.50255 time=53.98it/s +epoch=101 global_step=39650 loss=3.16883 loss_avg=3.48121 acc=0.71875 acc_top1_avg=0.67409 acc_top5_avg=0.89215 lr=0.00010 gn=33.45255 time=53.06it/s +epoch=101 global_step=39700 loss=2.97788 loss_avg=3.46487 acc=0.74219 acc_top1_avg=0.67595 acc_top5_avg=0.89253 lr=0.00010 gn=38.33131 time=58.55it/s +epoch=101 global_step=39750 loss=3.23639 loss_avg=3.44028 acc=0.71094 acc_top1_avg=0.67881 acc_top5_avg=0.89440 lr=0.00010 gn=40.88409 time=52.09it/s +epoch=101 global_step=39800 loss=3.78813 loss_avg=3.43827 acc=0.64062 acc_top1_avg=0.67893 acc_top5_avg=0.89341 lr=0.00010 gn=31.65396 time=54.54it/s +epoch=101 global_step=39850 loss=3.70117 loss_avg=3.43481 acc=0.64844 acc_top1_avg=0.67936 acc_top5_avg=0.89313 lr=0.00010 gn=29.67400 time=53.78it/s +====================Eval==================== +epoch=101 global_step=39882 loss=1.21413 test_loss_avg=0.95678 acc=0.69531 test_acc_avg=0.72842 test_acc_top5_avg=0.96726 time=236.51it/s +epoch=101 global_step=39882 loss=0.15769 test_loss_avg=0.94661 acc=0.95312 test_acc_avg=0.73261 test_acc_top5_avg=0.96556 time=241.47it/s +epoch=101 global_step=39882 loss=0.27094 test_loss_avg=0.86684 acc=0.93750 test_acc_avg=0.75475 test_acc_top5_avg=0.96885 time=502.13it/s +curr_acc 0.7547 +BEST_ACC 0.8158 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=3.27382 loss_avg=3.58053 acc=0.69531 acc_top1_avg=0.66493 acc_top5_avg=0.88845 lr=0.00010 gn=31.67904 time=60.14it/s +epoch=102 global_step=39950 loss=2.98890 loss_avg=3.49124 acc=0.72656 acc_top1_avg=0.67222 acc_top5_avg=0.89258 lr=0.00010 gn=37.69285 time=60.99it/s +epoch=102 global_step=40000 loss=3.25945 loss_avg=3.45377 acc=0.71094 acc_top1_avg=0.67711 acc_top5_avg=0.89625 lr=0.00010 gn=31.19851 time=58.22it/s +epoch=102 global_step=40050 loss=3.18126 loss_avg=3.44771 acc=0.73438 acc_top1_avg=0.67885 acc_top5_avg=0.89486 lr=0.00010 gn=38.04058 time=54.37it/s +epoch=102 global_step=40100 loss=3.09458 loss_avg=3.44501 acc=0.74219 acc_top1_avg=0.67915 acc_top5_avg=0.89432 lr=0.00010 gn=39.45365 time=62.70it/s +epoch=102 global_step=40150 loss=3.39854 loss_avg=3.43164 acc=0.70312 acc_top1_avg=0.68050 acc_top5_avg=0.89441 lr=0.00010 gn=38.87750 time=51.39it/s +epoch=102 global_step=40200 loss=3.91696 loss_avg=3.43782 acc=0.64844 acc_top1_avg=0.67971 acc_top5_avg=0.89362 lr=0.00010 gn=36.16067 time=55.07it/s +epoch=102 global_step=40250 loss=3.46863 loss_avg=3.43294 acc=0.67188 acc_top1_avg=0.68037 acc_top5_avg=0.89345 lr=0.00010 gn=36.67651 time=55.60it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.92342 test_loss_avg=0.94721 acc=0.71875 test_acc_avg=0.72321 test_acc_top5_avg=0.96391 time=235.74it/s +epoch=102 global_step=40273 loss=0.27964 test_loss_avg=0.86274 acc=0.93750 test_acc_avg=0.75465 test_acc_top5_avg=0.96905 time=802.28it/s +curr_acc 0.7546 +BEST_ACC 0.8158 +curr_acc_top5 0.9690 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=3.40414 loss_avg=3.33334 acc=0.68750 acc_top1_avg=0.69155 acc_top5_avg=0.90307 lr=0.00010 gn=37.42940 time=56.16it/s +epoch=103 global_step=40350 loss=3.75107 loss_avg=3.39756 acc=0.64062 acc_top1_avg=0.68618 acc_top5_avg=0.89864 lr=0.00010 gn=24.30278 time=58.75it/s +epoch=103 global_step=40400 loss=3.47347 loss_avg=3.41631 acc=0.65625 acc_top1_avg=0.68246 acc_top5_avg=0.89573 lr=0.00010 gn=38.21126 time=62.14it/s +epoch=103 global_step=40450 loss=3.77086 loss_avg=3.40269 acc=0.64062 acc_top1_avg=0.68384 acc_top5_avg=0.89610 lr=0.00010 gn=32.96658 time=54.59it/s +epoch=103 global_step=40500 loss=3.94492 loss_avg=3.41860 acc=0.63281 acc_top1_avg=0.68251 acc_top5_avg=0.89496 lr=0.00010 gn=35.16969 time=51.09it/s +epoch=103 global_step=40550 loss=3.60462 loss_avg=3.41339 acc=0.66406 acc_top1_avg=0.68299 acc_top5_avg=0.89477 lr=0.00010 gn=35.11638 time=59.70it/s +epoch=103 global_step=40600 loss=3.52652 loss_avg=3.41916 acc=0.69531 acc_top1_avg=0.68222 acc_top5_avg=0.89450 lr=0.00010 gn=40.51948 time=56.37it/s +epoch=103 global_step=40650 loss=3.20633 loss_avg=3.42504 acc=0.71094 acc_top1_avg=0.68074 acc_top5_avg=0.89382 lr=0.00010 gn=35.29593 time=54.97it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.17978 test_loss_avg=0.95182 acc=0.96094 test_acc_avg=0.72115 test_acc_top5_avg=0.96875 time=223.98it/s +epoch=103 global_step=40664 loss=0.14678 test_loss_avg=1.02454 acc=0.94531 test_acc_avg=0.70610 test_acc_top5_avg=0.95995 time=72.59it/s +epoch=103 global_step=40664 loss=0.22674 test_loss_avg=0.84688 acc=0.93750 test_acc_avg=0.75633 test_acc_top5_avg=0.96766 time=868.39it/s +curr_acc 0.7563 +BEST_ACC 0.8158 +curr_acc_top5 0.9677 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=3.74473 loss_avg=3.32149 acc=0.65625 acc_top1_avg=0.69314 acc_top5_avg=0.89084 lr=0.00010 gn=38.27524 time=55.24it/s +epoch=104 global_step=40750 loss=3.42347 loss_avg=3.41262 acc=0.70312 acc_top1_avg=0.68405 acc_top5_avg=0.89326 lr=0.00010 gn=41.03331 time=54.68it/s +epoch=104 global_step=40800 loss=2.95584 loss_avg=3.43830 acc=0.73438 acc_top1_avg=0.68020 acc_top5_avg=0.89350 lr=0.00010 gn=36.04527 time=61.34it/s +epoch=104 global_step=40850 loss=3.35726 loss_avg=3.43353 acc=0.67188 acc_top1_avg=0.68162 acc_top5_avg=0.89352 lr=0.00010 gn=27.62284 time=54.81it/s +epoch=104 global_step=40900 loss=3.88229 loss_avg=3.44255 acc=0.63281 acc_top1_avg=0.67965 acc_top5_avg=0.89294 lr=0.00010 gn=30.78235 time=59.18it/s +epoch=104 global_step=40950 loss=3.68528 loss_avg=3.44131 acc=0.64062 acc_top1_avg=0.67985 acc_top5_avg=0.89390 lr=0.00010 gn=31.21651 time=63.55it/s +epoch=104 global_step=41000 loss=2.76715 loss_avg=3.42816 acc=0.75000 acc_top1_avg=0.68132 acc_top5_avg=0.89409 lr=0.00010 gn=36.17231 time=54.09it/s +epoch=104 global_step=41050 loss=2.95296 loss_avg=3.42736 acc=0.71875 acc_top1_avg=0.68141 acc_top5_avg=0.89398 lr=0.00010 gn=31.72437 time=62.38it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.40975 test_loss_avg=1.07195 acc=0.86719 test_acc_avg=0.68612 test_acc_top5_avg=0.95542 time=238.95it/s +epoch=104 global_step=41055 loss=0.22115 test_loss_avg=0.89612 acc=0.93750 test_acc_avg=0.74753 test_acc_top5_avg=0.96509 time=543.94it/s +curr_acc 0.7475 +BEST_ACC 0.8158 +curr_acc_top5 0.9651 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=3.24351 loss_avg=3.39993 acc=0.68750 acc_top1_avg=0.68194 acc_top5_avg=0.89826 lr=0.00010 gn=29.90171 time=53.49it/s +epoch=105 global_step=41150 loss=3.68939 loss_avg=3.42321 acc=0.64844 acc_top1_avg=0.68117 acc_top5_avg=0.89465 lr=0.00010 gn=32.65159 time=58.81it/s +epoch=105 global_step=41200 loss=2.77913 loss_avg=3.44558 acc=0.76562 acc_top1_avg=0.67834 acc_top5_avg=0.89380 lr=0.00010 gn=35.68883 time=47.80it/s +epoch=105 global_step=41250 loss=3.43662 loss_avg=3.41747 acc=0.66406 acc_top1_avg=0.68189 acc_top5_avg=0.89399 lr=0.00010 gn=38.59665 time=52.27it/s +epoch=105 global_step=41300 loss=3.46527 loss_avg=3.43128 acc=0.67188 acc_top1_avg=0.68026 acc_top5_avg=0.89346 lr=0.00010 gn=32.72103 time=56.00it/s +epoch=105 global_step=41350 loss=2.96929 loss_avg=3.41648 acc=0.73438 acc_top1_avg=0.68210 acc_top5_avg=0.89417 lr=0.00010 gn=28.82220 time=51.13it/s +epoch=105 global_step=41400 loss=3.28351 loss_avg=3.41017 acc=0.69531 acc_top1_avg=0.68265 acc_top5_avg=0.89509 lr=0.00010 gn=29.32049 time=61.81it/s +====================Eval==================== +epoch=105 global_step=41446 loss=1.33748 test_loss_avg=1.52248 acc=0.58594 test_acc_avg=0.56719 test_acc_top5_avg=0.94375 time=218.84it/s +epoch=105 global_step=41446 loss=1.57112 test_loss_avg=1.16943 acc=0.55469 test_acc_avg=0.66974 test_acc_top5_avg=0.95440 time=236.41it/s +epoch=105 global_step=41446 loss=0.27170 test_loss_avg=0.87812 acc=0.93750 test_acc_avg=0.75059 test_acc_top5_avg=0.96677 time=813.16it/s +curr_acc 0.7506 +BEST_ACC 0.8158 +curr_acc_top5 0.9668 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=3.62848 loss_avg=3.35807 acc=0.64844 acc_top1_avg=0.68555 acc_top5_avg=0.88867 lr=0.00010 gn=28.43372 time=56.94it/s +epoch=106 global_step=41500 loss=2.76803 loss_avg=3.31266 acc=0.75000 acc_top1_avg=0.68909 acc_top5_avg=0.90234 lr=0.00010 gn=32.23114 time=58.69it/s +epoch=106 global_step=41550 loss=3.70426 loss_avg=3.39903 acc=0.64844 acc_top1_avg=0.68337 acc_top5_avg=0.89686 lr=0.00010 gn=34.83131 time=51.34it/s +epoch=106 global_step=41600 loss=3.26128 loss_avg=3.42711 acc=0.70312 acc_top1_avg=0.68091 acc_top5_avg=0.89473 lr=0.00010 gn=34.48569 time=54.29it/s +epoch=106 global_step=41650 loss=3.40794 loss_avg=3.39103 acc=0.67969 acc_top1_avg=0.68478 acc_top5_avg=0.89495 lr=0.00010 gn=37.62097 time=55.64it/s +epoch=106 global_step=41700 loss=3.81961 loss_avg=3.40933 acc=0.64062 acc_top1_avg=0.68286 acc_top5_avg=0.89475 lr=0.00010 gn=33.22407 time=56.04it/s +epoch=106 global_step=41750 loss=3.07009 loss_avg=3.41070 acc=0.72656 acc_top1_avg=0.68298 acc_top5_avg=0.89489 lr=0.00010 gn=35.25350 time=56.45it/s +epoch=106 global_step=41800 loss=3.97337 loss_avg=3.41980 acc=0.64062 acc_top1_avg=0.68192 acc_top5_avg=0.89455 lr=0.00010 gn=38.74584 time=57.73it/s +====================Eval==================== +epoch=106 global_step=41837 loss=1.44974 test_loss_avg=1.06269 acc=0.60938 test_acc_avg=0.69201 test_acc_top5_avg=0.95733 time=107.93it/s +epoch=106 global_step=41837 loss=0.09396 test_loss_avg=0.90657 acc=0.96875 test_acc_avg=0.74198 test_acc_top5_avg=0.96536 time=242.94it/s +epoch=106 global_step=41837 loss=0.29932 test_loss_avg=0.87896 acc=0.93750 test_acc_avg=0.75000 test_acc_top5_avg=0.96667 time=509.08it/s +curr_acc 0.7500 +BEST_ACC 0.8158 +curr_acc_top5 0.9667 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=3.48627 loss_avg=3.41075 acc=0.67969 acc_top1_avg=0.68269 acc_top5_avg=0.89123 lr=0.00010 gn=32.14013 time=52.76it/s +epoch=107 global_step=41900 loss=3.52550 loss_avg=3.46352 acc=0.66406 acc_top1_avg=0.67721 acc_top5_avg=0.89348 lr=0.00010 gn=36.62292 time=51.26it/s +epoch=107 global_step=41950 loss=3.79303 loss_avg=3.42367 acc=0.63281 acc_top1_avg=0.68059 acc_top5_avg=0.89360 lr=0.00010 gn=39.74342 time=53.30it/s +epoch=107 global_step=42000 loss=3.05015 loss_avg=3.40414 acc=0.71875 acc_top1_avg=0.68285 acc_top5_avg=0.89422 lr=0.00010 gn=33.39549 time=32.41it/s +epoch=107 global_step=42050 loss=3.39488 loss_avg=3.41087 acc=0.67969 acc_top1_avg=0.68192 acc_top5_avg=0.89393 lr=0.00010 gn=26.47046 time=55.79it/s +epoch=107 global_step=42100 loss=3.95605 loss_avg=3.40415 acc=0.63281 acc_top1_avg=0.68287 acc_top5_avg=0.89374 lr=0.00010 gn=40.94079 time=57.38it/s +epoch=107 global_step=42150 loss=3.76044 loss_avg=3.40508 acc=0.64844 acc_top1_avg=0.68268 acc_top5_avg=0.89429 lr=0.00010 gn=36.28823 time=59.96it/s +epoch=107 global_step=42200 loss=3.45318 loss_avg=3.41338 acc=0.67969 acc_top1_avg=0.68197 acc_top5_avg=0.89510 lr=0.00010 gn=36.17817 time=62.42it/s +====================Eval==================== +epoch=107 global_step=42228 loss=1.23056 test_loss_avg=0.93742 acc=0.65625 test_acc_avg=0.72756 test_acc_top5_avg=0.96376 time=228.39it/s +epoch=107 global_step=42228 loss=0.29966 test_loss_avg=0.84798 acc=0.93750 test_acc_avg=0.75939 test_acc_top5_avg=0.96766 time=509.57it/s +curr_acc 0.7594 +BEST_ACC 0.8158 +curr_acc_top5 0.9677 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=3.61759 loss_avg=3.35599 acc=0.62500 acc_top1_avg=0.68501 acc_top5_avg=0.89666 lr=0.00010 gn=31.19284 time=54.06it/s +epoch=108 global_step=42300 loss=3.51034 loss_avg=3.40196 acc=0.67969 acc_top1_avg=0.68218 acc_top5_avg=0.89214 lr=0.00010 gn=41.95811 time=53.83it/s +epoch=108 global_step=42350 loss=3.54476 loss_avg=3.43573 acc=0.67188 acc_top1_avg=0.67982 acc_top5_avg=0.89107 lr=0.00010 gn=39.45041 time=61.27it/s +epoch=108 global_step=42400 loss=3.55533 loss_avg=3.41240 acc=0.66406 acc_top1_avg=0.68237 acc_top5_avg=0.89267 lr=0.00010 gn=33.29853 time=53.45it/s +epoch=108 global_step=42450 loss=3.63218 loss_avg=3.39548 acc=0.68750 acc_top1_avg=0.68409 acc_top5_avg=0.89330 lr=0.00010 gn=45.52861 time=60.29it/s +epoch=108 global_step=42500 loss=3.81693 loss_avg=3.40110 acc=0.63281 acc_top1_avg=0.68385 acc_top5_avg=0.89347 lr=0.00010 gn=32.86089 time=53.91it/s +epoch=108 global_step=42550 loss=3.58980 loss_avg=3.40303 acc=0.67188 acc_top1_avg=0.68386 acc_top5_avg=0.89422 lr=0.00010 gn=38.09372 time=56.48it/s +epoch=108 global_step=42600 loss=3.19104 loss_avg=3.40820 acc=0.72656 acc_top1_avg=0.68324 acc_top5_avg=0.89432 lr=0.00010 gn=35.17844 time=55.37it/s +====================Eval==================== +epoch=108 global_step=42619 loss=1.40850 test_loss_avg=0.93527 acc=0.67188 test_acc_avg=0.73524 test_acc_top5_avg=0.96571 time=237.37it/s +epoch=108 global_step=42619 loss=0.21855 test_loss_avg=1.02761 acc=0.92188 test_acc_avg=0.71220 test_acc_top5_avg=0.95910 time=241.18it/s +epoch=108 global_step=42619 loss=0.30817 test_loss_avg=0.90628 acc=0.93750 test_acc_avg=0.74575 test_acc_top5_avg=0.96440 time=824.19it/s +curr_acc 0.7457 +BEST_ACC 0.8158 +curr_acc_top5 0.9644 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=3.30115 loss_avg=3.24054 acc=0.69531 acc_top1_avg=0.70312 acc_top5_avg=0.89037 lr=0.00010 gn=33.39055 time=58.34it/s +epoch=109 global_step=42700 loss=3.21843 loss_avg=3.31645 acc=0.71094 acc_top1_avg=0.69348 acc_top5_avg=0.89477 lr=0.00010 gn=40.99522 time=54.09it/s +epoch=109 global_step=42750 loss=3.96599 loss_avg=3.37826 acc=0.61719 acc_top1_avg=0.68595 acc_top5_avg=0.89307 lr=0.00010 gn=28.72063 time=63.02it/s +epoch=109 global_step=42800 loss=3.67336 loss_avg=3.37321 acc=0.66406 acc_top1_avg=0.68651 acc_top5_avg=0.89416 lr=0.00010 gn=48.97482 time=53.72it/s +epoch=109 global_step=42850 loss=3.23558 loss_avg=3.37927 acc=0.68750 acc_top1_avg=0.68615 acc_top5_avg=0.89462 lr=0.00010 gn=35.98170 time=57.21it/s +epoch=109 global_step=42900 loss=3.37881 loss_avg=3.39775 acc=0.67969 acc_top1_avg=0.68411 acc_top5_avg=0.89379 lr=0.00010 gn=29.30838 time=54.73it/s +epoch=109 global_step=42950 loss=3.10173 loss_avg=3.40507 acc=0.71094 acc_top1_avg=0.68325 acc_top5_avg=0.89379 lr=0.00010 gn=32.06700 time=61.59it/s +epoch=109 global_step=43000 loss=3.86248 loss_avg=3.39960 acc=0.60938 acc_top1_avg=0.68399 acc_top5_avg=0.89446 lr=0.00010 gn=37.25901 time=56.79it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.41816 test_loss_avg=1.02201 acc=0.87500 test_acc_avg=0.70393 test_acc_top5_avg=0.95793 time=108.26it/s +epoch=109 global_step=43010 loss=0.32795 test_loss_avg=0.92952 acc=0.93750 test_acc_avg=0.74120 test_acc_top5_avg=0.96311 time=829.08it/s +curr_acc 0.7412 +BEST_ACC 0.8158 +curr_acc_top5 0.9631 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=3.24025 loss_avg=3.38242 acc=0.70312 acc_top1_avg=0.68535 acc_top5_avg=0.89785 lr=0.00010 gn=34.82535 time=61.41it/s +epoch=110 global_step=43100 loss=3.16677 loss_avg=3.37738 acc=0.70312 acc_top1_avg=0.68550 acc_top5_avg=0.90000 lr=0.00010 gn=33.94033 time=57.08it/s +epoch=110 global_step=43150 loss=2.75907 loss_avg=3.38650 acc=0.73438 acc_top1_avg=0.68365 acc_top5_avg=0.89715 lr=0.00010 gn=24.96999 time=61.10it/s +epoch=110 global_step=43200 loss=3.44699 loss_avg=3.40872 acc=0.65625 acc_top1_avg=0.68100 acc_top5_avg=0.89322 lr=0.00010 gn=36.00816 time=62.90it/s +epoch=110 global_step=43250 loss=2.78030 loss_avg=3.39201 acc=0.74219 acc_top1_avg=0.68265 acc_top5_avg=0.89437 lr=0.00010 gn=32.18208 time=54.78it/s +epoch=110 global_step=43300 loss=3.49502 loss_avg=3.37609 acc=0.67188 acc_top1_avg=0.68486 acc_top5_avg=0.89483 lr=0.00010 gn=34.24839 time=59.75it/s +epoch=110 global_step=43350 loss=3.31864 loss_avg=3.36730 acc=0.68750 acc_top1_avg=0.68594 acc_top5_avg=0.89515 lr=0.00010 gn=31.36071 time=60.09it/s +epoch=110 global_step=43400 loss=3.40481 loss_avg=3.38458 acc=0.70312 acc_top1_avg=0.68417 acc_top5_avg=0.89433 lr=0.00010 gn=42.58303 time=54.01it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.27341 test_loss_avg=1.14149 acc=0.94531 test_acc_avg=0.67031 test_acc_top5_avg=0.96016 time=221.63it/s +epoch=110 global_step=43401 loss=0.23555 test_loss_avg=1.13365 acc=0.91406 test_acc_avg=0.68542 test_acc_top5_avg=0.95221 time=232.85it/s +epoch=110 global_step=43401 loss=0.26981 test_loss_avg=0.90323 acc=0.93750 test_acc_avg=0.74822 test_acc_top5_avg=0.96282 time=621.65it/s +curr_acc 0.7482 +BEST_ACC 0.8158 +curr_acc_top5 0.9628 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=3.47423 loss_avg=3.40229 acc=0.66406 acc_top1_avg=0.68176 acc_top5_avg=0.88967 lr=0.00010 gn=36.77017 time=47.24it/s +epoch=111 global_step=43500 loss=3.13735 loss_avg=3.39574 acc=0.70312 acc_top1_avg=0.68395 acc_top5_avg=0.89110 lr=0.00010 gn=33.21007 time=56.32it/s +epoch=111 global_step=43550 loss=3.58124 loss_avg=3.37977 acc=0.65625 acc_top1_avg=0.68656 acc_top5_avg=0.89461 lr=0.00010 gn=30.34059 time=55.23it/s +epoch=111 global_step=43600 loss=3.90154 loss_avg=3.39392 acc=0.64062 acc_top1_avg=0.68436 acc_top5_avg=0.89380 lr=0.00010 gn=35.33820 time=59.20it/s +epoch=111 global_step=43650 loss=3.06619 loss_avg=3.39036 acc=0.71094 acc_top1_avg=0.68464 acc_top5_avg=0.89404 lr=0.00010 gn=35.10938 time=55.67it/s +epoch=111 global_step=43700 loss=3.53615 loss_avg=3.39239 acc=0.67969 acc_top1_avg=0.68444 acc_top5_avg=0.89363 lr=0.00010 gn=38.62335 time=55.39it/s +epoch=111 global_step=43750 loss=3.46654 loss_avg=3.39920 acc=0.67969 acc_top1_avg=0.68414 acc_top5_avg=0.89430 lr=0.00010 gn=42.41876 time=57.53it/s +====================Eval==================== +epoch=111 global_step=43792 loss=1.46081 test_loss_avg=1.11844 acc=0.53125 test_acc_avg=0.67061 test_acc_top5_avg=0.95413 time=189.85it/s +epoch=111 global_step=43792 loss=0.25614 test_loss_avg=0.90147 acc=0.93750 test_acc_avg=0.74397 test_acc_top5_avg=0.96272 time=502.79it/s +curr_acc 0.7440 +BEST_ACC 0.8158 +curr_acc_top5 0.9627 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=4.42033 loss_avg=3.31469 acc=0.57031 acc_top1_avg=0.69629 acc_top5_avg=0.89648 lr=0.00010 gn=34.37690 time=52.30it/s +epoch=112 global_step=43850 loss=2.83279 loss_avg=3.34977 acc=0.74219 acc_top1_avg=0.68979 acc_top5_avg=0.89655 lr=0.00010 gn=36.23059 time=52.93it/s +epoch=112 global_step=43900 loss=3.10177 loss_avg=3.39912 acc=0.72656 acc_top1_avg=0.68374 acc_top5_avg=0.89431 lr=0.00010 gn=42.26910 time=44.68it/s +epoch=112 global_step=43950 loss=3.00727 loss_avg=3.38239 acc=0.72656 acc_top1_avg=0.68587 acc_top5_avg=0.89527 lr=0.00010 gn=33.15747 time=48.81it/s +epoch=112 global_step=44000 loss=3.04638 loss_avg=3.36882 acc=0.71875 acc_top1_avg=0.68716 acc_top5_avg=0.89543 lr=0.00010 gn=31.60059 time=53.93it/s +epoch=112 global_step=44050 loss=3.04546 loss_avg=3.38090 acc=0.72656 acc_top1_avg=0.68596 acc_top5_avg=0.89496 lr=0.00010 gn=32.96434 time=57.42it/s +epoch=112 global_step=44100 loss=3.55798 loss_avg=3.39335 acc=0.65625 acc_top1_avg=0.68453 acc_top5_avg=0.89446 lr=0.00010 gn=32.05769 time=57.15it/s +epoch=112 global_step=44150 loss=3.29616 loss_avg=3.40547 acc=0.69531 acc_top1_avg=0.68333 acc_top5_avg=0.89394 lr=0.00010 gn=30.30961 time=51.13it/s +====================Eval==================== +epoch=112 global_step=44183 loss=1.56657 test_loss_avg=1.56647 acc=0.57812 test_acc_avg=0.55859 test_acc_top5_avg=0.94141 time=228.86it/s +epoch=112 global_step=44183 loss=2.34051 test_loss_avg=1.09344 acc=0.39844 test_acc_avg=0.68975 test_acc_top5_avg=0.95853 time=230.08it/s +epoch=112 global_step=44183 loss=0.26634 test_loss_avg=0.86435 acc=0.93750 test_acc_avg=0.75425 test_acc_top5_avg=0.96796 time=495.08it/s +curr_acc 0.7543 +BEST_ACC 0.8158 +curr_acc_top5 0.9680 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=3.57974 loss_avg=3.36558 acc=0.65625 acc_top1_avg=0.68336 acc_top5_avg=0.90119 lr=0.00010 gn=33.24527 time=55.17it/s +epoch=113 global_step=44250 loss=3.63917 loss_avg=3.36007 acc=0.64844 acc_top1_avg=0.68552 acc_top5_avg=0.89797 lr=0.00010 gn=39.76353 time=58.14it/s +epoch=113 global_step=44300 loss=3.35578 loss_avg=3.38807 acc=0.68750 acc_top1_avg=0.68403 acc_top5_avg=0.89597 lr=0.00010 gn=37.59352 time=54.59it/s +epoch=113 global_step=44350 loss=3.58122 loss_avg=3.37695 acc=0.67188 acc_top1_avg=0.68493 acc_top5_avg=0.89512 lr=0.00010 gn=41.84859 time=59.21it/s +epoch=113 global_step=44400 loss=3.28817 loss_avg=3.37945 acc=0.68750 acc_top1_avg=0.68548 acc_top5_avg=0.89538 lr=0.00010 gn=40.70837 time=60.76it/s +epoch=113 global_step=44450 loss=3.59395 loss_avg=3.38239 acc=0.65625 acc_top1_avg=0.68539 acc_top5_avg=0.89537 lr=0.00010 gn=36.72344 time=54.79it/s +epoch=113 global_step=44500 loss=3.19372 loss_avg=3.38827 acc=0.71094 acc_top1_avg=0.68484 acc_top5_avg=0.89597 lr=0.00010 gn=37.15536 time=55.29it/s +epoch=113 global_step=44550 loss=3.96008 loss_avg=3.37624 acc=0.61719 acc_top1_avg=0.68610 acc_top5_avg=0.89633 lr=0.00010 gn=40.20033 time=55.53it/s +====================Eval==================== +epoch=113 global_step=44574 loss=1.32781 test_loss_avg=1.00754 acc=0.57812 test_acc_avg=0.71501 test_acc_top5_avg=0.95992 time=156.83it/s +epoch=113 global_step=44574 loss=0.15690 test_loss_avg=1.00679 acc=0.95312 test_acc_avg=0.72474 test_acc_top5_avg=0.95880 time=249.51it/s +epoch=113 global_step=44574 loss=0.29776 test_loss_avg=0.94218 acc=0.93750 test_acc_avg=0.74199 test_acc_top5_avg=0.96183 time=507.85it/s +curr_acc 0.7420 +BEST_ACC 0.8158 +curr_acc_top5 0.9618 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=3.84076 loss_avg=3.37896 acc=0.63281 acc_top1_avg=0.68660 acc_top5_avg=0.89093 lr=0.00010 gn=36.10681 time=53.38it/s +epoch=114 global_step=44650 loss=2.88380 loss_avg=3.35771 acc=0.73438 acc_top1_avg=0.68822 acc_top5_avg=0.89597 lr=0.00010 gn=31.96796 time=61.80it/s +epoch=114 global_step=44700 loss=3.45775 loss_avg=3.34199 acc=0.68750 acc_top1_avg=0.69054 acc_top5_avg=0.89695 lr=0.00010 gn=40.87165 time=61.68it/s +epoch=114 global_step=44750 loss=3.28449 loss_avg=3.35677 acc=0.71875 acc_top1_avg=0.68857 acc_top5_avg=0.89560 lr=0.00010 gn=44.14771 time=59.18it/s +epoch=114 global_step=44800 loss=3.13344 loss_avg=3.35115 acc=0.71875 acc_top1_avg=0.68878 acc_top5_avg=0.89636 lr=0.00010 gn=33.88591 time=58.12it/s +epoch=114 global_step=44850 loss=3.43233 loss_avg=3.35143 acc=0.67969 acc_top1_avg=0.68886 acc_top5_avg=0.89688 lr=0.00010 gn=36.79473 time=51.83it/s +epoch=114 global_step=44900 loss=3.60184 loss_avg=3.36776 acc=0.67969 acc_top1_avg=0.68755 acc_top5_avg=0.89566 lr=0.00010 gn=42.42748 time=57.17it/s +epoch=114 global_step=44950 loss=2.89240 loss_avg=3.37484 acc=0.72656 acc_top1_avg=0.68671 acc_top5_avg=0.89505 lr=0.00010 gn=28.62020 time=61.17it/s +====================Eval==================== +epoch=114 global_step=44965 loss=1.03229 test_loss_avg=1.00338 acc=0.70312 test_acc_avg=0.70810 test_acc_top5_avg=0.96076 time=236.86it/s +epoch=114 global_step=44965 loss=0.32147 test_loss_avg=0.92255 acc=0.93750 test_acc_avg=0.74258 test_acc_top5_avg=0.96460 time=533.83it/s +curr_acc 0.7426 +BEST_ACC 0.8158 +curr_acc_top5 0.9646 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=3.28087 loss_avg=3.35941 acc=0.69531 acc_top1_avg=0.68862 acc_top5_avg=0.89196 lr=0.00010 gn=32.00821 time=53.46it/s +epoch=115 global_step=45050 loss=3.08953 loss_avg=3.36980 acc=0.71094 acc_top1_avg=0.68722 acc_top5_avg=0.88925 lr=0.00010 gn=28.09174 time=57.80it/s +epoch=115 global_step=45100 loss=2.78790 loss_avg=3.36689 acc=0.75000 acc_top1_avg=0.68785 acc_top5_avg=0.88953 lr=0.00010 gn=34.84405 time=56.51it/s +epoch=115 global_step=45150 loss=3.22150 loss_avg=3.37116 acc=0.70312 acc_top1_avg=0.68809 acc_top5_avg=0.89029 lr=0.00010 gn=41.71046 time=58.03it/s +epoch=115 global_step=45200 loss=3.43928 loss_avg=3.35962 acc=0.68750 acc_top1_avg=0.68916 acc_top5_avg=0.89072 lr=0.00010 gn=42.80691 time=54.20it/s +epoch=115 global_step=45250 loss=3.57190 loss_avg=3.37042 acc=0.66406 acc_top1_avg=0.68772 acc_top5_avg=0.89073 lr=0.00010 gn=42.96472 time=52.61it/s +epoch=115 global_step=45300 loss=3.26340 loss_avg=3.36808 acc=0.69531 acc_top1_avg=0.68787 acc_top5_avg=0.89205 lr=0.00010 gn=41.35172 time=55.72it/s +epoch=115 global_step=45350 loss=4.04055 loss_avg=3.37003 acc=0.59375 acc_top1_avg=0.68742 acc_top5_avg=0.89341 lr=0.00010 gn=39.10474 time=62.90it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.28109 test_loss_avg=0.94528 acc=0.93750 test_acc_avg=0.72552 test_acc_top5_avg=0.96979 time=239.65it/s +epoch=115 global_step=45356 loss=0.13589 test_loss_avg=1.07991 acc=0.95312 test_acc_avg=0.69543 test_acc_top5_avg=0.95433 time=233.08it/s +epoch=115 global_step=45356 loss=0.31729 test_loss_avg=0.91840 acc=0.93750 test_acc_avg=0.74061 test_acc_top5_avg=0.96212 time=503.34it/s +curr_acc 0.7406 +BEST_ACC 0.8158 +curr_acc_top5 0.9621 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=3.52775 loss_avg=3.43878 acc=0.67969 acc_top1_avg=0.67898 acc_top5_avg=0.88991 lr=0.00010 gn=32.76911 time=59.38it/s +epoch=116 global_step=45450 loss=2.67312 loss_avg=3.40393 acc=0.75000 acc_top1_avg=0.68409 acc_top5_avg=0.88938 lr=0.00010 gn=31.77647 time=58.65it/s +epoch=116 global_step=45500 loss=3.99980 loss_avg=3.40922 acc=0.62500 acc_top1_avg=0.68370 acc_top5_avg=0.89133 lr=0.00010 gn=35.09521 time=61.49it/s +epoch=116 global_step=45550 loss=2.83841 loss_avg=3.38293 acc=0.73438 acc_top1_avg=0.68585 acc_top5_avg=0.89256 lr=0.00010 gn=36.98792 time=55.31it/s +epoch=116 global_step=45600 loss=3.06733 loss_avg=3.38513 acc=0.70312 acc_top1_avg=0.68590 acc_top5_avg=0.89296 lr=0.00010 gn=29.00927 time=52.02it/s +epoch=116 global_step=45650 loss=3.52867 loss_avg=3.37905 acc=0.67969 acc_top1_avg=0.68678 acc_top5_avg=0.89395 lr=0.00010 gn=35.78875 time=59.75it/s +epoch=116 global_step=45700 loss=3.56795 loss_avg=3.37445 acc=0.65625 acc_top1_avg=0.68734 acc_top5_avg=0.89390 lr=0.00010 gn=28.98980 time=55.30it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.41291 test_loss_avg=1.08814 acc=0.88281 test_acc_avg=0.68576 test_acc_top5_avg=0.95530 time=237.37it/s +epoch=116 global_step=45747 loss=0.29728 test_loss_avg=0.90975 acc=0.93750 test_acc_avg=0.74041 test_acc_top5_avg=0.96499 time=862.67it/s +curr_acc 0.7404 +BEST_ACC 0.8158 +curr_acc_top5 0.9650 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=3.25154 loss_avg=3.44689 acc=0.71094 acc_top1_avg=0.68490 acc_top5_avg=0.89323 lr=0.00010 gn=36.18856 time=55.61it/s +epoch=117 global_step=45800 loss=3.55956 loss_avg=3.31971 acc=0.65625 acc_top1_avg=0.69015 acc_top5_avg=0.90006 lr=0.00010 gn=38.76900 time=52.68it/s +epoch=117 global_step=45850 loss=3.97524 loss_avg=3.35983 acc=0.61719 acc_top1_avg=0.68750 acc_top5_avg=0.89639 lr=0.00010 gn=31.12458 time=50.18it/s +epoch=117 global_step=45900 loss=2.64638 loss_avg=3.34474 acc=0.76562 acc_top1_avg=0.68959 acc_top5_avg=0.89721 lr=0.00010 gn=37.00576 time=59.24it/s +epoch=117 global_step=45950 loss=3.56544 loss_avg=3.34969 acc=0.67188 acc_top1_avg=0.68915 acc_top5_avg=0.89532 lr=0.00010 gn=34.83466 time=53.21it/s +epoch=117 global_step=46000 loss=3.14673 loss_avg=3.34675 acc=0.72656 acc_top1_avg=0.68960 acc_top5_avg=0.89581 lr=0.00010 gn=37.63649 time=62.95it/s +epoch=117 global_step=46050 loss=3.14812 loss_avg=3.35712 acc=0.69531 acc_top1_avg=0.68827 acc_top5_avg=0.89560 lr=0.00010 gn=30.55299 time=60.46it/s +epoch=117 global_step=46100 loss=3.43301 loss_avg=3.35900 acc=0.67969 acc_top1_avg=0.68852 acc_top5_avg=0.89525 lr=0.00010 gn=31.50240 time=54.20it/s +====================Eval==================== +epoch=117 global_step=46138 loss=1.68344 test_loss_avg=1.53671 acc=0.52344 test_acc_avg=0.55580 test_acc_top5_avg=0.95536 time=246.75it/s +epoch=117 global_step=46138 loss=0.38636 test_loss_avg=1.17395 acc=0.89844 test_acc_avg=0.66927 test_acc_top5_avg=0.95271 time=245.27it/s +epoch=117 global_step=46138 loss=0.33323 test_loss_avg=0.90632 acc=0.93750 test_acc_avg=0.74367 test_acc_top5_avg=0.96479 time=885.06it/s +curr_acc 0.7437 +BEST_ACC 0.8158 +curr_acc_top5 0.9648 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=3.62581 loss_avg=3.36486 acc=0.64844 acc_top1_avg=0.68555 acc_top5_avg=0.88542 lr=0.00010 gn=38.08988 time=54.50it/s +epoch=118 global_step=46200 loss=2.82197 loss_avg=3.30993 acc=0.75781 acc_top1_avg=0.69292 acc_top5_avg=0.89100 lr=0.00010 gn=36.32675 time=52.79it/s +epoch=118 global_step=46250 loss=3.34132 loss_avg=3.33159 acc=0.68750 acc_top1_avg=0.69120 acc_top5_avg=0.89153 lr=0.00010 gn=32.97162 time=58.77it/s +epoch=118 global_step=46300 loss=2.70498 loss_avg=3.37861 acc=0.75781 acc_top1_avg=0.68625 acc_top5_avg=0.89149 lr=0.00010 gn=37.64325 time=53.79it/s +epoch=118 global_step=46350 loss=3.87816 loss_avg=3.35956 acc=0.64062 acc_top1_avg=0.68816 acc_top5_avg=0.89269 lr=0.00010 gn=37.41483 time=54.87it/s +epoch=118 global_step=46400 loss=3.54533 loss_avg=3.36397 acc=0.66406 acc_top1_avg=0.68777 acc_top5_avg=0.89346 lr=0.00010 gn=34.48038 time=63.92it/s +epoch=118 global_step=46450 loss=3.33974 loss_avg=3.36704 acc=0.67969 acc_top1_avg=0.68717 acc_top5_avg=0.89323 lr=0.00010 gn=36.22674 time=63.81it/s +epoch=118 global_step=46500 loss=2.59899 loss_avg=3.36718 acc=0.75000 acc_top1_avg=0.68698 acc_top5_avg=0.89401 lr=0.00010 gn=36.93132 time=64.23it/s +====================Eval==================== +epoch=118 global_step=46529 loss=1.75364 test_loss_avg=1.10433 acc=0.48438 test_acc_avg=0.68136 test_acc_top5_avg=0.95396 time=237.13it/s +epoch=118 global_step=46529 loss=0.11195 test_loss_avg=0.91641 acc=0.95312 test_acc_avg=0.74018 test_acc_top5_avg=0.96314 time=257.79it/s +epoch=118 global_step=46529 loss=0.30020 test_loss_avg=0.90861 acc=0.93750 test_acc_avg=0.74268 test_acc_top5_avg=0.96361 time=885.43it/s +curr_acc 0.7427 +BEST_ACC 0.8158 +curr_acc_top5 0.9636 +BEST_ACC_top5 0.9833 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=2.98073 loss_avg=3.47517 acc=0.71875 acc_top1_avg=0.67783 acc_top5_avg=0.88728 lr=0.00010 gn=35.05257 time=55.77it/s +epoch=119 global_step=46600 loss=2.82023 loss_avg=3.36501 acc=0.75000 acc_top1_avg=0.68717 acc_top5_avg=0.89029 lr=0.00010 gn=34.48162 time=60.16it/s +epoch=119 global_step=46650 loss=3.34822 loss_avg=3.36679 acc=0.69531 acc_top1_avg=0.68705 acc_top5_avg=0.89360 lr=0.00010 gn=34.29317 time=55.67it/s +epoch=119 global_step=46700 loss=3.05445 loss_avg=3.36055 acc=0.71875 acc_top1_avg=0.68841 acc_top5_avg=0.89501 lr=0.00010 gn=29.47948 time=61.40it/s +epoch=119 global_step=46750 loss=3.22047 loss_avg=3.38033 acc=0.69531 acc_top1_avg=0.68640 acc_top5_avg=0.89381 lr=0.00010 gn=35.71973 time=47.35it/s +epoch=119 global_step=46800 loss=3.68101 loss_avg=3.38391 acc=0.65625 acc_top1_avg=0.68606 acc_top5_avg=0.89432 lr=0.00010 gn=42.25788 time=62.90it/s +epoch=119 global_step=46850 loss=3.28921 loss_avg=3.38437 acc=0.68750 acc_top1_avg=0.68638 acc_top5_avg=0.89452 lr=0.00010 gn=38.95018 time=60.82it/s +epoch=119 global_step=46900 loss=3.22912 loss_avg=3.36910 acc=0.70312 acc_top1_avg=0.68817 acc_top5_avg=0.89505 lr=0.00010 gn=37.13454 time=61.86it/s +====================Eval==================== +epoch=119 global_step=46920 loss=2.27789 test_loss_avg=1.07249 acc=0.40625 test_acc_avg=0.68766 test_acc_top5_avg=0.95504 time=245.45it/s +epoch=119 global_step=46920 loss=0.31881 test_loss_avg=0.90792 acc=0.93750 test_acc_avg=0.74100 test_acc_top5_avg=0.96301 time=897.37it/s +curr_acc 0.7410 +BEST_ACC 0.8158 +curr_acc_top5 0.9630 +BEST_ACC_top5 0.9833 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_2_6.log b/other_methods/sceloss/sceloss_results/out_2_6.log new file mode 100644 index 0000000..7339970 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_2_6.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.2__noise_amount__0.6.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.81884 loss_avg=7.92294 acc=0.18750 acc_top1_avg=0.18797 acc_top5_avg=0.61813 lr=0.01000 gn=6.07946 time=55.17it/s +epoch=0 global_step=100 loss=7.46832 loss_avg=7.75544 acc=0.24219 acc_top1_avg=0.20789 acc_top5_avg=0.63945 lr=0.01000 gn=3.63682 time=65.48it/s +epoch=0 global_step=150 loss=7.32510 loss_avg=7.67589 acc=0.22656 acc_top1_avg=0.21568 acc_top5_avg=0.65063 lr=0.01000 gn=2.75923 time=66.12it/s +epoch=0 global_step=200 loss=7.61753 loss_avg=7.62272 acc=0.21875 acc_top1_avg=0.22227 acc_top5_avg=0.65570 lr=0.01000 gn=2.91001 time=65.54it/s +epoch=0 global_step=250 loss=7.71541 loss_avg=7.57813 acc=0.21094 acc_top1_avg=0.22681 acc_top5_avg=0.66063 lr=0.01000 gn=2.46730 time=64.25it/s +epoch=0 global_step=300 loss=7.40554 loss_avg=7.53841 acc=0.23438 acc_top1_avg=0.23091 acc_top5_avg=0.66401 lr=0.01000 gn=2.48363 time=64.80it/s +epoch=0 global_step=350 loss=8.02885 loss_avg=7.51791 acc=0.18750 acc_top1_avg=0.23321 acc_top5_avg=0.66705 lr=0.01000 gn=3.94142 time=53.67it/s +====================Eval==================== +epoch=0 global_step=391 loss=5.20480 test_loss_avg=4.35885 acc=0.00000 test_acc_avg=0.02906 test_acc_top5_avg=0.65563 time=235.95it/s +epoch=0 global_step=391 loss=4.92310 test_loss_avg=3.71620 acc=0.00000 test_acc_avg=0.18562 test_acc_top5_avg=0.65032 time=31.18it/s +curr_acc 0.1856 +BEST_ACC 0.0000 +curr_acc_top5 0.6503 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=7.46921 loss_avg=7.29482 acc=0.23438 acc_top1_avg=0.25347 acc_top5_avg=0.68576 lr=0.01000 gn=2.59509 time=63.32it/s +epoch=1 global_step=450 loss=7.95455 loss_avg=7.34704 acc=0.18750 acc_top1_avg=0.25358 acc_top5_avg=0.67624 lr=0.01000 gn=2.70117 time=61.80it/s +epoch=1 global_step=500 loss=7.81208 loss_avg=7.35493 acc=0.17969 acc_top1_avg=0.25244 acc_top5_avg=0.68270 lr=0.01000 gn=3.38598 time=64.06it/s +epoch=1 global_step=550 loss=7.26631 loss_avg=7.31049 acc=0.25000 acc_top1_avg=0.25634 acc_top5_avg=0.68902 lr=0.01000 gn=3.75952 time=65.58it/s +epoch=1 global_step=600 loss=7.20291 loss_avg=7.30085 acc=0.26562 acc_top1_avg=0.25759 acc_top5_avg=0.69243 lr=0.01000 gn=2.88635 time=57.62it/s +epoch=1 global_step=650 loss=7.02625 loss_avg=7.29028 acc=0.28906 acc_top1_avg=0.25884 acc_top5_avg=0.69202 lr=0.01000 gn=2.60079 time=59.38it/s +epoch=1 global_step=700 loss=7.20215 loss_avg=7.28889 acc=0.26562 acc_top1_avg=0.25900 acc_top5_avg=0.69377 lr=0.01000 gn=2.65442 time=59.09it/s +epoch=1 global_step=750 loss=7.65417 loss_avg=7.28311 acc=0.21094 acc_top1_avg=0.25964 acc_top5_avg=0.69420 lr=0.01000 gn=2.30655 time=62.41it/s +====================Eval==================== +epoch=1 global_step=782 loss=4.33738 test_loss_avg=4.45424 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.82403 time=171.50it/s +epoch=1 global_step=782 loss=3.89572 test_loss_avg=3.64968 acc=0.25000 test_acc_avg=0.21380 test_acc_top5_avg=0.73404 time=238.19it/s +epoch=1 global_step=782 loss=5.38710 test_loss_avg=3.80323 acc=0.00000 test_acc_avg=0.19215 test_acc_top5_avg=0.67682 time=458.39it/s +curr_acc 0.1921 +BEST_ACC 0.1856 +curr_acc_top5 0.6768 +BEST_ACC_top5 0.6503 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=7.32250 loss_avg=7.21261 acc=0.26562 acc_top1_avg=0.26042 acc_top5_avg=0.72526 lr=0.01000 gn=2.40691 time=62.09it/s +epoch=2 global_step=850 loss=7.06470 loss_avg=7.14439 acc=0.26562 acc_top1_avg=0.27114 acc_top5_avg=0.71473 lr=0.01000 gn=4.17553 time=53.00it/s +epoch=2 global_step=900 loss=8.01555 loss_avg=7.15225 acc=0.17969 acc_top1_avg=0.27119 acc_top5_avg=0.71299 lr=0.01000 gn=2.16316 time=55.01it/s +epoch=2 global_step=950 loss=7.34981 loss_avg=7.16364 acc=0.25000 acc_top1_avg=0.27028 acc_top5_avg=0.71108 lr=0.01000 gn=2.97044 time=60.63it/s +epoch=2 global_step=1000 loss=6.45294 loss_avg=7.15995 acc=0.34375 acc_top1_avg=0.27036 acc_top5_avg=0.71051 lr=0.01000 gn=3.20631 time=60.13it/s +epoch=2 global_step=1050 loss=7.51286 loss_avg=7.16639 acc=0.21094 acc_top1_avg=0.27026 acc_top5_avg=0.71056 lr=0.01000 gn=3.45854 time=56.35it/s +epoch=2 global_step=1100 loss=6.72000 loss_avg=7.16680 acc=0.31250 acc_top1_avg=0.27044 acc_top5_avg=0.71111 lr=0.01000 gn=2.39583 time=54.92it/s +epoch=2 global_step=1150 loss=7.60987 loss_avg=7.16486 acc=0.21875 acc_top1_avg=0.27068 acc_top5_avg=0.71172 lr=0.01000 gn=3.75903 time=59.31it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.86841 test_loss_avg=4.22320 acc=0.47656 test_acc_avg=0.03497 test_acc_top5_avg=0.64528 time=103.94it/s +epoch=2 global_step=1173 loss=4.49577 test_loss_avg=3.46653 acc=0.00000 test_acc_avg=0.21420 test_acc_top5_avg=0.64725 time=469.69it/s +curr_acc 0.2142 +BEST_ACC 0.1921 +curr_acc_top5 0.6473 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=7.01650 loss_avg=7.08123 acc=0.29688 acc_top1_avg=0.28009 acc_top5_avg=0.70978 lr=0.01000 gn=2.29122 time=59.26it/s +epoch=3 global_step=1250 loss=6.70060 loss_avg=7.04268 acc=0.33594 acc_top1_avg=0.28470 acc_top5_avg=0.71753 lr=0.01000 gn=3.03291 time=64.98it/s +epoch=3 global_step=1300 loss=7.82607 loss_avg=7.06802 acc=0.20312 acc_top1_avg=0.28119 acc_top5_avg=0.72004 lr=0.01000 gn=2.62339 time=52.21it/s +epoch=3 global_step=1350 loss=7.34850 loss_avg=7.08060 acc=0.26562 acc_top1_avg=0.28028 acc_top5_avg=0.71871 lr=0.01000 gn=2.25631 time=61.26it/s +epoch=3 global_step=1400 loss=7.32485 loss_avg=7.07885 acc=0.25000 acc_top1_avg=0.28053 acc_top5_avg=0.71672 lr=0.01000 gn=2.69716 time=55.47it/s +epoch=3 global_step=1450 loss=7.17073 loss_avg=7.09097 acc=0.26562 acc_top1_avg=0.27897 acc_top5_avg=0.71635 lr=0.01000 gn=2.65603 time=61.48it/s +epoch=3 global_step=1500 loss=7.37197 loss_avg=7.09832 acc=0.25781 acc_top1_avg=0.27817 acc_top5_avg=0.71562 lr=0.01000 gn=3.74339 time=61.82it/s +epoch=3 global_step=1550 loss=6.36999 loss_avg=7.10514 acc=0.35156 acc_top1_avg=0.27766 acc_top5_avg=0.71535 lr=0.01000 gn=3.25533 time=65.33it/s +====================Eval==================== +epoch=3 global_step=1564 loss=3.22034 test_loss_avg=4.74158 acc=0.01562 test_acc_avg=0.00541 test_acc_top5_avg=0.83954 time=235.62it/s +epoch=3 global_step=1564 loss=0.29162 test_loss_avg=3.99256 acc=0.92188 test_acc_avg=0.16481 test_acc_top5_avg=0.71106 time=223.54it/s +epoch=3 global_step=1564 loss=5.04839 test_loss_avg=3.77781 acc=0.00000 test_acc_avg=0.21865 test_acc_top5_avg=0.67633 time=813.01it/s +curr_acc 0.2187 +BEST_ACC 0.2142 +curr_acc_top5 0.6763 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=6.87665 loss_avg=7.17356 acc=0.32031 acc_top1_avg=0.27235 acc_top5_avg=0.71311 lr=0.01000 gn=3.46054 time=54.62it/s +epoch=4 global_step=1650 loss=6.83719 loss_avg=7.08720 acc=0.31250 acc_top1_avg=0.28043 acc_top5_avg=0.71593 lr=0.01000 gn=3.02913 time=27.79it/s +epoch=4 global_step=1700 loss=7.36892 loss_avg=7.07987 acc=0.24219 acc_top1_avg=0.28108 acc_top5_avg=0.71829 lr=0.01000 gn=3.27261 time=53.75it/s +epoch=4 global_step=1750 loss=6.95827 loss_avg=7.08571 acc=0.29688 acc_top1_avg=0.27982 acc_top5_avg=0.71871 lr=0.01000 gn=3.33905 time=56.23it/s +epoch=4 global_step=1800 loss=6.90463 loss_avg=7.08620 acc=0.32031 acc_top1_avg=0.27940 acc_top5_avg=0.71792 lr=0.01000 gn=3.35725 time=60.77it/s +epoch=4 global_step=1850 loss=6.93725 loss_avg=7.07482 acc=0.29688 acc_top1_avg=0.28087 acc_top5_avg=0.71878 lr=0.01000 gn=3.78856 time=58.32it/s +epoch=4 global_step=1900 loss=6.83749 loss_avg=7.06590 acc=0.31250 acc_top1_avg=0.28172 acc_top5_avg=0.71908 lr=0.01000 gn=2.90858 time=63.14it/s +epoch=4 global_step=1950 loss=7.29112 loss_avg=7.06791 acc=0.27344 acc_top1_avg=0.28153 acc_top5_avg=0.71905 lr=0.01000 gn=3.71566 time=56.43it/s +====================Eval==================== +epoch=4 global_step=1955 loss=5.42324 test_loss_avg=4.47263 acc=0.00000 test_acc_avg=0.00184 test_acc_top5_avg=0.66981 time=234.00it/s +epoch=4 global_step=1955 loss=4.77627 test_loss_avg=3.65057 acc=0.00000 test_acc_avg=0.21994 test_acc_top5_avg=0.66901 time=552.61it/s +curr_acc 0.2199 +BEST_ACC 0.2187 +curr_acc_top5 0.6690 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=6.51529 loss_avg=7.07589 acc=0.32812 acc_top1_avg=0.27917 acc_top5_avg=0.71910 lr=0.01000 gn=3.19346 time=57.04it/s +epoch=5 global_step=2050 loss=7.29722 loss_avg=7.08996 acc=0.25781 acc_top1_avg=0.27887 acc_top5_avg=0.71834 lr=0.01000 gn=3.82991 time=58.38it/s +epoch=5 global_step=2100 loss=7.01398 loss_avg=7.06494 acc=0.26562 acc_top1_avg=0.28087 acc_top5_avg=0.72392 lr=0.01000 gn=2.87666 time=51.02it/s +epoch=5 global_step=2150 loss=6.83014 loss_avg=7.06834 acc=0.32812 acc_top1_avg=0.28041 acc_top5_avg=0.72208 lr=0.01000 gn=2.78271 time=38.38it/s +epoch=5 global_step=2200 loss=6.34747 loss_avg=7.04598 acc=0.36719 acc_top1_avg=0.28310 acc_top5_avg=0.72261 lr=0.01000 gn=3.02355 time=60.28it/s +epoch=5 global_step=2250 loss=7.31423 loss_avg=7.02665 acc=0.25000 acc_top1_avg=0.28554 acc_top5_avg=0.72307 lr=0.01000 gn=2.52890 time=56.56it/s +epoch=5 global_step=2300 loss=7.02247 loss_avg=7.03068 acc=0.30469 acc_top1_avg=0.28501 acc_top5_avg=0.72323 lr=0.01000 gn=2.90067 time=52.78it/s +====================Eval==================== +epoch=5 global_step=2346 loss=5.12171 test_loss_avg=5.07876 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.81875 time=244.08it/s +epoch=5 global_step=2346 loss=3.60958 test_loss_avg=4.20665 acc=0.17969 test_acc_avg=0.09702 test_acc_top5_avg=0.64901 time=242.11it/s +epoch=5 global_step=2346 loss=5.73462 test_loss_avg=3.68255 acc=0.00000 test_acc_avg=0.22290 test_acc_top5_avg=0.67375 time=533.36it/s +curr_acc 0.2229 +BEST_ACC 0.2199 +curr_acc_top5 0.6738 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=6.88069 loss_avg=7.21030 acc=0.30469 acc_top1_avg=0.26953 acc_top5_avg=0.70312 lr=0.01000 gn=2.52331 time=53.88it/s +epoch=6 global_step=2400 loss=7.45494 loss_avg=7.00555 acc=0.23438 acc_top1_avg=0.28877 acc_top5_avg=0.71962 lr=0.01000 gn=2.72810 time=56.61it/s +epoch=6 global_step=2450 loss=6.65538 loss_avg=6.99136 acc=0.32812 acc_top1_avg=0.29026 acc_top5_avg=0.72589 lr=0.01000 gn=4.39417 time=60.03it/s +epoch=6 global_step=2500 loss=6.98452 loss_avg=6.99980 acc=0.29688 acc_top1_avg=0.28957 acc_top5_avg=0.72727 lr=0.01000 gn=3.39737 time=31.85it/s +epoch=6 global_step=2550 loss=6.74902 loss_avg=6.99589 acc=0.30469 acc_top1_avg=0.28979 acc_top5_avg=0.72672 lr=0.01000 gn=4.17607 time=53.93it/s +epoch=6 global_step=2600 loss=7.06026 loss_avg=7.01923 acc=0.27344 acc_top1_avg=0.28734 acc_top5_avg=0.72413 lr=0.01000 gn=2.95814 time=56.44it/s +epoch=6 global_step=2650 loss=7.54194 loss_avg=7.01710 acc=0.21875 acc_top1_avg=0.28729 acc_top5_avg=0.72289 lr=0.01000 gn=2.76280 time=54.59it/s +epoch=6 global_step=2700 loss=7.47368 loss_avg=7.01867 acc=0.25000 acc_top1_avg=0.28716 acc_top5_avg=0.72272 lr=0.01000 gn=3.48779 time=59.04it/s +====================Eval==================== +epoch=6 global_step=2737 loss=5.08808 test_loss_avg=4.17816 acc=0.00000 test_acc_avg=0.00150 test_acc_top5_avg=0.75992 time=70.86it/s +epoch=6 global_step=2737 loss=4.92675 test_loss_avg=3.57730 acc=0.00000 test_acc_avg=0.22502 test_acc_top5_avg=0.68061 time=243.80it/s +epoch=6 global_step=2737 loss=4.43203 test_loss_avg=3.62322 acc=0.00000 test_acc_avg=0.21648 test_acc_top5_avg=0.65932 time=517.88it/s +curr_acc 0.2165 +BEST_ACC 0.2229 +curr_acc_top5 0.6593 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=6.97686 loss_avg=7.04093 acc=0.31250 acc_top1_avg=0.28486 acc_top5_avg=0.69712 lr=0.01000 gn=3.60863 time=61.33it/s +epoch=7 global_step=2800 loss=7.02394 loss_avg=7.04671 acc=0.27344 acc_top1_avg=0.28187 acc_top5_avg=0.72160 lr=0.01000 gn=3.74897 time=62.84it/s +epoch=7 global_step=2850 loss=6.72855 loss_avg=7.03611 acc=0.32031 acc_top1_avg=0.28312 acc_top5_avg=0.72470 lr=0.01000 gn=3.10031 time=56.02it/s +epoch=7 global_step=2900 loss=7.22818 loss_avg=7.01095 acc=0.27344 acc_top1_avg=0.28676 acc_top5_avg=0.72656 lr=0.01000 gn=3.00237 time=56.08it/s +epoch=7 global_step=2950 loss=7.45106 loss_avg=7.01667 acc=0.23438 acc_top1_avg=0.28635 acc_top5_avg=0.72686 lr=0.01000 gn=2.73130 time=58.14it/s +epoch=7 global_step=3000 loss=6.60833 loss_avg=7.01231 acc=0.33594 acc_top1_avg=0.28713 acc_top5_avg=0.72710 lr=0.01000 gn=2.58655 time=57.11it/s +epoch=7 global_step=3050 loss=7.28595 loss_avg=6.99807 acc=0.25781 acc_top1_avg=0.28894 acc_top5_avg=0.72838 lr=0.01000 gn=2.74024 time=57.40it/s +epoch=7 global_step=3100 loss=6.54563 loss_avg=6.99654 acc=0.35938 acc_top1_avg=0.28921 acc_top5_avg=0.72850 lr=0.01000 gn=3.63058 time=60.20it/s +====================Eval==================== +epoch=7 global_step=3128 loss=2.53412 test_loss_avg=4.39305 acc=0.45312 test_acc_avg=0.09109 test_acc_top5_avg=0.68617 time=226.45it/s +epoch=7 global_step=3128 loss=4.30185 test_loss_avg=3.71575 acc=0.00000 test_acc_avg=0.22320 test_acc_top5_avg=0.65279 time=491.25it/s +curr_acc 0.2232 +BEST_ACC 0.2229 +curr_acc_top5 0.6528 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=6.79055 loss_avg=6.95652 acc=0.32812 acc_top1_avg=0.29510 acc_top5_avg=0.71378 lr=0.01000 gn=3.41278 time=56.33it/s +epoch=8 global_step=3200 loss=7.39457 loss_avg=6.93923 acc=0.24219 acc_top1_avg=0.29579 acc_top5_avg=0.72005 lr=0.01000 gn=2.80238 time=65.24it/s +epoch=8 global_step=3250 loss=6.84999 loss_avg=6.96303 acc=0.29688 acc_top1_avg=0.29258 acc_top5_avg=0.72349 lr=0.01000 gn=3.33446 time=58.06it/s +epoch=8 global_step=3300 loss=7.19939 loss_avg=6.94520 acc=0.26562 acc_top1_avg=0.29479 acc_top5_avg=0.72615 lr=0.01000 gn=3.84004 time=55.29it/s +epoch=8 global_step=3350 loss=6.96943 loss_avg=6.96044 acc=0.28906 acc_top1_avg=0.29212 acc_top5_avg=0.72540 lr=0.01000 gn=3.58594 time=54.22it/s +epoch=8 global_step=3400 loss=7.24580 loss_avg=6.96947 acc=0.25781 acc_top1_avg=0.29053 acc_top5_avg=0.72524 lr=0.01000 gn=3.25436 time=58.45it/s +epoch=8 global_step=3450 loss=7.34070 loss_avg=6.97335 acc=0.23438 acc_top1_avg=0.29054 acc_top5_avg=0.72397 lr=0.01000 gn=5.59868 time=52.18it/s +epoch=8 global_step=3500 loss=6.84850 loss_avg=6.98712 acc=0.29688 acc_top1_avg=0.28902 acc_top5_avg=0.72486 lr=0.01000 gn=3.73728 time=39.10it/s +====================Eval==================== +epoch=8 global_step=3519 loss=4.66862 test_loss_avg=5.00888 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.84201 time=198.95it/s +epoch=8 global_step=3519 loss=0.58381 test_loss_avg=3.66280 acc=0.85156 test_acc_avg=0.23323 test_acc_top5_avg=0.71473 time=241.45it/s +epoch=8 global_step=3519 loss=6.13473 test_loss_avg=3.86111 acc=0.00000 test_acc_avg=0.22686 test_acc_top5_avg=0.67336 time=842.91it/s +curr_acc 0.2269 +BEST_ACC 0.2232 +curr_acc_top5 0.6734 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=6.99145 loss_avg=6.89251 acc=0.28906 acc_top1_avg=0.30318 acc_top5_avg=0.73009 lr=0.01000 gn=2.68781 time=57.27it/s +epoch=9 global_step=3600 loss=7.15792 loss_avg=6.95744 acc=0.28125 acc_top1_avg=0.29389 acc_top5_avg=0.72907 lr=0.01000 gn=3.05370 time=52.97it/s +epoch=9 global_step=3650 loss=6.46284 loss_avg=6.96868 acc=0.35156 acc_top1_avg=0.29324 acc_top5_avg=0.72811 lr=0.01000 gn=4.44568 time=59.86it/s +epoch=9 global_step=3700 loss=7.14451 loss_avg=6.99492 acc=0.25781 acc_top1_avg=0.28962 acc_top5_avg=0.72794 lr=0.01000 gn=3.23546 time=47.94it/s +epoch=9 global_step=3750 loss=7.02918 loss_avg=6.97800 acc=0.27344 acc_top1_avg=0.29102 acc_top5_avg=0.72978 lr=0.01000 gn=2.69670 time=58.11it/s +epoch=9 global_step=3800 loss=6.80954 loss_avg=6.98010 acc=0.30469 acc_top1_avg=0.29090 acc_top5_avg=0.72712 lr=0.01000 gn=3.64676 time=60.16it/s +epoch=9 global_step=3850 loss=6.79652 loss_avg=6.98055 acc=0.31250 acc_top1_avg=0.29064 acc_top5_avg=0.72682 lr=0.01000 gn=2.62148 time=61.45it/s +epoch=9 global_step=3900 loss=7.36444 loss_avg=6.98019 acc=0.25000 acc_top1_avg=0.29031 acc_top5_avg=0.72625 lr=0.01000 gn=3.37523 time=59.58it/s +====================Eval==================== +epoch=9 global_step=3910 loss=4.85948 test_loss_avg=4.45042 acc=0.00000 test_acc_avg=0.00441 test_acc_top5_avg=0.65946 time=150.28it/s +epoch=9 global_step=3910 loss=5.74343 test_loss_avg=3.63656 acc=0.00000 test_acc_avg=0.20560 test_acc_top5_avg=0.66466 time=856.16it/s +curr_acc 0.2056 +BEST_ACC 0.2269 +curr_acc_top5 0.6647 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=6.67413 loss_avg=6.98948 acc=0.33594 acc_top1_avg=0.28887 acc_top5_avg=0.72363 lr=0.01000 gn=4.09019 time=56.55it/s +epoch=10 global_step=4000 loss=6.44105 loss_avg=7.00458 acc=0.33594 acc_top1_avg=0.28655 acc_top5_avg=0.72118 lr=0.01000 gn=3.18902 time=56.56it/s +epoch=10 global_step=4050 loss=7.33887 loss_avg=6.99150 acc=0.25000 acc_top1_avg=0.28912 acc_top5_avg=0.72009 lr=0.01000 gn=3.28989 time=58.94it/s +epoch=10 global_step=4100 loss=6.97159 loss_avg=6.98305 acc=0.28125 acc_top1_avg=0.29021 acc_top5_avg=0.72418 lr=0.01000 gn=2.98552 time=52.01it/s +epoch=10 global_step=4150 loss=7.14336 loss_avg=6.98818 acc=0.28125 acc_top1_avg=0.29010 acc_top5_avg=0.72490 lr=0.01000 gn=3.71521 time=62.06it/s +epoch=10 global_step=4200 loss=6.42558 loss_avg=6.97859 acc=0.33594 acc_top1_avg=0.29098 acc_top5_avg=0.72476 lr=0.01000 gn=3.08249 time=59.57it/s +epoch=10 global_step=4250 loss=7.10522 loss_avg=6.97720 acc=0.28125 acc_top1_avg=0.29118 acc_top5_avg=0.72406 lr=0.01000 gn=4.24857 time=63.19it/s +epoch=10 global_step=4300 loss=7.35454 loss_avg=6.97389 acc=0.26562 acc_top1_avg=0.29111 acc_top5_avg=0.72382 lr=0.01000 gn=3.37591 time=59.73it/s +====================Eval==================== +epoch=10 global_step=4301 loss=1.88247 test_loss_avg=4.07020 acc=0.17188 test_acc_avg=0.03281 test_acc_top5_avg=0.74922 time=225.38it/s +epoch=10 global_step=4301 loss=0.11804 test_loss_avg=3.79396 acc=0.96094 test_acc_avg=0.12500 test_acc_top5_avg=0.70273 time=242.00it/s +epoch=10 global_step=4301 loss=4.82211 test_loss_avg=3.46031 acc=0.00000 test_acc_avg=0.21776 test_acc_top5_avg=0.66980 time=850.77it/s +curr_acc 0.2178 +BEST_ACC 0.2269 +curr_acc_top5 0.6698 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=7.25263 loss_avg=6.89715 acc=0.25000 acc_top1_avg=0.30070 acc_top5_avg=0.72624 lr=0.01000 gn=3.43617 time=56.05it/s +epoch=11 global_step=4400 loss=7.10738 loss_avg=6.92839 acc=0.28125 acc_top1_avg=0.29585 acc_top5_avg=0.72467 lr=0.01000 gn=4.42423 time=58.83it/s +epoch=11 global_step=4450 loss=6.82661 loss_avg=6.92613 acc=0.31250 acc_top1_avg=0.29640 acc_top5_avg=0.72724 lr=0.01000 gn=3.12747 time=64.03it/s +epoch=11 global_step=4500 loss=7.23443 loss_avg=6.95375 acc=0.25781 acc_top1_avg=0.29318 acc_top5_avg=0.72637 lr=0.01000 gn=3.56179 time=51.25it/s +epoch=11 global_step=4550 loss=6.99573 loss_avg=6.95699 acc=0.27344 acc_top1_avg=0.29258 acc_top5_avg=0.72603 lr=0.01000 gn=3.00700 time=62.33it/s +epoch=11 global_step=4600 loss=7.46681 loss_avg=6.96562 acc=0.23438 acc_top1_avg=0.29183 acc_top5_avg=0.72526 lr=0.01000 gn=3.22376 time=59.54it/s +epoch=11 global_step=4650 loss=7.33554 loss_avg=6.96137 acc=0.26562 acc_top1_avg=0.29258 acc_top5_avg=0.72589 lr=0.01000 gn=4.01302 time=53.63it/s +====================Eval==================== +epoch=11 global_step=4692 loss=5.69625 test_loss_avg=5.30218 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.61946 time=243.27it/s +epoch=11 global_step=4692 loss=5.00446 test_loss_avg=4.06434 acc=0.00000 test_acc_avg=0.21608 test_acc_top5_avg=0.66950 time=495.60it/s +curr_acc 0.2161 +BEST_ACC 0.2269 +curr_acc_top5 0.6695 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=7.25521 loss_avg=7.05766 acc=0.27344 acc_top1_avg=0.27930 acc_top5_avg=0.71777 lr=0.01000 gn=3.66055 time=64.46it/s +epoch=12 global_step=4750 loss=7.14386 loss_avg=6.95516 acc=0.27344 acc_top1_avg=0.29310 acc_top5_avg=0.72697 lr=0.01000 gn=3.27173 time=52.25it/s +epoch=12 global_step=4800 loss=6.70087 loss_avg=6.96895 acc=0.32812 acc_top1_avg=0.29188 acc_top5_avg=0.72381 lr=0.01000 gn=3.99087 time=57.43it/s +epoch=12 global_step=4850 loss=6.47289 loss_avg=6.98569 acc=0.32031 acc_top1_avg=0.28966 acc_top5_avg=0.71840 lr=0.01000 gn=2.84673 time=56.57it/s +epoch=12 global_step=4900 loss=6.76654 loss_avg=6.98132 acc=0.32031 acc_top1_avg=0.29026 acc_top5_avg=0.72089 lr=0.01000 gn=4.00413 time=54.70it/s +epoch=12 global_step=4950 loss=6.63883 loss_avg=6.98518 acc=0.32812 acc_top1_avg=0.29003 acc_top5_avg=0.72244 lr=0.01000 gn=3.77225 time=60.77it/s +epoch=12 global_step=5000 loss=6.88699 loss_avg=6.98559 acc=0.30469 acc_top1_avg=0.29031 acc_top5_avg=0.72255 lr=0.01000 gn=4.55513 time=54.65it/s +epoch=12 global_step=5050 loss=6.61777 loss_avg=6.97561 acc=0.32031 acc_top1_avg=0.29148 acc_top5_avg=0.72327 lr=0.01000 gn=3.69930 time=56.75it/s +====================Eval==================== +epoch=12 global_step=5083 loss=4.83325 test_loss_avg=4.89038 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.76172 time=90.64it/s +epoch=12 global_step=5083 loss=4.98842 test_loss_avg=4.32960 acc=0.00000 test_acc_avg=0.05619 test_acc_top5_avg=0.65234 time=252.81it/s +epoch=12 global_step=5083 loss=4.62842 test_loss_avg=3.58771 acc=0.00000 test_acc_avg=0.21915 test_acc_top5_avg=0.64448 time=492.64it/s +curr_acc 0.2191 +BEST_ACC 0.2269 +curr_acc_top5 0.6445 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=6.37350 loss_avg=6.70674 acc=0.35156 acc_top1_avg=0.32169 acc_top5_avg=0.73483 lr=0.01000 gn=2.58268 time=59.61it/s +epoch=13 global_step=5150 loss=6.81975 loss_avg=6.89269 acc=0.32031 acc_top1_avg=0.30154 acc_top5_avg=0.73123 lr=0.01000 gn=3.67590 time=54.94it/s +epoch=13 global_step=5200 loss=6.61265 loss_avg=6.90694 acc=0.31250 acc_top1_avg=0.30001 acc_top5_avg=0.73264 lr=0.01000 gn=2.55612 time=62.04it/s +epoch=13 global_step=5250 loss=7.35604 loss_avg=6.90996 acc=0.24219 acc_top1_avg=0.29847 acc_top5_avg=0.72984 lr=0.01000 gn=2.75166 time=60.23it/s +epoch=13 global_step=5300 loss=7.51582 loss_avg=6.93658 acc=0.23438 acc_top1_avg=0.29547 acc_top5_avg=0.72804 lr=0.01000 gn=3.90349 time=51.11it/s +epoch=13 global_step=5350 loss=6.65104 loss_avg=6.92939 acc=0.32031 acc_top1_avg=0.29614 acc_top5_avg=0.72893 lr=0.01000 gn=4.76869 time=64.75it/s +epoch=13 global_step=5400 loss=7.03983 loss_avg=6.94044 acc=0.28125 acc_top1_avg=0.29493 acc_top5_avg=0.72681 lr=0.01000 gn=3.15726 time=54.97it/s +epoch=13 global_step=5450 loss=7.19770 loss_avg=6.94957 acc=0.26562 acc_top1_avg=0.29377 acc_top5_avg=0.72516 lr=0.01000 gn=4.08946 time=62.78it/s +====================Eval==================== +epoch=13 global_step=5474 loss=4.65424 test_loss_avg=4.29626 acc=0.00000 test_acc_avg=0.00136 test_acc_top5_avg=0.83016 time=235.42it/s +epoch=13 global_step=5474 loss=4.61511 test_loss_avg=3.55754 acc=0.00000 test_acc_avg=0.24048 test_acc_top5_avg=0.68301 time=241.37it/s +epoch=13 global_step=5474 loss=4.73786 test_loss_avg=3.64292 acc=0.00000 test_acc_avg=0.22221 test_acc_top5_avg=0.65101 time=821.93it/s +curr_acc 0.2222 +BEST_ACC 0.2269 +curr_acc_top5 0.6510 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=6.94206 loss_avg=6.81914 acc=0.29688 acc_top1_avg=0.31160 acc_top5_avg=0.73528 lr=0.01000 gn=3.40256 time=60.52it/s +epoch=14 global_step=5550 loss=6.81082 loss_avg=6.90602 acc=0.29688 acc_top1_avg=0.29934 acc_top5_avg=0.72492 lr=0.01000 gn=4.18530 time=59.69it/s +epoch=14 global_step=5600 loss=7.13099 loss_avg=6.97690 acc=0.28906 acc_top1_avg=0.29117 acc_top5_avg=0.71869 lr=0.01000 gn=3.79159 time=57.73it/s +epoch=14 global_step=5650 loss=7.08607 loss_avg=6.95521 acc=0.27344 acc_top1_avg=0.29310 acc_top5_avg=0.72337 lr=0.01000 gn=3.40249 time=62.55it/s +epoch=14 global_step=5700 loss=6.22442 loss_avg=6.92567 acc=0.38281 acc_top1_avg=0.29667 acc_top5_avg=0.72577 lr=0.01000 gn=4.00371 time=61.82it/s +epoch=14 global_step=5750 loss=6.77463 loss_avg=6.93897 acc=0.32031 acc_top1_avg=0.29481 acc_top5_avg=0.72532 lr=0.01000 gn=4.83524 time=55.17it/s +epoch=14 global_step=5800 loss=6.43760 loss_avg=6.94300 acc=0.35156 acc_top1_avg=0.29486 acc_top5_avg=0.72534 lr=0.01000 gn=3.36383 time=58.48it/s +epoch=14 global_step=5850 loss=7.12520 loss_avg=6.95642 acc=0.26562 acc_top1_avg=0.29328 acc_top5_avg=0.72453 lr=0.01000 gn=3.84566 time=59.05it/s +====================Eval==================== +epoch=14 global_step=5865 loss=2.11558 test_loss_avg=4.28018 acc=0.40625 test_acc_avg=0.06481 test_acc_top5_avg=0.67738 time=234.46it/s +epoch=14 global_step=5865 loss=4.76099 test_loss_avg=3.55102 acc=0.00000 test_acc_avg=0.23002 test_acc_top5_avg=0.69136 time=633.10it/s +curr_acc 0.2300 +BEST_ACC 0.2269 +curr_acc_top5 0.6914 +BEST_ACC_top5 0.6768 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=7.25044 loss_avg=6.98187 acc=0.25000 acc_top1_avg=0.29107 acc_top5_avg=0.71853 lr=0.01000 gn=3.95043 time=60.71it/s +epoch=15 global_step=5950 loss=6.73079 loss_avg=6.99137 acc=0.32031 acc_top1_avg=0.28925 acc_top5_avg=0.72261 lr=0.01000 gn=3.78014 time=60.04it/s +epoch=15 global_step=6000 loss=6.70995 loss_avg=6.99072 acc=0.30469 acc_top1_avg=0.28866 acc_top5_avg=0.72031 lr=0.01000 gn=3.02682 time=60.31it/s +epoch=15 global_step=6050 loss=6.26410 loss_avg=6.96700 acc=0.35938 acc_top1_avg=0.29198 acc_top5_avg=0.72204 lr=0.01000 gn=4.53653 time=55.55it/s +epoch=15 global_step=6100 loss=7.21211 loss_avg=6.97273 acc=0.26562 acc_top1_avg=0.29126 acc_top5_avg=0.72101 lr=0.01000 gn=3.51228 time=56.57it/s +epoch=15 global_step=6150 loss=7.01927 loss_avg=6.96063 acc=0.29688 acc_top1_avg=0.29367 acc_top5_avg=0.72067 lr=0.01000 gn=3.32450 time=49.15it/s +epoch=15 global_step=6200 loss=7.28112 loss_avg=6.95044 acc=0.25781 acc_top1_avg=0.29450 acc_top5_avg=0.72164 lr=0.01000 gn=4.00005 time=56.67it/s +epoch=15 global_step=6250 loss=6.79676 loss_avg=6.94763 acc=0.30469 acc_top1_avg=0.29476 acc_top5_avg=0.72259 lr=0.01000 gn=2.83274 time=54.87it/s +====================Eval==================== +epoch=15 global_step=6256 loss=4.59538 test_loss_avg=5.07802 acc=0.00000 test_acc_avg=0.00052 test_acc_top5_avg=0.74115 time=228.95it/s +epoch=15 global_step=6256 loss=0.02382 test_loss_avg=4.21387 acc=0.98438 test_acc_avg=0.17091 test_acc_top5_avg=0.65060 time=240.38it/s +epoch=15 global_step=6256 loss=5.20729 test_loss_avg=4.05968 acc=0.00000 test_acc_avg=0.20708 test_acc_top5_avg=0.61610 time=792.13it/s +curr_acc 0.2071 +BEST_ACC 0.2300 +curr_acc_top5 0.6161 +BEST_ACC_top5 0.6914 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=7.52121 loss_avg=7.01974 acc=0.22656 acc_top1_avg=0.28374 acc_top5_avg=0.71822 lr=0.01000 gn=3.63073 time=62.35it/s +epoch=16 global_step=6350 loss=7.11280 loss_avg=6.98661 acc=0.27344 acc_top1_avg=0.28848 acc_top5_avg=0.72365 lr=0.01000 gn=2.95348 time=61.74it/s +epoch=16 global_step=6400 loss=7.40895 loss_avg=6.96704 acc=0.23438 acc_top1_avg=0.29107 acc_top5_avg=0.72385 lr=0.01000 gn=3.81737 time=56.07it/s +epoch=16 global_step=6450 loss=6.90211 loss_avg=6.95365 acc=0.30469 acc_top1_avg=0.29277 acc_top5_avg=0.72411 lr=0.01000 gn=3.61835 time=61.56it/s +epoch=16 global_step=6500 loss=6.73955 loss_avg=6.96490 acc=0.32031 acc_top1_avg=0.29127 acc_top5_avg=0.72227 lr=0.01000 gn=3.11703 time=54.67it/s +epoch=16 global_step=6550 loss=7.35646 loss_avg=6.97138 acc=0.23438 acc_top1_avg=0.29050 acc_top5_avg=0.72258 lr=0.01000 gn=3.44106 time=53.45it/s +epoch=16 global_step=6600 loss=6.96793 loss_avg=6.96822 acc=0.32031 acc_top1_avg=0.29158 acc_top5_avg=0.72184 lr=0.01000 gn=4.33102 time=59.94it/s +====================Eval==================== +epoch=16 global_step=6647 loss=5.15394 test_loss_avg=4.82441 acc=0.00000 test_acc_avg=0.00326 test_acc_top5_avg=0.65538 time=80.88it/s +epoch=16 global_step=6647 loss=7.01059 test_loss_avg=3.87439 acc=0.00000 test_acc_avg=0.22528 test_acc_top5_avg=0.67751 time=862.32it/s +curr_acc 0.2253 +BEST_ACC 0.2300 +curr_acc_top5 0.6775 +BEST_ACC_top5 0.6914 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=6.40704 loss_avg=6.50362 acc=0.35156 acc_top1_avg=0.34375 acc_top5_avg=0.72396 lr=0.01000 gn=4.25102 time=52.49it/s +epoch=17 global_step=6700 loss=7.28628 loss_avg=6.90319 acc=0.25000 acc_top1_avg=0.29923 acc_top5_avg=0.71963 lr=0.01000 gn=3.74962 time=55.49it/s +epoch=17 global_step=6750 loss=6.63240 loss_avg=6.93275 acc=0.32812 acc_top1_avg=0.29604 acc_top5_avg=0.72497 lr=0.01000 gn=3.89030 time=59.37it/s +epoch=17 global_step=6800 loss=7.09064 loss_avg=6.94169 acc=0.28906 acc_top1_avg=0.29514 acc_top5_avg=0.72457 lr=0.01000 gn=3.39673 time=53.13it/s +epoch=17 global_step=6850 loss=7.52124 loss_avg=6.93149 acc=0.22656 acc_top1_avg=0.29568 acc_top5_avg=0.72649 lr=0.01000 gn=3.79732 time=62.38it/s +epoch=17 global_step=6900 loss=6.87483 loss_avg=6.94034 acc=0.28906 acc_top1_avg=0.29447 acc_top5_avg=0.72462 lr=0.01000 gn=4.62930 time=64.81it/s +epoch=17 global_step=6950 loss=6.81055 loss_avg=6.93834 acc=0.32031 acc_top1_avg=0.29468 acc_top5_avg=0.72481 lr=0.01000 gn=3.17271 time=51.61it/s +epoch=17 global_step=7000 loss=7.06226 loss_avg=6.93608 acc=0.28906 acc_top1_avg=0.29513 acc_top5_avg=0.72495 lr=0.01000 gn=3.12989 time=56.15it/s +====================Eval==================== +epoch=17 global_step=7038 loss=4.97825 test_loss_avg=4.96374 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.48996 time=239.98it/s +epoch=17 global_step=7038 loss=0.31394 test_loss_avg=4.01468 acc=0.89062 test_acc_avg=0.13103 test_acc_top5_avg=0.66530 time=61.77it/s +epoch=17 global_step=7038 loss=4.78520 test_loss_avg=3.45370 acc=0.00000 test_acc_avg=0.25336 test_acc_top5_avg=0.66851 time=826.95it/s +curr_acc 0.2534 +BEST_ACC 0.2300 +curr_acc_top5 0.6685 +BEST_ACC_top5 0.6914 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=6.91939 loss_avg=6.89596 acc=0.29688 acc_top1_avg=0.30208 acc_top5_avg=0.71419 lr=0.01000 gn=4.63877 time=54.80it/s +epoch=18 global_step=7100 loss=6.59580 loss_avg=6.86633 acc=0.34375 acc_top1_avg=0.30456 acc_top5_avg=0.72644 lr=0.01000 gn=3.58749 time=55.97it/s +epoch=18 global_step=7150 loss=6.64912 loss_avg=6.87785 acc=0.32812 acc_top1_avg=0.30239 acc_top5_avg=0.72663 lr=0.01000 gn=3.73147 time=47.18it/s +epoch=18 global_step=7200 loss=6.82623 loss_avg=6.91464 acc=0.31250 acc_top1_avg=0.29856 acc_top5_avg=0.72473 lr=0.01000 gn=2.68838 time=61.16it/s +epoch=18 global_step=7250 loss=6.70929 loss_avg=6.91228 acc=0.32031 acc_top1_avg=0.29831 acc_top5_avg=0.72557 lr=0.01000 gn=3.51522 time=50.74it/s +epoch=18 global_step=7300 loss=6.83237 loss_avg=6.90168 acc=0.30469 acc_top1_avg=0.29950 acc_top5_avg=0.72534 lr=0.01000 gn=5.44805 time=59.77it/s +epoch=18 global_step=7350 loss=6.72197 loss_avg=6.91021 acc=0.31250 acc_top1_avg=0.29813 acc_top5_avg=0.72531 lr=0.01000 gn=2.95709 time=57.70it/s +epoch=18 global_step=7400 loss=6.45058 loss_avg=6.92096 acc=0.34375 acc_top1_avg=0.29683 acc_top5_avg=0.72328 lr=0.01000 gn=2.95978 time=61.38it/s +====================Eval==================== +epoch=18 global_step=7429 loss=5.06843 test_loss_avg=4.35779 acc=0.00000 test_acc_avg=0.02818 test_acc_top5_avg=0.70703 time=242.95it/s +epoch=18 global_step=7429 loss=4.48921 test_loss_avg=3.38770 acc=0.00000 test_acc_avg=0.26452 test_acc_top5_avg=0.67808 time=255.35it/s +epoch=18 global_step=7429 loss=4.57130 test_loss_avg=3.40268 acc=0.00000 test_acc_avg=0.26117 test_acc_top5_avg=0.67108 time=844.26it/s +curr_acc 0.2612 +BEST_ACC 0.2534 +curr_acc_top5 0.6711 +BEST_ACC_top5 0.6914 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=7.01411 loss_avg=7.00531 acc=0.29688 acc_top1_avg=0.29018 acc_top5_avg=0.73289 lr=0.01000 gn=2.92494 time=62.42it/s +epoch=19 global_step=7500 loss=6.11320 loss_avg=6.90651 acc=0.35938 acc_top1_avg=0.30073 acc_top5_avg=0.73008 lr=0.01000 gn=3.05477 time=55.91it/s +epoch=19 global_step=7550 loss=6.69033 loss_avg=6.92627 acc=0.32031 acc_top1_avg=0.29759 acc_top5_avg=0.72811 lr=0.01000 gn=3.42244 time=55.99it/s +epoch=19 global_step=7600 loss=7.37694 loss_avg=6.91132 acc=0.25000 acc_top1_avg=0.29870 acc_top5_avg=0.72848 lr=0.01000 gn=3.70648 time=58.80it/s +epoch=19 global_step=7650 loss=6.00777 loss_avg=6.90651 acc=0.40625 acc_top1_avg=0.29871 acc_top5_avg=0.72904 lr=0.01000 gn=3.57924 time=63.51it/s +epoch=19 global_step=7700 loss=7.79040 loss_avg=6.93980 acc=0.20312 acc_top1_avg=0.29494 acc_top5_avg=0.72694 lr=0.01000 gn=3.89950 time=58.27it/s +epoch=19 global_step=7750 loss=6.63048 loss_avg=6.94140 acc=0.32031 acc_top1_avg=0.29464 acc_top5_avg=0.72656 lr=0.01000 gn=3.42355 time=57.17it/s +epoch=19 global_step=7800 loss=7.11608 loss_avg=6.92942 acc=0.26562 acc_top1_avg=0.29549 acc_top5_avg=0.72703 lr=0.01000 gn=3.27427 time=56.20it/s +====================Eval==================== +epoch=19 global_step=7820 loss=5.60152 test_loss_avg=4.94912 acc=0.00000 test_acc_avg=0.16247 test_acc_top5_avg=0.67841 time=238.91it/s +epoch=19 global_step=7820 loss=6.41956 test_loss_avg=4.77579 acc=0.00000 test_acc_avg=0.18345 test_acc_top5_avg=0.71746 time=834.19it/s +curr_acc 0.1834 +BEST_ACC 0.2612 +curr_acc_top5 0.7175 +BEST_ACC_top5 0.6914 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=6.51544 loss_avg=6.93161 acc=0.37500 acc_top1_avg=0.29740 acc_top5_avg=0.72630 lr=0.01000 gn=4.23522 time=64.02it/s +epoch=20 global_step=7900 loss=7.07526 loss_avg=6.89115 acc=0.26562 acc_top1_avg=0.29922 acc_top5_avg=0.73027 lr=0.01000 gn=4.90147 time=56.49it/s +epoch=20 global_step=7950 loss=6.10225 loss_avg=6.89198 acc=0.39844 acc_top1_avg=0.29958 acc_top5_avg=0.72440 lr=0.01000 gn=4.29397 time=53.17it/s +epoch=20 global_step=8000 loss=6.69113 loss_avg=6.89473 acc=0.31250 acc_top1_avg=0.29931 acc_top5_avg=0.72591 lr=0.01000 gn=3.12449 time=53.07it/s +epoch=20 global_step=8050 loss=7.15075 loss_avg=6.90732 acc=0.28125 acc_top1_avg=0.29796 acc_top5_avg=0.72582 lr=0.01000 gn=3.19533 time=61.90it/s +epoch=20 global_step=8100 loss=6.64432 loss_avg=6.91093 acc=0.32812 acc_top1_avg=0.29799 acc_top5_avg=0.72528 lr=0.01000 gn=4.65190 time=53.22it/s +epoch=20 global_step=8150 loss=7.30239 loss_avg=6.92928 acc=0.27344 acc_top1_avg=0.29640 acc_top5_avg=0.72483 lr=0.01000 gn=3.24239 time=63.10it/s +epoch=20 global_step=8200 loss=6.52702 loss_avg=6.93729 acc=0.32812 acc_top1_avg=0.29529 acc_top5_avg=0.72465 lr=0.01000 gn=4.82510 time=61.94it/s +====================Eval==================== +epoch=20 global_step=8211 loss=4.52026 test_loss_avg=3.99288 acc=0.00000 test_acc_avg=0.03164 test_acc_top5_avg=0.77109 time=236.10it/s +epoch=20 global_step=8211 loss=0.32167 test_loss_avg=3.12095 acc=0.89062 test_acc_avg=0.28627 test_acc_top5_avg=0.79196 time=227.12it/s +epoch=20 global_step=8211 loss=4.41679 test_loss_avg=3.24901 acc=0.00000 test_acc_avg=0.25722 test_acc_top5_avg=0.77581 time=779.47it/s +curr_acc 0.2572 +BEST_ACC 0.2612 +curr_acc_top5 0.7758 +BEST_ACC_top5 0.7175 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=7.40828 loss_avg=6.99238 acc=0.23438 acc_top1_avg=0.28826 acc_top5_avg=0.72236 lr=0.01000 gn=2.80136 time=61.10it/s +epoch=21 global_step=8300 loss=6.79298 loss_avg=6.95841 acc=0.30469 acc_top1_avg=0.29398 acc_top5_avg=0.72437 lr=0.01000 gn=3.61561 time=54.07it/s +epoch=21 global_step=8350 loss=6.90837 loss_avg=6.90518 acc=0.31250 acc_top1_avg=0.29985 acc_top5_avg=0.72611 lr=0.01000 gn=3.88138 time=57.88it/s +epoch=21 global_step=8400 loss=6.50459 loss_avg=6.89793 acc=0.32812 acc_top1_avg=0.30035 acc_top5_avg=0.72693 lr=0.01000 gn=3.41563 time=59.58it/s +epoch=21 global_step=8450 loss=6.80877 loss_avg=6.89686 acc=0.29688 acc_top1_avg=0.29998 acc_top5_avg=0.72702 lr=0.01000 gn=3.75033 time=56.33it/s +epoch=21 global_step=8500 loss=6.72332 loss_avg=6.90929 acc=0.32812 acc_top1_avg=0.29844 acc_top5_avg=0.72643 lr=0.01000 gn=3.80901 time=58.33it/s +epoch=21 global_step=8550 loss=6.88308 loss_avg=6.90880 acc=0.29688 acc_top1_avg=0.29833 acc_top5_avg=0.72569 lr=0.01000 gn=4.34390 time=63.52it/s +epoch=21 global_step=8600 loss=6.91478 loss_avg=6.90911 acc=0.31250 acc_top1_avg=0.29856 acc_top5_avg=0.72494 lr=0.01000 gn=2.97008 time=57.06it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.10308 test_loss_avg=4.55754 acc=0.68750 test_acc_avg=0.03849 test_acc_top5_avg=0.62748 time=235.26it/s +epoch=21 global_step=8602 loss=5.84428 test_loss_avg=3.63245 acc=0.00000 test_acc_avg=0.23962 test_acc_top5_avg=0.65348 time=858.08it/s +curr_acc 0.2396 +BEST_ACC 0.2612 +curr_acc_top5 0.6535 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=6.83715 loss_avg=6.81387 acc=0.30469 acc_top1_avg=0.30892 acc_top5_avg=0.72461 lr=0.01000 gn=4.16108 time=57.29it/s +epoch=22 global_step=8700 loss=6.60794 loss_avg=6.86464 acc=0.33594 acc_top1_avg=0.30325 acc_top5_avg=0.72848 lr=0.01000 gn=3.54379 time=61.03it/s +epoch=22 global_step=8750 loss=7.06295 loss_avg=6.87912 acc=0.29688 acc_top1_avg=0.30136 acc_top5_avg=0.73031 lr=0.01000 gn=4.40754 time=59.44it/s +epoch=22 global_step=8800 loss=7.08715 loss_avg=6.88098 acc=0.28906 acc_top1_avg=0.30118 acc_top5_avg=0.73082 lr=0.01000 gn=4.03559 time=57.81it/s +epoch=22 global_step=8850 loss=6.68366 loss_avg=6.87980 acc=0.31250 acc_top1_avg=0.30116 acc_top5_avg=0.73072 lr=0.01000 gn=4.08389 time=56.00it/s +epoch=22 global_step=8900 loss=7.22256 loss_avg=6.89771 acc=0.25781 acc_top1_avg=0.29905 acc_top5_avg=0.72947 lr=0.01000 gn=3.64998 time=51.11it/s +epoch=22 global_step=8950 loss=6.65031 loss_avg=6.89982 acc=0.32031 acc_top1_avg=0.29887 acc_top5_avg=0.72807 lr=0.01000 gn=3.58163 time=59.05it/s +====================Eval==================== +epoch=22 global_step=8993 loss=1.39504 test_loss_avg=3.74543 acc=0.39844 test_acc_avg=0.14518 test_acc_top5_avg=0.78906 time=235.11it/s +epoch=22 global_step=8993 loss=0.61694 test_loss_avg=3.54581 acc=0.78906 test_acc_avg=0.21762 test_acc_top5_avg=0.69141 time=187.87it/s +epoch=22 global_step=8993 loss=5.28415 test_loss_avg=3.42831 acc=0.00000 test_acc_avg=0.26820 test_acc_top5_avg=0.65140 time=494.61it/s +curr_acc 0.2682 +BEST_ACC 0.2612 +curr_acc_top5 0.6514 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=7.25659 loss_avg=6.91623 acc=0.25781 acc_top1_avg=0.30804 acc_top5_avg=0.70647 lr=0.01000 gn=4.09129 time=57.45it/s +epoch=23 global_step=9050 loss=6.96214 loss_avg=6.88868 acc=0.28906 acc_top1_avg=0.30044 acc_top5_avg=0.73465 lr=0.01000 gn=3.56568 time=55.52it/s +epoch=23 global_step=9100 loss=7.04478 loss_avg=6.92668 acc=0.27344 acc_top1_avg=0.29636 acc_top5_avg=0.72963 lr=0.01000 gn=3.47514 time=60.20it/s +epoch=23 global_step=9150 loss=7.16187 loss_avg=6.90307 acc=0.26562 acc_top1_avg=0.29896 acc_top5_avg=0.73114 lr=0.01000 gn=3.40420 time=56.89it/s +epoch=23 global_step=9200 loss=6.82997 loss_avg=6.91956 acc=0.28906 acc_top1_avg=0.29706 acc_top5_avg=0.72698 lr=0.01000 gn=3.18167 time=54.83it/s +epoch=23 global_step=9250 loss=6.91339 loss_avg=6.92148 acc=0.28906 acc_top1_avg=0.29657 acc_top5_avg=0.72763 lr=0.01000 gn=2.41012 time=56.92it/s +epoch=23 global_step=9300 loss=7.23173 loss_avg=6.92174 acc=0.26562 acc_top1_avg=0.29667 acc_top5_avg=0.72712 lr=0.01000 gn=3.58769 time=58.45it/s +epoch=23 global_step=9350 loss=7.50117 loss_avg=6.91857 acc=0.23438 acc_top1_avg=0.29701 acc_top5_avg=0.72772 lr=0.01000 gn=3.68715 time=59.08it/s +====================Eval==================== +epoch=23 global_step=9384 loss=5.13206 test_loss_avg=4.12255 acc=0.00000 test_acc_avg=0.03788 test_acc_top5_avg=0.65223 time=235.93it/s +epoch=23 global_step=9384 loss=5.49468 test_loss_avg=3.37260 acc=0.00000 test_acc_avg=0.25791 test_acc_top5_avg=0.66673 time=834.69it/s +curr_acc 0.2579 +BEST_ACC 0.2682 +curr_acc_top5 0.6667 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=6.78121 loss_avg=6.92235 acc=0.29688 acc_top1_avg=0.29785 acc_top5_avg=0.72168 lr=0.01000 gn=3.30654 time=58.71it/s +epoch=24 global_step=9450 loss=7.19632 loss_avg=6.89413 acc=0.27344 acc_top1_avg=0.30161 acc_top5_avg=0.72976 lr=0.01000 gn=3.16976 time=55.24it/s +epoch=24 global_step=9500 loss=6.87299 loss_avg=6.92753 acc=0.28906 acc_top1_avg=0.29755 acc_top5_avg=0.73040 lr=0.01000 gn=4.55049 time=55.32it/s +epoch=24 global_step=9550 loss=7.34726 loss_avg=6.93831 acc=0.25000 acc_top1_avg=0.29532 acc_top5_avg=0.73033 lr=0.01000 gn=3.70356 time=56.66it/s +epoch=24 global_step=9600 loss=6.98172 loss_avg=6.92839 acc=0.28906 acc_top1_avg=0.29640 acc_top5_avg=0.73076 lr=0.01000 gn=5.00104 time=54.74it/s +epoch=24 global_step=9650 loss=6.98563 loss_avg=6.92829 acc=0.28906 acc_top1_avg=0.29673 acc_top5_avg=0.72959 lr=0.01000 gn=3.87838 time=64.05it/s +epoch=24 global_step=9700 loss=6.96967 loss_avg=6.92083 acc=0.28125 acc_top1_avg=0.29710 acc_top5_avg=0.73074 lr=0.01000 gn=2.70594 time=54.99it/s +epoch=24 global_step=9750 loss=6.20320 loss_avg=6.91018 acc=0.37500 acc_top1_avg=0.29828 acc_top5_avg=0.73064 lr=0.01000 gn=3.41643 time=61.53it/s +====================Eval==================== +epoch=24 global_step=9775 loss=5.20664 test_loss_avg=5.12381 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.73438 time=242.88it/s +epoch=24 global_step=9775 loss=4.74807 test_loss_avg=4.22500 acc=0.00000 test_acc_avg=0.07769 test_acc_top5_avg=0.58304 time=236.74it/s +epoch=24 global_step=9775 loss=5.87681 test_loss_avg=3.70826 acc=0.00000 test_acc_avg=0.21044 test_acc_top5_avg=0.59385 time=644.78it/s +curr_acc 0.2104 +BEST_ACC 0.2682 +curr_acc_top5 0.5938 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=7.19708 loss_avg=6.93092 acc=0.26562 acc_top1_avg=0.29906 acc_top5_avg=0.72094 lr=0.01000 gn=3.21536 time=63.12it/s +epoch=25 global_step=9850 loss=7.11583 loss_avg=6.95383 acc=0.27344 acc_top1_avg=0.29469 acc_top5_avg=0.72604 lr=0.01000 gn=3.57901 time=58.96it/s +epoch=25 global_step=9900 loss=7.12218 loss_avg=6.92030 acc=0.28906 acc_top1_avg=0.29775 acc_top5_avg=0.72856 lr=0.01000 gn=4.34815 time=57.38it/s +epoch=25 global_step=9950 loss=7.13535 loss_avg=6.92508 acc=0.27344 acc_top1_avg=0.29723 acc_top5_avg=0.72875 lr=0.01000 gn=2.89062 time=61.25it/s +epoch=25 global_step=10000 loss=7.10999 loss_avg=6.92344 acc=0.27344 acc_top1_avg=0.29753 acc_top5_avg=0.72979 lr=0.01000 gn=3.50705 time=63.10it/s +epoch=25 global_step=10050 loss=7.25600 loss_avg=6.92944 acc=0.25781 acc_top1_avg=0.29659 acc_top5_avg=0.72815 lr=0.01000 gn=4.58339 time=65.10it/s +epoch=25 global_step=10100 loss=7.47929 loss_avg=6.93164 acc=0.23438 acc_top1_avg=0.29630 acc_top5_avg=0.72651 lr=0.01000 gn=3.24294 time=57.60it/s +epoch=25 global_step=10150 loss=6.91121 loss_avg=6.91568 acc=0.30469 acc_top1_avg=0.29817 acc_top5_avg=0.72848 lr=0.01000 gn=4.12627 time=59.18it/s +====================Eval==================== +epoch=25 global_step=10166 loss=4.96331 test_loss_avg=4.58893 acc=0.00000 test_acc_avg=0.00438 test_acc_top5_avg=0.74969 time=234.78it/s +epoch=25 global_step=10166 loss=5.00825 test_loss_avg=3.48097 acc=0.00000 test_acc_avg=0.25031 test_acc_top5_avg=0.70490 time=240.04it/s +epoch=25 global_step=10166 loss=5.04729 test_loss_avg=3.56465 acc=0.00000 test_acc_avg=0.23764 test_acc_top5_avg=0.68335 time=612.22it/s +curr_acc 0.2376 +BEST_ACC 0.2682 +curr_acc_top5 0.6833 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=7.16691 loss_avg=6.85862 acc=0.27344 acc_top1_avg=0.30446 acc_top5_avg=0.73070 lr=0.01000 gn=3.81308 time=56.83it/s +epoch=26 global_step=10250 loss=6.47879 loss_avg=6.84380 acc=0.37500 acc_top1_avg=0.30655 acc_top5_avg=0.72945 lr=0.01000 gn=4.23485 time=62.70it/s +epoch=26 global_step=10300 loss=6.39018 loss_avg=6.84791 acc=0.34375 acc_top1_avg=0.30626 acc_top5_avg=0.72761 lr=0.01000 gn=3.43692 time=56.94it/s +epoch=26 global_step=10350 loss=7.00090 loss_avg=6.87974 acc=0.29688 acc_top1_avg=0.30222 acc_top5_avg=0.72754 lr=0.01000 gn=4.30446 time=51.95it/s +epoch=26 global_step=10400 loss=6.91727 loss_avg=6.89390 acc=0.29688 acc_top1_avg=0.30051 acc_top5_avg=0.72750 lr=0.01000 gn=4.65702 time=57.63it/s +epoch=26 global_step=10450 loss=7.51801 loss_avg=6.89587 acc=0.25000 acc_top1_avg=0.29998 acc_top5_avg=0.72959 lr=0.01000 gn=3.60063 time=57.35it/s +epoch=26 global_step=10500 loss=7.35238 loss_avg=6.90290 acc=0.23438 acc_top1_avg=0.29884 acc_top5_avg=0.72841 lr=0.01000 gn=3.09442 time=51.20it/s +epoch=26 global_step=10550 loss=7.59778 loss_avg=6.90148 acc=0.22656 acc_top1_avg=0.29919 acc_top5_avg=0.72833 lr=0.01000 gn=2.63053 time=61.91it/s +====================Eval==================== +epoch=26 global_step=10557 loss=2.43594 test_loss_avg=4.42264 acc=0.49219 test_acc_avg=0.08781 test_acc_top5_avg=0.65999 time=237.57it/s +epoch=26 global_step=10557 loss=4.74420 test_loss_avg=3.72296 acc=0.00000 test_acc_avg=0.23655 test_acc_top5_avg=0.68799 time=496.72it/s +curr_acc 0.2366 +BEST_ACC 0.2682 +curr_acc_top5 0.6880 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=7.35453 loss_avg=6.92652 acc=0.27344 acc_top1_avg=0.29524 acc_top5_avg=0.72456 lr=0.01000 gn=3.29413 time=56.06it/s +epoch=27 global_step=10650 loss=6.95310 loss_avg=6.88924 acc=0.31250 acc_top1_avg=0.30015 acc_top5_avg=0.73101 lr=0.01000 gn=4.08860 time=63.39it/s +epoch=27 global_step=10700 loss=6.47107 loss_avg=6.89857 acc=0.33594 acc_top1_avg=0.29928 acc_top5_avg=0.72820 lr=0.01000 gn=4.70739 time=56.41it/s +epoch=27 global_step=10750 loss=6.02053 loss_avg=6.89125 acc=0.42188 acc_top1_avg=0.29943 acc_top5_avg=0.72806 lr=0.01000 gn=4.27157 time=56.82it/s +epoch=27 global_step=10800 loss=7.23448 loss_avg=6.87966 acc=0.25781 acc_top1_avg=0.30138 acc_top5_avg=0.73036 lr=0.01000 gn=4.34635 time=54.10it/s +epoch=27 global_step=10850 loss=7.10665 loss_avg=6.89253 acc=0.28125 acc_top1_avg=0.30002 acc_top5_avg=0.72934 lr=0.01000 gn=4.03424 time=56.73it/s +epoch=27 global_step=10900 loss=6.71848 loss_avg=6.89363 acc=0.31250 acc_top1_avg=0.29981 acc_top5_avg=0.72934 lr=0.01000 gn=3.00483 time=54.21it/s +====================Eval==================== +epoch=27 global_step=10948 loss=4.62962 test_loss_avg=3.95849 acc=0.00000 test_acc_avg=0.03493 test_acc_top5_avg=0.85708 time=236.99it/s +epoch=27 global_step=10948 loss=0.13156 test_loss_avg=3.48974 acc=0.95312 test_acc_avg=0.23169 test_acc_top5_avg=0.75140 time=229.12it/s +epoch=27 global_step=10948 loss=5.05377 test_loss_avg=3.51130 acc=0.00000 test_acc_avg=0.23596 test_acc_top5_avg=0.69956 time=822.74it/s +curr_acc 0.2360 +BEST_ACC 0.2682 +curr_acc_top5 0.6996 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=7.59655 loss_avg=7.10591 acc=0.21094 acc_top1_avg=0.27344 acc_top5_avg=0.66406 lr=0.01000 gn=2.97848 time=58.03it/s +epoch=28 global_step=11000 loss=7.51433 loss_avg=6.90072 acc=0.22656 acc_top1_avg=0.29988 acc_top5_avg=0.71559 lr=0.01000 gn=2.76832 time=58.12it/s +epoch=28 global_step=11050 loss=6.45344 loss_avg=6.87677 acc=0.32812 acc_top1_avg=0.30224 acc_top5_avg=0.72465 lr=0.01000 gn=3.50046 time=56.32it/s +epoch=28 global_step=11100 loss=6.92727 loss_avg=6.90063 acc=0.28906 acc_top1_avg=0.29960 acc_top5_avg=0.72600 lr=0.01000 gn=3.60110 time=53.29it/s +epoch=28 global_step=11150 loss=6.60548 loss_avg=6.89525 acc=0.30469 acc_top1_avg=0.30028 acc_top5_avg=0.72621 lr=0.01000 gn=4.11924 time=56.62it/s +epoch=28 global_step=11200 loss=6.66047 loss_avg=6.89650 acc=0.31250 acc_top1_avg=0.29982 acc_top5_avg=0.72613 lr=0.01000 gn=4.65625 time=59.08it/s +epoch=28 global_step=11250 loss=6.74907 loss_avg=6.90165 acc=0.31250 acc_top1_avg=0.29907 acc_top5_avg=0.72680 lr=0.01000 gn=3.69137 time=56.81it/s +epoch=28 global_step=11300 loss=6.95838 loss_avg=6.90543 acc=0.30469 acc_top1_avg=0.29885 acc_top5_avg=0.72627 lr=0.01000 gn=4.69840 time=59.89it/s +====================Eval==================== +epoch=28 global_step=11339 loss=5.31875 test_loss_avg=4.91830 acc=0.00000 test_acc_avg=0.01398 test_acc_top5_avg=0.63836 time=244.88it/s +epoch=28 global_step=11339 loss=5.84135 test_loss_avg=3.75085 acc=0.00000 test_acc_avg=0.23675 test_acc_top5_avg=0.66011 time=847.16it/s +curr_acc 0.2367 +BEST_ACC 0.2682 +curr_acc_top5 0.6601 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=6.82115 loss_avg=6.91376 acc=0.32812 acc_top1_avg=0.29830 acc_top5_avg=0.71165 lr=0.01000 gn=3.38060 time=60.22it/s +epoch=29 global_step=11400 loss=6.98515 loss_avg=6.89676 acc=0.28906 acc_top1_avg=0.29752 acc_top5_avg=0.72374 lr=0.01000 gn=3.53142 time=56.03it/s +epoch=29 global_step=11450 loss=6.93065 loss_avg=6.88893 acc=0.29688 acc_top1_avg=0.29990 acc_top5_avg=0.72494 lr=0.01000 gn=4.20592 time=57.62it/s +epoch=29 global_step=11500 loss=6.85144 loss_avg=6.89584 acc=0.30469 acc_top1_avg=0.29862 acc_top5_avg=0.72535 lr=0.01000 gn=4.00167 time=62.23it/s +epoch=29 global_step=11550 loss=7.15927 loss_avg=6.89266 acc=0.28125 acc_top1_avg=0.29917 acc_top5_avg=0.72686 lr=0.01000 gn=4.37777 time=56.79it/s +epoch=29 global_step=11600 loss=7.39436 loss_avg=6.88805 acc=0.24219 acc_top1_avg=0.29999 acc_top5_avg=0.72785 lr=0.01000 gn=3.75373 time=56.71it/s +epoch=29 global_step=11650 loss=7.35460 loss_avg=6.90116 acc=0.25781 acc_top1_avg=0.29848 acc_top5_avg=0.72847 lr=0.01000 gn=2.85472 time=58.72it/s +epoch=29 global_step=11700 loss=7.03171 loss_avg=6.89817 acc=0.28906 acc_top1_avg=0.29923 acc_top5_avg=0.72819 lr=0.01000 gn=3.82383 time=55.39it/s +====================Eval==================== +epoch=29 global_step=11730 loss=4.07242 test_loss_avg=5.17768 acc=0.00781 test_acc_avg=0.00087 test_acc_top5_avg=0.84896 time=236.70it/s +epoch=29 global_step=11730 loss=3.01231 test_loss_avg=4.29817 acc=0.32812 test_acc_avg=0.14936 test_acc_top5_avg=0.65506 time=239.50it/s +epoch=29 global_step=11730 loss=5.46186 test_loss_avg=3.97112 acc=0.00000 test_acc_avg=0.21934 test_acc_top5_avg=0.67277 time=508.71it/s +curr_acc 0.2193 +BEST_ACC 0.2682 +curr_acc_top5 0.6728 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=6.29571 loss_avg=6.73886 acc=0.36719 acc_top1_avg=0.31562 acc_top5_avg=0.74648 lr=0.01000 gn=3.08321 time=61.94it/s +epoch=30 global_step=11800 loss=6.77571 loss_avg=6.87190 acc=0.32031 acc_top1_avg=0.30290 acc_top5_avg=0.73549 lr=0.01000 gn=4.34674 time=56.73it/s +epoch=30 global_step=11850 loss=7.00819 loss_avg=6.89847 acc=0.29688 acc_top1_avg=0.29896 acc_top5_avg=0.73145 lr=0.01000 gn=3.43490 time=60.23it/s +epoch=30 global_step=11900 loss=6.87238 loss_avg=6.90054 acc=0.31250 acc_top1_avg=0.29949 acc_top5_avg=0.72904 lr=0.01000 gn=3.86542 time=55.13it/s +epoch=30 global_step=11950 loss=6.63845 loss_avg=6.91651 acc=0.32812 acc_top1_avg=0.29773 acc_top5_avg=0.72884 lr=0.01000 gn=3.99891 time=60.96it/s +epoch=30 global_step=12000 loss=7.16393 loss_avg=6.92712 acc=0.27344 acc_top1_avg=0.29612 acc_top5_avg=0.72633 lr=0.01000 gn=3.89910 time=59.65it/s +epoch=30 global_step=12050 loss=7.20423 loss_avg=6.91372 acc=0.25781 acc_top1_avg=0.29771 acc_top5_avg=0.72849 lr=0.01000 gn=3.27915 time=60.17it/s +epoch=30 global_step=12100 loss=6.84706 loss_avg=6.91029 acc=0.30469 acc_top1_avg=0.29818 acc_top5_avg=0.72836 lr=0.01000 gn=4.73359 time=55.73it/s +====================Eval==================== +epoch=30 global_step=12121 loss=5.81538 test_loss_avg=4.82865 acc=0.00000 test_acc_avg=0.00651 test_acc_top5_avg=0.64687 time=238.58it/s +epoch=30 global_step=12121 loss=5.84855 test_loss_avg=3.73238 acc=0.00000 test_acc_avg=0.25148 test_acc_top5_avg=0.66564 time=812.22it/s +curr_acc 0.2515 +BEST_ACC 0.2682 +curr_acc_top5 0.6656 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=7.16791 loss_avg=6.95817 acc=0.26562 acc_top1_avg=0.29122 acc_top5_avg=0.71848 lr=0.01000 gn=3.07129 time=61.29it/s +epoch=31 global_step=12200 loss=7.28568 loss_avg=6.91659 acc=0.25000 acc_top1_avg=0.29608 acc_top5_avg=0.72518 lr=0.01000 gn=4.30298 time=57.48it/s +epoch=31 global_step=12250 loss=6.27808 loss_avg=6.90339 acc=0.35938 acc_top1_avg=0.29784 acc_top5_avg=0.72432 lr=0.01000 gn=3.27919 time=56.89it/s +epoch=31 global_step=12300 loss=6.71933 loss_avg=6.91972 acc=0.32812 acc_top1_avg=0.29574 acc_top5_avg=0.72512 lr=0.01000 gn=3.90364 time=60.07it/s +epoch=31 global_step=12350 loss=7.08053 loss_avg=6.91663 acc=0.27344 acc_top1_avg=0.29575 acc_top5_avg=0.72704 lr=0.01000 gn=4.93059 time=47.90it/s +epoch=31 global_step=12400 loss=7.06808 loss_avg=6.91825 acc=0.28125 acc_top1_avg=0.29598 acc_top5_avg=0.72729 lr=0.01000 gn=2.54477 time=54.57it/s +epoch=31 global_step=12450 loss=6.78317 loss_avg=6.91023 acc=0.32031 acc_top1_avg=0.29725 acc_top5_avg=0.72652 lr=0.01000 gn=3.95567 time=56.51it/s +epoch=31 global_step=12500 loss=6.61065 loss_avg=6.90854 acc=0.32812 acc_top1_avg=0.29751 acc_top5_avg=0.72730 lr=0.01000 gn=2.98219 time=55.92it/s +====================Eval==================== +epoch=31 global_step=12512 loss=4.93260 test_loss_avg=4.93260 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.78906 time=213.52it/s +epoch=31 global_step=12512 loss=4.96876 test_loss_avg=4.04954 acc=0.00000 test_acc_avg=0.12194 test_acc_top5_avg=0.68153 time=232.13it/s +epoch=31 global_step=12512 loss=5.30229 test_loss_avg=3.51095 acc=0.00000 test_acc_avg=0.25415 test_acc_top5_avg=0.66139 time=829.41it/s +curr_acc 0.2542 +BEST_ACC 0.2682 +curr_acc_top5 0.6614 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=6.91645 loss_avg=6.96072 acc=0.29688 acc_top1_avg=0.28865 acc_top5_avg=0.73232 lr=0.01000 gn=3.90171 time=60.89it/s +epoch=32 global_step=12600 loss=6.86551 loss_avg=6.93636 acc=0.30469 acc_top1_avg=0.29368 acc_top5_avg=0.72914 lr=0.01000 gn=4.62945 time=52.83it/s +epoch=32 global_step=12650 loss=6.83323 loss_avg=6.92508 acc=0.32031 acc_top1_avg=0.29563 acc_top5_avg=0.72747 lr=0.01000 gn=3.59482 time=52.78it/s +epoch=32 global_step=12700 loss=7.27824 loss_avg=6.92662 acc=0.25781 acc_top1_avg=0.29525 acc_top5_avg=0.72839 lr=0.01000 gn=3.25738 time=56.72it/s +epoch=32 global_step=12750 loss=7.08925 loss_avg=6.90741 acc=0.28125 acc_top1_avg=0.29737 acc_top5_avg=0.73030 lr=0.01000 gn=3.97878 time=56.79it/s +epoch=32 global_step=12800 loss=6.75344 loss_avg=6.88907 acc=0.31250 acc_top1_avg=0.29991 acc_top5_avg=0.73229 lr=0.01000 gn=4.22065 time=59.14it/s +epoch=32 global_step=12850 loss=6.27575 loss_avg=6.88076 acc=0.37500 acc_top1_avg=0.30106 acc_top5_avg=0.73186 lr=0.01000 gn=4.51028 time=58.99it/s +epoch=32 global_step=12900 loss=6.78635 loss_avg=6.89360 acc=0.32031 acc_top1_avg=0.29959 acc_top5_avg=0.73079 lr=0.01000 gn=3.26981 time=61.14it/s +====================Eval==================== +epoch=32 global_step=12903 loss=4.39174 test_loss_avg=4.18575 acc=0.00000 test_acc_avg=0.00213 test_acc_top5_avg=0.72834 time=241.59it/s +epoch=32 global_step=12903 loss=4.59548 test_loss_avg=3.29361 acc=0.00000 test_acc_avg=0.24523 test_acc_top5_avg=0.75109 time=250.62it/s +epoch=32 global_step=12903 loss=4.88741 test_loss_avg=3.42425 acc=0.00000 test_acc_avg=0.22350 test_acc_top5_avg=0.74229 time=501.11it/s +curr_acc 0.2235 +BEST_ACC 0.2682 +curr_acc_top5 0.7423 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=6.79884 loss_avg=6.90925 acc=0.31250 acc_top1_avg=0.29970 acc_top5_avg=0.72557 lr=0.01000 gn=3.31262 time=51.31it/s +epoch=33 global_step=13000 loss=6.72369 loss_avg=6.88905 acc=0.32031 acc_top1_avg=0.30284 acc_top5_avg=0.72914 lr=0.01000 gn=4.44109 time=56.18it/s +epoch=33 global_step=13050 loss=6.80701 loss_avg=6.89966 acc=0.29688 acc_top1_avg=0.30150 acc_top5_avg=0.73028 lr=0.01000 gn=4.45083 time=56.35it/s +epoch=33 global_step=13100 loss=7.11355 loss_avg=6.90153 acc=0.27344 acc_top1_avg=0.30068 acc_top5_avg=0.72934 lr=0.01000 gn=2.59707 time=58.96it/s +epoch=33 global_step=13150 loss=7.09012 loss_avg=6.91661 acc=0.27344 acc_top1_avg=0.29849 acc_top5_avg=0.72821 lr=0.01000 gn=3.45025 time=61.60it/s +epoch=33 global_step=13200 loss=6.82039 loss_avg=6.89704 acc=0.32031 acc_top1_avg=0.30035 acc_top5_avg=0.72943 lr=0.01000 gn=3.76377 time=63.44it/s +epoch=33 global_step=13250 loss=6.99568 loss_avg=6.88629 acc=0.28125 acc_top1_avg=0.30115 acc_top5_avg=0.73007 lr=0.01000 gn=4.01057 time=55.04it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.37099 test_loss_avg=4.53849 acc=0.91406 test_acc_avg=0.08085 test_acc_top5_avg=0.58812 time=239.55it/s +epoch=33 global_step=13294 loss=5.43606 test_loss_avg=3.65123 acc=0.00000 test_acc_avg=0.24248 test_acc_top5_avg=0.67267 time=469.58it/s +curr_acc 0.2425 +BEST_ACC 0.2682 +curr_acc_top5 0.6727 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=7.25368 loss_avg=6.95918 acc=0.23438 acc_top1_avg=0.28516 acc_top5_avg=0.72526 lr=0.01000 gn=4.15191 time=56.59it/s +epoch=34 global_step=13350 loss=6.43647 loss_avg=6.90330 acc=0.36719 acc_top1_avg=0.29674 acc_top5_avg=0.72475 lr=0.01000 gn=4.31047 time=53.90it/s +epoch=34 global_step=13400 loss=6.85323 loss_avg=6.87665 acc=0.28906 acc_top1_avg=0.30115 acc_top5_avg=0.73069 lr=0.01000 gn=5.02272 time=53.09it/s +epoch=34 global_step=13450 loss=6.78597 loss_avg=6.90332 acc=0.29688 acc_top1_avg=0.29753 acc_top5_avg=0.72781 lr=0.01000 gn=3.91188 time=60.55it/s +epoch=34 global_step=13500 loss=6.61431 loss_avg=6.90287 acc=0.32812 acc_top1_avg=0.29775 acc_top5_avg=0.72633 lr=0.01000 gn=3.82637 time=57.67it/s +epoch=34 global_step=13550 loss=6.91475 loss_avg=6.89268 acc=0.29688 acc_top1_avg=0.29941 acc_top5_avg=0.72723 lr=0.01000 gn=3.16577 time=56.71it/s +epoch=34 global_step=13600 loss=6.85002 loss_avg=6.89509 acc=0.32031 acc_top1_avg=0.29910 acc_top5_avg=0.72863 lr=0.01000 gn=4.47222 time=64.53it/s +epoch=34 global_step=13650 loss=6.18020 loss_avg=6.88855 acc=0.37500 acc_top1_avg=0.30025 acc_top5_avg=0.73003 lr=0.01000 gn=3.52771 time=55.79it/s +====================Eval==================== +epoch=34 global_step=13685 loss=1.05349 test_loss_avg=3.20443 acc=0.58594 test_acc_avg=0.29129 test_acc_top5_avg=0.72042 time=239.59it/s +epoch=34 global_step=13685 loss=0.56940 test_loss_avg=3.74347 acc=0.86719 test_acc_avg=0.27295 test_acc_top5_avg=0.71045 time=254.77it/s +epoch=34 global_step=13685 loss=5.34247 test_loss_avg=3.69479 acc=0.00000 test_acc_avg=0.29045 test_acc_top5_avg=0.65783 time=658.96it/s +curr_acc 0.2904 +BEST_ACC 0.2682 +curr_acc_top5 0.6578 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=6.83888 loss_avg=6.87422 acc=0.29688 acc_top1_avg=0.30000 acc_top5_avg=0.72969 lr=0.01000 gn=3.60053 time=55.62it/s +epoch=35 global_step=13750 loss=7.24770 loss_avg=6.94029 acc=0.25781 acc_top1_avg=0.29507 acc_top5_avg=0.72308 lr=0.01000 gn=4.93059 time=62.26it/s +epoch=35 global_step=13800 loss=6.62717 loss_avg=6.91091 acc=0.32031 acc_top1_avg=0.29844 acc_top5_avg=0.72113 lr=0.01000 gn=3.37009 time=55.94it/s +epoch=35 global_step=13850 loss=6.58169 loss_avg=6.92569 acc=0.35156 acc_top1_avg=0.29683 acc_top5_avg=0.72235 lr=0.01000 gn=3.96855 time=53.87it/s +epoch=35 global_step=13900 loss=6.57861 loss_avg=6.90709 acc=0.33594 acc_top1_avg=0.29880 acc_top5_avg=0.72478 lr=0.01000 gn=3.43108 time=61.14it/s +epoch=35 global_step=13950 loss=6.25682 loss_avg=6.89867 acc=0.35938 acc_top1_avg=0.29982 acc_top5_avg=0.72509 lr=0.01000 gn=3.88324 time=53.40it/s +epoch=35 global_step=14000 loss=7.72565 loss_avg=6.89797 acc=0.21094 acc_top1_avg=0.30010 acc_top5_avg=0.72587 lr=0.01000 gn=4.25202 time=64.44it/s +epoch=35 global_step=14050 loss=6.63736 loss_avg=6.90500 acc=0.33594 acc_top1_avg=0.29940 acc_top5_avg=0.72671 lr=0.01000 gn=3.26567 time=49.95it/s +====================Eval==================== +epoch=35 global_step=14076 loss=5.71614 test_loss_avg=5.10899 acc=0.00000 test_acc_avg=0.01451 test_acc_top5_avg=0.65513 time=236.43it/s +epoch=35 global_step=14076 loss=6.17783 test_loss_avg=4.08042 acc=0.00000 test_acc_avg=0.22231 test_acc_top5_avg=0.67029 time=849.22it/s +curr_acc 0.2223 +BEST_ACC 0.2904 +curr_acc_top5 0.6703 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=6.14408 loss_avg=6.84708 acc=0.39062 acc_top1_avg=0.30306 acc_top5_avg=0.73763 lr=0.01000 gn=4.48497 time=59.95it/s +epoch=36 global_step=14150 loss=6.76879 loss_avg=6.84682 acc=0.32031 acc_top1_avg=0.30416 acc_top5_avg=0.73796 lr=0.01000 gn=3.71912 time=56.18it/s +epoch=36 global_step=14200 loss=7.07266 loss_avg=6.84660 acc=0.28906 acc_top1_avg=0.30481 acc_top5_avg=0.73425 lr=0.01000 gn=3.69141 time=55.19it/s +epoch=36 global_step=14250 loss=6.86090 loss_avg=6.86652 acc=0.29688 acc_top1_avg=0.30172 acc_top5_avg=0.73424 lr=0.01000 gn=3.64967 time=58.59it/s +epoch=36 global_step=14300 loss=6.55498 loss_avg=6.87076 acc=0.35156 acc_top1_avg=0.30193 acc_top5_avg=0.73427 lr=0.01000 gn=3.79257 time=61.50it/s +epoch=36 global_step=14350 loss=7.05278 loss_avg=6.88397 acc=0.28125 acc_top1_avg=0.30081 acc_top5_avg=0.73332 lr=0.01000 gn=4.89264 time=59.94it/s +epoch=36 global_step=14400 loss=6.71450 loss_avg=6.87789 acc=0.32812 acc_top1_avg=0.30138 acc_top5_avg=0.73293 lr=0.01000 gn=4.12573 time=55.52it/s +epoch=36 global_step=14450 loss=7.07511 loss_avg=6.88362 acc=0.28906 acc_top1_avg=0.30059 acc_top5_avg=0.73256 lr=0.01000 gn=3.94929 time=63.96it/s +====================Eval==================== +epoch=36 global_step=14467 loss=4.95999 test_loss_avg=4.96871 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.78776 time=201.09it/s +epoch=36 global_step=14467 loss=0.12884 test_loss_avg=4.18678 acc=0.96094 test_acc_avg=0.10282 test_acc_top5_avg=0.66699 time=219.08it/s +epoch=36 global_step=14467 loss=5.05603 test_loss_avg=3.57960 acc=0.00000 test_acc_avg=0.23714 test_acc_top5_avg=0.65981 time=855.98it/s +curr_acc 0.2371 +BEST_ACC 0.2904 +curr_acc_top5 0.6598 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=7.20114 loss_avg=6.88837 acc=0.27344 acc_top1_avg=0.29924 acc_top5_avg=0.72704 lr=0.01000 gn=3.61666 time=55.87it/s +epoch=37 global_step=14550 loss=6.76100 loss_avg=6.87418 acc=0.31250 acc_top1_avg=0.30130 acc_top5_avg=0.72487 lr=0.01000 gn=3.62771 time=56.12it/s +epoch=37 global_step=14600 loss=6.85534 loss_avg=6.88706 acc=0.29688 acc_top1_avg=0.30011 acc_top5_avg=0.72809 lr=0.01000 gn=3.67918 time=58.97it/s +epoch=37 global_step=14650 loss=6.61629 loss_avg=6.86718 acc=0.33594 acc_top1_avg=0.30281 acc_top5_avg=0.73032 lr=0.01000 gn=2.54260 time=57.37it/s +epoch=37 global_step=14700 loss=6.45948 loss_avg=6.87893 acc=0.32812 acc_top1_avg=0.30154 acc_top5_avg=0.73002 lr=0.01000 gn=3.68138 time=59.48it/s +epoch=37 global_step=14750 loss=6.82152 loss_avg=6.89113 acc=0.31250 acc_top1_avg=0.30024 acc_top5_avg=0.72979 lr=0.01000 gn=4.18070 time=60.40it/s +epoch=37 global_step=14800 loss=6.75813 loss_avg=6.89274 acc=0.30469 acc_top1_avg=0.30000 acc_top5_avg=0.72867 lr=0.01000 gn=3.46304 time=54.80it/s +epoch=37 global_step=14850 loss=6.70246 loss_avg=6.89443 acc=0.30469 acc_top1_avg=0.29983 acc_top5_avg=0.72885 lr=0.01000 gn=3.28263 time=56.52it/s +====================Eval==================== +epoch=37 global_step=14858 loss=5.25893 test_loss_avg=5.02465 acc=0.00000 test_acc_avg=0.00550 test_acc_top5_avg=0.70891 time=159.86it/s +epoch=37 global_step=14858 loss=6.34109 test_loss_avg=3.86573 acc=0.00000 test_acc_avg=0.23153 test_acc_top5_avg=0.65442 time=241.14it/s +epoch=37 global_step=14858 loss=6.23071 test_loss_avg=3.92603 acc=0.00000 test_acc_avg=0.22567 test_acc_top5_avg=0.63805 time=492.06it/s +curr_acc 0.2257 +BEST_ACC 0.2904 +curr_acc_top5 0.6381 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=7.28560 loss_avg=6.80759 acc=0.24219 acc_top1_avg=0.30934 acc_top5_avg=0.73456 lr=0.01000 gn=3.44484 time=53.81it/s +epoch=38 global_step=14950 loss=6.43816 loss_avg=6.86295 acc=0.33594 acc_top1_avg=0.30375 acc_top5_avg=0.72953 lr=0.01000 gn=4.28422 time=59.86it/s +epoch=38 global_step=15000 loss=6.92994 loss_avg=6.86952 acc=0.28906 acc_top1_avg=0.30287 acc_top5_avg=0.72931 lr=0.01000 gn=3.92583 time=56.27it/s +epoch=38 global_step=15050 loss=6.95745 loss_avg=6.86704 acc=0.28906 acc_top1_avg=0.30326 acc_top5_avg=0.72778 lr=0.01000 gn=3.79001 time=56.78it/s +epoch=38 global_step=15100 loss=6.28445 loss_avg=6.88915 acc=0.36719 acc_top1_avg=0.30101 acc_top5_avg=0.72760 lr=0.01000 gn=4.61184 time=62.74it/s +epoch=38 global_step=15150 loss=7.51240 loss_avg=6.89880 acc=0.24219 acc_top1_avg=0.29984 acc_top5_avg=0.72924 lr=0.01000 gn=4.70334 time=60.14it/s +epoch=38 global_step=15200 loss=6.56537 loss_avg=6.89794 acc=0.34375 acc_top1_avg=0.29994 acc_top5_avg=0.72850 lr=0.01000 gn=4.02504 time=56.06it/s +====================Eval==================== +epoch=38 global_step=15249 loss=4.93102 test_loss_avg=4.44956 acc=0.00000 test_acc_avg=0.06868 test_acc_top5_avg=0.71094 time=229.57it/s +epoch=38 global_step=15249 loss=6.25287 test_loss_avg=3.90183 acc=0.00000 test_acc_avg=0.20995 test_acc_top5_avg=0.66851 time=511.63it/s +curr_acc 0.2099 +BEST_ACC 0.2904 +curr_acc_top5 0.6685 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=6.56025 loss_avg=6.56025 acc=0.32812 acc_top1_avg=0.32812 acc_top5_avg=0.71875 lr=0.01000 gn=4.36661 time=51.77it/s +epoch=39 global_step=15300 loss=7.09510 loss_avg=6.86390 acc=0.27344 acc_top1_avg=0.29963 acc_top5_avg=0.72488 lr=0.01000 gn=3.08130 time=52.20it/s +epoch=39 global_step=15350 loss=6.53848 loss_avg=6.87259 acc=0.34375 acc_top1_avg=0.30020 acc_top5_avg=0.72579 lr=0.01000 gn=4.61298 time=55.82it/s +epoch=39 global_step=15400 loss=6.63248 loss_avg=6.86363 acc=0.32812 acc_top1_avg=0.30231 acc_top5_avg=0.72729 lr=0.01000 gn=3.99924 time=61.02it/s +epoch=39 global_step=15450 loss=7.40112 loss_avg=6.88263 acc=0.25000 acc_top1_avg=0.29975 acc_top5_avg=0.72683 lr=0.01000 gn=3.94331 time=56.57it/s +epoch=39 global_step=15500 loss=6.97135 loss_avg=6.88208 acc=0.28906 acc_top1_avg=0.29996 acc_top5_avg=0.72827 lr=0.01000 gn=3.84568 time=57.21it/s +epoch=39 global_step=15550 loss=6.82904 loss_avg=6.88053 acc=0.30469 acc_top1_avg=0.30004 acc_top5_avg=0.72833 lr=0.01000 gn=3.58386 time=59.66it/s +epoch=39 global_step=15600 loss=6.35102 loss_avg=6.88486 acc=0.35156 acc_top1_avg=0.29959 acc_top5_avg=0.72986 lr=0.01000 gn=3.84253 time=60.00it/s +====================Eval==================== +epoch=39 global_step=15640 loss=4.62527 test_loss_avg=4.37246 acc=0.00000 test_acc_avg=0.01604 test_acc_top5_avg=0.84498 time=239.41it/s +epoch=39 global_step=15640 loss=1.52641 test_loss_avg=3.53115 acc=0.59375 test_acc_avg=0.25159 test_acc_top5_avg=0.74706 time=219.30it/s +epoch=39 global_step=15640 loss=4.94230 test_loss_avg=3.67098 acc=0.00000 test_acc_avg=0.23012 test_acc_top5_avg=0.69254 time=473.02it/s +curr_acc 0.2301 +BEST_ACC 0.2904 +curr_acc_top5 0.6925 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=6.84122 loss_avg=6.89661 acc=0.29688 acc_top1_avg=0.30469 acc_top5_avg=0.72109 lr=0.00100 gn=2.77650 time=61.99it/s +epoch=40 global_step=15700 loss=6.65200 loss_avg=6.82541 acc=0.32812 acc_top1_avg=0.30898 acc_top5_avg=0.72995 lr=0.00100 gn=3.36699 time=53.88it/s +epoch=40 global_step=15750 loss=6.86453 loss_avg=6.80881 acc=0.30469 acc_top1_avg=0.30945 acc_top5_avg=0.73104 lr=0.00100 gn=4.14416 time=53.44it/s +epoch=40 global_step=15800 loss=7.27386 loss_avg=6.77741 acc=0.26562 acc_top1_avg=0.31245 acc_top5_avg=0.73604 lr=0.00100 gn=3.72342 time=63.14it/s +epoch=40 global_step=15850 loss=7.22321 loss_avg=6.75099 acc=0.25000 acc_top1_avg=0.31525 acc_top5_avg=0.73724 lr=0.00100 gn=3.56317 time=50.97it/s +epoch=40 global_step=15900 loss=6.97245 loss_avg=6.74547 acc=0.30469 acc_top1_avg=0.31617 acc_top5_avg=0.73897 lr=0.00100 gn=4.18265 time=64.53it/s +epoch=40 global_step=15950 loss=6.15196 loss_avg=6.74621 acc=0.38281 acc_top1_avg=0.31626 acc_top5_avg=0.74088 lr=0.00100 gn=2.82123 time=56.35it/s +epoch=40 global_step=16000 loss=6.41321 loss_avg=6.74119 acc=0.34375 acc_top1_avg=0.31701 acc_top5_avg=0.74167 lr=0.00100 gn=4.55651 time=55.96it/s +====================Eval==================== +epoch=40 global_step=16031 loss=1.15463 test_loss_avg=4.36156 acc=0.70312 test_acc_avg=0.04063 test_acc_top5_avg=0.67910 time=233.89it/s +epoch=40 global_step=16031 loss=5.26569 test_loss_avg=3.36543 acc=0.00000 test_acc_avg=0.27235 test_acc_top5_avg=0.68246 time=800.75it/s +curr_acc 0.2723 +BEST_ACC 0.2904 +curr_acc_top5 0.6825 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=6.93200 loss_avg=6.65516 acc=0.29688 acc_top1_avg=0.32525 acc_top5_avg=0.74301 lr=0.00100 gn=3.01755 time=63.81it/s +epoch=41 global_step=16100 loss=6.33405 loss_avg=6.67340 acc=0.37500 acc_top1_avg=0.32495 acc_top5_avg=0.74060 lr=0.00100 gn=5.06035 time=54.31it/s +epoch=41 global_step=16150 loss=6.72924 loss_avg=6.66861 acc=0.32031 acc_top1_avg=0.32543 acc_top5_avg=0.74409 lr=0.00100 gn=4.28678 time=59.90it/s +epoch=41 global_step=16200 loss=6.44578 loss_avg=6.67684 acc=0.35156 acc_top1_avg=0.32355 acc_top5_avg=0.74450 lr=0.00100 gn=3.10730 time=58.72it/s +epoch=41 global_step=16250 loss=7.24470 loss_avg=6.67550 acc=0.24219 acc_top1_avg=0.32374 acc_top5_avg=0.74700 lr=0.00100 gn=3.82724 time=57.89it/s +epoch=41 global_step=16300 loss=7.27840 loss_avg=6.67470 acc=0.26562 acc_top1_avg=0.32388 acc_top5_avg=0.74649 lr=0.00100 gn=4.23646 time=57.04it/s +epoch=41 global_step=16350 loss=6.38047 loss_avg=6.67020 acc=0.38281 acc_top1_avg=0.32421 acc_top5_avg=0.74647 lr=0.00100 gn=4.56067 time=56.08it/s +epoch=41 global_step=16400 loss=6.40807 loss_avg=6.67090 acc=0.35156 acc_top1_avg=0.32400 acc_top5_avg=0.74695 lr=0.00100 gn=4.01322 time=48.04it/s +====================Eval==================== +epoch=41 global_step=16422 loss=2.79959 test_loss_avg=4.27768 acc=0.11719 test_acc_avg=0.04830 test_acc_top5_avg=0.83736 time=223.94it/s +epoch=41 global_step=16422 loss=0.11586 test_loss_avg=3.60305 acc=0.96875 test_acc_avg=0.20876 test_acc_top5_avg=0.72349 time=236.51it/s +epoch=41 global_step=16422 loss=5.25869 test_loss_avg=3.37102 acc=0.00000 test_acc_avg=0.27354 test_acc_top5_avg=0.68631 time=839.20it/s +curr_acc 0.2735 +BEST_ACC 0.2904 +curr_acc_top5 0.6863 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=6.73054 loss_avg=6.63878 acc=0.32812 acc_top1_avg=0.32952 acc_top5_avg=0.74135 lr=0.00100 gn=3.82376 time=63.86it/s +epoch=42 global_step=16500 loss=5.78500 loss_avg=6.65413 acc=0.41406 acc_top1_avg=0.32582 acc_top5_avg=0.74900 lr=0.00100 gn=4.53632 time=60.10it/s +epoch=42 global_step=16550 loss=6.63247 loss_avg=6.65558 acc=0.33594 acc_top1_avg=0.32562 acc_top5_avg=0.75018 lr=0.00100 gn=4.58464 time=56.46it/s +epoch=42 global_step=16600 loss=6.79927 loss_avg=6.65606 acc=0.32812 acc_top1_avg=0.32606 acc_top5_avg=0.74873 lr=0.00100 gn=6.78349 time=55.69it/s +epoch=42 global_step=16650 loss=6.43297 loss_avg=6.66101 acc=0.32812 acc_top1_avg=0.32528 acc_top5_avg=0.74805 lr=0.00100 gn=4.18167 time=59.44it/s +epoch=42 global_step=16700 loss=6.72569 loss_avg=6.64902 acc=0.32031 acc_top1_avg=0.32664 acc_top5_avg=0.74947 lr=0.00100 gn=3.78843 time=60.63it/s +epoch=42 global_step=16750 loss=6.80699 loss_avg=6.64769 acc=0.31250 acc_top1_avg=0.32696 acc_top5_avg=0.75007 lr=0.00100 gn=3.98664 time=60.99it/s +epoch=42 global_step=16800 loss=6.61846 loss_avg=6.64653 acc=0.32031 acc_top1_avg=0.32699 acc_top5_avg=0.75041 lr=0.00100 gn=3.75690 time=59.90it/s +====================Eval==================== +epoch=42 global_step=16813 loss=5.31020 test_loss_avg=4.50210 acc=0.00000 test_acc_avg=0.02661 test_acc_top5_avg=0.68530 time=230.17it/s +epoch=42 global_step=16813 loss=5.09173 test_loss_avg=3.44124 acc=0.00000 test_acc_avg=0.26919 test_acc_top5_avg=0.69670 time=596.21it/s +curr_acc 0.2692 +BEST_ACC 0.2904 +curr_acc_top5 0.6967 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=6.85583 loss_avg=6.62377 acc=0.30469 acc_top1_avg=0.33150 acc_top5_avg=0.75338 lr=0.00100 gn=4.26927 time=47.16it/s +epoch=43 global_step=16900 loss=6.02404 loss_avg=6.61032 acc=0.40625 acc_top1_avg=0.33145 acc_top5_avg=0.75629 lr=0.00100 gn=4.29780 time=64.72it/s +epoch=43 global_step=16950 loss=6.85892 loss_avg=6.61458 acc=0.29688 acc_top1_avg=0.33115 acc_top5_avg=0.75530 lr=0.00100 gn=4.07019 time=57.66it/s +epoch=43 global_step=17000 loss=6.64682 loss_avg=6.61673 acc=0.30469 acc_top1_avg=0.33147 acc_top5_avg=0.75201 lr=0.00100 gn=4.42016 time=56.37it/s +epoch=43 global_step=17050 loss=6.19587 loss_avg=6.61436 acc=0.38281 acc_top1_avg=0.33155 acc_top5_avg=0.75297 lr=0.00100 gn=5.06793 time=55.89it/s +epoch=43 global_step=17100 loss=6.37640 loss_avg=6.60767 acc=0.34375 acc_top1_avg=0.33221 acc_top5_avg=0.75457 lr=0.00100 gn=5.08686 time=54.71it/s +epoch=43 global_step=17150 loss=6.66685 loss_avg=6.61374 acc=0.32812 acc_top1_avg=0.33144 acc_top5_avg=0.75290 lr=0.00100 gn=6.28972 time=52.24it/s +epoch=43 global_step=17200 loss=6.53396 loss_avg=6.61753 acc=0.34375 acc_top1_avg=0.33057 acc_top5_avg=0.75281 lr=0.00100 gn=4.39447 time=63.39it/s +====================Eval==================== +epoch=43 global_step=17204 loss=5.24886 test_loss_avg=5.26286 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.83854 time=240.32it/s +epoch=43 global_step=17204 loss=4.75366 test_loss_avg=4.03394 acc=0.00000 test_acc_avg=0.14062 test_acc_top5_avg=0.70430 time=238.03it/s +epoch=43 global_step=17204 loss=5.20934 test_loss_avg=3.41499 acc=0.00000 test_acc_avg=0.27977 test_acc_top5_avg=0.68691 time=453.68it/s +curr_acc 0.2798 +BEST_ACC 0.2904 +curr_acc_top5 0.6869 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=6.46110 loss_avg=6.63089 acc=0.34375 acc_top1_avg=0.32931 acc_top5_avg=0.75543 lr=0.00100 gn=5.43135 time=55.51it/s +epoch=44 global_step=17300 loss=6.98364 loss_avg=6.61605 acc=0.28906 acc_top1_avg=0.33089 acc_top5_avg=0.75326 lr=0.00100 gn=4.73229 time=55.36it/s +epoch=44 global_step=17350 loss=6.53438 loss_avg=6.59925 acc=0.35156 acc_top1_avg=0.33219 acc_top5_avg=0.75251 lr=0.00100 gn=6.03702 time=55.72it/s +epoch=44 global_step=17400 loss=6.33365 loss_avg=6.59886 acc=0.35938 acc_top1_avg=0.33251 acc_top5_avg=0.75279 lr=0.00100 gn=4.20457 time=57.38it/s +epoch=44 global_step=17450 loss=6.75489 loss_avg=6.58701 acc=0.31250 acc_top1_avg=0.33346 acc_top5_avg=0.75359 lr=0.00100 gn=4.11175 time=56.50it/s +epoch=44 global_step=17500 loss=7.04462 loss_avg=6.58902 acc=0.27344 acc_top1_avg=0.33301 acc_top5_avg=0.75351 lr=0.00100 gn=4.48404 time=56.02it/s +epoch=44 global_step=17550 loss=6.82369 loss_avg=6.60038 acc=0.30469 acc_top1_avg=0.33160 acc_top5_avg=0.75345 lr=0.00100 gn=5.90182 time=51.81it/s +====================Eval==================== +epoch=44 global_step=17595 loss=4.90927 test_loss_avg=4.02665 acc=0.00000 test_acc_avg=0.07910 test_acc_top5_avg=0.85482 time=234.15it/s +epoch=44 global_step=17595 loss=5.33503 test_loss_avg=3.19191 acc=0.00000 test_acc_avg=0.30680 test_acc_top5_avg=0.72076 time=246.40it/s +epoch=44 global_step=17595 loss=5.25802 test_loss_avg=3.31121 acc=0.00000 test_acc_avg=0.28738 test_acc_top5_avg=0.68374 time=740.00it/s +curr_acc 0.2874 +BEST_ACC 0.2904 +curr_acc_top5 0.6837 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=7.25637 loss_avg=6.72196 acc=0.25000 acc_top1_avg=0.31406 acc_top5_avg=0.74687 lr=0.00100 gn=4.42174 time=57.64it/s +epoch=45 global_step=17650 loss=7.09146 loss_avg=6.62520 acc=0.30469 acc_top1_avg=0.32855 acc_top5_avg=0.75099 lr=0.00100 gn=6.62283 time=59.84it/s +epoch=45 global_step=17700 loss=6.55905 loss_avg=6.60715 acc=0.35156 acc_top1_avg=0.33177 acc_top5_avg=0.75342 lr=0.00100 gn=4.64946 time=55.24it/s +epoch=45 global_step=17750 loss=6.91743 loss_avg=6.58343 acc=0.30469 acc_top1_avg=0.33427 acc_top5_avg=0.75439 lr=0.00100 gn=5.79784 time=54.66it/s +epoch=45 global_step=17800 loss=6.55663 loss_avg=6.58989 acc=0.33594 acc_top1_avg=0.33319 acc_top5_avg=0.75469 lr=0.00100 gn=3.70612 time=56.03it/s +epoch=45 global_step=17850 loss=6.54660 loss_avg=6.57539 acc=0.32031 acc_top1_avg=0.33474 acc_top5_avg=0.75662 lr=0.00100 gn=6.38572 time=61.50it/s +epoch=45 global_step=17900 loss=7.31133 loss_avg=6.59222 acc=0.25781 acc_top1_avg=0.33320 acc_top5_avg=0.75561 lr=0.00100 gn=6.39234 time=55.63it/s +epoch=45 global_step=17950 loss=6.57354 loss_avg=6.59275 acc=0.32031 acc_top1_avg=0.33310 acc_top5_avg=0.75517 lr=0.00100 gn=4.62448 time=61.55it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.71309 test_loss_avg=3.99071 acc=0.80469 test_acc_avg=0.13941 test_acc_top5_avg=0.71927 time=241.18it/s +epoch=45 global_step=17986 loss=5.17673 test_loss_avg=3.31250 acc=0.00000 test_acc_avg=0.28501 test_acc_top5_avg=0.69432 time=501.05it/s +curr_acc 0.2850 +BEST_ACC 0.2904 +curr_acc_top5 0.6943 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=6.35102 loss_avg=6.54928 acc=0.35938 acc_top1_avg=0.34040 acc_top5_avg=0.75614 lr=0.00100 gn=6.66704 time=57.14it/s +epoch=46 global_step=18050 loss=5.46858 loss_avg=6.45797 acc=0.46875 acc_top1_avg=0.34875 acc_top5_avg=0.76196 lr=0.00100 gn=7.20983 time=56.90it/s +epoch=46 global_step=18100 loss=6.76703 loss_avg=6.50862 acc=0.30469 acc_top1_avg=0.34197 acc_top5_avg=0.75795 lr=0.00100 gn=6.88794 time=54.71it/s +epoch=46 global_step=18150 loss=7.21420 loss_avg=6.54172 acc=0.27344 acc_top1_avg=0.33875 acc_top5_avg=0.75615 lr=0.00100 gn=5.06773 time=55.19it/s +epoch=46 global_step=18200 loss=6.23321 loss_avg=6.52741 acc=0.36719 acc_top1_avg=0.34061 acc_top5_avg=0.75719 lr=0.00100 gn=6.42321 time=53.80it/s +epoch=46 global_step=18250 loss=6.71541 loss_avg=6.54367 acc=0.29688 acc_top1_avg=0.33848 acc_top5_avg=0.75731 lr=0.00100 gn=6.51088 time=62.70it/s +epoch=46 global_step=18300 loss=6.52667 loss_avg=6.55030 acc=0.34375 acc_top1_avg=0.33773 acc_top5_avg=0.75759 lr=0.00100 gn=7.22252 time=48.64it/s +epoch=46 global_step=18350 loss=7.34959 loss_avg=6.56786 acc=0.25781 acc_top1_avg=0.33579 acc_top5_avg=0.75543 lr=0.00100 gn=5.44671 time=51.44it/s +====================Eval==================== +epoch=46 global_step=18377 loss=3.15338 test_loss_avg=3.48220 acc=0.25000 test_acc_avg=0.15820 test_acc_top5_avg=0.89893 time=208.14it/s +epoch=46 global_step=18377 loss=0.15144 test_loss_avg=3.24157 acc=0.93750 test_acc_avg=0.29238 test_acc_top5_avg=0.77356 time=243.68it/s +epoch=46 global_step=18377 loss=4.91902 test_loss_avg=3.26417 acc=0.00000 test_acc_avg=0.29598 test_acc_top5_avg=0.72172 time=494.55it/s +curr_acc 0.2960 +BEST_ACC 0.2904 +curr_acc_top5 0.7217 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=6.49338 loss_avg=6.48493 acc=0.35156 acc_top1_avg=0.34545 acc_top5_avg=0.73947 lr=0.00100 gn=5.51089 time=52.82it/s +epoch=47 global_step=18450 loss=6.40252 loss_avg=6.55777 acc=0.35938 acc_top1_avg=0.33744 acc_top5_avg=0.75086 lr=0.00100 gn=6.48370 time=60.27it/s +epoch=47 global_step=18500 loss=6.17904 loss_avg=6.55984 acc=0.37500 acc_top1_avg=0.33606 acc_top5_avg=0.75495 lr=0.00100 gn=5.81843 time=54.95it/s +epoch=47 global_step=18550 loss=6.75915 loss_avg=6.57286 acc=0.31250 acc_top1_avg=0.33494 acc_top5_avg=0.75253 lr=0.00100 gn=6.48351 time=61.98it/s +epoch=47 global_step=18600 loss=6.81912 loss_avg=6.54875 acc=0.30469 acc_top1_avg=0.33804 acc_top5_avg=0.75554 lr=0.00100 gn=5.45310 time=56.08it/s +epoch=47 global_step=18650 loss=6.74552 loss_avg=6.55735 acc=0.31250 acc_top1_avg=0.33691 acc_top5_avg=0.75632 lr=0.00100 gn=6.25940 time=55.88it/s +epoch=47 global_step=18700 loss=6.39699 loss_avg=6.55833 acc=0.35938 acc_top1_avg=0.33707 acc_top5_avg=0.75687 lr=0.00100 gn=5.99000 time=52.47it/s +epoch=47 global_step=18750 loss=7.24407 loss_avg=6.55920 acc=0.27344 acc_top1_avg=0.33688 acc_top5_avg=0.75689 lr=0.00100 gn=7.54265 time=53.90it/s +====================Eval==================== +epoch=47 global_step=18768 loss=5.17470 test_loss_avg=4.32824 acc=0.00000 test_acc_avg=0.08066 test_acc_top5_avg=0.68138 time=240.76it/s +epoch=47 global_step=18768 loss=5.16347 test_loss_avg=3.31451 acc=0.00000 test_acc_avg=0.30152 test_acc_top5_avg=0.68384 time=647.57it/s +curr_acc 0.3015 +BEST_ACC 0.2960 +curr_acc_top5 0.6838 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=6.58695 loss_avg=6.50283 acc=0.32031 acc_top1_avg=0.34570 acc_top5_avg=0.76611 lr=0.00100 gn=6.79721 time=54.82it/s +epoch=48 global_step=18850 loss=6.74641 loss_avg=6.57129 acc=0.31250 acc_top1_avg=0.33479 acc_top5_avg=0.75638 lr=0.00100 gn=6.16242 time=63.17it/s +epoch=48 global_step=18900 loss=6.54105 loss_avg=6.53722 acc=0.32812 acc_top1_avg=0.33937 acc_top5_avg=0.75870 lr=0.00100 gn=6.70890 time=55.37it/s +epoch=48 global_step=18950 loss=6.27077 loss_avg=6.53144 acc=0.36719 acc_top1_avg=0.33954 acc_top5_avg=0.75893 lr=0.00100 gn=6.42378 time=55.27it/s +epoch=48 global_step=19000 loss=6.49685 loss_avg=6.54259 acc=0.34375 acc_top1_avg=0.33803 acc_top5_avg=0.75899 lr=0.00100 gn=5.65405 time=51.69it/s +epoch=48 global_step=19050 loss=6.57975 loss_avg=6.56219 acc=0.34375 acc_top1_avg=0.33555 acc_top5_avg=0.75695 lr=0.00100 gn=6.37515 time=52.31it/s +epoch=48 global_step=19100 loss=6.22717 loss_avg=6.55424 acc=0.38281 acc_top1_avg=0.33648 acc_top5_avg=0.75642 lr=0.00100 gn=7.58723 time=55.40it/s +epoch=48 global_step=19150 loss=5.63465 loss_avg=6.53549 acc=0.42188 acc_top1_avg=0.33878 acc_top5_avg=0.75765 lr=0.00100 gn=6.33890 time=58.24it/s +====================Eval==================== +epoch=48 global_step=19159 loss=4.79518 test_loss_avg=5.13112 acc=0.02344 test_acc_avg=0.00293 test_acc_top5_avg=0.84570 time=115.76it/s +epoch=48 global_step=19159 loss=0.24077 test_loss_avg=3.81265 acc=0.92969 test_acc_avg=0.18279 test_acc_top5_avg=0.73411 time=222.89it/s +epoch=48 global_step=19159 loss=5.16118 test_loss_avg=3.39978 acc=0.00000 test_acc_avg=0.28293 test_acc_top5_avg=0.71311 time=502.85it/s +curr_acc 0.2829 +BEST_ACC 0.3015 +curr_acc_top5 0.7131 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=6.93137 loss_avg=6.44393 acc=0.31250 acc_top1_avg=0.34870 acc_top5_avg=0.75553 lr=0.00100 gn=6.53571 time=57.09it/s +epoch=49 global_step=19250 loss=6.57788 loss_avg=6.48510 acc=0.34375 acc_top1_avg=0.34564 acc_top5_avg=0.75618 lr=0.00100 gn=8.34636 time=57.00it/s +epoch=49 global_step=19300 loss=6.62864 loss_avg=6.49781 acc=0.32031 acc_top1_avg=0.34447 acc_top5_avg=0.75571 lr=0.00100 gn=7.16499 time=63.22it/s +epoch=49 global_step=19350 loss=6.80911 loss_avg=6.49623 acc=0.31250 acc_top1_avg=0.34490 acc_top5_avg=0.75659 lr=0.00100 gn=7.66453 time=56.67it/s +epoch=49 global_step=19400 loss=6.94865 loss_avg=6.51960 acc=0.30469 acc_top1_avg=0.34236 acc_top5_avg=0.75457 lr=0.00100 gn=7.73111 time=54.97it/s +epoch=49 global_step=19450 loss=5.91654 loss_avg=6.51341 acc=0.39844 acc_top1_avg=0.34257 acc_top5_avg=0.75626 lr=0.00100 gn=6.14561 time=57.12it/s +epoch=49 global_step=19500 loss=5.85286 loss_avg=6.52622 acc=0.41406 acc_top1_avg=0.34098 acc_top5_avg=0.75646 lr=0.00100 gn=8.21193 time=51.58it/s +epoch=49 global_step=19550 loss=6.70383 loss_avg=6.52037 acc=0.32500 acc_top1_avg=0.34172 acc_top5_avg=0.75697 lr=0.00100 gn=9.22742 time=72.82it/s +====================Eval==================== +epoch=49 global_step=19550 loss=4.95694 test_loss_avg=4.16642 acc=0.00000 test_acc_avg=0.08163 test_acc_top5_avg=0.73869 time=235.99it/s +epoch=49 global_step=19550 loss=5.43796 test_loss_avg=3.36389 acc=0.00000 test_acc_avg=0.28936 test_acc_top5_avg=0.70599 time=859.66it/s +epoch=49 global_step=19550 loss=5.43796 test_loss_avg=3.36389 acc=0.00000 test_acc_avg=0.28936 test_acc_top5_avg=0.70599 time=859.66it/s +curr_acc 0.2894 +BEST_ACC 0.3015 +curr_acc_top5 0.7060 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=6.39339 loss_avg=6.46903 acc=0.35938 acc_top1_avg=0.34656 acc_top5_avg=0.76109 lr=0.00100 gn=8.12131 time=53.11it/s +epoch=50 global_step=19650 loss=6.82832 loss_avg=6.50474 acc=0.31250 acc_top1_avg=0.34313 acc_top5_avg=0.76102 lr=0.00100 gn=7.40912 time=54.50it/s +epoch=50 global_step=19700 loss=7.02197 loss_avg=6.49216 acc=0.29688 acc_top1_avg=0.34510 acc_top5_avg=0.76083 lr=0.00100 gn=10.11995 time=61.34it/s +epoch=50 global_step=19750 loss=6.74083 loss_avg=6.49857 acc=0.32812 acc_top1_avg=0.34430 acc_top5_avg=0.76113 lr=0.00100 gn=6.75136 time=53.30it/s +epoch=50 global_step=19800 loss=6.06635 loss_avg=6.49056 acc=0.39844 acc_top1_avg=0.34506 acc_top5_avg=0.76094 lr=0.00100 gn=7.75232 time=62.01it/s +epoch=50 global_step=19850 loss=7.06742 loss_avg=6.49630 acc=0.28125 acc_top1_avg=0.34404 acc_top5_avg=0.75984 lr=0.00100 gn=7.08381 time=59.28it/s +epoch=50 global_step=19900 loss=6.67700 loss_avg=6.50080 acc=0.32812 acc_top1_avg=0.34328 acc_top5_avg=0.75964 lr=0.00100 gn=7.53296 time=54.19it/s +====================Eval==================== +epoch=50 global_step=19941 loss=4.62905 test_loss_avg=3.78455 acc=0.00000 test_acc_avg=0.17156 test_acc_top5_avg=0.73687 time=235.19it/s +epoch=50 global_step=19941 loss=5.70714 test_loss_avg=3.28125 acc=0.00000 test_acc_avg=0.29707 test_acc_top5_avg=0.71371 time=503.64it/s +curr_acc 0.2971 +BEST_ACC 0.3015 +curr_acc_top5 0.7137 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=6.80655 loss_avg=6.62769 acc=0.32031 acc_top1_avg=0.33247 acc_top5_avg=0.74479 lr=0.00100 gn=10.07409 time=58.64it/s +epoch=51 global_step=20000 loss=6.56100 loss_avg=6.49921 acc=0.33594 acc_top1_avg=0.34269 acc_top5_avg=0.75530 lr=0.00100 gn=8.72436 time=55.80it/s +epoch=51 global_step=20050 loss=6.98488 loss_avg=6.49447 acc=0.29688 acc_top1_avg=0.34375 acc_top5_avg=0.75939 lr=0.00100 gn=9.34707 time=36.52it/s +epoch=51 global_step=20100 loss=6.45824 loss_avg=6.48940 acc=0.35938 acc_top1_avg=0.34522 acc_top5_avg=0.76091 lr=0.00100 gn=7.15736 time=55.45it/s +epoch=51 global_step=20150 loss=6.33791 loss_avg=6.48428 acc=0.35938 acc_top1_avg=0.34566 acc_top5_avg=0.76013 lr=0.00100 gn=7.10246 time=55.41it/s +epoch=51 global_step=20200 loss=5.96438 loss_avg=6.48195 acc=0.41406 acc_top1_avg=0.34619 acc_top5_avg=0.75974 lr=0.00100 gn=8.94723 time=62.89it/s +epoch=51 global_step=20250 loss=6.58762 loss_avg=6.48160 acc=0.33594 acc_top1_avg=0.34613 acc_top5_avg=0.75989 lr=0.00100 gn=7.35112 time=53.05it/s +epoch=51 global_step=20300 loss=6.41812 loss_avg=6.47801 acc=0.35938 acc_top1_avg=0.34612 acc_top5_avg=0.76016 lr=0.00100 gn=8.82000 time=43.84it/s +====================Eval==================== +epoch=51 global_step=20332 loss=4.58563 test_loss_avg=3.89934 acc=0.00000 test_acc_avg=0.10714 test_acc_top5_avg=0.89993 time=240.96it/s +epoch=51 global_step=20332 loss=3.73340 test_loss_avg=3.19665 acc=0.29688 test_acc_avg=0.32009 test_acc_top5_avg=0.78136 time=253.05it/s +epoch=51 global_step=20332 loss=5.50786 test_loss_avg=3.40676 acc=0.00000 test_acc_avg=0.28768 test_acc_top5_avg=0.71608 time=597.82it/s +curr_acc 0.2877 +BEST_ACC 0.3015 +curr_acc_top5 0.7161 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=6.60999 loss_avg=6.40229 acc=0.31250 acc_top1_avg=0.35547 acc_top5_avg=0.75781 lr=0.00100 gn=10.87811 time=62.00it/s +epoch=52 global_step=20400 loss=6.54097 loss_avg=6.47330 acc=0.32812 acc_top1_avg=0.34524 acc_top5_avg=0.76011 lr=0.00100 gn=5.96766 time=59.66it/s +epoch=52 global_step=20450 loss=6.12516 loss_avg=6.49168 acc=0.39062 acc_top1_avg=0.34428 acc_top5_avg=0.75715 lr=0.00100 gn=8.18405 time=57.13it/s +epoch=52 global_step=20500 loss=6.21819 loss_avg=6.48840 acc=0.37500 acc_top1_avg=0.34496 acc_top5_avg=0.75484 lr=0.00100 gn=7.91894 time=35.33it/s +epoch=52 global_step=20550 loss=6.48415 loss_avg=6.49061 acc=0.35156 acc_top1_avg=0.34461 acc_top5_avg=0.75548 lr=0.00100 gn=6.92078 time=54.49it/s +epoch=52 global_step=20600 loss=6.08963 loss_avg=6.47381 acc=0.38281 acc_top1_avg=0.34667 acc_top5_avg=0.75673 lr=0.00100 gn=8.04010 time=54.44it/s +epoch=52 global_step=20650 loss=6.15527 loss_avg=6.46807 acc=0.38281 acc_top1_avg=0.34751 acc_top5_avg=0.75808 lr=0.00100 gn=9.50913 time=58.61it/s +epoch=52 global_step=20700 loss=6.93214 loss_avg=6.46167 acc=0.29688 acc_top1_avg=0.34838 acc_top5_avg=0.75866 lr=0.00100 gn=7.78355 time=55.11it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.67704 test_loss_avg=4.26294 acc=0.83594 test_acc_avg=0.08966 test_acc_top5_avg=0.73121 time=234.82it/s +epoch=52 global_step=20723 loss=5.38849 test_loss_avg=3.33616 acc=0.00000 test_acc_avg=0.28273 test_acc_top5_avg=0.72735 time=506.01it/s +curr_acc 0.2827 +BEST_ACC 0.3015 +curr_acc_top5 0.7274 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=6.73199 loss_avg=6.39015 acc=0.32031 acc_top1_avg=0.35214 acc_top5_avg=0.77633 lr=0.00100 gn=9.75499 time=53.35it/s +epoch=53 global_step=20800 loss=6.28355 loss_avg=6.40903 acc=0.35938 acc_top1_avg=0.35227 acc_top5_avg=0.77039 lr=0.00100 gn=9.53384 time=54.18it/s +epoch=53 global_step=20850 loss=6.36292 loss_avg=6.41224 acc=0.35156 acc_top1_avg=0.35242 acc_top5_avg=0.76852 lr=0.00100 gn=9.25573 time=54.75it/s +epoch=53 global_step=20900 loss=6.64150 loss_avg=6.41947 acc=0.32031 acc_top1_avg=0.35152 acc_top5_avg=0.76810 lr=0.00100 gn=8.71341 time=59.92it/s +epoch=53 global_step=20950 loss=6.59142 loss_avg=6.42040 acc=0.33594 acc_top1_avg=0.35163 acc_top5_avg=0.76618 lr=0.00100 gn=10.55522 time=56.39it/s +epoch=53 global_step=21000 loss=6.60198 loss_avg=6.42514 acc=0.32812 acc_top1_avg=0.35148 acc_top5_avg=0.76484 lr=0.00100 gn=8.52035 time=51.16it/s +epoch=53 global_step=21050 loss=6.34907 loss_avg=6.44247 acc=0.35938 acc_top1_avg=0.34936 acc_top5_avg=0.76218 lr=0.00100 gn=9.65974 time=53.65it/s +epoch=53 global_step=21100 loss=5.80024 loss_avg=6.44445 acc=0.42969 acc_top1_avg=0.34901 acc_top5_avg=0.76080 lr=0.00100 gn=10.49819 time=55.62it/s +====================Eval==================== +epoch=53 global_step=21114 loss=2.06329 test_loss_avg=3.87769 acc=0.33594 test_acc_avg=0.13221 test_acc_top5_avg=0.87500 time=221.28it/s +epoch=53 global_step=21114 loss=0.13166 test_loss_avg=3.42581 acc=0.96094 test_acc_avg=0.26637 test_acc_top5_avg=0.73748 time=139.68it/s +epoch=53 global_step=21114 loss=4.99740 test_loss_avg=3.29828 acc=0.00000 test_acc_avg=0.30044 test_acc_top5_avg=0.69709 time=493.91it/s +curr_acc 0.3004 +BEST_ACC 0.3015 +curr_acc_top5 0.6971 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=6.82031 loss_avg=6.46576 acc=0.32031 acc_top1_avg=0.34505 acc_top5_avg=0.74848 lr=0.00100 gn=11.69104 time=54.90it/s +epoch=54 global_step=21200 loss=6.29377 loss_avg=6.45074 acc=0.35156 acc_top1_avg=0.34947 acc_top5_avg=0.75836 lr=0.00100 gn=10.73209 time=63.19it/s +epoch=54 global_step=21250 loss=6.58697 loss_avg=6.43434 acc=0.33594 acc_top1_avg=0.35162 acc_top5_avg=0.75936 lr=0.00100 gn=9.10005 time=60.57it/s +epoch=54 global_step=21300 loss=6.22728 loss_avg=6.44173 acc=0.35938 acc_top1_avg=0.35064 acc_top5_avg=0.76016 lr=0.00100 gn=10.17646 time=55.90it/s +epoch=54 global_step=21350 loss=7.01516 loss_avg=6.44238 acc=0.28125 acc_top1_avg=0.35054 acc_top5_avg=0.75990 lr=0.00100 gn=10.44903 time=54.47it/s +epoch=54 global_step=21400 loss=5.98253 loss_avg=6.43189 acc=0.39844 acc_top1_avg=0.35123 acc_top5_avg=0.76106 lr=0.00100 gn=11.87131 time=42.31it/s +epoch=54 global_step=21450 loss=7.12740 loss_avg=6.42748 acc=0.27344 acc_top1_avg=0.35207 acc_top5_avg=0.76135 lr=0.00100 gn=8.13272 time=55.86it/s +epoch=54 global_step=21500 loss=6.64172 loss_avg=6.43245 acc=0.32031 acc_top1_avg=0.35126 acc_top5_avg=0.76010 lr=0.00100 gn=9.39879 time=59.95it/s +====================Eval==================== +epoch=54 global_step=21505 loss=5.08039 test_loss_avg=4.35875 acc=0.00000 test_acc_avg=0.06204 test_acc_top5_avg=0.70565 time=238.68it/s +epoch=54 global_step=21505 loss=5.64845 test_loss_avg=3.41696 acc=0.00000 test_acc_avg=0.28244 test_acc_top5_avg=0.72162 time=652.91it/s +curr_acc 0.2824 +BEST_ACC 0.3015 +curr_acc_top5 0.7216 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=7.20500 loss_avg=6.58002 acc=0.25781 acc_top1_avg=0.33316 acc_top5_avg=0.75434 lr=0.00100 gn=8.01333 time=53.70it/s +epoch=55 global_step=21600 loss=6.30006 loss_avg=6.48461 acc=0.38281 acc_top1_avg=0.34391 acc_top5_avg=0.75609 lr=0.00100 gn=9.95155 time=57.98it/s +epoch=55 global_step=21650 loss=6.27459 loss_avg=6.42923 acc=0.35938 acc_top1_avg=0.35081 acc_top5_avg=0.75851 lr=0.00100 gn=11.53412 time=62.22it/s +epoch=55 global_step=21700 loss=6.75108 loss_avg=6.41335 acc=0.32031 acc_top1_avg=0.35296 acc_top5_avg=0.76086 lr=0.00100 gn=10.88117 time=59.65it/s +epoch=55 global_step=21750 loss=6.45916 loss_avg=6.40809 acc=0.35938 acc_top1_avg=0.35386 acc_top5_avg=0.76103 lr=0.00100 gn=11.52058 time=58.38it/s +epoch=55 global_step=21800 loss=6.30600 loss_avg=6.40582 acc=0.35156 acc_top1_avg=0.35429 acc_top5_avg=0.76186 lr=0.00100 gn=9.06575 time=54.24it/s +epoch=55 global_step=21850 loss=6.78639 loss_avg=6.40368 acc=0.32812 acc_top1_avg=0.35503 acc_top5_avg=0.76316 lr=0.00100 gn=13.42913 time=55.67it/s +====================Eval==================== +epoch=55 global_step=21896 loss=5.00565 test_loss_avg=5.02470 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.83906 time=240.46it/s +epoch=55 global_step=21896 loss=3.38776 test_loss_avg=3.96275 acc=0.30469 test_acc_avg=0.15881 test_acc_top5_avg=0.70753 time=233.39it/s +epoch=55 global_step=21896 loss=5.94666 test_loss_avg=3.40913 acc=0.00000 test_acc_avg=0.29440 test_acc_top5_avg=0.69947 time=527.32it/s +curr_acc 0.2944 +BEST_ACC 0.3015 +curr_acc_top5 0.6995 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=6.52684 loss_avg=6.63747 acc=0.34375 acc_top1_avg=0.33594 acc_top5_avg=0.78516 lr=0.00100 gn=12.07073 time=56.39it/s +epoch=56 global_step=21950 loss=6.18277 loss_avg=6.35671 acc=0.37500 acc_top1_avg=0.36024 acc_top5_avg=0.77358 lr=0.00100 gn=12.34765 time=46.36it/s +epoch=56 global_step=22000 loss=6.33993 loss_avg=6.38422 acc=0.36719 acc_top1_avg=0.35705 acc_top5_avg=0.76645 lr=0.00100 gn=9.48384 time=54.07it/s +epoch=56 global_step=22050 loss=6.75933 loss_avg=6.41567 acc=0.30469 acc_top1_avg=0.35379 acc_top5_avg=0.76527 lr=0.00100 gn=11.52880 time=57.76it/s +epoch=56 global_step=22100 loss=6.33615 loss_avg=6.40309 acc=0.35938 acc_top1_avg=0.35547 acc_top5_avg=0.76417 lr=0.00100 gn=12.29290 time=53.79it/s +epoch=56 global_step=22150 loss=6.56064 loss_avg=6.39941 acc=0.34375 acc_top1_avg=0.35584 acc_top5_avg=0.76366 lr=0.00100 gn=10.53816 time=57.21it/s +epoch=56 global_step=22200 loss=6.64866 loss_avg=6.39980 acc=0.34375 acc_top1_avg=0.35555 acc_top5_avg=0.76203 lr=0.00100 gn=12.82685 time=59.96it/s +epoch=56 global_step=22250 loss=6.25937 loss_avg=6.39676 acc=0.35156 acc_top1_avg=0.35584 acc_top5_avg=0.76053 lr=0.00100 gn=11.72772 time=57.38it/s +====================Eval==================== +epoch=56 global_step=22287 loss=4.82853 test_loss_avg=4.10909 acc=0.00000 test_acc_avg=0.09285 test_acc_top5_avg=0.81190 time=228.91it/s +epoch=56 global_step=22287 loss=4.78573 test_loss_avg=3.32900 acc=0.00000 test_acc_avg=0.29616 test_acc_top5_avg=0.73715 time=247.20it/s +epoch=56 global_step=22287 loss=5.04651 test_loss_avg=3.38836 acc=0.00000 test_acc_avg=0.28491 test_acc_top5_avg=0.71717 time=787.81it/s +curr_acc 0.2849 +BEST_ACC 0.3015 +curr_acc_top5 0.7172 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=6.22848 loss_avg=6.33792 acc=0.37500 acc_top1_avg=0.36058 acc_top5_avg=0.76262 lr=0.00100 gn=12.40151 time=58.08it/s +epoch=57 global_step=22350 loss=6.48585 loss_avg=6.41025 acc=0.35156 acc_top1_avg=0.35404 acc_top5_avg=0.75955 lr=0.00100 gn=9.71255 time=56.52it/s +epoch=57 global_step=22400 loss=6.56622 loss_avg=6.37222 acc=0.34375 acc_top1_avg=0.35827 acc_top5_avg=0.75816 lr=0.00100 gn=10.28047 time=43.13it/s +epoch=57 global_step=22450 loss=6.39165 loss_avg=6.37901 acc=0.35938 acc_top1_avg=0.35727 acc_top5_avg=0.75877 lr=0.00100 gn=11.47471 time=52.78it/s +epoch=57 global_step=22500 loss=6.19245 loss_avg=6.38592 acc=0.37500 acc_top1_avg=0.35681 acc_top5_avg=0.75829 lr=0.00100 gn=13.39911 time=56.28it/s +epoch=57 global_step=22550 loss=5.88637 loss_avg=6.38435 acc=0.42188 acc_top1_avg=0.35673 acc_top5_avg=0.75977 lr=0.00100 gn=13.10047 time=58.46it/s +epoch=57 global_step=22600 loss=7.43507 loss_avg=6.37925 acc=0.24219 acc_top1_avg=0.35793 acc_top5_avg=0.75993 lr=0.00100 gn=12.32036 time=53.73it/s +epoch=57 global_step=22650 loss=5.93065 loss_avg=6.36952 acc=0.40625 acc_top1_avg=0.35854 acc_top5_avg=0.76031 lr=0.00100 gn=12.01937 time=57.55it/s +====================Eval==================== +epoch=57 global_step=22678 loss=1.65973 test_loss_avg=3.85742 acc=0.57812 test_acc_avg=0.19066 test_acc_top5_avg=0.73321 time=87.01it/s +epoch=57 global_step=22678 loss=5.08878 test_loss_avg=3.33249 acc=0.00000 test_acc_avg=0.30083 test_acc_top5_avg=0.68790 time=795.43it/s +curr_acc 0.3008 +BEST_ACC 0.3015 +curr_acc_top5 0.6879 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=6.95909 loss_avg=6.47401 acc=0.29688 acc_top1_avg=0.34304 acc_top5_avg=0.74822 lr=0.00100 gn=10.60241 time=55.56it/s +epoch=58 global_step=22750 loss=6.54561 loss_avg=6.42208 acc=0.32812 acc_top1_avg=0.35069 acc_top5_avg=0.75694 lr=0.00100 gn=12.22030 time=56.77it/s +epoch=58 global_step=22800 loss=6.92183 loss_avg=6.37666 acc=0.28125 acc_top1_avg=0.35656 acc_top5_avg=0.75858 lr=0.00100 gn=11.71826 time=55.05it/s +epoch=58 global_step=22850 loss=5.93600 loss_avg=6.34570 acc=0.39844 acc_top1_avg=0.35978 acc_top5_avg=0.75858 lr=0.00100 gn=11.39233 time=61.76it/s +epoch=58 global_step=22900 loss=6.57018 loss_avg=6.34183 acc=0.32812 acc_top1_avg=0.36043 acc_top5_avg=0.76021 lr=0.00100 gn=9.87072 time=59.59it/s +epoch=58 global_step=22950 loss=5.95442 loss_avg=6.34035 acc=0.40625 acc_top1_avg=0.36087 acc_top5_avg=0.75988 lr=0.00100 gn=10.92818 time=63.77it/s +epoch=58 global_step=23000 loss=6.52415 loss_avg=6.35084 acc=0.33594 acc_top1_avg=0.35964 acc_top5_avg=0.75951 lr=0.00100 gn=12.90194 time=59.25it/s +epoch=58 global_step=23050 loss=6.58271 loss_avg=6.35313 acc=0.35156 acc_top1_avg=0.35938 acc_top5_avg=0.75872 lr=0.00100 gn=12.39182 time=60.76it/s +====================Eval==================== +epoch=58 global_step=23069 loss=4.89226 test_loss_avg=4.20552 acc=0.00000 test_acc_avg=0.10113 test_acc_top5_avg=0.87760 time=235.73it/s +epoch=58 global_step=23069 loss=0.19038 test_loss_avg=3.40696 acc=0.94531 test_acc_avg=0.28837 test_acc_top5_avg=0.78171 time=241.16it/s +epoch=58 global_step=23069 loss=4.79944 test_loss_avg=3.45286 acc=0.00000 test_acc_avg=0.27631 test_acc_top5_avg=0.73566 time=812.06it/s +curr_acc 0.2763 +BEST_ACC 0.3015 +curr_acc_top5 0.7357 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=6.08574 loss_avg=6.28742 acc=0.39062 acc_top1_avg=0.36694 acc_top5_avg=0.76663 lr=0.00100 gn=13.16305 time=53.16it/s +epoch=59 global_step=23150 loss=5.96439 loss_avg=6.31141 acc=0.39844 acc_top1_avg=0.36516 acc_top5_avg=0.76341 lr=0.00100 gn=10.87633 time=60.78it/s +epoch=59 global_step=23200 loss=6.63303 loss_avg=6.33911 acc=0.32812 acc_top1_avg=0.36218 acc_top5_avg=0.76002 lr=0.00100 gn=14.86857 time=61.89it/s +epoch=59 global_step=23250 loss=6.07975 loss_avg=6.32498 acc=0.38281 acc_top1_avg=0.36343 acc_top5_avg=0.75911 lr=0.00100 gn=11.05360 time=54.63it/s +epoch=59 global_step=23300 loss=7.22441 loss_avg=6.32262 acc=0.25781 acc_top1_avg=0.36374 acc_top5_avg=0.76042 lr=0.00100 gn=16.10679 time=56.42it/s +epoch=59 global_step=23350 loss=6.66039 loss_avg=6.32222 acc=0.32812 acc_top1_avg=0.36380 acc_top5_avg=0.76212 lr=0.00100 gn=12.83254 time=55.84it/s +epoch=59 global_step=23400 loss=5.20650 loss_avg=6.32007 acc=0.46875 acc_top1_avg=0.36402 acc_top5_avg=0.76244 lr=0.00100 gn=14.40977 time=59.67it/s +epoch=59 global_step=23450 loss=7.01996 loss_avg=6.33582 acc=0.28125 acc_top1_avg=0.36194 acc_top5_avg=0.76171 lr=0.00100 gn=13.25039 time=62.71it/s +====================Eval==================== +epoch=59 global_step=23460 loss=5.16309 test_loss_avg=4.07829 acc=0.00000 test_acc_avg=0.14904 test_acc_top5_avg=0.68990 time=240.84it/s +epoch=59 global_step=23460 loss=6.13938 test_loss_avg=3.26074 acc=0.00000 test_acc_avg=0.33178 test_acc_top5_avg=0.69620 time=818.24it/s +curr_acc 0.3318 +BEST_ACC 0.3015 +curr_acc_top5 0.6962 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=6.34129 loss_avg=6.31699 acc=0.36719 acc_top1_avg=0.36621 acc_top5_avg=0.76699 lr=0.00100 gn=12.78675 time=53.42it/s +epoch=60 global_step=23550 loss=6.33169 loss_avg=6.33173 acc=0.36719 acc_top1_avg=0.36345 acc_top5_avg=0.76146 lr=0.00100 gn=15.69322 time=59.56it/s +epoch=60 global_step=23600 loss=6.40558 loss_avg=6.33043 acc=0.35156 acc_top1_avg=0.36323 acc_top5_avg=0.75921 lr=0.00100 gn=10.87913 time=50.42it/s +epoch=60 global_step=23650 loss=6.75552 loss_avg=6.31739 acc=0.32031 acc_top1_avg=0.36472 acc_top5_avg=0.76003 lr=0.00100 gn=14.92234 time=60.30it/s +epoch=60 global_step=23700 loss=6.18982 loss_avg=6.32635 acc=0.36719 acc_top1_avg=0.36367 acc_top5_avg=0.75905 lr=0.00100 gn=8.27694 time=60.09it/s +epoch=60 global_step=23750 loss=6.16206 loss_avg=6.33086 acc=0.39062 acc_top1_avg=0.36304 acc_top5_avg=0.75938 lr=0.00100 gn=14.71611 time=57.84it/s +epoch=60 global_step=23800 loss=6.65606 loss_avg=6.32958 acc=0.33594 acc_top1_avg=0.36298 acc_top5_avg=0.75970 lr=0.00100 gn=13.74050 time=59.21it/s +epoch=60 global_step=23850 loss=6.78680 loss_avg=6.33216 acc=0.31250 acc_top1_avg=0.36246 acc_top5_avg=0.75921 lr=0.00100 gn=15.68362 time=54.48it/s +====================Eval==================== +epoch=60 global_step=23851 loss=3.21543 test_loss_avg=4.57977 acc=0.17188 test_acc_avg=0.04844 test_acc_top5_avg=0.81953 time=245.96it/s +epoch=60 global_step=23851 loss=0.14300 test_loss_avg=3.83871 acc=0.95312 test_acc_avg=0.21003 test_acc_top5_avg=0.72813 time=225.65it/s +epoch=60 global_step=23851 loss=4.66069 test_loss_avg=3.42287 acc=0.00000 test_acc_avg=0.28372 test_acc_top5_avg=0.71934 time=537.32it/s +curr_acc 0.2837 +BEST_ACC 0.3318 +curr_acc_top5 0.7193 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=6.35261 loss_avg=6.26706 acc=0.35938 acc_top1_avg=0.37085 acc_top5_avg=0.76770 lr=0.00100 gn=10.54091 time=52.04it/s +epoch=61 global_step=23950 loss=6.35195 loss_avg=6.28785 acc=0.36719 acc_top1_avg=0.36727 acc_top5_avg=0.76128 lr=0.00100 gn=17.43720 time=63.51it/s +epoch=61 global_step=24000 loss=5.53786 loss_avg=6.28558 acc=0.44531 acc_top1_avg=0.36761 acc_top5_avg=0.76064 lr=0.00100 gn=15.11697 time=56.68it/s +epoch=61 global_step=24050 loss=6.28667 loss_avg=6.30420 acc=0.36719 acc_top1_avg=0.36550 acc_top5_avg=0.76033 lr=0.00100 gn=14.72978 time=53.24it/s +epoch=61 global_step=24100 loss=6.76914 loss_avg=6.30706 acc=0.31250 acc_top1_avg=0.36521 acc_top5_avg=0.75875 lr=0.00100 gn=11.54748 time=62.56it/s +epoch=61 global_step=24150 loss=5.44610 loss_avg=6.30111 acc=0.48438 acc_top1_avg=0.36570 acc_top5_avg=0.75920 lr=0.00100 gn=18.50109 time=61.02it/s +epoch=61 global_step=24200 loss=6.19372 loss_avg=6.31304 acc=0.35938 acc_top1_avg=0.36448 acc_top5_avg=0.75972 lr=0.00100 gn=13.53186 time=51.97it/s +====================Eval==================== +epoch=61 global_step=24242 loss=4.94894 test_loss_avg=4.44106 acc=0.00000 test_acc_avg=0.06351 test_acc_top5_avg=0.68750 time=223.21it/s +epoch=61 global_step=24242 loss=5.18589 test_loss_avg=3.41421 acc=0.00000 test_acc_avg=0.28085 test_acc_top5_avg=0.70876 time=495.14it/s +curr_acc 0.2809 +BEST_ACC 0.3318 +curr_acc_top5 0.7088 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=5.89978 loss_avg=6.45808 acc=0.41406 acc_top1_avg=0.35156 acc_top5_avg=0.75098 lr=0.00100 gn=20.36315 time=59.41it/s +epoch=62 global_step=24300 loss=6.01561 loss_avg=6.26246 acc=0.38281 acc_top1_avg=0.36853 acc_top5_avg=0.76320 lr=0.00100 gn=12.58110 time=55.05it/s +epoch=62 global_step=24350 loss=6.12483 loss_avg=6.28009 acc=0.39062 acc_top1_avg=0.36646 acc_top5_avg=0.76005 lr=0.00100 gn=18.29947 time=54.37it/s +epoch=62 global_step=24400 loss=6.45197 loss_avg=6.28739 acc=0.35156 acc_top1_avg=0.36674 acc_top5_avg=0.75564 lr=0.00100 gn=15.39763 time=55.42it/s +epoch=62 global_step=24450 loss=6.59843 loss_avg=6.29981 acc=0.32812 acc_top1_avg=0.36569 acc_top5_avg=0.75571 lr=0.00100 gn=17.84109 time=54.04it/s +epoch=62 global_step=24500 loss=6.43367 loss_avg=6.31532 acc=0.35938 acc_top1_avg=0.36395 acc_top5_avg=0.75560 lr=0.00100 gn=14.76149 time=52.65it/s +epoch=62 global_step=24550 loss=6.25057 loss_avg=6.31913 acc=0.37500 acc_top1_avg=0.36414 acc_top5_avg=0.75632 lr=0.00100 gn=16.83838 time=46.59it/s +epoch=62 global_step=24600 loss=6.19110 loss_avg=6.31025 acc=0.37500 acc_top1_avg=0.36511 acc_top5_avg=0.75659 lr=0.00100 gn=13.67389 time=51.70it/s +====================Eval==================== +epoch=62 global_step=24633 loss=4.78579 test_loss_avg=4.72529 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.70312 time=224.57it/s +epoch=62 global_step=24633 loss=4.90041 test_loss_avg=3.81958 acc=0.00000 test_acc_avg=0.18149 test_acc_top5_avg=0.72776 time=230.39it/s +epoch=62 global_step=24633 loss=5.20751 test_loss_avg=3.24979 acc=0.00000 test_acc_avg=0.30914 test_acc_top5_avg=0.72557 time=500.27it/s +curr_acc 0.3091 +BEST_ACC 0.3318 +curr_acc_top5 0.7256 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=6.44254 loss_avg=6.41300 acc=0.35156 acc_top1_avg=0.35340 acc_top5_avg=0.75781 lr=0.00100 gn=15.55660 time=55.28it/s +epoch=63 global_step=24700 loss=6.20126 loss_avg=6.27992 acc=0.37500 acc_top1_avg=0.36847 acc_top5_avg=0.75968 lr=0.00100 gn=19.77955 time=57.23it/s +epoch=63 global_step=24750 loss=5.97514 loss_avg=6.28607 acc=0.39062 acc_top1_avg=0.36739 acc_top5_avg=0.75821 lr=0.00100 gn=15.46440 time=56.62it/s +epoch=63 global_step=24800 loss=6.20957 loss_avg=6.28723 acc=0.39062 acc_top1_avg=0.36709 acc_top5_avg=0.75875 lr=0.00100 gn=16.01294 time=52.34it/s +epoch=63 global_step=24850 loss=5.77277 loss_avg=6.27366 acc=0.42969 acc_top1_avg=0.36834 acc_top5_avg=0.76012 lr=0.00100 gn=12.94147 time=54.45it/s +epoch=63 global_step=24900 loss=5.88348 loss_avg=6.27978 acc=0.41406 acc_top1_avg=0.36839 acc_top5_avg=0.75963 lr=0.00100 gn=14.29491 time=55.06it/s +epoch=63 global_step=24950 loss=6.72900 loss_avg=6.28535 acc=0.32812 acc_top1_avg=0.36795 acc_top5_avg=0.75969 lr=0.00100 gn=15.73312 time=58.54it/s +epoch=63 global_step=25000 loss=6.20286 loss_avg=6.28926 acc=0.37500 acc_top1_avg=0.36714 acc_top5_avg=0.75886 lr=0.00100 gn=16.43926 time=55.52it/s +====================Eval==================== +epoch=63 global_step=25024 loss=4.76523 test_loss_avg=4.02905 acc=0.00000 test_acc_avg=0.12228 test_acc_top5_avg=0.88485 time=138.97it/s +epoch=63 global_step=25024 loss=4.74604 test_loss_avg=3.17818 acc=0.00000 test_acc_avg=0.32053 test_acc_top5_avg=0.76445 time=250.11it/s +epoch=63 global_step=25024 loss=4.92296 test_loss_avg=3.30328 acc=0.00000 test_acc_avg=0.29618 test_acc_top5_avg=0.72884 time=815.22it/s +curr_acc 0.2962 +BEST_ACC 0.3318 +curr_acc_top5 0.7288 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=5.96972 loss_avg=6.10428 acc=0.41406 acc_top1_avg=0.38642 acc_top5_avg=0.76142 lr=0.00100 gn=17.19925 time=63.18it/s +epoch=64 global_step=25100 loss=6.47179 loss_avg=6.17343 acc=0.35156 acc_top1_avg=0.38045 acc_top5_avg=0.76007 lr=0.00100 gn=16.01463 time=59.05it/s +epoch=64 global_step=25150 loss=6.22914 loss_avg=6.16112 acc=0.37500 acc_top1_avg=0.38182 acc_top5_avg=0.76184 lr=0.00100 gn=14.27624 time=60.03it/s +epoch=64 global_step=25200 loss=6.11673 loss_avg=6.17942 acc=0.39062 acc_top1_avg=0.37948 acc_top5_avg=0.76225 lr=0.00100 gn=17.19333 time=53.36it/s +epoch=64 global_step=25250 loss=6.13455 loss_avg=6.21083 acc=0.36719 acc_top1_avg=0.37583 acc_top5_avg=0.76044 lr=0.00100 gn=14.99114 time=57.41it/s +epoch=64 global_step=25300 loss=6.81903 loss_avg=6.23406 acc=0.29688 acc_top1_avg=0.37288 acc_top5_avg=0.76061 lr=0.00100 gn=15.80325 time=63.13it/s +epoch=64 global_step=25350 loss=6.56631 loss_avg=6.25580 acc=0.32031 acc_top1_avg=0.37037 acc_top5_avg=0.75999 lr=0.00100 gn=12.72841 time=54.78it/s +epoch=64 global_step=25400 loss=7.03282 loss_avg=6.26512 acc=0.28125 acc_top1_avg=0.36927 acc_top5_avg=0.75966 lr=0.00100 gn=11.28090 time=62.70it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.93394 test_loss_avg=3.98439 acc=0.76562 test_acc_avg=0.14631 test_acc_top5_avg=0.73615 time=226.69it/s +epoch=64 global_step=25415 loss=5.20257 test_loss_avg=3.31229 acc=0.00000 test_acc_avg=0.29460 test_acc_top5_avg=0.70273 time=844.77it/s +curr_acc 0.2946 +BEST_ACC 0.3318 +curr_acc_top5 0.7027 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=6.45921 loss_avg=6.23424 acc=0.32812 acc_top1_avg=0.37254 acc_top5_avg=0.76071 lr=0.00100 gn=13.99223 time=59.16it/s +epoch=65 global_step=25500 loss=5.77460 loss_avg=6.18177 acc=0.41406 acc_top1_avg=0.37675 acc_top5_avg=0.76379 lr=0.00100 gn=15.26893 time=54.69it/s +epoch=65 global_step=25550 loss=5.60700 loss_avg=6.18633 acc=0.41406 acc_top1_avg=0.37726 acc_top5_avg=0.76256 lr=0.00100 gn=18.74571 time=60.36it/s +epoch=65 global_step=25600 loss=6.76183 loss_avg=6.21716 acc=0.30469 acc_top1_avg=0.37382 acc_top5_avg=0.76014 lr=0.00100 gn=12.49454 time=57.29it/s +epoch=65 global_step=25650 loss=6.01683 loss_avg=6.23809 acc=0.39062 acc_top1_avg=0.37171 acc_top5_avg=0.75991 lr=0.00100 gn=13.63227 time=43.48it/s +epoch=65 global_step=25700 loss=6.41316 loss_avg=6.24651 acc=0.35156 acc_top1_avg=0.37094 acc_top5_avg=0.76025 lr=0.00100 gn=15.67554 time=60.10it/s +epoch=65 global_step=25750 loss=6.33528 loss_avg=6.24139 acc=0.35938 acc_top1_avg=0.37136 acc_top5_avg=0.75968 lr=0.00100 gn=16.12790 time=55.81it/s +epoch=65 global_step=25800 loss=5.73577 loss_avg=6.24745 acc=0.42969 acc_top1_avg=0.37078 acc_top5_avg=0.76090 lr=0.00100 gn=18.11829 time=62.09it/s +====================Eval==================== +epoch=65 global_step=25806 loss=3.08981 test_loss_avg=4.06700 acc=0.25781 test_acc_avg=0.12500 test_acc_top5_avg=0.90365 time=238.90it/s +epoch=65 global_step=25806 loss=0.12359 test_loss_avg=3.46045 acc=0.97656 test_acc_avg=0.27091 test_acc_top5_avg=0.75409 time=237.97it/s +epoch=65 global_step=25806 loss=5.00357 test_loss_avg=3.40867 acc=0.00000 test_acc_avg=0.28610 test_acc_top5_avg=0.72458 time=501.11it/s +curr_acc 0.2861 +BEST_ACC 0.3318 +curr_acc_top5 0.7246 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=5.57074 loss_avg=6.16603 acc=0.44531 acc_top1_avg=0.37926 acc_top5_avg=0.76438 lr=0.00100 gn=13.47377 time=57.41it/s +epoch=66 global_step=25900 loss=6.33912 loss_avg=6.22220 acc=0.35156 acc_top1_avg=0.37434 acc_top5_avg=0.75831 lr=0.00100 gn=18.36418 time=61.00it/s +epoch=66 global_step=25950 loss=6.21381 loss_avg=6.22477 acc=0.37500 acc_top1_avg=0.37516 acc_top5_avg=0.75987 lr=0.00100 gn=19.76933 time=57.93it/s +epoch=66 global_step=26000 loss=6.15079 loss_avg=6.25632 acc=0.38281 acc_top1_avg=0.37130 acc_top5_avg=0.75765 lr=0.00100 gn=24.56299 time=57.91it/s +epoch=66 global_step=26050 loss=6.37983 loss_avg=6.24653 acc=0.35156 acc_top1_avg=0.37212 acc_top5_avg=0.75752 lr=0.00100 gn=16.46626 time=57.23it/s +epoch=66 global_step=26100 loss=6.70603 loss_avg=6.23071 acc=0.32812 acc_top1_avg=0.37388 acc_top5_avg=0.75946 lr=0.00100 gn=19.27275 time=54.25it/s +epoch=66 global_step=26150 loss=6.17415 loss_avg=6.22947 acc=0.39844 acc_top1_avg=0.37396 acc_top5_avg=0.75856 lr=0.00100 gn=23.63037 time=55.05it/s +====================Eval==================== +epoch=66 global_step=26197 loss=5.08892 test_loss_avg=4.66538 acc=0.00000 test_acc_avg=0.04839 test_acc_top5_avg=0.67057 time=114.93it/s +epoch=66 global_step=26197 loss=4.34741 test_loss_avg=3.44068 acc=0.00000 test_acc_avg=0.27809 test_acc_top5_avg=0.71005 time=837.69it/s +curr_acc 0.2781 +BEST_ACC 0.3318 +curr_acc_top5 0.7100 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=6.36211 loss_avg=6.08484 acc=0.38281 acc_top1_avg=0.39062 acc_top5_avg=0.72396 lr=0.00100 gn=20.57468 time=51.54it/s +epoch=67 global_step=26250 loss=6.78098 loss_avg=6.17994 acc=0.33594 acc_top1_avg=0.38075 acc_top5_avg=0.76047 lr=0.00100 gn=19.76727 time=56.08it/s +epoch=67 global_step=26300 loss=5.91159 loss_avg=6.20854 acc=0.42188 acc_top1_avg=0.37773 acc_top5_avg=0.76054 lr=0.00100 gn=21.10677 time=50.55it/s +epoch=67 global_step=26350 loss=5.88911 loss_avg=6.22528 acc=0.39844 acc_top1_avg=0.37546 acc_top5_avg=0.75914 lr=0.00100 gn=15.73923 time=54.99it/s +epoch=67 global_step=26400 loss=5.90547 loss_avg=6.21716 acc=0.41406 acc_top1_avg=0.37631 acc_top5_avg=0.76085 lr=0.00100 gn=16.48306 time=55.53it/s +epoch=67 global_step=26450 loss=6.39510 loss_avg=6.19285 acc=0.36719 acc_top1_avg=0.37858 acc_top5_avg=0.76217 lr=0.00100 gn=17.17733 time=61.59it/s +epoch=67 global_step=26500 loss=6.39273 loss_avg=6.20952 acc=0.35938 acc_top1_avg=0.37652 acc_top5_avg=0.76055 lr=0.00100 gn=19.59074 time=57.85it/s +epoch=67 global_step=26550 loss=5.58597 loss_avg=6.21915 acc=0.44531 acc_top1_avg=0.37560 acc_top5_avg=0.75956 lr=0.00100 gn=21.36757 time=57.69it/s +====================Eval==================== +epoch=67 global_step=26588 loss=4.99896 test_loss_avg=5.04750 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.82924 time=92.11it/s +epoch=67 global_step=26588 loss=0.17563 test_loss_avg=3.84726 acc=0.94531 test_acc_avg=0.18298 test_acc_top5_avg=0.71902 time=198.39it/s +epoch=67 global_step=26588 loss=4.48126 test_loss_avg=3.34474 acc=0.00000 test_acc_avg=0.29124 test_acc_top5_avg=0.71954 time=507.54it/s +curr_acc 0.2912 +BEST_ACC 0.3318 +curr_acc_top5 0.7195 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=6.44345 loss_avg=6.30682 acc=0.35156 acc_top1_avg=0.36523 acc_top5_avg=0.77474 lr=0.00100 gn=18.36873 time=57.18it/s +epoch=68 global_step=26650 loss=5.83792 loss_avg=6.20665 acc=0.41406 acc_top1_avg=0.37475 acc_top5_avg=0.76096 lr=0.00100 gn=24.69309 time=58.62it/s +epoch=68 global_step=26700 loss=6.26197 loss_avg=6.17957 acc=0.36719 acc_top1_avg=0.37814 acc_top5_avg=0.76535 lr=0.00100 gn=21.49922 time=57.14it/s +epoch=68 global_step=26750 loss=6.53227 loss_avg=6.19215 acc=0.32812 acc_top1_avg=0.37727 acc_top5_avg=0.76056 lr=0.00100 gn=12.61721 time=60.54it/s +epoch=68 global_step=26800 loss=5.65061 loss_avg=6.20090 acc=0.43750 acc_top1_avg=0.37592 acc_top5_avg=0.76013 lr=0.00100 gn=16.62380 time=53.88it/s +epoch=68 global_step=26850 loss=6.50293 loss_avg=6.19582 acc=0.35938 acc_top1_avg=0.37661 acc_top5_avg=0.76020 lr=0.00100 gn=21.76918 time=57.20it/s +epoch=68 global_step=26900 loss=6.38735 loss_avg=6.21293 acc=0.35156 acc_top1_avg=0.37538 acc_top5_avg=0.76057 lr=0.00100 gn=14.77576 time=54.21it/s +epoch=68 global_step=26950 loss=6.12296 loss_avg=6.21515 acc=0.37500 acc_top1_avg=0.37541 acc_top5_avg=0.75986 lr=0.00100 gn=23.81435 time=53.64it/s +====================Eval==================== +epoch=68 global_step=26979 loss=5.14029 test_loss_avg=3.93709 acc=0.00000 test_acc_avg=0.16602 test_acc_top5_avg=0.74721 time=238.11it/s +epoch=68 global_step=26979 loss=5.47928 test_loss_avg=3.28604 acc=0.00000 test_acc_avg=0.32212 test_acc_top5_avg=0.70553 time=244.59it/s +epoch=68 global_step=26979 loss=5.46090 test_loss_avg=3.31357 acc=0.00000 test_acc_avg=0.31804 test_acc_top5_avg=0.69739 time=781.94it/s +curr_acc 0.3180 +BEST_ACC 0.3318 +curr_acc_top5 0.6974 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=5.97041 loss_avg=6.21711 acc=0.41406 acc_top1_avg=0.37723 acc_top5_avg=0.75223 lr=0.00100 gn=22.67127 time=56.13it/s +epoch=69 global_step=27050 loss=6.18604 loss_avg=6.16636 acc=0.41406 acc_top1_avg=0.38248 acc_top5_avg=0.76045 lr=0.00100 gn=23.84202 time=57.47it/s +epoch=69 global_step=27100 loss=6.02127 loss_avg=6.21625 acc=0.39844 acc_top1_avg=0.37610 acc_top5_avg=0.76020 lr=0.00100 gn=17.17551 time=56.78it/s +epoch=69 global_step=27150 loss=5.92275 loss_avg=6.17866 acc=0.40625 acc_top1_avg=0.38003 acc_top5_avg=0.76165 lr=0.00100 gn=19.00473 time=54.31it/s +epoch=69 global_step=27200 loss=6.01110 loss_avg=6.20168 acc=0.40625 acc_top1_avg=0.37793 acc_top5_avg=0.75993 lr=0.00100 gn=20.02284 time=32.79it/s +epoch=69 global_step=27250 loss=6.47897 loss_avg=6.20291 acc=0.34375 acc_top1_avg=0.37745 acc_top5_avg=0.75957 lr=0.00100 gn=21.81999 time=54.41it/s +epoch=69 global_step=27300 loss=6.67154 loss_avg=6.20908 acc=0.32031 acc_top1_avg=0.37678 acc_top5_avg=0.75978 lr=0.00100 gn=20.15339 time=54.42it/s +epoch=69 global_step=27350 loss=5.79232 loss_avg=6.19712 acc=0.43750 acc_top1_avg=0.37810 acc_top5_avg=0.76051 lr=0.00100 gn=18.60799 time=56.27it/s +====================Eval==================== +epoch=69 global_step=27370 loss=4.80435 test_loss_avg=3.95576 acc=0.00000 test_acc_avg=0.15896 test_acc_top5_avg=0.72130 time=199.76it/s +epoch=69 global_step=27370 loss=4.67782 test_loss_avg=3.37367 acc=0.00000 test_acc_avg=0.28817 test_acc_top5_avg=0.69828 time=467.28it/s +curr_acc 0.2882 +BEST_ACC 0.3318 +curr_acc_top5 0.6983 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=5.99414 loss_avg=6.15985 acc=0.40625 acc_top1_avg=0.38177 acc_top5_avg=0.75365 lr=0.00100 gn=21.23551 time=52.91it/s +epoch=70 global_step=27450 loss=5.76509 loss_avg=6.16445 acc=0.42188 acc_top1_avg=0.37920 acc_top5_avg=0.75742 lr=0.00100 gn=21.89787 time=62.89it/s +epoch=70 global_step=27500 loss=6.26478 loss_avg=6.16870 acc=0.36719 acc_top1_avg=0.38035 acc_top5_avg=0.75727 lr=0.00100 gn=24.42854 time=59.65it/s +epoch=70 global_step=27550 loss=6.08863 loss_avg=6.15657 acc=0.38281 acc_top1_avg=0.38155 acc_top5_avg=0.75764 lr=0.00100 gn=18.99106 time=56.10it/s +epoch=70 global_step=27600 loss=6.02688 loss_avg=6.15628 acc=0.39062 acc_top1_avg=0.38166 acc_top5_avg=0.75686 lr=0.00100 gn=16.87371 time=57.65it/s +epoch=70 global_step=27650 loss=5.44720 loss_avg=6.16719 acc=0.46875 acc_top1_avg=0.38136 acc_top5_avg=0.75689 lr=0.00100 gn=28.57140 time=55.15it/s +epoch=70 global_step=27700 loss=6.86608 loss_avg=6.17200 acc=0.29688 acc_top1_avg=0.38080 acc_top5_avg=0.75675 lr=0.00100 gn=21.63003 time=60.27it/s +epoch=70 global_step=27750 loss=6.63246 loss_avg=6.16901 acc=0.33594 acc_top1_avg=0.38102 acc_top5_avg=0.75794 lr=0.00100 gn=26.06932 time=53.51it/s +====================Eval==================== +epoch=70 global_step=27761 loss=4.51138 test_loss_avg=4.44929 acc=0.00000 test_acc_avg=0.07891 test_acc_top5_avg=0.87539 time=234.74it/s +epoch=70 global_step=27761 loss=0.21172 test_loss_avg=3.29752 acc=0.95312 test_acc_avg=0.30859 test_acc_top5_avg=0.76719 time=241.66it/s +epoch=70 global_step=27761 loss=4.25412 test_loss_avg=3.43619 acc=0.00000 test_acc_avg=0.27700 test_acc_top5_avg=0.73853 time=506.74it/s +curr_acc 0.2770 +BEST_ACC 0.3318 +curr_acc_top5 0.7385 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=6.84504 loss_avg=6.09982 acc=0.28906 acc_top1_avg=0.38782 acc_top5_avg=0.75300 lr=0.00100 gn=24.30772 time=59.53it/s +epoch=71 global_step=27850 loss=5.64459 loss_avg=6.11224 acc=0.44531 acc_top1_avg=0.38694 acc_top5_avg=0.75816 lr=0.00100 gn=21.66833 time=63.29it/s +epoch=71 global_step=27900 loss=6.38406 loss_avg=6.14748 acc=0.35938 acc_top1_avg=0.38427 acc_top5_avg=0.75629 lr=0.00100 gn=22.74510 time=57.85it/s +epoch=71 global_step=27950 loss=6.07113 loss_avg=6.17529 acc=0.39062 acc_top1_avg=0.38137 acc_top5_avg=0.75595 lr=0.00100 gn=21.20935 time=53.16it/s +epoch=71 global_step=28000 loss=5.70532 loss_avg=6.16895 acc=0.42188 acc_top1_avg=0.38154 acc_top5_avg=0.75481 lr=0.00100 gn=20.79977 time=55.50it/s +epoch=71 global_step=28050 loss=5.84683 loss_avg=6.16328 acc=0.41406 acc_top1_avg=0.38227 acc_top5_avg=0.75727 lr=0.00100 gn=18.71980 time=54.63it/s +epoch=71 global_step=28100 loss=6.73638 loss_avg=6.16784 acc=0.32812 acc_top1_avg=0.38152 acc_top5_avg=0.75862 lr=0.00100 gn=23.90233 time=50.39it/s +epoch=71 global_step=28150 loss=5.87720 loss_avg=6.16289 acc=0.39844 acc_top1_avg=0.38215 acc_top5_avg=0.75870 lr=0.00100 gn=16.51976 time=58.70it/s +====================Eval==================== +epoch=71 global_step=28152 loss=1.62294 test_loss_avg=4.40835 acc=0.62500 test_acc_avg=0.07622 test_acc_top5_avg=0.69912 time=230.98it/s +epoch=71 global_step=28152 loss=4.39352 test_loss_avg=3.38556 acc=0.00000 test_acc_avg=0.27621 test_acc_top5_avg=0.71934 time=849.39it/s +curr_acc 0.2762 +BEST_ACC 0.3318 +curr_acc_top5 0.7193 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=6.55129 loss_avg=6.15944 acc=0.32812 acc_top1_avg=0.38460 acc_top5_avg=0.76270 lr=0.00100 gn=20.12924 time=56.47it/s +epoch=72 global_step=28250 loss=5.34033 loss_avg=6.15712 acc=0.48438 acc_top1_avg=0.38377 acc_top5_avg=0.76156 lr=0.00100 gn=17.36613 time=55.53it/s +epoch=72 global_step=28300 loss=6.76247 loss_avg=6.14136 acc=0.31250 acc_top1_avg=0.38461 acc_top5_avg=0.75940 lr=0.00100 gn=25.92358 time=26.45it/s +epoch=72 global_step=28350 loss=5.95284 loss_avg=6.14779 acc=0.39844 acc_top1_avg=0.38360 acc_top5_avg=0.76065 lr=0.00100 gn=21.90316 time=54.64it/s +epoch=72 global_step=28400 loss=6.73527 loss_avg=6.15051 acc=0.32031 acc_top1_avg=0.38376 acc_top5_avg=0.75914 lr=0.00100 gn=20.24515 time=51.69it/s +epoch=72 global_step=28450 loss=6.48579 loss_avg=6.17055 acc=0.34375 acc_top1_avg=0.38148 acc_top5_avg=0.75891 lr=0.00100 gn=18.65744 time=60.68it/s +epoch=72 global_step=28500 loss=6.58788 loss_avg=6.16350 acc=0.32812 acc_top1_avg=0.38196 acc_top5_avg=0.75871 lr=0.00100 gn=19.86319 time=46.45it/s +====================Eval==================== +epoch=72 global_step=28543 loss=2.79441 test_loss_avg=4.19161 acc=0.33594 test_acc_avg=0.11003 test_acc_top5_avg=0.81706 time=247.61it/s +epoch=72 global_step=28543 loss=0.22070 test_loss_avg=3.65274 acc=0.93750 test_acc_avg=0.23400 test_acc_top5_avg=0.72089 time=236.81it/s +epoch=72 global_step=28543 loss=4.43909 test_loss_avg=3.39264 acc=0.00000 test_acc_avg=0.28511 test_acc_top5_avg=0.70807 time=518.33it/s +curr_acc 0.2851 +BEST_ACC 0.3318 +curr_acc_top5 0.7081 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=6.17026 loss_avg=6.19635 acc=0.36719 acc_top1_avg=0.37388 acc_top5_avg=0.74777 lr=0.00100 gn=12.63718 time=51.55it/s +epoch=73 global_step=28600 loss=6.66515 loss_avg=6.08954 acc=0.32812 acc_top1_avg=0.39282 acc_top5_avg=0.75630 lr=0.00100 gn=25.73097 time=32.17it/s +epoch=73 global_step=28650 loss=6.00027 loss_avg=6.06001 acc=0.39062 acc_top1_avg=0.39486 acc_top5_avg=0.76212 lr=0.00100 gn=15.85320 time=55.00it/s +epoch=73 global_step=28700 loss=6.14161 loss_avg=6.07025 acc=0.39062 acc_top1_avg=0.39286 acc_top5_avg=0.76224 lr=0.00100 gn=25.08398 time=56.04it/s +epoch=73 global_step=28750 loss=5.75982 loss_avg=6.09549 acc=0.41406 acc_top1_avg=0.38919 acc_top5_avg=0.76227 lr=0.00100 gn=17.30425 time=55.35it/s +epoch=73 global_step=28800 loss=6.19751 loss_avg=6.11287 acc=0.37500 acc_top1_avg=0.38707 acc_top5_avg=0.76161 lr=0.00100 gn=20.99639 time=50.93it/s +epoch=73 global_step=28850 loss=6.19042 loss_avg=6.12122 acc=0.38281 acc_top1_avg=0.38622 acc_top5_avg=0.75998 lr=0.00100 gn=22.88485 time=56.46it/s +epoch=73 global_step=28900 loss=6.01544 loss_avg=6.13480 acc=0.40625 acc_top1_avg=0.38467 acc_top5_avg=0.75956 lr=0.00100 gn=21.06800 time=62.35it/s +====================Eval==================== +epoch=73 global_step=28934 loss=5.09878 test_loss_avg=4.30249 acc=0.00000 test_acc_avg=0.07197 test_acc_top5_avg=0.65696 time=224.63it/s +epoch=73 global_step=28934 loss=4.77852 test_loss_avg=3.34453 acc=0.00000 test_acc_avg=0.28679 test_acc_top5_avg=0.71608 time=657.83it/s +curr_acc 0.2868 +BEST_ACC 0.3318 +curr_acc_top5 0.7161 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=6.06758 loss_avg=5.90436 acc=0.38281 acc_top1_avg=0.41357 acc_top5_avg=0.77979 lr=0.00100 gn=20.94450 time=48.86it/s +epoch=74 global_step=29000 loss=6.00272 loss_avg=6.01072 acc=0.39844 acc_top1_avg=0.39950 acc_top5_avg=0.75829 lr=0.00100 gn=20.15656 time=52.35it/s +epoch=74 global_step=29050 loss=5.92548 loss_avg=6.06573 acc=0.41406 acc_top1_avg=0.39312 acc_top5_avg=0.75566 lr=0.00100 gn=22.21465 time=45.07it/s +epoch=74 global_step=29100 loss=6.12722 loss_avg=6.07837 acc=0.38281 acc_top1_avg=0.39171 acc_top5_avg=0.75772 lr=0.00100 gn=22.89183 time=53.90it/s +epoch=74 global_step=29150 loss=5.80831 loss_avg=6.08195 acc=0.42188 acc_top1_avg=0.39110 acc_top5_avg=0.75781 lr=0.00100 gn=19.89268 time=48.61it/s +epoch=74 global_step=29200 loss=6.08920 loss_avg=6.09699 acc=0.38281 acc_top1_avg=0.38954 acc_top5_avg=0.75805 lr=0.00100 gn=27.61916 time=45.51it/s +epoch=74 global_step=29250 loss=6.42778 loss_avg=6.12086 acc=0.35156 acc_top1_avg=0.38652 acc_top5_avg=0.75645 lr=0.00100 gn=20.90885 time=52.26it/s +epoch=74 global_step=29300 loss=6.31760 loss_avg=6.11278 acc=0.37500 acc_top1_avg=0.38723 acc_top5_avg=0.75651 lr=0.00100 gn=22.73181 time=62.27it/s +====================Eval==================== +epoch=74 global_step=29325 loss=5.19170 test_loss_avg=5.13657 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.79297 time=240.15it/s +epoch=74 global_step=29325 loss=4.34623 test_loss_avg=4.00198 acc=0.00000 test_acc_avg=0.15712 test_acc_top5_avg=0.67593 time=213.29it/s +epoch=74 global_step=29325 loss=4.62025 test_loss_avg=3.32204 acc=0.00000 test_acc_avg=0.29183 test_acc_top5_avg=0.71539 time=505.76it/s +curr_acc 0.2918 +BEST_ACC 0.3318 +curr_acc_top5 0.7154 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=6.24005 loss_avg=6.03928 acc=0.36719 acc_top1_avg=0.39750 acc_top5_avg=0.76313 lr=0.00100 gn=20.29604 time=54.38it/s +epoch=75 global_step=29400 loss=5.64019 loss_avg=6.07922 acc=0.43750 acc_top1_avg=0.39146 acc_top5_avg=0.75698 lr=0.00100 gn=24.11748 time=52.09it/s +epoch=75 global_step=29450 loss=6.01328 loss_avg=6.07459 acc=0.39844 acc_top1_avg=0.39131 acc_top5_avg=0.75725 lr=0.00100 gn=21.15949 time=58.92it/s +epoch=75 global_step=29500 loss=6.09219 loss_avg=6.09236 acc=0.38281 acc_top1_avg=0.38871 acc_top5_avg=0.75656 lr=0.00100 gn=22.99520 time=60.45it/s +epoch=75 global_step=29550 loss=6.07828 loss_avg=6.06760 acc=0.36719 acc_top1_avg=0.39170 acc_top5_avg=0.75851 lr=0.00100 gn=23.71200 time=51.00it/s +epoch=75 global_step=29600 loss=6.79164 loss_avg=6.07494 acc=0.29688 acc_top1_avg=0.39153 acc_top5_avg=0.75892 lr=0.00100 gn=23.97699 time=58.62it/s +epoch=75 global_step=29650 loss=5.47095 loss_avg=6.09177 acc=0.46875 acc_top1_avg=0.38983 acc_top5_avg=0.75769 lr=0.00100 gn=20.80903 time=57.68it/s +epoch=75 global_step=29700 loss=6.37736 loss_avg=6.09977 acc=0.35156 acc_top1_avg=0.38896 acc_top5_avg=0.75731 lr=0.00100 gn=22.85862 time=52.52it/s +====================Eval==================== +epoch=75 global_step=29716 loss=5.03105 test_loss_avg=4.48853 acc=0.00000 test_acc_avg=0.06125 test_acc_top5_avg=0.78031 time=239.06it/s +epoch=75 global_step=29716 loss=4.78164 test_loss_avg=3.35874 acc=0.00000 test_acc_avg=0.29604 test_acc_top5_avg=0.71875 time=254.83it/s +epoch=75 global_step=29716 loss=4.44041 test_loss_avg=3.42414 acc=0.00000 test_acc_avg=0.28105 test_acc_top5_avg=0.71025 time=844.60it/s +curr_acc 0.2811 +BEST_ACC 0.3318 +curr_acc_top5 0.7102 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=5.87451 loss_avg=6.09772 acc=0.39844 acc_top1_avg=0.38925 acc_top5_avg=0.75988 lr=0.00100 gn=22.68415 time=60.64it/s +epoch=76 global_step=29800 loss=5.77784 loss_avg=6.14008 acc=0.42969 acc_top1_avg=0.38421 acc_top5_avg=0.75530 lr=0.00100 gn=22.93301 time=56.65it/s +epoch=76 global_step=29850 loss=6.34616 loss_avg=6.10966 acc=0.35156 acc_top1_avg=0.38853 acc_top5_avg=0.75321 lr=0.00100 gn=14.51909 time=53.36it/s +epoch=76 global_step=29900 loss=6.29718 loss_avg=6.10134 acc=0.37500 acc_top1_avg=0.38914 acc_top5_avg=0.75539 lr=0.00100 gn=22.04381 time=59.70it/s +epoch=76 global_step=29950 loss=5.44289 loss_avg=6.11068 acc=0.47656 acc_top1_avg=0.38795 acc_top5_avg=0.75331 lr=0.00100 gn=25.23246 time=62.46it/s +epoch=76 global_step=30000 loss=6.10409 loss_avg=6.10426 acc=0.40625 acc_top1_avg=0.38853 acc_top5_avg=0.75512 lr=0.00100 gn=26.49605 time=58.53it/s +epoch=76 global_step=30050 loss=5.25807 loss_avg=6.10930 acc=0.46875 acc_top1_avg=0.38772 acc_top5_avg=0.75529 lr=0.00100 gn=27.87667 time=52.13it/s +epoch=76 global_step=30100 loss=5.83723 loss_avg=6.09856 acc=0.42969 acc_top1_avg=0.38877 acc_top5_avg=0.75629 lr=0.00100 gn=26.44182 time=57.91it/s +====================Eval==================== +epoch=76 global_step=30107 loss=1.10282 test_loss_avg=4.10251 acc=0.73438 test_acc_avg=0.15489 test_acc_top5_avg=0.71620 time=246.10it/s +epoch=76 global_step=30107 loss=4.20554 test_loss_avg=3.42428 acc=0.00000 test_acc_avg=0.28254 test_acc_top5_avg=0.71311 time=831.38it/s +curr_acc 0.2825 +BEST_ACC 0.3318 +curr_acc_top5 0.7131 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=5.38165 loss_avg=5.96486 acc=0.48438 acc_top1_avg=0.40280 acc_top5_avg=0.76417 lr=0.00100 gn=27.75084 time=55.19it/s +epoch=77 global_step=30200 loss=5.62280 loss_avg=6.04271 acc=0.43750 acc_top1_avg=0.39390 acc_top5_avg=0.76184 lr=0.00100 gn=20.37101 time=59.52it/s +epoch=77 global_step=30250 loss=6.39107 loss_avg=6.06830 acc=0.38281 acc_top1_avg=0.39183 acc_top5_avg=0.76125 lr=0.00100 gn=30.15779 time=55.03it/s +epoch=77 global_step=30300 loss=5.80106 loss_avg=6.05500 acc=0.42969 acc_top1_avg=0.39334 acc_top5_avg=0.76113 lr=0.00100 gn=33.52341 time=59.78it/s +epoch=77 global_step=30350 loss=5.62825 loss_avg=6.06009 acc=0.43750 acc_top1_avg=0.39294 acc_top5_avg=0.75955 lr=0.00100 gn=21.40265 time=57.70it/s +epoch=77 global_step=30400 loss=6.16704 loss_avg=6.08763 acc=0.38281 acc_top1_avg=0.38988 acc_top5_avg=0.75768 lr=0.00100 gn=27.03011 time=58.05it/s +epoch=77 global_step=30450 loss=6.65493 loss_avg=6.09060 acc=0.33594 acc_top1_avg=0.38955 acc_top5_avg=0.75790 lr=0.00100 gn=28.26539 time=54.39it/s +====================Eval==================== +epoch=77 global_step=30498 loss=4.49266 test_loss_avg=3.62709 acc=0.00000 test_acc_avg=0.21461 test_acc_top5_avg=0.88695 time=235.16it/s +epoch=77 global_step=30498 loss=0.25211 test_loss_avg=3.19270 acc=0.92969 test_acc_avg=0.31437 test_acc_top5_avg=0.74382 time=237.45it/s +epoch=77 global_step=30498 loss=4.74017 test_loss_avg=3.25198 acc=0.00000 test_acc_avg=0.30657 test_acc_top5_avg=0.69927 time=807.37it/s +curr_acc 0.3066 +BEST_ACC 0.3318 +curr_acc_top5 0.6993 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=5.46363 loss_avg=5.85070 acc=0.47656 acc_top1_avg=0.42578 acc_top5_avg=0.76172 lr=0.00100 gn=25.09368 time=58.95it/s +epoch=78 global_step=30550 loss=6.31917 loss_avg=6.02542 acc=0.36719 acc_top1_avg=0.39573 acc_top5_avg=0.75992 lr=0.00100 gn=16.98237 time=43.60it/s +epoch=78 global_step=30600 loss=5.69980 loss_avg=6.02831 acc=0.42969 acc_top1_avg=0.39591 acc_top5_avg=0.76065 lr=0.00100 gn=27.62025 time=58.01it/s +epoch=78 global_step=30650 loss=6.17556 loss_avg=6.03342 acc=0.37500 acc_top1_avg=0.39525 acc_top5_avg=0.75966 lr=0.00100 gn=21.27821 time=63.50it/s +epoch=78 global_step=30700 loss=6.21682 loss_avg=6.06831 acc=0.37500 acc_top1_avg=0.39155 acc_top5_avg=0.75700 lr=0.00100 gn=22.52880 time=57.50it/s +epoch=78 global_step=30750 loss=6.36904 loss_avg=6.06430 acc=0.35938 acc_top1_avg=0.39252 acc_top5_avg=0.75704 lr=0.00100 gn=24.51544 time=62.40it/s +epoch=78 global_step=30800 loss=5.48763 loss_avg=6.05478 acc=0.46875 acc_top1_avg=0.39360 acc_top5_avg=0.75776 lr=0.00100 gn=28.91459 time=55.85it/s +epoch=78 global_step=30850 loss=6.23790 loss_avg=6.06300 acc=0.35938 acc_top1_avg=0.39302 acc_top5_avg=0.75777 lr=0.00100 gn=21.61125 time=57.59it/s +====================Eval==================== +epoch=78 global_step=30889 loss=4.96229 test_loss_avg=4.42078 acc=0.00000 test_acc_avg=0.09108 test_acc_top5_avg=0.66591 time=237.27it/s +epoch=78 global_step=30889 loss=4.51985 test_loss_avg=3.31316 acc=0.00000 test_acc_avg=0.30202 test_acc_top5_avg=0.70293 time=491.02it/s +curr_acc 0.3020 +BEST_ACC 0.3318 +curr_acc_top5 0.7029 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=6.03759 loss_avg=6.17980 acc=0.39062 acc_top1_avg=0.37713 acc_top5_avg=0.76634 lr=0.00100 gn=22.34578 time=55.17it/s +epoch=79 global_step=30950 loss=6.18511 loss_avg=6.06024 acc=0.37500 acc_top1_avg=0.39255 acc_top5_avg=0.76524 lr=0.00100 gn=19.96448 time=55.07it/s +epoch=79 global_step=31000 loss=6.83285 loss_avg=6.02063 acc=0.28906 acc_top1_avg=0.39640 acc_top5_avg=0.76309 lr=0.00100 gn=24.65175 time=51.34it/s +epoch=79 global_step=31050 loss=6.16674 loss_avg=6.04493 acc=0.38281 acc_top1_avg=0.39388 acc_top5_avg=0.76092 lr=0.00100 gn=20.66948 time=57.75it/s +epoch=79 global_step=31100 loss=5.72237 loss_avg=6.03215 acc=0.42188 acc_top1_avg=0.39581 acc_top5_avg=0.76126 lr=0.00100 gn=18.76091 time=44.78it/s +epoch=79 global_step=31150 loss=5.93432 loss_avg=6.04714 acc=0.40625 acc_top1_avg=0.39458 acc_top5_avg=0.75988 lr=0.00100 gn=30.23528 time=54.70it/s +epoch=79 global_step=31200 loss=6.26757 loss_avg=6.05608 acc=0.38281 acc_top1_avg=0.39394 acc_top5_avg=0.75862 lr=0.00100 gn=23.11715 time=54.17it/s +epoch=79 global_step=31250 loss=5.80994 loss_avg=6.06257 acc=0.42188 acc_top1_avg=0.39348 acc_top5_avg=0.75792 lr=0.00100 gn=27.23904 time=43.71it/s +====================Eval==================== +epoch=79 global_step=31280 loss=1.83006 test_loss_avg=4.44146 acc=0.46875 test_acc_avg=0.06163 test_acc_top5_avg=0.81076 time=234.65it/s +epoch=79 global_step=31280 loss=0.23571 test_loss_avg=3.61676 acc=0.92969 test_acc_avg=0.21915 test_acc_top5_avg=0.70458 time=240.78it/s +epoch=79 global_step=31280 loss=4.42457 test_loss_avg=3.28537 acc=0.00000 test_acc_avg=0.29885 test_acc_top5_avg=0.68977 time=489.87it/s +curr_acc 0.2989 +BEST_ACC 0.3318 +curr_acc_top5 0.6898 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=5.23237 loss_avg=5.98991 acc=0.48438 acc_top1_avg=0.40273 acc_top5_avg=0.75156 lr=0.00010 gn=22.73133 time=62.02it/s +epoch=80 global_step=31350 loss=5.45082 loss_avg=5.95629 acc=0.46875 acc_top1_avg=0.40324 acc_top5_avg=0.76105 lr=0.00010 gn=28.50102 time=56.93it/s +epoch=80 global_step=31400 loss=5.73612 loss_avg=5.96008 acc=0.41406 acc_top1_avg=0.40286 acc_top5_avg=0.76100 lr=0.00010 gn=19.98942 time=54.16it/s +epoch=80 global_step=31450 loss=6.05511 loss_avg=5.91522 acc=0.40625 acc_top1_avg=0.40813 acc_top5_avg=0.76319 lr=0.00010 gn=22.09281 time=50.78it/s +epoch=80 global_step=31500 loss=5.95969 loss_avg=5.90644 acc=0.39844 acc_top1_avg=0.40909 acc_top5_avg=0.76293 lr=0.00010 gn=31.74625 time=41.79it/s +epoch=80 global_step=31550 loss=5.37791 loss_avg=5.91467 acc=0.46875 acc_top1_avg=0.40813 acc_top5_avg=0.76143 lr=0.00010 gn=28.59237 time=60.32it/s +epoch=80 global_step=31600 loss=6.32022 loss_avg=5.92909 acc=0.35938 acc_top1_avg=0.40632 acc_top5_avg=0.76042 lr=0.00010 gn=23.62649 time=58.18it/s +epoch=80 global_step=31650 loss=6.08103 loss_avg=5.93770 acc=0.39062 acc_top1_avg=0.40500 acc_top5_avg=0.75988 lr=0.00010 gn=19.74312 time=61.38it/s +====================Eval==================== +epoch=80 global_step=31671 loss=4.75787 test_loss_avg=4.10107 acc=0.00000 test_acc_avg=0.11042 test_acc_top5_avg=0.68802 time=196.70it/s +epoch=80 global_step=31671 loss=4.41192 test_loss_avg=3.22875 acc=0.00000 test_acc_avg=0.30004 test_acc_top5_avg=0.70896 time=825.98it/s +curr_acc 0.3000 +BEST_ACC 0.3318 +curr_acc_top5 0.7090 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=6.27556 loss_avg=5.75167 acc=0.36719 acc_top1_avg=0.42672 acc_top5_avg=0.76212 lr=0.00010 gn=26.45481 time=60.05it/s +epoch=81 global_step=31750 loss=5.75934 loss_avg=5.88641 acc=0.42188 acc_top1_avg=0.41090 acc_top5_avg=0.75287 lr=0.00010 gn=23.79048 time=60.72it/s +epoch=81 global_step=31800 loss=6.31472 loss_avg=5.89652 acc=0.36719 acc_top1_avg=0.40940 acc_top5_avg=0.75581 lr=0.00010 gn=30.14305 time=54.22it/s +epoch=81 global_step=31850 loss=6.13227 loss_avg=5.92064 acc=0.37500 acc_top1_avg=0.40695 acc_top5_avg=0.75698 lr=0.00010 gn=26.25610 time=57.45it/s +epoch=81 global_step=31900 loss=5.40864 loss_avg=5.90814 acc=0.46875 acc_top1_avg=0.40850 acc_top5_avg=0.75699 lr=0.00010 gn=30.30200 time=54.80it/s +epoch=81 global_step=31950 loss=5.95536 loss_avg=5.89994 acc=0.41406 acc_top1_avg=0.40927 acc_top5_avg=0.75756 lr=0.00010 gn=29.93383 time=46.46it/s +epoch=81 global_step=32000 loss=6.23020 loss_avg=5.89113 acc=0.38281 acc_top1_avg=0.41005 acc_top5_avg=0.75812 lr=0.00010 gn=29.59468 time=60.24it/s +epoch=81 global_step=32050 loss=6.01880 loss_avg=5.89578 acc=0.40625 acc_top1_avg=0.40961 acc_top5_avg=0.75771 lr=0.00010 gn=32.34214 time=59.05it/s +====================Eval==================== +epoch=81 global_step=32062 loss=4.96745 test_loss_avg=4.96745 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.78906 time=199.71it/s +epoch=81 global_step=32062 loss=4.24847 test_loss_avg=3.89738 acc=0.00000 test_acc_avg=0.16973 test_acc_top5_avg=0.70205 time=238.86it/s +epoch=81 global_step=32062 loss=4.26267 test_loss_avg=3.27121 acc=0.00000 test_acc_avg=0.29490 test_acc_top5_avg=0.71173 time=486.58it/s +curr_acc 0.2949 +BEST_ACC 0.3318 +curr_acc_top5 0.7117 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=5.75808 loss_avg=5.90565 acc=0.42188 acc_top1_avg=0.40851 acc_top5_avg=0.76419 lr=0.00010 gn=19.03874 time=55.42it/s +epoch=82 global_step=32150 loss=5.71401 loss_avg=5.88881 acc=0.43750 acc_top1_avg=0.41131 acc_top5_avg=0.75630 lr=0.00010 gn=27.75916 time=59.41it/s +epoch=82 global_step=32200 loss=5.96715 loss_avg=5.86721 acc=0.39844 acc_top1_avg=0.41321 acc_top5_avg=0.75810 lr=0.00010 gn=21.57089 time=54.81it/s +epoch=82 global_step=32250 loss=5.79320 loss_avg=5.87184 acc=0.42969 acc_top1_avg=0.41223 acc_top5_avg=0.75827 lr=0.00010 gn=24.09763 time=59.56it/s +epoch=82 global_step=32300 loss=5.78124 loss_avg=5.86228 acc=0.42188 acc_top1_avg=0.41370 acc_top5_avg=0.75991 lr=0.00010 gn=27.93584 time=59.23it/s +epoch=82 global_step=32350 loss=6.13190 loss_avg=5.86860 acc=0.38281 acc_top1_avg=0.41254 acc_top5_avg=0.75884 lr=0.00010 gn=26.85350 time=58.34it/s +epoch=82 global_step=32400 loss=6.36354 loss_avg=5.87594 acc=0.35938 acc_top1_avg=0.41164 acc_top5_avg=0.75800 lr=0.00010 gn=24.24812 time=55.33it/s +epoch=82 global_step=32450 loss=6.27886 loss_avg=5.87039 acc=0.36719 acc_top1_avg=0.41233 acc_top5_avg=0.75789 lr=0.00010 gn=19.95733 time=55.77it/s +====================Eval==================== +epoch=82 global_step=32453 loss=4.50579 test_loss_avg=3.90753 acc=0.00000 test_acc_avg=0.14524 test_acc_top5_avg=0.85831 time=241.20it/s +epoch=82 global_step=32453 loss=4.54603 test_loss_avg=3.12683 acc=0.00000 test_acc_avg=0.32671 test_acc_top5_avg=0.74544 time=242.33it/s +epoch=82 global_step=32453 loss=4.19722 test_loss_avg=3.25728 acc=0.00000 test_acc_avg=0.29777 test_acc_top5_avg=0.71143 time=576.30it/s +curr_acc 0.2978 +BEST_ACC 0.3318 +curr_acc_top5 0.7114 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=6.27605 loss_avg=5.99676 acc=0.37500 acc_top1_avg=0.39727 acc_top5_avg=0.75249 lr=0.00010 gn=27.48054 time=54.98it/s +epoch=83 global_step=32550 loss=5.96482 loss_avg=5.90905 acc=0.39844 acc_top1_avg=0.40665 acc_top5_avg=0.75685 lr=0.00010 gn=31.68984 time=50.66it/s +epoch=83 global_step=32600 loss=5.68218 loss_avg=5.87809 acc=0.42969 acc_top1_avg=0.41040 acc_top5_avg=0.75866 lr=0.00010 gn=25.65958 time=55.17it/s +epoch=83 global_step=32650 loss=6.19071 loss_avg=5.86809 acc=0.39844 acc_top1_avg=0.41248 acc_top5_avg=0.76035 lr=0.00010 gn=33.85264 time=59.66it/s +epoch=83 global_step=32700 loss=6.24367 loss_avg=5.85969 acc=0.36719 acc_top1_avg=0.41378 acc_top5_avg=0.76094 lr=0.00010 gn=27.29885 time=63.26it/s +epoch=83 global_step=32750 loss=5.78517 loss_avg=5.85757 acc=0.42969 acc_top1_avg=0.41385 acc_top5_avg=0.75997 lr=0.00010 gn=23.69117 time=60.03it/s +epoch=83 global_step=32800 loss=6.25245 loss_avg=5.86022 acc=0.37500 acc_top1_avg=0.41325 acc_top5_avg=0.75957 lr=0.00010 gn=28.30011 time=57.08it/s +====================Eval==================== +epoch=83 global_step=32844 loss=1.07769 test_loss_avg=4.01748 acc=0.65625 test_acc_avg=0.13899 test_acc_top5_avg=0.69241 time=241.52it/s +epoch=83 global_step=32844 loss=4.12622 test_loss_avg=3.20426 acc=0.00000 test_acc_avg=0.29866 test_acc_top5_avg=0.70609 time=541.76it/s +curr_acc 0.2987 +BEST_ACC 0.3318 +curr_acc_top5 0.7061 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=5.25983 loss_avg=5.55398 acc=0.47656 acc_top1_avg=0.45312 acc_top5_avg=0.77214 lr=0.00010 gn=22.72725 time=55.69it/s +epoch=84 global_step=32900 loss=5.69513 loss_avg=5.85569 acc=0.42188 acc_top1_avg=0.41267 acc_top5_avg=0.76074 lr=0.00010 gn=21.51489 time=50.64it/s +epoch=84 global_step=32950 loss=5.61787 loss_avg=5.85737 acc=0.44531 acc_top1_avg=0.41296 acc_top5_avg=0.75693 lr=0.00010 gn=24.15531 time=54.33it/s +epoch=84 global_step=33000 loss=6.26229 loss_avg=5.87940 acc=0.36719 acc_top1_avg=0.41051 acc_top5_avg=0.75626 lr=0.00010 gn=21.12170 time=54.13it/s +epoch=84 global_step=33050 loss=6.31553 loss_avg=5.85313 acc=0.35938 acc_top1_avg=0.41285 acc_top5_avg=0.75614 lr=0.00010 gn=37.01310 time=59.19it/s +epoch=84 global_step=33100 loss=6.12987 loss_avg=5.86001 acc=0.39062 acc_top1_avg=0.41235 acc_top5_avg=0.75656 lr=0.00010 gn=28.84630 time=54.49it/s +epoch=84 global_step=33150 loss=6.02423 loss_avg=5.85160 acc=0.38281 acc_top1_avg=0.41350 acc_top5_avg=0.75702 lr=0.00010 gn=23.54340 time=49.44it/s +epoch=84 global_step=33200 loss=5.53922 loss_avg=5.84190 acc=0.45312 acc_top1_avg=0.41470 acc_top5_avg=0.75834 lr=0.00010 gn=28.53153 time=52.01it/s +====================Eval==================== +epoch=84 global_step=33235 loss=2.75002 test_loss_avg=3.80118 acc=0.38281 test_acc_avg=0.18415 test_acc_top5_avg=0.84933 time=54.69it/s +epoch=84 global_step=33235 loss=0.29533 test_loss_avg=3.31377 acc=0.92188 test_acc_avg=0.27368 test_acc_top5_avg=0.72522 time=240.53it/s +epoch=84 global_step=33235 loss=4.27899 test_loss_avg=3.21909 acc=0.00000 test_acc_avg=0.29757 test_acc_top5_avg=0.70243 time=511.81it/s +curr_acc 0.2976 +BEST_ACC 0.3318 +curr_acc_top5 0.7024 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=5.22970 loss_avg=5.86880 acc=0.46875 acc_top1_avg=0.40937 acc_top5_avg=0.76562 lr=0.00010 gn=25.76069 time=46.26it/s +epoch=85 global_step=33300 loss=5.95668 loss_avg=5.76990 acc=0.40625 acc_top1_avg=0.42188 acc_top5_avg=0.75998 lr=0.00010 gn=26.76793 time=51.07it/s +epoch=85 global_step=33350 loss=6.00296 loss_avg=5.80761 acc=0.39844 acc_top1_avg=0.41800 acc_top5_avg=0.75849 lr=0.00010 gn=31.28467 time=55.40it/s +epoch=85 global_step=33400 loss=5.68335 loss_avg=5.82467 acc=0.43750 acc_top1_avg=0.41572 acc_top5_avg=0.75833 lr=0.00010 gn=28.96909 time=58.76it/s +epoch=85 global_step=33450 loss=5.48206 loss_avg=5.84579 acc=0.44531 acc_top1_avg=0.41366 acc_top5_avg=0.75956 lr=0.00010 gn=26.48297 time=53.86it/s +epoch=85 global_step=33500 loss=6.23583 loss_avg=5.84304 acc=0.36719 acc_top1_avg=0.41397 acc_top5_avg=0.75908 lr=0.00010 gn=27.10447 time=46.20it/s +epoch=85 global_step=33550 loss=6.34485 loss_avg=5.84572 acc=0.35938 acc_top1_avg=0.41369 acc_top5_avg=0.75935 lr=0.00010 gn=24.33007 time=54.88it/s +epoch=85 global_step=33600 loss=6.07182 loss_avg=5.83880 acc=0.38281 acc_top1_avg=0.41421 acc_top5_avg=0.75976 lr=0.00010 gn=31.12687 time=58.67it/s +====================Eval==================== +epoch=85 global_step=33626 loss=5.12985 test_loss_avg=4.32735 acc=0.00000 test_acc_avg=0.08571 test_acc_top5_avg=0.66317 time=235.65it/s +epoch=85 global_step=33626 loss=4.29322 test_loss_avg=3.26958 acc=0.00000 test_acc_avg=0.29282 test_acc_top5_avg=0.70817 time=852.15it/s +curr_acc 0.2928 +BEST_ACC 0.3318 +curr_acc_top5 0.7082 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=5.52593 loss_avg=5.86586 acc=0.44531 acc_top1_avg=0.41406 acc_top5_avg=0.74935 lr=0.00010 gn=26.02266 time=61.43it/s +epoch=86 global_step=33700 loss=5.64100 loss_avg=5.84774 acc=0.43750 acc_top1_avg=0.41512 acc_top5_avg=0.75401 lr=0.00010 gn=24.95907 time=49.65it/s +epoch=86 global_step=33750 loss=5.76331 loss_avg=5.85532 acc=0.43750 acc_top1_avg=0.41337 acc_top5_avg=0.75819 lr=0.00010 gn=32.60566 time=58.47it/s +epoch=86 global_step=33800 loss=5.57207 loss_avg=5.85828 acc=0.45312 acc_top1_avg=0.41281 acc_top5_avg=0.75938 lr=0.00010 gn=26.11145 time=52.28it/s +epoch=86 global_step=33850 loss=5.74267 loss_avg=5.84858 acc=0.42969 acc_top1_avg=0.41434 acc_top5_avg=0.75882 lr=0.00010 gn=33.84195 time=54.15it/s +epoch=86 global_step=33900 loss=5.81912 loss_avg=5.82668 acc=0.41406 acc_top1_avg=0.41674 acc_top5_avg=0.75838 lr=0.00010 gn=27.07279 time=56.46it/s +epoch=86 global_step=33950 loss=5.64662 loss_avg=5.81988 acc=0.44531 acc_top1_avg=0.41751 acc_top5_avg=0.75815 lr=0.00010 gn=25.88852 time=58.87it/s +epoch=86 global_step=34000 loss=5.72480 loss_avg=5.82228 acc=0.42969 acc_top1_avg=0.41724 acc_top5_avg=0.75923 lr=0.00010 gn=25.44585 time=54.96it/s +====================Eval==================== +epoch=86 global_step=34017 loss=4.90106 test_loss_avg=4.95054 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.79557 time=239.66it/s +epoch=86 global_step=34017 loss=0.36291 test_loss_avg=3.84573 acc=0.90625 test_acc_avg=0.17146 test_acc_top5_avg=0.68471 time=222.60it/s +epoch=86 global_step=34017 loss=4.15222 test_loss_avg=3.26334 acc=0.00000 test_acc_avg=0.29302 test_acc_top5_avg=0.70767 time=855.63it/s +curr_acc 0.2930 +BEST_ACC 0.3318 +curr_acc_top5 0.7077 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=5.69001 loss_avg=5.78676 acc=0.43750 acc_top1_avg=0.42069 acc_top5_avg=0.74645 lr=0.00010 gn=30.43405 time=62.32it/s +epoch=87 global_step=34100 loss=5.68914 loss_avg=5.82673 acc=0.43750 acc_top1_avg=0.41698 acc_top5_avg=0.75226 lr=0.00010 gn=31.46985 time=61.36it/s +epoch=87 global_step=34150 loss=5.67950 loss_avg=5.83925 acc=0.43750 acc_top1_avg=0.41559 acc_top5_avg=0.75517 lr=0.00010 gn=26.13450 time=56.16it/s +epoch=87 global_step=34200 loss=6.22576 loss_avg=5.84407 acc=0.37500 acc_top1_avg=0.41492 acc_top5_avg=0.75393 lr=0.00010 gn=29.26221 time=57.98it/s +epoch=87 global_step=34250 loss=5.28299 loss_avg=5.81104 acc=0.48438 acc_top1_avg=0.41859 acc_top5_avg=0.75751 lr=0.00010 gn=26.16855 time=54.59it/s +epoch=87 global_step=34300 loss=6.55950 loss_avg=5.81370 acc=0.34375 acc_top1_avg=0.41804 acc_top5_avg=0.75748 lr=0.00010 gn=24.49641 time=55.60it/s +epoch=87 global_step=34350 loss=5.87776 loss_avg=5.81276 acc=0.40625 acc_top1_avg=0.41824 acc_top5_avg=0.75861 lr=0.00010 gn=27.59371 time=59.37it/s +epoch=87 global_step=34400 loss=6.15300 loss_avg=5.81310 acc=0.37500 acc_top1_avg=0.41806 acc_top5_avg=0.75787 lr=0.00010 gn=22.59355 time=56.89it/s +====================Eval==================== +epoch=87 global_step=34408 loss=4.81506 test_loss_avg=4.08711 acc=0.00000 test_acc_avg=0.11227 test_acc_top5_avg=0.74450 time=137.19it/s +epoch=87 global_step=34408 loss=4.73289 test_loss_avg=3.20490 acc=0.00000 test_acc_avg=0.30043 test_acc_top5_avg=0.71530 time=253.42it/s +epoch=87 global_step=34408 loss=4.14579 test_loss_avg=3.23617 acc=0.00000 test_acc_avg=0.29282 test_acc_top5_avg=0.70896 time=830.56it/s +curr_acc 0.2928 +BEST_ACC 0.3318 +curr_acc_top5 0.7090 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=5.69165 loss_avg=5.71619 acc=0.42969 acc_top1_avg=0.42764 acc_top5_avg=0.76656 lr=0.00010 gn=25.29404 time=62.95it/s +epoch=88 global_step=34500 loss=5.32826 loss_avg=5.74310 acc=0.47656 acc_top1_avg=0.42493 acc_top5_avg=0.76486 lr=0.00010 gn=32.75447 time=58.62it/s +epoch=88 global_step=34550 loss=5.44303 loss_avg=5.76843 acc=0.44531 acc_top1_avg=0.42165 acc_top5_avg=0.76276 lr=0.00010 gn=23.97444 time=54.63it/s +epoch=88 global_step=34600 loss=6.00305 loss_avg=5.79000 acc=0.39844 acc_top1_avg=0.41903 acc_top5_avg=0.75981 lr=0.00010 gn=32.72162 time=57.71it/s +epoch=88 global_step=34650 loss=6.60298 loss_avg=5.78621 acc=0.34375 acc_top1_avg=0.41991 acc_top5_avg=0.75936 lr=0.00010 gn=31.83517 time=62.56it/s +epoch=88 global_step=34700 loss=5.75244 loss_avg=5.80304 acc=0.42188 acc_top1_avg=0.41789 acc_top5_avg=0.75904 lr=0.00010 gn=25.47277 time=49.03it/s +epoch=88 global_step=34750 loss=5.73535 loss_avg=5.80205 acc=0.43750 acc_top1_avg=0.41833 acc_top5_avg=0.75749 lr=0.00010 gn=33.49716 time=57.68it/s +====================Eval==================== +epoch=88 global_step=34799 loss=4.27533 test_loss_avg=3.84094 acc=0.00000 test_acc_avg=0.17692 test_acc_top5_avg=0.70964 time=241.33it/s +epoch=88 global_step=34799 loss=4.22521 test_loss_avg=3.24530 acc=0.00000 test_acc_avg=0.29322 test_acc_top5_avg=0.71232 time=662.71it/s +curr_acc 0.2932 +BEST_ACC 0.3318 +curr_acc_top5 0.7123 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=5.28284 loss_avg=5.28284 acc=0.46875 acc_top1_avg=0.46875 acc_top5_avg=0.75000 lr=0.00010 gn=32.50645 time=49.59it/s +epoch=89 global_step=34850 loss=5.20164 loss_avg=5.80032 acc=0.48438 acc_top1_avg=0.41942 acc_top5_avg=0.75444 lr=0.00010 gn=28.93797 time=57.30it/s +epoch=89 global_step=34900 loss=5.66934 loss_avg=5.82617 acc=0.43750 acc_top1_avg=0.41615 acc_top5_avg=0.75998 lr=0.00010 gn=24.76066 time=63.34it/s +epoch=89 global_step=34950 loss=6.13478 loss_avg=5.82478 acc=0.38281 acc_top1_avg=0.41608 acc_top5_avg=0.75849 lr=0.00010 gn=20.95924 time=53.26it/s +epoch=89 global_step=35000 loss=5.25061 loss_avg=5.81052 acc=0.48438 acc_top1_avg=0.41857 acc_top5_avg=0.75902 lr=0.00010 gn=32.93114 time=57.49it/s +epoch=89 global_step=35050 loss=6.25908 loss_avg=5.80262 acc=0.35938 acc_top1_avg=0.41914 acc_top5_avg=0.75856 lr=0.00010 gn=22.55857 time=61.85it/s +epoch=89 global_step=35100 loss=5.75013 loss_avg=5.80345 acc=0.41406 acc_top1_avg=0.41910 acc_top5_avg=0.75794 lr=0.00010 gn=21.02855 time=52.97it/s +epoch=89 global_step=35150 loss=5.97962 loss_avg=5.81011 acc=0.39062 acc_top1_avg=0.41798 acc_top5_avg=0.75839 lr=0.00010 gn=20.06044 time=59.90it/s +====================Eval==================== +epoch=89 global_step=35190 loss=4.44228 test_loss_avg=3.79064 acc=0.00000 test_acc_avg=0.17146 test_acc_top5_avg=0.85403 time=235.93it/s +epoch=89 global_step=35190 loss=0.17814 test_loss_avg=3.09699 acc=0.95312 test_acc_avg=0.32054 test_acc_top5_avg=0.74083 time=233.12it/s +epoch=89 global_step=35190 loss=4.19323 test_loss_avg=3.20635 acc=0.00000 test_acc_avg=0.29598 test_acc_top5_avg=0.70481 time=479.73it/s +curr_acc 0.2960 +BEST_ACC 0.3318 +curr_acc_top5 0.7048 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=5.62519 loss_avg=5.74421 acc=0.43750 acc_top1_avg=0.42578 acc_top5_avg=0.76719 lr=0.00010 gn=28.70351 time=60.79it/s +epoch=90 global_step=35250 loss=6.38390 loss_avg=5.81554 acc=0.35156 acc_top1_avg=0.41628 acc_top5_avg=0.76615 lr=0.00010 gn=28.53840 time=54.37it/s +epoch=90 global_step=35300 loss=4.98714 loss_avg=5.74606 acc=0.50000 acc_top1_avg=0.42457 acc_top5_avg=0.76662 lr=0.00010 gn=26.70332 time=55.54it/s +epoch=90 global_step=35350 loss=6.22851 loss_avg=5.77690 acc=0.36719 acc_top1_avg=0.42104 acc_top5_avg=0.76372 lr=0.00010 gn=25.11970 time=54.90it/s +epoch=90 global_step=35400 loss=5.67849 loss_avg=5.80206 acc=0.43750 acc_top1_avg=0.41860 acc_top5_avg=0.76202 lr=0.00010 gn=26.84706 time=49.89it/s +epoch=90 global_step=35450 loss=5.74314 loss_avg=5.79135 acc=0.41406 acc_top1_avg=0.42028 acc_top5_avg=0.76088 lr=0.00010 gn=27.61988 time=54.77it/s +epoch=90 global_step=35500 loss=5.57832 loss_avg=5.78139 acc=0.44531 acc_top1_avg=0.42157 acc_top5_avg=0.76169 lr=0.00010 gn=25.96207 time=53.58it/s +epoch=90 global_step=35550 loss=5.69778 loss_avg=5.79063 acc=0.44531 acc_top1_avg=0.42059 acc_top5_avg=0.76039 lr=0.00010 gn=35.89523 time=54.36it/s +====================Eval==================== +epoch=90 global_step=35581 loss=1.67698 test_loss_avg=4.29631 acc=0.57031 test_acc_avg=0.09316 test_acc_top5_avg=0.67109 time=224.79it/s +epoch=90 global_step=35581 loss=4.19954 test_loss_avg=3.25565 acc=0.00000 test_acc_avg=0.29262 test_acc_top5_avg=0.70708 time=501.17it/s +curr_acc 0.2926 +BEST_ACC 0.3318 +curr_acc_top5 0.7071 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=5.48421 loss_avg=5.66936 acc=0.45312 acc_top1_avg=0.43462 acc_top5_avg=0.76686 lr=0.00010 gn=26.89626 time=60.90it/s +epoch=91 global_step=35650 loss=6.17360 loss_avg=5.78824 acc=0.36719 acc_top1_avg=0.42176 acc_top5_avg=0.76019 lr=0.00010 gn=25.66721 time=54.80it/s +epoch=91 global_step=35700 loss=5.74330 loss_avg=5.77669 acc=0.42969 acc_top1_avg=0.42220 acc_top5_avg=0.75985 lr=0.00010 gn=30.77357 time=54.87it/s +epoch=91 global_step=35750 loss=5.49000 loss_avg=5.76965 acc=0.44531 acc_top1_avg=0.42285 acc_top5_avg=0.76151 lr=0.00010 gn=27.11062 time=62.09it/s +epoch=91 global_step=35800 loss=5.53934 loss_avg=5.78216 acc=0.42969 acc_top1_avg=0.42159 acc_top5_avg=0.76184 lr=0.00010 gn=27.45854 time=51.97it/s +epoch=91 global_step=35850 loss=6.04041 loss_avg=5.77026 acc=0.40625 acc_top1_avg=0.42298 acc_top5_avg=0.76115 lr=0.00010 gn=37.89577 time=51.52it/s +epoch=91 global_step=35900 loss=5.52683 loss_avg=5.77498 acc=0.45312 acc_top1_avg=0.42236 acc_top5_avg=0.75909 lr=0.00010 gn=28.88328 time=57.59it/s +epoch=91 global_step=35950 loss=5.94382 loss_avg=5.77795 acc=0.39062 acc_top1_avg=0.42200 acc_top5_avg=0.75915 lr=0.00010 gn=20.34740 time=56.96it/s +====================Eval==================== +epoch=91 global_step=35972 loss=2.40591 test_loss_avg=4.20454 acc=0.40625 test_acc_avg=0.11506 test_acc_top5_avg=0.81960 time=240.06it/s +epoch=91 global_step=35972 loss=0.14093 test_loss_avg=3.53551 acc=0.96094 test_acc_avg=0.23297 test_acc_top5_avg=0.71196 time=235.12it/s +epoch=91 global_step=35972 loss=4.11538 test_loss_avg=3.24418 acc=0.00000 test_acc_avg=0.29114 test_acc_top5_avg=0.71341 time=558.27it/s +curr_acc 0.2911 +BEST_ACC 0.3318 +curr_acc_top5 0.7134 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=5.89935 loss_avg=5.80946 acc=0.40625 acc_top1_avg=0.41490 acc_top5_avg=0.76144 lr=0.00010 gn=27.04267 time=63.67it/s +epoch=92 global_step=36050 loss=6.12617 loss_avg=5.73533 acc=0.38281 acc_top1_avg=0.42458 acc_top5_avg=0.76132 lr=0.00010 gn=24.33765 time=61.59it/s +epoch=92 global_step=36100 loss=6.24805 loss_avg=5.74564 acc=0.36719 acc_top1_avg=0.42468 acc_top5_avg=0.76031 lr=0.00010 gn=24.25985 time=54.58it/s +epoch=92 global_step=36150 loss=5.87641 loss_avg=5.74528 acc=0.40625 acc_top1_avg=0.42499 acc_top5_avg=0.76053 lr=0.00010 gn=28.87815 time=55.54it/s +epoch=92 global_step=36200 loss=5.81557 loss_avg=5.76250 acc=0.41406 acc_top1_avg=0.42318 acc_top5_avg=0.75812 lr=0.00010 gn=25.56374 time=43.32it/s +epoch=92 global_step=36250 loss=5.97316 loss_avg=5.78279 acc=0.39844 acc_top1_avg=0.42092 acc_top5_avg=0.75807 lr=0.00010 gn=22.99225 time=61.40it/s +epoch=92 global_step=36300 loss=5.81433 loss_avg=5.76857 acc=0.40625 acc_top1_avg=0.42216 acc_top5_avg=0.75855 lr=0.00010 gn=22.77130 time=54.08it/s +epoch=92 global_step=36350 loss=5.53716 loss_avg=5.76975 acc=0.44531 acc_top1_avg=0.42208 acc_top5_avg=0.75916 lr=0.00010 gn=26.91886 time=63.63it/s +====================Eval==================== +epoch=92 global_step=36363 loss=5.05091 test_loss_avg=4.17834 acc=0.00000 test_acc_avg=0.10107 test_acc_top5_avg=0.64966 time=239.56it/s +epoch=92 global_step=36363 loss=4.11264 test_loss_avg=3.23613 acc=0.00000 test_acc_avg=0.29312 test_acc_top5_avg=0.71292 time=855.46it/s +curr_acc 0.2931 +BEST_ACC 0.3318 +curr_acc_top5 0.7129 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=5.71988 loss_avg=5.74296 acc=0.42188 acc_top1_avg=0.42694 acc_top5_avg=0.74979 lr=0.00010 gn=21.33224 time=62.30it/s +epoch=93 global_step=36450 loss=5.13947 loss_avg=5.72915 acc=0.49219 acc_top1_avg=0.42726 acc_top5_avg=0.76149 lr=0.00010 gn=33.77220 time=46.38it/s +epoch=93 global_step=36500 loss=6.10588 loss_avg=5.72550 acc=0.39844 acc_top1_avg=0.42758 acc_top5_avg=0.76135 lr=0.00010 gn=34.80119 time=53.96it/s +epoch=93 global_step=36550 loss=6.44516 loss_avg=5.76071 acc=0.35938 acc_top1_avg=0.42359 acc_top5_avg=0.75978 lr=0.00010 gn=27.11298 time=62.04it/s +epoch=93 global_step=36600 loss=5.56220 loss_avg=5.76064 acc=0.45312 acc_top1_avg=0.42352 acc_top5_avg=0.75943 lr=0.00010 gn=24.37685 time=54.64it/s +epoch=93 global_step=36650 loss=6.25000 loss_avg=5.77550 acc=0.36719 acc_top1_avg=0.42174 acc_top5_avg=0.75806 lr=0.00010 gn=31.00560 time=57.53it/s +epoch=93 global_step=36700 loss=5.41458 loss_avg=5.76769 acc=0.45312 acc_top1_avg=0.42271 acc_top5_avg=0.75839 lr=0.00010 gn=19.08680 time=55.20it/s +epoch=93 global_step=36750 loss=6.17289 loss_avg=5.78496 acc=0.37500 acc_top1_avg=0.42093 acc_top5_avg=0.75751 lr=0.00010 gn=27.43404 time=63.59it/s +====================Eval==================== +epoch=93 global_step=36754 loss=5.00130 test_loss_avg=4.96184 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.78385 time=239.66it/s +epoch=93 global_step=36754 loss=4.21442 test_loss_avg=3.86042 acc=0.00000 test_acc_avg=0.16450 test_acc_top5_avg=0.68116 time=230.28it/s +epoch=93 global_step=36754 loss=4.11636 test_loss_avg=3.23196 acc=0.00000 test_acc_avg=0.29411 test_acc_top5_avg=0.70253 time=493.39it/s +curr_acc 0.2941 +BEST_ACC 0.3318 +curr_acc_top5 0.7025 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=5.44891 loss_avg=5.75459 acc=0.44531 acc_top1_avg=0.42391 acc_top5_avg=0.75391 lr=0.00010 gn=19.84408 time=63.44it/s +epoch=94 global_step=36850 loss=5.64913 loss_avg=5.75007 acc=0.44531 acc_top1_avg=0.42529 acc_top5_avg=0.75960 lr=0.00010 gn=33.91102 time=55.36it/s +epoch=94 global_step=36900 loss=5.96889 loss_avg=5.76350 acc=0.39844 acc_top1_avg=0.42369 acc_top5_avg=0.75765 lr=0.00010 gn=28.65698 time=59.10it/s +epoch=94 global_step=36950 loss=5.45753 loss_avg=5.77175 acc=0.45312 acc_top1_avg=0.42287 acc_top5_avg=0.75498 lr=0.00010 gn=24.95374 time=59.40it/s +epoch=94 global_step=37000 loss=5.68162 loss_avg=5.77967 acc=0.43750 acc_top1_avg=0.42168 acc_top5_avg=0.75622 lr=0.00010 gn=28.31004 time=47.58it/s +epoch=94 global_step=37050 loss=5.86111 loss_avg=5.78434 acc=0.40625 acc_top1_avg=0.42087 acc_top5_avg=0.75818 lr=0.00010 gn=28.19837 time=47.49it/s +epoch=94 global_step=37100 loss=5.53154 loss_avg=5.78015 acc=0.44531 acc_top1_avg=0.42140 acc_top5_avg=0.75808 lr=0.00010 gn=18.89024 time=58.64it/s +====================Eval==================== +epoch=94 global_step=37145 loss=4.61441 test_loss_avg=3.93311 acc=0.00000 test_acc_avg=0.13639 test_acc_top5_avg=0.82422 time=242.28it/s +epoch=94 global_step=37145 loss=4.84390 test_loss_avg=3.12480 acc=0.00000 test_acc_avg=0.31588 test_acc_top5_avg=0.72361 time=252.75it/s +epoch=94 global_step=37145 loss=4.24379 test_loss_avg=3.22070 acc=0.00000 test_acc_avg=0.29589 test_acc_top5_avg=0.70134 time=821.45it/s +curr_acc 0.2959 +BEST_ACC 0.3318 +curr_acc_top5 0.7013 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=5.30929 loss_avg=5.68886 acc=0.46875 acc_top1_avg=0.42188 acc_top5_avg=0.74062 lr=0.00010 gn=26.97873 time=56.09it/s +epoch=95 global_step=37200 loss=6.28107 loss_avg=5.83792 acc=0.36719 acc_top1_avg=0.41278 acc_top5_avg=0.75313 lr=0.00010 gn=35.79439 time=56.44it/s +epoch=95 global_step=37250 loss=5.25571 loss_avg=5.78203 acc=0.48438 acc_top1_avg=0.41920 acc_top5_avg=0.75632 lr=0.00010 gn=28.96790 time=57.74it/s +epoch=95 global_step=37300 loss=6.60733 loss_avg=5.78048 acc=0.32812 acc_top1_avg=0.42077 acc_top5_avg=0.75680 lr=0.00010 gn=24.54237 time=53.67it/s +epoch=95 global_step=37350 loss=6.02509 loss_avg=5.77884 acc=0.40625 acc_top1_avg=0.42138 acc_top5_avg=0.75789 lr=0.00010 gn=29.84556 time=52.39it/s +epoch=95 global_step=37400 loss=5.96818 loss_avg=5.77596 acc=0.39844 acc_top1_avg=0.42157 acc_top5_avg=0.75885 lr=0.00010 gn=33.41534 time=50.16it/s +epoch=95 global_step=37450 loss=5.65043 loss_avg=5.75447 acc=0.42969 acc_top1_avg=0.42413 acc_top5_avg=0.76030 lr=0.00010 gn=27.97739 time=56.64it/s +epoch=95 global_step=37500 loss=5.92493 loss_avg=5.75674 acc=0.40625 acc_top1_avg=0.42392 acc_top5_avg=0.75889 lr=0.00010 gn=32.71778 time=57.04it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.96914 test_loss_avg=3.84662 acc=0.69531 test_acc_avg=0.16858 test_acc_top5_avg=0.69948 time=62.77it/s +epoch=95 global_step=37536 loss=4.21605 test_loss_avg=3.19839 acc=0.00000 test_acc_avg=0.29806 test_acc_top5_avg=0.69769 time=790.04it/s +curr_acc 0.2981 +BEST_ACC 0.3318 +curr_acc_top5 0.6977 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=5.54525 loss_avg=5.65964 acc=0.45312 acc_top1_avg=0.43750 acc_top5_avg=0.75056 lr=0.00010 gn=28.95442 time=63.50it/s +epoch=96 global_step=37600 loss=6.24832 loss_avg=5.71903 acc=0.37500 acc_top1_avg=0.43042 acc_top5_avg=0.75671 lr=0.00010 gn=25.36712 time=54.46it/s +epoch=96 global_step=37650 loss=6.18318 loss_avg=5.73784 acc=0.37500 acc_top1_avg=0.42695 acc_top5_avg=0.75487 lr=0.00010 gn=26.37765 time=55.40it/s +epoch=96 global_step=37700 loss=5.54003 loss_avg=5.73047 acc=0.43750 acc_top1_avg=0.42726 acc_top5_avg=0.75534 lr=0.00010 gn=24.21931 time=53.03it/s +epoch=96 global_step=37750 loss=4.93767 loss_avg=5.75254 acc=0.50781 acc_top1_avg=0.42494 acc_top5_avg=0.75526 lr=0.00010 gn=36.04367 time=62.35it/s +epoch=96 global_step=37800 loss=4.93455 loss_avg=5.74636 acc=0.51562 acc_top1_avg=0.42546 acc_top5_avg=0.75512 lr=0.00010 gn=33.95961 time=56.19it/s +epoch=96 global_step=37850 loss=5.40804 loss_avg=5.75605 acc=0.46094 acc_top1_avg=0.42421 acc_top5_avg=0.75542 lr=0.00010 gn=26.12652 time=59.26it/s +epoch=96 global_step=37900 loss=5.65972 loss_avg=5.75270 acc=0.42188 acc_top1_avg=0.42443 acc_top5_avg=0.75657 lr=0.00010 gn=30.50081 time=60.01it/s +====================Eval==================== +epoch=96 global_step=37927 loss=3.11573 test_loss_avg=3.70656 acc=0.27344 test_acc_avg=0.19678 test_acc_top5_avg=0.84570 time=125.05it/s +epoch=96 global_step=37927 loss=0.12825 test_loss_avg=3.26746 acc=0.95312 test_acc_avg=0.28563 test_acc_top5_avg=0.72289 time=227.19it/s +epoch=96 global_step=37927 loss=4.16874 test_loss_avg=3.24333 acc=0.00000 test_acc_avg=0.29035 test_acc_top5_avg=0.69887 time=581.25it/s +curr_acc 0.2903 +BEST_ACC 0.3318 +curr_acc_top5 0.6989 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=5.81429 loss_avg=5.65046 acc=0.42969 acc_top1_avg=0.43580 acc_top5_avg=0.74932 lr=0.00010 gn=36.99097 time=63.56it/s +epoch=97 global_step=38000 loss=5.80001 loss_avg=5.77211 acc=0.43750 acc_top1_avg=0.42177 acc_top5_avg=0.75064 lr=0.00010 gn=31.60733 time=54.58it/s +epoch=97 global_step=38050 loss=5.62409 loss_avg=5.79402 acc=0.44531 acc_top1_avg=0.41927 acc_top5_avg=0.75216 lr=0.00010 gn=27.63633 time=53.06it/s +epoch=97 global_step=38100 loss=5.81165 loss_avg=5.77551 acc=0.41406 acc_top1_avg=0.42106 acc_top5_avg=0.75641 lr=0.00010 gn=33.31184 time=63.14it/s +epoch=97 global_step=38150 loss=5.46470 loss_avg=5.77021 acc=0.44531 acc_top1_avg=0.42145 acc_top5_avg=0.75511 lr=0.00010 gn=32.97935 time=54.03it/s +epoch=97 global_step=38200 loss=6.15614 loss_avg=5.76260 acc=0.38281 acc_top1_avg=0.42202 acc_top5_avg=0.75670 lr=0.00010 gn=32.38504 time=56.56it/s +epoch=97 global_step=38250 loss=6.14138 loss_avg=5.75710 acc=0.38281 acc_top1_avg=0.42265 acc_top5_avg=0.75757 lr=0.00010 gn=29.61223 time=55.18it/s +epoch=97 global_step=38300 loss=5.15814 loss_avg=5.75060 acc=0.49219 acc_top1_avg=0.42372 acc_top5_avg=0.75838 lr=0.00010 gn=35.75620 time=58.08it/s +====================Eval==================== +epoch=97 global_step=38318 loss=5.14385 test_loss_avg=4.26218 acc=0.00000 test_acc_avg=0.09185 test_acc_top5_avg=0.66343 time=198.00it/s +epoch=97 global_step=38318 loss=4.01315 test_loss_avg=3.20783 acc=0.00000 test_acc_avg=0.29480 test_acc_top5_avg=0.70303 time=490.28it/s +curr_acc 0.2948 +BEST_ACC 0.3318 +curr_acc_top5 0.7030 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=5.08160 loss_avg=5.62927 acc=0.50000 acc_top1_avg=0.43799 acc_top5_avg=0.75610 lr=0.00010 gn=31.66609 time=56.99it/s +epoch=98 global_step=38400 loss=5.99878 loss_avg=5.67940 acc=0.39844 acc_top1_avg=0.43159 acc_top5_avg=0.75915 lr=0.00010 gn=29.05521 time=55.29it/s +epoch=98 global_step=38450 loss=6.34020 loss_avg=5.73370 acc=0.35156 acc_top1_avg=0.42566 acc_top5_avg=0.75894 lr=0.00010 gn=31.05587 time=38.55it/s +epoch=98 global_step=38500 loss=5.90639 loss_avg=5.75879 acc=0.39844 acc_top1_avg=0.42312 acc_top5_avg=0.75760 lr=0.00010 gn=22.17607 time=53.02it/s +epoch=98 global_step=38550 loss=6.89091 loss_avg=5.75496 acc=0.29688 acc_top1_avg=0.42349 acc_top5_avg=0.75855 lr=0.00010 gn=27.77064 time=52.82it/s +epoch=98 global_step=38600 loss=4.88023 loss_avg=5.74533 acc=0.52344 acc_top1_avg=0.42453 acc_top5_avg=0.75720 lr=0.00010 gn=29.46458 time=28.48it/s +epoch=98 global_step=38650 loss=5.83723 loss_avg=5.74485 acc=0.41406 acc_top1_avg=0.42456 acc_top5_avg=0.75711 lr=0.00010 gn=32.82590 time=62.64it/s +epoch=98 global_step=38700 loss=5.06100 loss_avg=5.74239 acc=0.51562 acc_top1_avg=0.42507 acc_top5_avg=0.75683 lr=0.00010 gn=38.08149 time=61.60it/s +====================Eval==================== +epoch=98 global_step=38709 loss=4.46718 test_loss_avg=4.81238 acc=0.07031 test_acc_avg=0.00879 test_acc_top5_avg=0.77441 time=240.44it/s +epoch=98 global_step=38709 loss=0.33692 test_loss_avg=3.64496 acc=0.88281 test_acc_avg=0.20137 test_acc_top5_avg=0.68534 time=233.02it/s +epoch=98 global_step=38709 loss=3.97408 test_loss_avg=3.19400 acc=0.00000 test_acc_avg=0.29480 test_acc_top5_avg=0.70589 time=826.79it/s +curr_acc 0.2948 +BEST_ACC 0.3318 +curr_acc_top5 0.7059 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=5.56299 loss_avg=5.78977 acc=0.44531 acc_top1_avg=0.42111 acc_top5_avg=0.76010 lr=0.00010 gn=28.23476 time=58.68it/s +epoch=99 global_step=38800 loss=6.21727 loss_avg=5.75231 acc=0.37500 acc_top1_avg=0.42454 acc_top5_avg=0.75816 lr=0.00010 gn=31.14688 time=63.19it/s +epoch=99 global_step=38850 loss=5.84256 loss_avg=5.76538 acc=0.40625 acc_top1_avg=0.42260 acc_top5_avg=0.75609 lr=0.00010 gn=27.30723 time=51.16it/s +epoch=99 global_step=38900 loss=5.47723 loss_avg=5.78133 acc=0.46094 acc_top1_avg=0.42114 acc_top5_avg=0.75474 lr=0.00010 gn=34.76262 time=55.06it/s +epoch=99 global_step=38950 loss=5.59649 loss_avg=5.76748 acc=0.43750 acc_top1_avg=0.42236 acc_top5_avg=0.75593 lr=0.00010 gn=20.85432 time=56.00it/s +epoch=99 global_step=39000 loss=6.05581 loss_avg=5.74484 acc=0.39062 acc_top1_avg=0.42494 acc_top5_avg=0.75730 lr=0.00010 gn=30.69583 time=49.33it/s +epoch=99 global_step=39050 loss=6.57096 loss_avg=5.73864 acc=0.35156 acc_top1_avg=0.42577 acc_top5_avg=0.75726 lr=0.00010 gn=33.07200 time=57.16it/s +epoch=99 global_step=39100 loss=5.80586 loss_avg=5.74390 acc=0.42500 acc_top1_avg=0.42516 acc_top5_avg=0.75701 lr=0.00010 gn=35.80033 time=71.61it/s +====================Eval==================== +epoch=99 global_step=39100 loss=4.84043 test_loss_avg=4.11340 acc=0.00000 test_acc_avg=0.11207 test_acc_top5_avg=0.68804 time=226.56it/s +epoch=99 global_step=39100 loss=4.15474 test_loss_avg=3.20880 acc=0.00000 test_acc_avg=0.29341 test_acc_top5_avg=0.70105 time=562.92it/s +epoch=99 global_step=39100 loss=4.15474 test_loss_avg=3.20880 acc=0.00000 test_acc_avg=0.29341 test_acc_top5_avg=0.70105 time=562.92it/s +curr_acc 0.2934 +BEST_ACC 0.3318 +curr_acc_top5 0.7010 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=5.68935 loss_avg=5.71909 acc=0.42969 acc_top1_avg=0.42672 acc_top5_avg=0.77000 lr=0.00010 gn=31.43862 time=54.39it/s +epoch=100 global_step=39200 loss=6.39728 loss_avg=5.69732 acc=0.34375 acc_top1_avg=0.43039 acc_top5_avg=0.77148 lr=0.00010 gn=28.28050 time=58.73it/s +epoch=100 global_step=39250 loss=5.26010 loss_avg=5.67048 acc=0.46875 acc_top1_avg=0.43339 acc_top5_avg=0.76802 lr=0.00010 gn=32.16520 time=56.36it/s +epoch=100 global_step=39300 loss=5.32714 loss_avg=5.70287 acc=0.47656 acc_top1_avg=0.42977 acc_top5_avg=0.76449 lr=0.00010 gn=26.90908 time=60.61it/s +epoch=100 global_step=39350 loss=5.63212 loss_avg=5.70317 acc=0.46094 acc_top1_avg=0.43003 acc_top5_avg=0.76422 lr=0.00010 gn=38.36131 time=55.13it/s +epoch=100 global_step=39400 loss=5.32136 loss_avg=5.71207 acc=0.46875 acc_top1_avg=0.42878 acc_top5_avg=0.76354 lr=0.00010 gn=27.58148 time=55.05it/s +epoch=100 global_step=39450 loss=6.11070 loss_avg=5.72966 acc=0.40625 acc_top1_avg=0.42658 acc_top5_avg=0.76109 lr=0.00010 gn=38.35762 time=60.46it/s +====================Eval==================== +epoch=100 global_step=39491 loss=4.21107 test_loss_avg=3.84606 acc=0.00000 test_acc_avg=0.16953 test_acc_top5_avg=0.68406 time=108.25it/s +epoch=100 global_step=39491 loss=4.02440 test_loss_avg=3.23394 acc=0.00000 test_acc_avg=0.29183 test_acc_top5_avg=0.69877 time=838.02it/s +curr_acc 0.2918 +BEST_ACC 0.3318 +curr_acc_top5 0.6988 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=6.32119 loss_avg=5.65453 acc=0.35156 acc_top1_avg=0.43316 acc_top5_avg=0.77691 lr=0.00010 gn=20.37775 time=62.86it/s +epoch=101 global_step=39550 loss=5.13701 loss_avg=5.73331 acc=0.49219 acc_top1_avg=0.42492 acc_top5_avg=0.76046 lr=0.00010 gn=27.80530 time=53.72it/s +epoch=101 global_step=39600 loss=5.02433 loss_avg=5.68866 acc=0.50000 acc_top1_avg=0.43055 acc_top5_avg=0.76175 lr=0.00010 gn=32.58732 time=39.37it/s +epoch=101 global_step=39650 loss=5.70239 loss_avg=5.69042 acc=0.42969 acc_top1_avg=0.43042 acc_top5_avg=0.76233 lr=0.00010 gn=26.38676 time=53.33it/s +epoch=101 global_step=39700 loss=5.14837 loss_avg=5.70426 acc=0.48438 acc_top1_avg=0.42875 acc_top5_avg=0.76009 lr=0.00010 gn=28.05722 time=59.41it/s +epoch=101 global_step=39750 loss=6.17864 loss_avg=5.71298 acc=0.38281 acc_top1_avg=0.42800 acc_top5_avg=0.75950 lr=0.00010 gn=37.13714 time=62.65it/s +epoch=101 global_step=39800 loss=5.59386 loss_avg=5.72247 acc=0.43750 acc_top1_avg=0.42713 acc_top5_avg=0.75860 lr=0.00010 gn=31.87604 time=53.89it/s +epoch=101 global_step=39850 loss=5.23303 loss_avg=5.72820 acc=0.48438 acc_top1_avg=0.42644 acc_top5_avg=0.75910 lr=0.00010 gn=29.09402 time=63.39it/s +====================Eval==================== +epoch=101 global_step=39882 loss=4.44986 test_loss_avg=3.73652 acc=0.00000 test_acc_avg=0.16890 test_acc_top5_avg=0.85677 time=238.87it/s +epoch=101 global_step=39882 loss=3.27989 test_loss_avg=3.03359 acc=0.29688 test_acc_avg=0.32989 test_acc_top5_avg=0.73834 time=233.78it/s +epoch=101 global_step=39882 loss=4.09451 test_loss_avg=3.19296 acc=0.00000 test_acc_avg=0.29648 test_acc_top5_avg=0.69798 time=503.82it/s +curr_acc 0.2965 +BEST_ACC 0.3318 +curr_acc_top5 0.6980 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=6.21899 loss_avg=5.82799 acc=0.36719 acc_top1_avg=0.41233 acc_top5_avg=0.74132 lr=0.00010 gn=23.75784 time=54.32it/s +epoch=102 global_step=39950 loss=5.40014 loss_avg=5.78416 acc=0.46094 acc_top1_avg=0.41854 acc_top5_avg=0.75299 lr=0.00010 gn=33.25454 time=56.54it/s +epoch=102 global_step=40000 loss=5.35855 loss_avg=5.75594 acc=0.46875 acc_top1_avg=0.42327 acc_top5_avg=0.75563 lr=0.00010 gn=30.32688 time=38.65it/s +epoch=102 global_step=40050 loss=5.34553 loss_avg=5.73926 acc=0.47656 acc_top1_avg=0.42541 acc_top5_avg=0.75586 lr=0.00010 gn=33.56165 time=55.32it/s +epoch=102 global_step=40100 loss=5.51857 loss_avg=5.73115 acc=0.44531 acc_top1_avg=0.42653 acc_top5_avg=0.75699 lr=0.00010 gn=25.23767 time=54.13it/s +epoch=102 global_step=40150 loss=6.72899 loss_avg=5.72601 acc=0.32031 acc_top1_avg=0.42715 acc_top5_avg=0.75752 lr=0.00010 gn=28.02925 time=60.25it/s +epoch=102 global_step=40200 loss=6.02026 loss_avg=5.72237 acc=0.40625 acc_top1_avg=0.42748 acc_top5_avg=0.75816 lr=0.00010 gn=39.22727 time=55.03it/s +epoch=102 global_step=40250 loss=5.37218 loss_avg=5.72221 acc=0.46094 acc_top1_avg=0.42756 acc_top5_avg=0.75817 lr=0.00010 gn=33.25172 time=52.24it/s +====================Eval==================== +epoch=102 global_step=40273 loss=1.03942 test_loss_avg=4.09948 acc=0.71875 test_acc_avg=0.12109 test_acc_top5_avg=0.68025 time=216.16it/s +epoch=102 global_step=40273 loss=4.02741 test_loss_avg=3.21749 acc=0.00000 test_acc_avg=0.29084 test_acc_top5_avg=0.70500 time=828.26it/s +curr_acc 0.2908 +BEST_ACC 0.3318 +curr_acc_top5 0.7050 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=5.54783 loss_avg=5.66525 acc=0.43750 acc_top1_avg=0.43258 acc_top5_avg=0.76244 lr=0.00010 gn=16.39549 time=63.82it/s +epoch=103 global_step=40350 loss=5.63430 loss_avg=5.68135 acc=0.42188 acc_top1_avg=0.43202 acc_top5_avg=0.76075 lr=0.00010 gn=24.67339 time=58.06it/s +epoch=103 global_step=40400 loss=5.62709 loss_avg=5.69110 acc=0.42969 acc_top1_avg=0.43061 acc_top5_avg=0.75997 lr=0.00010 gn=26.71795 time=41.41it/s +epoch=103 global_step=40450 loss=5.64791 loss_avg=5.68082 acc=0.44531 acc_top1_avg=0.43163 acc_top5_avg=0.75883 lr=0.00010 gn=40.56259 time=30.65it/s +epoch=103 global_step=40500 loss=6.16758 loss_avg=5.69422 acc=0.36719 acc_top1_avg=0.43000 acc_top5_avg=0.75943 lr=0.00010 gn=21.13482 time=46.01it/s +epoch=103 global_step=40550 loss=6.07095 loss_avg=5.70832 acc=0.38281 acc_top1_avg=0.42864 acc_top5_avg=0.75784 lr=0.00010 gn=30.28658 time=52.07it/s +epoch=103 global_step=40600 loss=6.03595 loss_avg=5.71020 acc=0.39844 acc_top1_avg=0.42849 acc_top5_avg=0.75908 lr=0.00010 gn=36.80983 time=54.99it/s +epoch=103 global_step=40650 loss=5.22091 loss_avg=5.71595 acc=0.49219 acc_top1_avg=0.42793 acc_top5_avg=0.75792 lr=0.00010 gn=32.35694 time=54.15it/s +====================Eval==================== +epoch=103 global_step=40664 loss=2.35730 test_loss_avg=3.94458 acc=0.47656 test_acc_avg=0.17308 test_acc_top5_avg=0.85156 time=225.22it/s +epoch=103 global_step=40664 loss=0.25576 test_loss_avg=3.39098 acc=0.93750 test_acc_avg=0.25657 test_acc_top5_avg=0.70995 time=242.42it/s +epoch=103 global_step=40664 loss=4.10139 test_loss_avg=3.21599 acc=0.00000 test_acc_avg=0.29233 test_acc_top5_avg=0.70194 time=563.60it/s +curr_acc 0.2923 +BEST_ACC 0.3318 +curr_acc_top5 0.7019 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=5.81856 loss_avg=5.67249 acc=0.41406 acc_top1_avg=0.43446 acc_top5_avg=0.75673 lr=0.00010 gn=29.55661 time=60.98it/s +epoch=104 global_step=40750 loss=5.55013 loss_avg=5.69703 acc=0.43750 acc_top1_avg=0.42978 acc_top5_avg=0.75854 lr=0.00010 gn=28.31819 time=59.91it/s +epoch=104 global_step=40800 loss=6.13174 loss_avg=5.71925 acc=0.39062 acc_top1_avg=0.42785 acc_top5_avg=0.75977 lr=0.00010 gn=40.45379 time=54.54it/s +epoch=104 global_step=40850 loss=5.40831 loss_avg=5.72646 acc=0.46875 acc_top1_avg=0.42742 acc_top5_avg=0.75848 lr=0.00010 gn=33.76528 time=57.98it/s +epoch=104 global_step=40900 loss=5.21015 loss_avg=5.71386 acc=0.50000 acc_top1_avg=0.42899 acc_top5_avg=0.76099 lr=0.00010 gn=32.55768 time=51.95it/s +epoch=104 global_step=40950 loss=6.77637 loss_avg=5.70753 acc=0.31250 acc_top1_avg=0.42963 acc_top5_avg=0.76046 lr=0.00010 gn=25.59415 time=57.25it/s +epoch=104 global_step=41000 loss=5.85743 loss_avg=5.70423 acc=0.41406 acc_top1_avg=0.42962 acc_top5_avg=0.76093 lr=0.00010 gn=30.15338 time=53.54it/s +epoch=104 global_step=41050 loss=5.40208 loss_avg=5.70815 acc=0.44531 acc_top1_avg=0.42888 acc_top5_avg=0.76087 lr=0.00010 gn=22.67442 time=54.66it/s +====================Eval==================== +epoch=104 global_step=41055 loss=4.96821 test_loss_avg=4.21948 acc=0.00000 test_acc_avg=0.08961 test_acc_top5_avg=0.63971 time=239.40it/s +epoch=104 global_step=41055 loss=4.03777 test_loss_avg=3.19286 acc=0.00000 test_acc_avg=0.29193 test_acc_top5_avg=0.70362 time=438.37it/s +curr_acc 0.2919 +BEST_ACC 0.3318 +curr_acc_top5 0.7036 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=5.52915 loss_avg=5.68015 acc=0.44531 acc_top1_avg=0.43368 acc_top5_avg=0.76354 lr=0.00010 gn=29.62160 time=51.05it/s +epoch=105 global_step=41150 loss=6.06983 loss_avg=5.72866 acc=0.38281 acc_top1_avg=0.42582 acc_top5_avg=0.75732 lr=0.00010 gn=35.41283 time=55.64it/s +epoch=105 global_step=41200 loss=5.95813 loss_avg=5.68617 acc=0.39844 acc_top1_avg=0.43055 acc_top5_avg=0.75857 lr=0.00010 gn=27.35036 time=55.36it/s +epoch=105 global_step=41250 loss=5.68564 loss_avg=5.69265 acc=0.43750 acc_top1_avg=0.43001 acc_top5_avg=0.75837 lr=0.00010 gn=32.35506 time=63.42it/s +epoch=105 global_step=41300 loss=5.30343 loss_avg=5.70383 acc=0.46875 acc_top1_avg=0.42899 acc_top5_avg=0.75845 lr=0.00010 gn=27.47048 time=58.07it/s +epoch=105 global_step=41350 loss=4.92394 loss_avg=5.69777 acc=0.50781 acc_top1_avg=0.42985 acc_top5_avg=0.75900 lr=0.00010 gn=34.25466 time=55.51it/s +epoch=105 global_step=41400 loss=5.84476 loss_avg=5.70526 acc=0.40625 acc_top1_avg=0.42892 acc_top5_avg=0.75797 lr=0.00010 gn=28.15612 time=59.54it/s +====================Eval==================== +epoch=105 global_step=41446 loss=4.91347 test_loss_avg=4.90528 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.78594 time=231.97it/s +epoch=105 global_step=41446 loss=3.08942 test_loss_avg=3.78512 acc=0.27344 test_acc_avg=0.16520 test_acc_top5_avg=0.67102 time=180.87it/s +epoch=105 global_step=41446 loss=4.12946 test_loss_avg=3.18104 acc=0.00000 test_acc_avg=0.29509 test_acc_top5_avg=0.70461 time=834.52it/s +curr_acc 0.2951 +BEST_ACC 0.3318 +curr_acc_top5 0.7046 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=5.36089 loss_avg=5.51628 acc=0.46875 acc_top1_avg=0.44727 acc_top5_avg=0.77148 lr=0.00010 gn=30.49296 time=22.00it/s +epoch=106 global_step=41500 loss=6.19630 loss_avg=5.70114 acc=0.38281 acc_top1_avg=0.42766 acc_top5_avg=0.75637 lr=0.00010 gn=40.21417 time=42.76it/s +epoch=106 global_step=41550 loss=5.18324 loss_avg=5.64903 acc=0.48438 acc_top1_avg=0.43307 acc_top5_avg=0.75999 lr=0.00010 gn=23.54385 time=60.63it/s +epoch=106 global_step=41600 loss=5.29738 loss_avg=5.67753 acc=0.46875 acc_top1_avg=0.43035 acc_top5_avg=0.75802 lr=0.00010 gn=36.77271 time=63.46it/s +epoch=106 global_step=41650 loss=5.96839 loss_avg=5.70691 acc=0.39844 acc_top1_avg=0.42781 acc_top5_avg=0.75678 lr=0.00010 gn=38.35522 time=55.70it/s +epoch=106 global_step=41700 loss=5.95007 loss_avg=5.70436 acc=0.40625 acc_top1_avg=0.42855 acc_top5_avg=0.75667 lr=0.00010 gn=35.87803 time=54.14it/s +epoch=106 global_step=41750 loss=5.92948 loss_avg=5.69755 acc=0.39844 acc_top1_avg=0.42976 acc_top5_avg=0.75843 lr=0.00010 gn=23.44970 time=54.93it/s +epoch=106 global_step=41800 loss=6.06303 loss_avg=5.69244 acc=0.38281 acc_top1_avg=0.43026 acc_top5_avg=0.75856 lr=0.00010 gn=24.17305 time=56.39it/s +====================Eval==================== +epoch=106 global_step=41837 loss=4.77068 test_loss_avg=4.02298 acc=0.00000 test_acc_avg=0.12139 test_acc_top5_avg=0.75841 time=235.03it/s +epoch=106 global_step=41837 loss=4.57630 test_loss_avg=3.15969 acc=0.00000 test_acc_avg=0.30315 test_acc_top5_avg=0.70970 time=252.62it/s +epoch=106 global_step=41837 loss=4.10543 test_loss_avg=3.20889 acc=0.00000 test_acc_avg=0.29163 test_acc_top5_avg=0.69818 time=474.47it/s +curr_acc 0.2916 +BEST_ACC 0.3318 +curr_acc_top5 0.6982 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=4.85790 loss_avg=5.55607 acc=0.53125 acc_top1_avg=0.44892 acc_top5_avg=0.75962 lr=0.00010 gn=41.02484 time=54.67it/s +epoch=107 global_step=41900 loss=6.03334 loss_avg=5.63574 acc=0.39062 acc_top1_avg=0.43676 acc_top5_avg=0.75409 lr=0.00010 gn=27.66645 time=50.86it/s +epoch=107 global_step=41950 loss=6.10939 loss_avg=5.63703 acc=0.38281 acc_top1_avg=0.43605 acc_top5_avg=0.75878 lr=0.00010 gn=28.11288 time=55.75it/s +epoch=107 global_step=42000 loss=5.65635 loss_avg=5.67766 acc=0.43750 acc_top1_avg=0.43184 acc_top5_avg=0.75709 lr=0.00010 gn=28.79939 time=49.49it/s +epoch=107 global_step=42050 loss=5.64838 loss_avg=5.68581 acc=0.43750 acc_top1_avg=0.43064 acc_top5_avg=0.75895 lr=0.00010 gn=31.30033 time=54.74it/s +epoch=107 global_step=42100 loss=6.10127 loss_avg=5.68787 acc=0.38281 acc_top1_avg=0.43040 acc_top5_avg=0.75968 lr=0.00010 gn=28.16508 time=55.60it/s +epoch=107 global_step=42150 loss=6.16081 loss_avg=5.68970 acc=0.37500 acc_top1_avg=0.43006 acc_top5_avg=0.75861 lr=0.00010 gn=28.69976 time=60.33it/s +epoch=107 global_step=42200 loss=5.55410 loss_avg=5.68656 acc=0.45312 acc_top1_avg=0.43053 acc_top5_avg=0.75958 lr=0.00010 gn=35.28907 time=54.96it/s +====================Eval==================== +epoch=107 global_step=42228 loss=1.65416 test_loss_avg=3.81602 acc=0.61719 test_acc_avg=0.17337 test_acc_top5_avg=0.70013 time=127.59it/s +epoch=107 global_step=42228 loss=4.09412 test_loss_avg=3.23124 acc=0.00000 test_acc_avg=0.28847 test_acc_top5_avg=0.70184 time=857.91it/s +curr_acc 0.2885 +BEST_ACC 0.3318 +curr_acc_top5 0.7018 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=6.02977 loss_avg=5.59874 acc=0.39062 acc_top1_avg=0.43999 acc_top5_avg=0.75994 lr=0.00010 gn=31.81692 time=52.23it/s +epoch=108 global_step=42300 loss=5.48758 loss_avg=5.63569 acc=0.43750 acc_top1_avg=0.43446 acc_top5_avg=0.76280 lr=0.00010 gn=37.59790 time=54.39it/s +epoch=108 global_step=42350 loss=5.07326 loss_avg=5.66949 acc=0.50000 acc_top1_avg=0.43090 acc_top5_avg=0.75730 lr=0.00010 gn=34.40071 time=52.95it/s +epoch=108 global_step=42400 loss=5.91708 loss_avg=5.66807 acc=0.41406 acc_top1_avg=0.43110 acc_top5_avg=0.75777 lr=0.00010 gn=31.18712 time=60.86it/s +epoch=108 global_step=42450 loss=6.36903 loss_avg=5.68269 acc=0.35156 acc_top1_avg=0.43011 acc_top5_avg=0.75630 lr=0.00010 gn=28.61192 time=54.80it/s +epoch=108 global_step=42500 loss=4.90672 loss_avg=5.69672 acc=0.52344 acc_top1_avg=0.42911 acc_top5_avg=0.75753 lr=0.00010 gn=29.92416 time=54.44it/s +epoch=108 global_step=42550 loss=6.48617 loss_avg=5.68940 acc=0.33594 acc_top1_avg=0.42991 acc_top5_avg=0.75934 lr=0.00010 gn=24.45559 time=55.18it/s +epoch=108 global_step=42600 loss=5.99264 loss_avg=5.68975 acc=0.40625 acc_top1_avg=0.42967 acc_top5_avg=0.75949 lr=0.00010 gn=33.19932 time=51.65it/s +====================Eval==================== +epoch=108 global_step=42619 loss=4.67788 test_loss_avg=3.78408 acc=0.00000 test_acc_avg=0.18403 test_acc_top5_avg=0.83767 time=237.49it/s +epoch=108 global_step=42619 loss=0.34956 test_loss_avg=3.14973 acc=0.92969 test_acc_avg=0.30480 test_acc_top5_avg=0.73254 time=232.38it/s +epoch=108 global_step=42619 loss=4.26147 test_loss_avg=3.22650 acc=0.00000 test_acc_avg=0.29064 test_acc_top5_avg=0.70332 time=731.35it/s +curr_acc 0.2906 +BEST_ACC 0.3318 +curr_acc_top5 0.7033 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=5.96824 loss_avg=5.76260 acc=0.40625 acc_top1_avg=0.42288 acc_top5_avg=0.74622 lr=0.00010 gn=33.05636 time=57.25it/s +epoch=109 global_step=42700 loss=6.27823 loss_avg=5.70864 acc=0.35938 acc_top1_avg=0.42863 acc_top5_avg=0.75723 lr=0.00010 gn=28.05401 time=61.03it/s +epoch=109 global_step=42750 loss=5.69785 loss_avg=5.69959 acc=0.42969 acc_top1_avg=0.43010 acc_top5_avg=0.75716 lr=0.00010 gn=35.14066 time=53.03it/s +epoch=109 global_step=42800 loss=5.53064 loss_avg=5.70274 acc=0.46094 acc_top1_avg=0.42947 acc_top5_avg=0.75738 lr=0.00010 gn=31.63997 time=53.65it/s +epoch=109 global_step=42850 loss=5.24405 loss_avg=5.71288 acc=0.46875 acc_top1_avg=0.42823 acc_top5_avg=0.75747 lr=0.00010 gn=26.93358 time=54.54it/s +epoch=109 global_step=42900 loss=5.12257 loss_avg=5.69839 acc=0.49219 acc_top1_avg=0.42983 acc_top5_avg=0.75678 lr=0.00010 gn=30.42059 time=56.55it/s +epoch=109 global_step=42950 loss=5.93336 loss_avg=5.69528 acc=0.40625 acc_top1_avg=0.43009 acc_top5_avg=0.75810 lr=0.00010 gn=34.07896 time=55.63it/s +epoch=109 global_step=43000 loss=5.88497 loss_avg=5.69275 acc=0.41406 acc_top1_avg=0.43002 acc_top5_avg=0.75828 lr=0.00010 gn=29.99858 time=55.01it/s +====================Eval==================== +epoch=109 global_step=43010 loss=4.95197 test_loss_avg=4.25696 acc=0.00000 test_acc_avg=0.08634 test_acc_top5_avg=0.64884 time=150.89it/s +epoch=109 global_step=43010 loss=4.02594 test_loss_avg=3.17248 acc=0.00000 test_acc_avg=0.29539 test_acc_top5_avg=0.69502 time=735.71it/s +curr_acc 0.2954 +BEST_ACC 0.3318 +curr_acc_top5 0.6950 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=6.09726 loss_avg=5.60834 acc=0.39062 acc_top1_avg=0.43848 acc_top5_avg=0.76055 lr=0.00010 gn=32.21883 time=52.75it/s +epoch=110 global_step=43100 loss=6.09826 loss_avg=5.63110 acc=0.39062 acc_top1_avg=0.43689 acc_top5_avg=0.76102 lr=0.00010 gn=35.44484 time=59.61it/s +epoch=110 global_step=43150 loss=6.88466 loss_avg=5.64879 acc=0.29688 acc_top1_avg=0.43443 acc_top5_avg=0.75971 lr=0.00010 gn=29.64011 time=58.57it/s +epoch=110 global_step=43200 loss=5.29422 loss_avg=5.66040 acc=0.46875 acc_top1_avg=0.43359 acc_top5_avg=0.75966 lr=0.00010 gn=26.98576 time=60.83it/s +epoch=110 global_step=43250 loss=5.68410 loss_avg=5.67049 acc=0.42969 acc_top1_avg=0.43232 acc_top5_avg=0.75794 lr=0.00010 gn=37.46111 time=56.63it/s +epoch=110 global_step=43300 loss=5.99590 loss_avg=5.67952 acc=0.39844 acc_top1_avg=0.43122 acc_top5_avg=0.75749 lr=0.00010 gn=31.89996 time=52.57it/s +epoch=110 global_step=43350 loss=5.96280 loss_avg=5.66970 acc=0.40625 acc_top1_avg=0.43267 acc_top5_avg=0.75731 lr=0.00010 gn=27.25784 time=55.00it/s +epoch=110 global_step=43400 loss=6.06606 loss_avg=5.67613 acc=0.40625 acc_top1_avg=0.43213 acc_top5_avg=0.75797 lr=0.00010 gn=33.37928 time=59.24it/s +====================Eval==================== +epoch=110 global_step=43401 loss=2.84873 test_loss_avg=4.36027 acc=0.39062 test_acc_avg=0.08359 test_acc_top5_avg=0.79141 time=155.20it/s +epoch=110 global_step=43401 loss=0.15913 test_loss_avg=3.55376 acc=0.96094 test_acc_avg=0.22070 test_acc_top5_avg=0.69557 time=238.56it/s +epoch=110 global_step=43401 loss=4.08548 test_loss_avg=3.22685 acc=0.00000 test_acc_avg=0.29015 test_acc_top5_avg=0.69689 time=857.20it/s +curr_acc 0.2902 +BEST_ACC 0.3318 +curr_acc_top5 0.6969 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=5.66361 loss_avg=5.61184 acc=0.42188 acc_top1_avg=0.43718 acc_top5_avg=0.76148 lr=0.00010 gn=31.19583 time=41.32it/s +epoch=111 global_step=43500 loss=5.93189 loss_avg=5.62297 acc=0.39844 acc_top1_avg=0.43624 acc_top5_avg=0.76042 lr=0.00010 gn=19.94894 time=61.80it/s +epoch=111 global_step=43550 loss=5.97672 loss_avg=5.64665 acc=0.38281 acc_top1_avg=0.43372 acc_top5_avg=0.76064 lr=0.00010 gn=21.58823 time=45.15it/s +epoch=111 global_step=43600 loss=5.67404 loss_avg=5.64576 acc=0.44531 acc_top1_avg=0.43389 acc_top5_avg=0.76052 lr=0.00010 gn=34.09150 time=61.97it/s +epoch=111 global_step=43650 loss=5.63231 loss_avg=5.65741 acc=0.44531 acc_top1_avg=0.43286 acc_top5_avg=0.76048 lr=0.00010 gn=32.75398 time=58.49it/s +epoch=111 global_step=43700 loss=6.12953 loss_avg=5.66254 acc=0.37500 acc_top1_avg=0.43259 acc_top5_avg=0.75980 lr=0.00010 gn=25.49559 time=59.04it/s +epoch=111 global_step=43750 loss=6.03502 loss_avg=5.66640 acc=0.38281 acc_top1_avg=0.43231 acc_top5_avg=0.75873 lr=0.00010 gn=25.55457 time=54.11it/s +====================Eval==================== +epoch=111 global_step=43792 loss=4.88355 test_loss_avg=4.19801 acc=0.00000 test_acc_avg=0.09451 test_acc_top5_avg=0.62903 time=221.28it/s +epoch=111 global_step=43792 loss=4.00209 test_loss_avg=3.22226 acc=0.00000 test_acc_avg=0.28807 test_acc_top5_avg=0.69551 time=805.82it/s +curr_acc 0.2881 +BEST_ACC 0.3318 +curr_acc_top5 0.6955 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=6.14274 loss_avg=5.86723 acc=0.38281 acc_top1_avg=0.41602 acc_top5_avg=0.75684 lr=0.00010 gn=26.14661 time=61.82it/s +epoch=112 global_step=43850 loss=5.66206 loss_avg=5.63625 acc=0.42969 acc_top1_avg=0.43817 acc_top5_avg=0.75929 lr=0.00010 gn=31.16559 time=56.63it/s +epoch=112 global_step=43900 loss=5.82084 loss_avg=5.66756 acc=0.41406 acc_top1_avg=0.43381 acc_top5_avg=0.75760 lr=0.00010 gn=31.26917 time=55.08it/s +epoch=112 global_step=43950 loss=5.42914 loss_avg=5.65855 acc=0.47656 acc_top1_avg=0.43424 acc_top5_avg=0.75910 lr=0.00010 gn=35.06129 time=51.52it/s +epoch=112 global_step=44000 loss=6.07328 loss_avg=5.66990 acc=0.38281 acc_top1_avg=0.43314 acc_top5_avg=0.75909 lr=0.00010 gn=29.36641 time=60.43it/s +epoch=112 global_step=44050 loss=5.82691 loss_avg=5.66583 acc=0.40625 acc_top1_avg=0.43323 acc_top5_avg=0.75942 lr=0.00010 gn=22.94514 time=54.29it/s +epoch=112 global_step=44100 loss=5.96788 loss_avg=5.65964 acc=0.39062 acc_top1_avg=0.43387 acc_top5_avg=0.76045 lr=0.00010 gn=28.24875 time=55.59it/s +epoch=112 global_step=44150 loss=5.68188 loss_avg=5.66585 acc=0.43750 acc_top1_avg=0.43320 acc_top5_avg=0.75895 lr=0.00010 gn=35.83435 time=54.78it/s +====================Eval==================== +epoch=112 global_step=44183 loss=4.82289 test_loss_avg=4.80614 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.75391 time=236.18it/s +epoch=112 global_step=44183 loss=4.13369 test_loss_avg=3.79971 acc=0.00000 test_acc_avg=0.16496 test_acc_top5_avg=0.67698 time=239.18it/s +epoch=112 global_step=44183 loss=3.99066 test_loss_avg=3.18079 acc=0.00000 test_acc_avg=0.29302 test_acc_top5_avg=0.70372 time=512.94it/s +curr_acc 0.2930 +BEST_ACC 0.3318 +curr_acc_top5 0.7037 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=5.06305 loss_avg=5.77005 acc=0.50000 acc_top1_avg=0.42371 acc_top5_avg=0.74908 lr=0.00010 gn=31.81440 time=63.57it/s +epoch=113 global_step=44250 loss=6.16652 loss_avg=5.70458 acc=0.37500 acc_top1_avg=0.42922 acc_top5_avg=0.75816 lr=0.00010 gn=27.99996 time=57.55it/s +epoch=113 global_step=44300 loss=5.97182 loss_avg=5.69110 acc=0.39844 acc_top1_avg=0.43009 acc_top5_avg=0.76315 lr=0.00010 gn=28.40915 time=54.77it/s +epoch=113 global_step=44350 loss=5.78617 loss_avg=5.67539 acc=0.42969 acc_top1_avg=0.43231 acc_top5_avg=0.76179 lr=0.00010 gn=31.00312 time=56.11it/s +epoch=113 global_step=44400 loss=5.50319 loss_avg=5.66768 acc=0.46875 acc_top1_avg=0.43286 acc_top5_avg=0.76062 lr=0.00010 gn=36.92487 time=54.27it/s +epoch=113 global_step=44450 loss=5.99027 loss_avg=5.67131 acc=0.39062 acc_top1_avg=0.43258 acc_top5_avg=0.75767 lr=0.00010 gn=39.68246 time=55.92it/s +epoch=113 global_step=44500 loss=5.80779 loss_avg=5.66751 acc=0.41406 acc_top1_avg=0.43314 acc_top5_avg=0.75752 lr=0.00010 gn=27.37227 time=56.01it/s +epoch=113 global_step=44550 loss=6.02331 loss_avg=5.66710 acc=0.40625 acc_top1_avg=0.43307 acc_top5_avg=0.75896 lr=0.00010 gn=39.53840 time=55.02it/s +====================Eval==================== +epoch=113 global_step=44574 loss=4.30146 test_loss_avg=3.99905 acc=0.00000 test_acc_avg=0.13213 test_acc_top5_avg=0.84307 time=239.33it/s +epoch=113 global_step=44574 loss=4.53755 test_loss_avg=3.12149 acc=0.00000 test_acc_avg=0.31453 test_acc_top5_avg=0.72774 time=246.55it/s +epoch=113 global_step=44574 loss=4.03931 test_loss_avg=3.21925 acc=0.00000 test_acc_avg=0.29064 test_acc_top5_avg=0.70550 time=757.37it/s +curr_acc 0.2906 +BEST_ACC 0.3318 +curr_acc_top5 0.7055 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=4.86550 loss_avg=5.72642 acc=0.51562 acc_top1_avg=0.42578 acc_top5_avg=0.76262 lr=0.00010 gn=33.44020 time=53.95it/s +epoch=114 global_step=44650 loss=5.49207 loss_avg=5.69192 acc=0.46875 acc_top1_avg=0.43020 acc_top5_avg=0.76347 lr=0.00010 gn=38.17065 time=59.15it/s +epoch=114 global_step=44700 loss=5.99015 loss_avg=5.66682 acc=0.39062 acc_top1_avg=0.43310 acc_top5_avg=0.76166 lr=0.00010 gn=30.95758 time=55.29it/s +epoch=114 global_step=44750 loss=5.62230 loss_avg=5.63191 acc=0.42188 acc_top1_avg=0.43666 acc_top5_avg=0.76301 lr=0.00010 gn=32.32443 time=56.83it/s +epoch=114 global_step=44800 loss=6.25502 loss_avg=5.63037 acc=0.36719 acc_top1_avg=0.43660 acc_top5_avg=0.76162 lr=0.00010 gn=29.55238 time=54.99it/s +epoch=114 global_step=44850 loss=5.42650 loss_avg=5.63721 acc=0.45312 acc_top1_avg=0.43603 acc_top5_avg=0.76093 lr=0.00010 gn=34.31110 time=53.52it/s +epoch=114 global_step=44900 loss=5.57494 loss_avg=5.64186 acc=0.46094 acc_top1_avg=0.43573 acc_top5_avg=0.76122 lr=0.00010 gn=41.14022 time=60.41it/s +epoch=114 global_step=44950 loss=6.01767 loss_avg=5.64967 acc=0.39844 acc_top1_avg=0.43492 acc_top5_avg=0.75983 lr=0.00010 gn=36.06191 time=59.20it/s +====================Eval==================== +epoch=114 global_step=44965 loss=1.11586 test_loss_avg=3.97280 acc=0.72656 test_acc_avg=0.14329 test_acc_top5_avg=0.66974 time=153.80it/s +epoch=114 global_step=44965 loss=4.11068 test_loss_avg=3.22516 acc=0.00000 test_acc_avg=0.28728 test_acc_top5_avg=0.69769 time=573.23it/s +curr_acc 0.2873 +BEST_ACC 0.3318 +curr_acc_top5 0.6977 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=5.37221 loss_avg=5.64714 acc=0.46875 acc_top1_avg=0.43482 acc_top5_avg=0.76496 lr=0.00010 gn=38.38795 time=54.41it/s +epoch=115 global_step=45050 loss=5.76812 loss_avg=5.68583 acc=0.42969 acc_top1_avg=0.42978 acc_top5_avg=0.75588 lr=0.00010 gn=36.19225 time=56.91it/s +epoch=115 global_step=45100 loss=5.72733 loss_avg=5.69576 acc=0.42969 acc_top1_avg=0.42876 acc_top5_avg=0.75735 lr=0.00010 gn=31.45454 time=59.92it/s +epoch=115 global_step=45150 loss=6.32831 loss_avg=5.68243 acc=0.35938 acc_top1_avg=0.43036 acc_top5_avg=0.75655 lr=0.00010 gn=31.12149 time=58.89it/s +epoch=115 global_step=45200 loss=5.42590 loss_avg=5.66261 acc=0.46875 acc_top1_avg=0.43235 acc_top5_avg=0.75821 lr=0.00010 gn=36.98537 time=52.40it/s +epoch=115 global_step=45250 loss=5.25422 loss_avg=5.66734 acc=0.46094 acc_top1_avg=0.43188 acc_top5_avg=0.75877 lr=0.00010 gn=29.81799 time=57.81it/s +epoch=115 global_step=45300 loss=6.02516 loss_avg=5.65567 acc=0.39844 acc_top1_avg=0.43321 acc_top5_avg=0.76045 lr=0.00010 gn=34.45068 time=56.00it/s +epoch=115 global_step=45350 loss=5.39633 loss_avg=5.65924 acc=0.44531 acc_top1_avg=0.43287 acc_top5_avg=0.75883 lr=0.00010 gn=24.68300 time=61.90it/s +====================Eval==================== +epoch=115 global_step=45356 loss=2.83873 test_loss_avg=3.91184 acc=0.36719 test_acc_avg=0.16823 test_acc_top5_avg=0.82396 time=237.92it/s +epoch=115 global_step=45356 loss=0.13304 test_loss_avg=3.28769 acc=0.96875 test_acc_avg=0.27260 test_acc_top5_avg=0.71106 time=236.85it/s +epoch=115 global_step=45356 loss=4.04177 test_loss_avg=3.20855 acc=0.00000 test_acc_avg=0.28778 test_acc_top5_avg=0.69828 time=497.72it/s +curr_acc 0.2878 +BEST_ACC 0.3318 +curr_acc_top5 0.6983 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=6.02606 loss_avg=5.59608 acc=0.38281 acc_top1_avg=0.43999 acc_top5_avg=0.75160 lr=0.00010 gn=34.20347 time=53.61it/s +epoch=116 global_step=45450 loss=6.08002 loss_avg=5.62689 acc=0.38281 acc_top1_avg=0.43717 acc_top5_avg=0.75241 lr=0.00010 gn=29.62569 time=58.56it/s +epoch=116 global_step=45500 loss=4.64003 loss_avg=5.63862 acc=0.53125 acc_top1_avg=0.43544 acc_top5_avg=0.75635 lr=0.00010 gn=40.03633 time=54.97it/s +epoch=116 global_step=45550 loss=5.83146 loss_avg=5.65415 acc=0.42188 acc_top1_avg=0.43375 acc_top5_avg=0.75636 lr=0.00010 gn=37.95293 time=55.95it/s +epoch=116 global_step=45600 loss=5.36027 loss_avg=5.64971 acc=0.46094 acc_top1_avg=0.43433 acc_top5_avg=0.75743 lr=0.00010 gn=31.52530 time=56.18it/s +epoch=116 global_step=45650 loss=5.72480 loss_avg=5.65970 acc=0.41406 acc_top1_avg=0.43327 acc_top5_avg=0.75747 lr=0.00010 gn=33.52274 time=59.63it/s +epoch=116 global_step=45700 loss=4.97971 loss_avg=5.65540 acc=0.50781 acc_top1_avg=0.43368 acc_top5_avg=0.75833 lr=0.00010 gn=29.14312 time=55.09it/s +====================Eval==================== +epoch=116 global_step=45747 loss=4.80937 test_loss_avg=4.24753 acc=0.00000 test_acc_avg=0.08377 test_acc_top5_avg=0.64366 time=167.25it/s +epoch=116 global_step=45747 loss=4.15302 test_loss_avg=3.21862 acc=0.00000 test_acc_avg=0.28689 test_acc_top5_avg=0.70016 time=847.85it/s +curr_acc 0.2869 +BEST_ACC 0.3318 +curr_acc_top5 0.7002 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=5.22079 loss_avg=5.66143 acc=0.49219 acc_top1_avg=0.43750 acc_top5_avg=0.72656 lr=0.00010 gn=31.68837 time=38.90it/s +epoch=117 global_step=45800 loss=6.00856 loss_avg=5.73116 acc=0.39844 acc_top1_avg=0.42630 acc_top5_avg=0.75840 lr=0.00010 gn=33.30271 time=53.20it/s +epoch=117 global_step=45850 loss=5.81834 loss_avg=5.66948 acc=0.42188 acc_top1_avg=0.43318 acc_top5_avg=0.76138 lr=0.00010 gn=38.21418 time=56.23it/s +epoch=117 global_step=45900 loss=5.29781 loss_avg=5.64238 acc=0.46875 acc_top1_avg=0.43638 acc_top5_avg=0.76190 lr=0.00010 gn=24.50901 time=62.78it/s +epoch=117 global_step=45950 loss=6.06537 loss_avg=5.63289 acc=0.38281 acc_top1_avg=0.43819 acc_top5_avg=0.76093 lr=0.00010 gn=26.66462 time=50.35it/s +epoch=117 global_step=46000 loss=4.90880 loss_avg=5.63150 acc=0.52344 acc_top1_avg=0.43778 acc_top5_avg=0.75960 lr=0.00010 gn=35.30563 time=56.17it/s +epoch=117 global_step=46050 loss=5.29919 loss_avg=5.62334 acc=0.46094 acc_top1_avg=0.43848 acc_top5_avg=0.76000 lr=0.00010 gn=34.91491 time=53.50it/s +epoch=117 global_step=46100 loss=5.37127 loss_avg=5.63474 acc=0.46875 acc_top1_avg=0.43712 acc_top5_avg=0.75826 lr=0.00010 gn=38.09799 time=56.13it/s +====================Eval==================== +epoch=117 global_step=46138 loss=4.76772 test_loss_avg=4.77184 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.77790 time=237.50it/s +epoch=117 global_step=46138 loss=0.23572 test_loss_avg=3.68726 acc=0.93750 test_acc_avg=0.18517 test_acc_top5_avg=0.67654 time=235.01it/s +epoch=117 global_step=46138 loss=4.04061 test_loss_avg=3.18805 acc=0.00000 test_acc_avg=0.29173 test_acc_top5_avg=0.70312 time=727.55it/s +curr_acc 0.2917 +BEST_ACC 0.3318 +curr_acc_top5 0.7031 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=5.31934 loss_avg=5.46227 acc=0.45312 acc_top1_avg=0.45117 acc_top5_avg=0.76497 lr=0.00010 gn=32.56518 time=55.40it/s +epoch=118 global_step=46200 loss=5.08239 loss_avg=5.55575 acc=0.50000 acc_top1_avg=0.44292 acc_top5_avg=0.76537 lr=0.00010 gn=29.58166 time=53.54it/s +epoch=118 global_step=46250 loss=5.58410 loss_avg=5.63186 acc=0.43750 acc_top1_avg=0.43548 acc_top5_avg=0.76388 lr=0.00010 gn=36.32147 time=50.73it/s +epoch=118 global_step=46300 loss=5.71210 loss_avg=5.63132 acc=0.42969 acc_top1_avg=0.43601 acc_top5_avg=0.76196 lr=0.00010 gn=34.91773 time=64.07it/s +epoch=118 global_step=46350 loss=5.60504 loss_avg=5.62407 acc=0.43750 acc_top1_avg=0.43680 acc_top5_avg=0.76282 lr=0.00010 gn=35.77219 time=47.31it/s +epoch=118 global_step=46400 loss=5.84445 loss_avg=5.63550 acc=0.40625 acc_top1_avg=0.43544 acc_top5_avg=0.76032 lr=0.00010 gn=30.87465 time=59.17it/s +epoch=118 global_step=46450 loss=6.23258 loss_avg=5.63572 acc=0.36719 acc_top1_avg=0.43545 acc_top5_avg=0.76069 lr=0.00010 gn=29.89935 time=49.37it/s +epoch=118 global_step=46500 loss=5.35105 loss_avg=5.63937 acc=0.46094 acc_top1_avg=0.43497 acc_top5_avg=0.76001 lr=0.00010 gn=40.89530 time=56.05it/s +====================Eval==================== +epoch=118 global_step=46529 loss=4.68791 test_loss_avg=3.98138 acc=0.00000 test_acc_avg=0.12054 test_acc_top5_avg=0.70536 time=241.55it/s +epoch=118 global_step=46529 loss=4.62362 test_loss_avg=3.14340 acc=0.00000 test_acc_avg=0.29838 test_acc_top5_avg=0.70252 time=244.74it/s +epoch=118 global_step=46529 loss=4.09916 test_loss_avg=3.15550 acc=0.00000 test_acc_avg=0.29460 test_acc_top5_avg=0.70075 time=800.29it/s +curr_acc 0.2946 +BEST_ACC 0.3318 +curr_acc_top5 0.7008 +BEST_ACC_top5 0.7758 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=5.67907 loss_avg=5.66276 acc=0.44531 acc_top1_avg=0.42894 acc_top5_avg=0.77530 lr=0.00010 gn=33.94667 time=55.25it/s +epoch=119 global_step=46600 loss=5.59437 loss_avg=5.60474 acc=0.45312 acc_top1_avg=0.43849 acc_top5_avg=0.77036 lr=0.00010 gn=39.75146 time=57.26it/s +epoch=119 global_step=46650 loss=5.26491 loss_avg=5.61504 acc=0.47656 acc_top1_avg=0.43724 acc_top5_avg=0.76446 lr=0.00010 gn=28.23314 time=58.39it/s +epoch=119 global_step=46700 loss=6.01524 loss_avg=5.62407 acc=0.39844 acc_top1_avg=0.43645 acc_top5_avg=0.76275 lr=0.00010 gn=35.46992 time=61.09it/s +epoch=119 global_step=46750 loss=5.72487 loss_avg=5.62401 acc=0.42969 acc_top1_avg=0.43662 acc_top5_avg=0.76000 lr=0.00010 gn=33.99861 time=55.32it/s +epoch=119 global_step=46800 loss=6.09426 loss_avg=5.62173 acc=0.38281 acc_top1_avg=0.43689 acc_top5_avg=0.75951 lr=0.00010 gn=33.88697 time=62.69it/s +epoch=119 global_step=46850 loss=6.33782 loss_avg=5.63355 acc=0.36719 acc_top1_avg=0.43580 acc_top5_avg=0.75798 lr=0.00010 gn=36.05201 time=56.43it/s +epoch=119 global_step=46900 loss=6.15086 loss_avg=5.63060 acc=0.37500 acc_top1_avg=0.43609 acc_top5_avg=0.75840 lr=0.00010 gn=34.70853 time=64.09it/s +====================Eval==================== +epoch=119 global_step=46920 loss=4.35201 test_loss_avg=3.85584 acc=0.00000 test_acc_avg=0.16598 test_acc_top5_avg=0.68128 time=257.78it/s +epoch=119 global_step=46920 loss=4.15778 test_loss_avg=3.23372 acc=0.00000 test_acc_avg=0.28659 test_acc_top5_avg=0.70609 time=889.75it/s +curr_acc 0.2866 +BEST_ACC 0.3318 +curr_acc_top5 0.7061 +BEST_ACC_top5 0.7758 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_4_2.log b/other_methods/sceloss/sceloss_results/out_4_2.log new file mode 100644 index 0000000..c43dd31 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_4_2.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.4__noise_amount__0.2.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.35984 loss_avg=7.46088 acc=0.24219 acc_top1_avg=0.24156 acc_top5_avg=0.68984 lr=0.01000 gn=6.86869 time=57.39it/s +epoch=0 global_step=100 loss=6.22196 loss_avg=7.04927 acc=0.35156 acc_top1_avg=0.28734 acc_top5_avg=0.73297 lr=0.01000 gn=5.54802 time=61.86it/s +epoch=0 global_step=150 loss=6.59292 loss_avg=6.82104 acc=0.33594 acc_top1_avg=0.31255 acc_top5_avg=0.75354 lr=0.01000 gn=6.67772 time=57.70it/s +epoch=0 global_step=200 loss=5.78185 loss_avg=6.64326 acc=0.40625 acc_top1_avg=0.33094 acc_top5_avg=0.76750 lr=0.01000 gn=6.76361 time=64.68it/s +epoch=0 global_step=250 loss=5.58283 loss_avg=6.48838 acc=0.44531 acc_top1_avg=0.34819 acc_top5_avg=0.77806 lr=0.01000 gn=6.57475 time=61.17it/s +epoch=0 global_step=300 loss=5.81832 loss_avg=6.36889 acc=0.41406 acc_top1_avg=0.36120 acc_top5_avg=0.78656 lr=0.01000 gn=6.55697 time=56.06it/s +epoch=0 global_step=350 loss=5.40944 loss_avg=6.24190 acc=0.46875 acc_top1_avg=0.37518 acc_top5_avg=0.79321 lr=0.01000 gn=6.40011 time=61.43it/s +====================Eval==================== +epoch=0 global_step=391 loss=0.58756 test_loss_avg=2.97303 acc=0.88281 test_acc_avg=0.42250 test_acc_top5_avg=0.93781 time=250.11it/s +epoch=0 global_step=391 loss=0.61764 test_loss_avg=2.55279 acc=0.81250 test_acc_avg=0.49664 test_acc_top5_avg=0.93058 time=27.64it/s +curr_acc 0.4966 +BEST_ACC 0.0000 +curr_acc_top5 0.9306 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=5.28987 loss_avg=5.44915 acc=0.50781 acc_top1_avg=0.46267 acc_top5_avg=0.84635 lr=0.01000 gn=7.20645 time=56.03it/s +epoch=1 global_step=450 loss=5.07026 loss_avg=5.39077 acc=0.48438 acc_top1_avg=0.46663 acc_top5_avg=0.84243 lr=0.01000 gn=6.25139 time=64.07it/s +epoch=1 global_step=500 loss=5.04761 loss_avg=5.35531 acc=0.50781 acc_top1_avg=0.47018 acc_top5_avg=0.84518 lr=0.01000 gn=7.26374 time=61.04it/s +epoch=1 global_step=550 loss=4.46516 loss_avg=5.29226 acc=0.57031 acc_top1_avg=0.47637 acc_top5_avg=0.84680 lr=0.01000 gn=7.61062 time=57.76it/s +epoch=1 global_step=600 loss=4.10171 loss_avg=5.21644 acc=0.61719 acc_top1_avg=0.48400 acc_top5_avg=0.84902 lr=0.01000 gn=6.64788 time=55.85it/s +epoch=1 global_step=650 loss=4.69387 loss_avg=5.18155 acc=0.54688 acc_top1_avg=0.48757 acc_top5_avg=0.85066 lr=0.01000 gn=5.93007 time=63.50it/s +epoch=1 global_step=700 loss=4.84131 loss_avg=5.14130 acc=0.53125 acc_top1_avg=0.49168 acc_top5_avg=0.85237 lr=0.01000 gn=7.25148 time=59.78it/s +epoch=1 global_step=750 loss=4.94322 loss_avg=5.10681 acc=0.51562 acc_top1_avg=0.49539 acc_top5_avg=0.85472 lr=0.01000 gn=5.19999 time=50.00it/s +====================Eval==================== +epoch=1 global_step=782 loss=3.82811 test_loss_avg=1.59237 acc=0.17969 test_acc_avg=0.62426 test_acc_top5_avg=0.97396 time=260.32it/s +epoch=1 global_step=782 loss=0.72903 test_loss_avg=1.74742 acc=0.78906 test_acc_avg=0.60982 test_acc_top5_avg=0.95081 time=233.22it/s +epoch=1 global_step=782 loss=1.50734 test_loss_avg=1.67299 acc=0.50000 test_acc_avg=0.62065 test_acc_top5_avg=0.95372 time=834.02it/s +curr_acc 0.6206 +BEST_ACC 0.4966 +curr_acc_top5 0.9537 +BEST_ACC_top5 0.9306 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=4.63137 loss_avg=4.78448 acc=0.53906 acc_top1_avg=0.52865 acc_top5_avg=0.85807 lr=0.01000 gn=6.93422 time=58.49it/s +epoch=2 global_step=850 loss=4.15208 loss_avg=4.60420 acc=0.59375 acc_top1_avg=0.55009 acc_top5_avg=0.86811 lr=0.01000 gn=5.29101 time=58.58it/s +epoch=2 global_step=900 loss=4.56842 loss_avg=4.61264 acc=0.55469 acc_top1_avg=0.54952 acc_top5_avg=0.86805 lr=0.01000 gn=6.93819 time=51.75it/s +epoch=2 global_step=950 loss=4.70558 loss_avg=4.59513 acc=0.52344 acc_top1_avg=0.55106 acc_top5_avg=0.86765 lr=0.01000 gn=5.63845 time=60.16it/s +epoch=2 global_step=1000 loss=4.35877 loss_avg=4.59843 acc=0.57812 acc_top1_avg=0.55085 acc_top5_avg=0.86959 lr=0.01000 gn=5.63032 time=56.65it/s +epoch=2 global_step=1050 loss=4.71768 loss_avg=4.59966 acc=0.53125 acc_top1_avg=0.54999 acc_top5_avg=0.86926 lr=0.01000 gn=6.39684 time=60.38it/s +epoch=2 global_step=1100 loss=4.29797 loss_avg=4.58281 acc=0.58594 acc_top1_avg=0.55159 acc_top5_avg=0.86901 lr=0.01000 gn=5.79472 time=53.49it/s +epoch=2 global_step=1150 loss=4.09400 loss_avg=4.56406 acc=0.60938 acc_top1_avg=0.55354 acc_top5_avg=0.86952 lr=0.01000 gn=6.69669 time=56.92it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.67905 test_loss_avg=1.83120 acc=0.59375 test_acc_avg=0.59673 test_acc_top5_avg=0.95443 time=246.91it/s +epoch=2 global_step=1173 loss=0.67687 test_loss_avg=1.39126 acc=0.68750 test_acc_avg=0.67880 test_acc_top5_avg=0.96509 time=891.84it/s +curr_acc 0.6788 +BEST_ACC 0.6206 +curr_acc_top5 0.9651 +BEST_ACC_top5 0.9537 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=3.99776 loss_avg=4.39260 acc=0.60938 acc_top1_avg=0.57407 acc_top5_avg=0.87211 lr=0.01000 gn=6.02241 time=61.83it/s +epoch=3 global_step=1250 loss=3.74336 loss_avg=4.33177 acc=0.64062 acc_top1_avg=0.57853 acc_top5_avg=0.87683 lr=0.01000 gn=4.63863 time=64.69it/s +epoch=3 global_step=1300 loss=3.80564 loss_avg=4.30310 acc=0.63281 acc_top1_avg=0.58182 acc_top5_avg=0.88047 lr=0.01000 gn=6.60829 time=58.11it/s +epoch=3 global_step=1350 loss=4.15666 loss_avg=4.27678 acc=0.60156 acc_top1_avg=0.58351 acc_top5_avg=0.88056 lr=0.01000 gn=6.10303 time=59.54it/s +epoch=3 global_step=1400 loss=3.84477 loss_avg=4.28168 acc=0.64844 acc_top1_avg=0.58260 acc_top5_avg=0.88054 lr=0.01000 gn=7.03049 time=60.56it/s +epoch=3 global_step=1450 loss=4.31711 loss_avg=4.28507 acc=0.57031 acc_top1_avg=0.58241 acc_top5_avg=0.88073 lr=0.01000 gn=5.52372 time=60.58it/s +epoch=3 global_step=1500 loss=4.02014 loss_avg=4.27289 acc=0.62500 acc_top1_avg=0.58381 acc_top5_avg=0.88102 lr=0.01000 gn=7.51576 time=57.47it/s +epoch=3 global_step=1550 loss=4.20579 loss_avg=4.27345 acc=0.59375 acc_top1_avg=0.58333 acc_top5_avg=0.88101 lr=0.01000 gn=5.78846 time=64.67it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.33663 test_loss_avg=0.50834 acc=0.90625 test_acc_avg=0.84255 test_acc_top5_avg=0.99099 time=247.31it/s +epoch=3 global_step=1564 loss=0.42395 test_loss_avg=1.16631 acc=0.84375 test_acc_avg=0.69928 test_acc_top5_avg=0.96069 time=243.09it/s +epoch=3 global_step=1564 loss=2.38796 test_loss_avg=1.16211 acc=0.31250 test_acc_avg=0.69907 test_acc_top5_avg=0.96133 time=865.70it/s +curr_acc 0.6991 +BEST_ACC 0.6788 +curr_acc_top5 0.9613 +BEST_ACC_top5 0.9651 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.49720 loss_avg=4.20795 acc=0.56250 acc_top1_avg=0.58919 acc_top5_avg=0.87804 lr=0.01000 gn=5.96735 time=57.30it/s +epoch=4 global_step=1650 loss=4.98206 loss_avg=4.18237 acc=0.50000 acc_top1_avg=0.59375 acc_top5_avg=0.88172 lr=0.01000 gn=5.81308 time=52.46it/s +epoch=4 global_step=1700 loss=4.29557 loss_avg=4.17511 acc=0.60938 acc_top1_avg=0.59432 acc_top5_avg=0.88574 lr=0.01000 gn=6.58659 time=54.27it/s +epoch=4 global_step=1750 loss=3.71144 loss_avg=4.16074 acc=0.65625 acc_top1_avg=0.59556 acc_top5_avg=0.88554 lr=0.01000 gn=6.79302 time=64.23it/s +epoch=4 global_step=1800 loss=4.65740 loss_avg=4.15438 acc=0.53125 acc_top1_avg=0.59620 acc_top5_avg=0.88586 lr=0.01000 gn=5.63003 time=63.51it/s +epoch=4 global_step=1850 loss=4.59731 loss_avg=4.14261 acc=0.53906 acc_top1_avg=0.59716 acc_top5_avg=0.88623 lr=0.01000 gn=6.16977 time=55.41it/s +epoch=4 global_step=1900 loss=4.28213 loss_avg=4.11338 acc=0.57812 acc_top1_avg=0.60010 acc_top5_avg=0.88753 lr=0.01000 gn=5.33513 time=63.54it/s +epoch=4 global_step=1950 loss=3.92682 loss_avg=4.10467 acc=0.60938 acc_top1_avg=0.60085 acc_top5_avg=0.88688 lr=0.01000 gn=6.82704 time=64.31it/s +====================Eval==================== +epoch=4 global_step=1955 loss=1.15726 test_loss_avg=1.03724 acc=0.65625 test_acc_avg=0.73231 test_acc_top5_avg=0.97358 time=246.54it/s +epoch=4 global_step=1955 loss=1.23201 test_loss_avg=1.09371 acc=0.56250 test_acc_avg=0.71657 test_acc_top5_avg=0.97112 time=879.86it/s +curr_acc 0.7166 +BEST_ACC 0.6991 +curr_acc_top5 0.9711 +BEST_ACC_top5 0.9651 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=4.17850 loss_avg=4.01530 acc=0.59375 acc_top1_avg=0.61233 acc_top5_avg=0.88663 lr=0.01000 gn=7.30433 time=55.68it/s +epoch=5 global_step=2050 loss=3.98198 loss_avg=4.00382 acc=0.62500 acc_top1_avg=0.61176 acc_top5_avg=0.88479 lr=0.01000 gn=7.84149 time=60.16it/s +epoch=5 global_step=2100 loss=4.59796 loss_avg=4.03189 acc=0.57031 acc_top1_avg=0.60938 acc_top5_avg=0.88594 lr=0.01000 gn=5.85346 time=56.68it/s +epoch=5 global_step=2150 loss=3.38899 loss_avg=3.98779 acc=0.67969 acc_top1_avg=0.61466 acc_top5_avg=0.89022 lr=0.01000 gn=5.69985 time=60.56it/s +epoch=5 global_step=2200 loss=3.52867 loss_avg=3.98148 acc=0.65625 acc_top1_avg=0.61473 acc_top5_avg=0.89149 lr=0.01000 gn=6.15371 time=62.74it/s +epoch=5 global_step=2250 loss=3.79517 loss_avg=3.97439 acc=0.64062 acc_top1_avg=0.61552 acc_top5_avg=0.89195 lr=0.01000 gn=6.93232 time=53.01it/s +epoch=5 global_step=2300 loss=3.44109 loss_avg=3.96594 acc=0.67969 acc_top1_avg=0.61642 acc_top5_avg=0.89255 lr=0.01000 gn=7.61562 time=57.39it/s +====================Eval==================== +epoch=5 global_step=2346 loss=0.98043 test_loss_avg=1.27078 acc=0.69531 test_acc_avg=0.66719 test_acc_top5_avg=0.98438 time=250.81it/s +epoch=5 global_step=2346 loss=2.69455 test_loss_avg=1.77774 acc=0.45312 test_acc_avg=0.64418 test_acc_top5_avg=0.96534 time=247.70it/s +epoch=5 global_step=2346 loss=0.72575 test_loss_avg=1.63474 acc=0.81250 test_acc_avg=0.66812 test_acc_top5_avg=0.96875 time=694.88it/s +curr_acc 0.6681 +BEST_ACC 0.7166 +curr_acc_top5 0.9688 +BEST_ACC_top5 0.9711 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=3.70652 loss_avg=3.79452 acc=0.64844 acc_top1_avg=0.64453 acc_top5_avg=0.89453 lr=0.01000 gn=6.48216 time=53.25it/s +epoch=6 global_step=2400 loss=4.32823 loss_avg=3.85928 acc=0.59375 acc_top1_avg=0.62630 acc_top5_avg=0.89844 lr=0.01000 gn=8.63807 time=56.36it/s +epoch=6 global_step=2450 loss=3.96546 loss_avg=3.88829 acc=0.60938 acc_top1_avg=0.62237 acc_top5_avg=0.89528 lr=0.01000 gn=6.61076 time=61.30it/s +epoch=6 global_step=2500 loss=3.43151 loss_avg=3.89625 acc=0.69531 acc_top1_avg=0.62084 acc_top5_avg=0.89575 lr=0.01000 gn=7.09167 time=56.42it/s +epoch=6 global_step=2550 loss=3.60935 loss_avg=3.87273 acc=0.64844 acc_top1_avg=0.62381 acc_top5_avg=0.89491 lr=0.01000 gn=6.85669 time=61.55it/s +epoch=6 global_step=2600 loss=4.96172 loss_avg=3.87947 acc=0.53906 acc_top1_avg=0.62312 acc_top5_avg=0.89570 lr=0.01000 gn=7.49740 time=52.85it/s +epoch=6 global_step=2650 loss=3.75180 loss_avg=3.86080 acc=0.64844 acc_top1_avg=0.62554 acc_top5_avg=0.89656 lr=0.01000 gn=7.54512 time=60.52it/s +epoch=6 global_step=2700 loss=4.15762 loss_avg=3.86728 acc=0.59375 acc_top1_avg=0.62485 acc_top5_avg=0.89654 lr=0.01000 gn=8.64339 time=61.38it/s +====================Eval==================== +epoch=6 global_step=2737 loss=1.18024 test_loss_avg=1.16666 acc=0.71094 test_acc_avg=0.71064 test_acc_top5_avg=0.96394 time=231.19it/s +epoch=6 global_step=2737 loss=0.51655 test_loss_avg=0.93277 acc=0.85938 test_acc_avg=0.76326 test_acc_top5_avg=0.97296 time=248.07it/s +epoch=6 global_step=2737 loss=0.54967 test_loss_avg=0.91792 acc=0.87500 test_acc_avg=0.76711 test_acc_top5_avg=0.97399 time=420.31it/s +curr_acc 0.7671 +BEST_ACC 0.7166 +curr_acc_top5 0.9740 +BEST_ACC_top5 0.9711 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=3.62144 loss_avg=3.72671 acc=0.64844 acc_top1_avg=0.64423 acc_top5_avg=0.90805 lr=0.01000 gn=6.94098 time=52.64it/s +epoch=7 global_step=2800 loss=3.25350 loss_avg=3.76467 acc=0.68750 acc_top1_avg=0.63579 acc_top5_avg=0.89683 lr=0.01000 gn=5.87932 time=54.31it/s +epoch=7 global_step=2850 loss=4.21992 loss_avg=3.80270 acc=0.58594 acc_top1_avg=0.63122 acc_top5_avg=0.89830 lr=0.01000 gn=6.37105 time=61.14it/s +epoch=7 global_step=2900 loss=4.04334 loss_avg=3.81825 acc=0.59375 acc_top1_avg=0.62994 acc_top5_avg=0.89896 lr=0.01000 gn=8.15362 time=58.82it/s +epoch=7 global_step=2950 loss=3.42502 loss_avg=3.83112 acc=0.64062 acc_top1_avg=0.62804 acc_top5_avg=0.89811 lr=0.01000 gn=6.49287 time=54.21it/s +epoch=7 global_step=3000 loss=4.54849 loss_avg=3.84390 acc=0.53125 acc_top1_avg=0.62693 acc_top5_avg=0.89775 lr=0.01000 gn=7.58472 time=64.20it/s +epoch=7 global_step=3050 loss=3.68579 loss_avg=3.82187 acc=0.64844 acc_top1_avg=0.62924 acc_top5_avg=0.89824 lr=0.01000 gn=6.54334 time=60.21it/s +epoch=7 global_step=3100 loss=3.58349 loss_avg=3.80503 acc=0.64062 acc_top1_avg=0.63141 acc_top5_avg=0.89906 lr=0.01000 gn=8.45023 time=47.99it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.10650 test_loss_avg=1.08256 acc=0.71094 test_acc_avg=0.73604 test_acc_top5_avg=0.96393 time=254.80it/s +epoch=7 global_step=3128 loss=2.43704 test_loss_avg=1.00856 acc=0.31250 test_acc_avg=0.74812 test_acc_top5_avg=0.97350 time=819.20it/s +curr_acc 0.7481 +BEST_ACC 0.7671 +curr_acc_top5 0.9735 +BEST_ACC_top5 0.9740 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=3.63770 loss_avg=3.88958 acc=0.64062 acc_top1_avg=0.62358 acc_top5_avg=0.89240 lr=0.01000 gn=7.07231 time=54.51it/s +epoch=8 global_step=3200 loss=3.75821 loss_avg=3.72272 acc=0.63281 acc_top1_avg=0.64030 acc_top5_avg=0.89909 lr=0.01000 gn=5.03758 time=62.81it/s +epoch=8 global_step=3250 loss=3.86302 loss_avg=3.71172 acc=0.64062 acc_top1_avg=0.64171 acc_top5_avg=0.90106 lr=0.01000 gn=8.29127 time=62.98it/s +epoch=8 global_step=3300 loss=3.40848 loss_avg=3.71299 acc=0.66406 acc_top1_avg=0.64112 acc_top5_avg=0.90162 lr=0.01000 gn=7.21304 time=62.04it/s +epoch=8 global_step=3350 loss=3.38858 loss_avg=3.72614 acc=0.67188 acc_top1_avg=0.64003 acc_top5_avg=0.90132 lr=0.01000 gn=8.87651 time=61.24it/s +epoch=8 global_step=3400 loss=2.96417 loss_avg=3.72749 acc=0.73438 acc_top1_avg=0.63991 acc_top5_avg=0.90030 lr=0.01000 gn=8.49333 time=60.01it/s +epoch=8 global_step=3450 loss=3.48232 loss_avg=3.72785 acc=0.65625 acc_top1_avg=0.63985 acc_top5_avg=0.90060 lr=0.01000 gn=8.03481 time=60.53it/s +epoch=8 global_step=3500 loss=4.15094 loss_avg=3.72843 acc=0.60156 acc_top1_avg=0.63976 acc_top5_avg=0.90117 lr=0.01000 gn=8.24597 time=62.97it/s +====================Eval==================== +epoch=8 global_step=3519 loss=2.33686 test_loss_avg=1.00472 acc=0.52344 test_acc_avg=0.73351 test_acc_top5_avg=0.98785 time=238.64it/s +epoch=8 global_step=3519 loss=0.43594 test_loss_avg=1.13537 acc=0.85938 test_acc_avg=0.72829 test_acc_top5_avg=0.97139 time=246.33it/s +epoch=8 global_step=3519 loss=0.93003 test_loss_avg=1.13012 acc=0.75000 test_acc_avg=0.72805 test_acc_top5_avg=0.96944 time=557.46it/s +curr_acc 0.7280 +BEST_ACC 0.7671 +curr_acc_top5 0.9694 +BEST_ACC_top5 0.9740 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=3.48246 loss_avg=3.68093 acc=0.70312 acc_top1_avg=0.64516 acc_top5_avg=0.90675 lr=0.01000 gn=9.35478 time=56.96it/s +epoch=9 global_step=3600 loss=4.63890 loss_avg=3.65824 acc=0.53125 acc_top1_avg=0.64844 acc_top5_avg=0.90876 lr=0.01000 gn=7.20097 time=56.62it/s +epoch=9 global_step=3650 loss=4.02791 loss_avg=3.66444 acc=0.58594 acc_top1_avg=0.64701 acc_top5_avg=0.90655 lr=0.01000 gn=7.76884 time=60.41it/s +epoch=9 global_step=3700 loss=3.88768 loss_avg=3.67684 acc=0.61719 acc_top1_avg=0.64580 acc_top5_avg=0.90470 lr=0.01000 gn=8.09489 time=60.67it/s +epoch=9 global_step=3750 loss=3.74482 loss_avg=3.69671 acc=0.66406 acc_top1_avg=0.64370 acc_top5_avg=0.90368 lr=0.01000 gn=10.32650 time=55.02it/s +epoch=9 global_step=3800 loss=4.13591 loss_avg=3.69918 acc=0.59375 acc_top1_avg=0.64354 acc_top5_avg=0.90353 lr=0.01000 gn=11.28930 time=53.60it/s +epoch=9 global_step=3850 loss=3.50033 loss_avg=3.68028 acc=0.66406 acc_top1_avg=0.64572 acc_top5_avg=0.90498 lr=0.01000 gn=8.32894 time=56.54it/s +epoch=9 global_step=3900 loss=4.62010 loss_avg=3.69030 acc=0.51562 acc_top1_avg=0.64448 acc_top5_avg=0.90406 lr=0.01000 gn=9.22717 time=63.76it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.29089 test_loss_avg=1.14091 acc=0.92969 test_acc_avg=0.71274 test_acc_top5_avg=0.96875 time=243.76it/s +epoch=9 global_step=3910 loss=0.16507 test_loss_avg=0.94815 acc=0.93750 test_acc_avg=0.75880 test_acc_top5_avg=0.97528 time=881.16it/s +curr_acc 0.7588 +BEST_ACC 0.7671 +curr_acc_top5 0.9753 +BEST_ACC_top5 0.9740 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=3.79092 loss_avg=3.67330 acc=0.62500 acc_top1_avg=0.64551 acc_top5_avg=0.90078 lr=0.01000 gn=9.06092 time=54.44it/s +epoch=10 global_step=4000 loss=3.85412 loss_avg=3.63643 acc=0.62500 acc_top1_avg=0.65000 acc_top5_avg=0.90078 lr=0.01000 gn=9.20153 time=53.84it/s +epoch=10 global_step=4050 loss=3.26602 loss_avg=3.61354 acc=0.68750 acc_top1_avg=0.65223 acc_top5_avg=0.90374 lr=0.01000 gn=8.54415 time=63.11it/s +epoch=10 global_step=4100 loss=3.75883 loss_avg=3.60213 acc=0.64062 acc_top1_avg=0.65366 acc_top5_avg=0.90469 lr=0.01000 gn=8.40780 time=58.04it/s +epoch=10 global_step=4150 loss=3.25730 loss_avg=3.63060 acc=0.71094 acc_top1_avg=0.65111 acc_top5_avg=0.90482 lr=0.01000 gn=9.59849 time=52.80it/s +epoch=10 global_step=4200 loss=3.69675 loss_avg=3.63366 acc=0.62500 acc_top1_avg=0.65108 acc_top5_avg=0.90469 lr=0.01000 gn=6.60727 time=51.82it/s +epoch=10 global_step=4250 loss=3.31688 loss_avg=3.64815 acc=0.67188 acc_top1_avg=0.64961 acc_top5_avg=0.90411 lr=0.01000 gn=5.59251 time=56.46it/s +epoch=10 global_step=4300 loss=4.17945 loss_avg=3.64706 acc=0.60156 acc_top1_avg=0.64942 acc_top5_avg=0.90543 lr=0.01000 gn=10.16062 time=61.29it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.29510 test_loss_avg=2.26110 acc=0.89062 test_acc_avg=0.50938 test_acc_top5_avg=0.95078 time=228.50it/s +epoch=10 global_step=4301 loss=0.56706 test_loss_avg=1.27909 acc=0.84375 test_acc_avg=0.69479 test_acc_top5_avg=0.96719 time=248.21it/s +epoch=10 global_step=4301 loss=0.58017 test_loss_avg=1.07536 acc=0.87500 test_acc_avg=0.73981 test_acc_top5_avg=0.97330 time=597.73it/s +curr_acc 0.7398 +BEST_ACC 0.7671 +curr_acc_top5 0.9733 +BEST_ACC_top5 0.9753 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=3.96147 loss_avg=3.51953 acc=0.60938 acc_top1_avg=0.66422 acc_top5_avg=0.90561 lr=0.01000 gn=9.00974 time=63.35it/s +epoch=11 global_step=4400 loss=3.78287 loss_avg=3.54225 acc=0.63281 acc_top1_avg=0.65996 acc_top5_avg=0.90625 lr=0.01000 gn=9.74942 time=51.60it/s +epoch=11 global_step=4450 loss=2.90997 loss_avg=3.57438 acc=0.72656 acc_top1_avg=0.65662 acc_top5_avg=0.90672 lr=0.01000 gn=7.41737 time=56.91it/s +epoch=11 global_step=4500 loss=3.79489 loss_avg=3.57408 acc=0.63281 acc_top1_avg=0.65601 acc_top5_avg=0.90664 lr=0.01000 gn=8.03908 time=56.88it/s +epoch=11 global_step=4550 loss=2.76941 loss_avg=3.58958 acc=0.73438 acc_top1_avg=0.65421 acc_top5_avg=0.90660 lr=0.01000 gn=8.98325 time=53.05it/s +epoch=11 global_step=4600 loss=3.25476 loss_avg=3.58954 acc=0.68750 acc_top1_avg=0.65445 acc_top5_avg=0.90672 lr=0.01000 gn=7.28525 time=57.00it/s +epoch=11 global_step=4650 loss=3.34246 loss_avg=3.59689 acc=0.67188 acc_top1_avg=0.65412 acc_top5_avg=0.90582 lr=0.01000 gn=10.38867 time=44.97it/s +====================Eval==================== +epoch=11 global_step=4692 loss=1.88486 test_loss_avg=1.55031 acc=0.55469 test_acc_avg=0.63861 test_acc_top5_avg=0.95136 time=243.02it/s +epoch=11 global_step=4692 loss=0.46142 test_loss_avg=0.96951 acc=0.81250 test_acc_avg=0.76325 test_acc_top5_avg=0.97369 time=861.96it/s +curr_acc 0.7633 +BEST_ACC 0.7671 +curr_acc_top5 0.9737 +BEST_ACC_top5 0.9753 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=3.75985 loss_avg=3.79323 acc=0.63281 acc_top1_avg=0.63574 acc_top5_avg=0.90332 lr=0.01000 gn=8.75098 time=60.49it/s +epoch=12 global_step=4750 loss=3.90877 loss_avg=3.61993 acc=0.62500 acc_top1_avg=0.65356 acc_top5_avg=0.90679 lr=0.01000 gn=7.66867 time=56.14it/s +epoch=12 global_step=4800 loss=4.00210 loss_avg=3.63236 acc=0.62500 acc_top1_avg=0.65336 acc_top5_avg=0.90524 lr=0.01000 gn=9.99999 time=53.49it/s +epoch=12 global_step=4850 loss=3.70755 loss_avg=3.62575 acc=0.65625 acc_top1_avg=0.65249 acc_top5_avg=0.90398 lr=0.01000 gn=8.87883 time=57.17it/s +epoch=12 global_step=4900 loss=3.87159 loss_avg=3.64218 acc=0.63281 acc_top1_avg=0.65110 acc_top5_avg=0.90467 lr=0.01000 gn=8.41570 time=59.23it/s +epoch=12 global_step=4950 loss=3.89687 loss_avg=3.62706 acc=0.63281 acc_top1_avg=0.65265 acc_top5_avg=0.90440 lr=0.01000 gn=8.32496 time=55.31it/s +epoch=12 global_step=5000 loss=3.10251 loss_avg=3.62636 acc=0.71094 acc_top1_avg=0.65316 acc_top5_avg=0.90559 lr=0.01000 gn=7.54306 time=61.91it/s +epoch=12 global_step=5050 loss=3.12168 loss_avg=3.61194 acc=0.67969 acc_top1_avg=0.65461 acc_top5_avg=0.90666 lr=0.01000 gn=9.52941 time=53.51it/s +====================Eval==================== +epoch=12 global_step=5083 loss=0.69430 test_loss_avg=0.69251 acc=0.82812 test_acc_avg=0.82422 test_acc_top5_avg=0.98438 time=254.14it/s +epoch=12 global_step=5083 loss=0.90181 test_loss_avg=0.90960 acc=0.78125 test_acc_avg=0.75916 test_acc_top5_avg=0.97761 time=249.05it/s +epoch=12 global_step=5083 loss=0.00949 test_loss_avg=0.74591 acc=1.00000 test_acc_avg=0.79856 test_acc_top5_avg=0.98101 time=775.29it/s +curr_acc 0.7986 +BEST_ACC 0.7671 +curr_acc_top5 0.9810 +BEST_ACC_top5 0.9753 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=3.36479 loss_avg=3.52935 acc=0.67969 acc_top1_avg=0.65947 acc_top5_avg=0.90257 lr=0.01000 gn=6.46821 time=60.58it/s +epoch=13 global_step=5150 loss=4.04842 loss_avg=3.60086 acc=0.60156 acc_top1_avg=0.65368 acc_top5_avg=0.90683 lr=0.01000 gn=9.87623 time=62.43it/s +epoch=13 global_step=5200 loss=3.86529 loss_avg=3.61215 acc=0.61719 acc_top1_avg=0.65224 acc_top5_avg=0.90772 lr=0.01000 gn=8.27508 time=56.79it/s +epoch=13 global_step=5250 loss=3.32252 loss_avg=3.62685 acc=0.67188 acc_top1_avg=0.65068 acc_top5_avg=0.90854 lr=0.01000 gn=7.84243 time=62.43it/s +epoch=13 global_step=5300 loss=3.48513 loss_avg=3.61737 acc=0.67188 acc_top1_avg=0.65179 acc_top5_avg=0.90920 lr=0.01000 gn=9.20423 time=62.08it/s +epoch=13 global_step=5350 loss=4.15743 loss_avg=3.61807 acc=0.59375 acc_top1_avg=0.65218 acc_top5_avg=0.90842 lr=0.01000 gn=8.88743 time=55.34it/s +epoch=13 global_step=5400 loss=3.02102 loss_avg=3.61838 acc=0.71875 acc_top1_avg=0.65233 acc_top5_avg=0.90874 lr=0.01000 gn=7.79867 time=61.51it/s +epoch=13 global_step=5450 loss=3.77073 loss_avg=3.61802 acc=0.65625 acc_top1_avg=0.65189 acc_top5_avg=0.90776 lr=0.01000 gn=7.38627 time=60.31it/s +====================Eval==================== +epoch=13 global_step=5474 loss=0.59904 test_loss_avg=0.44736 acc=0.85156 test_acc_avg=0.86447 test_acc_top5_avg=0.99423 time=241.19it/s +epoch=13 global_step=5474 loss=1.08334 test_loss_avg=0.84879 acc=0.69531 test_acc_avg=0.77825 test_acc_top5_avg=0.97506 time=249.22it/s +epoch=13 global_step=5474 loss=0.68004 test_loss_avg=0.84786 acc=0.68750 test_acc_avg=0.77453 test_acc_top5_avg=0.97518 time=542.95it/s +curr_acc 0.7745 +BEST_ACC 0.7986 +curr_acc_top5 0.9752 +BEST_ACC_top5 0.9810 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=3.11937 loss_avg=3.55443 acc=0.68750 acc_top1_avg=0.65805 acc_top5_avg=0.90895 lr=0.01000 gn=7.92147 time=62.61it/s +epoch=14 global_step=5550 loss=3.26782 loss_avg=3.59249 acc=0.69531 acc_top1_avg=0.65563 acc_top5_avg=0.90471 lr=0.01000 gn=7.90874 time=60.00it/s +epoch=14 global_step=5600 loss=3.54033 loss_avg=3.60180 acc=0.64062 acc_top1_avg=0.65439 acc_top5_avg=0.90600 lr=0.01000 gn=9.09568 time=60.03it/s +epoch=14 global_step=5650 loss=3.73886 loss_avg=3.61153 acc=0.64844 acc_top1_avg=0.65332 acc_top5_avg=0.90536 lr=0.01000 gn=8.20573 time=49.94it/s +epoch=14 global_step=5700 loss=3.34730 loss_avg=3.59477 acc=0.68750 acc_top1_avg=0.65473 acc_top5_avg=0.90580 lr=0.01000 gn=11.14098 time=52.19it/s +epoch=14 global_step=5750 loss=3.30740 loss_avg=3.59115 acc=0.67969 acc_top1_avg=0.65466 acc_top5_avg=0.90642 lr=0.01000 gn=9.83585 time=47.21it/s +epoch=14 global_step=5800 loss=3.97450 loss_avg=3.58828 acc=0.60156 acc_top1_avg=0.65455 acc_top5_avg=0.90642 lr=0.01000 gn=8.22156 time=54.39it/s +epoch=14 global_step=5850 loss=2.99289 loss_avg=3.57325 acc=0.72656 acc_top1_avg=0.65621 acc_top5_avg=0.90673 lr=0.01000 gn=8.55383 time=56.51it/s +====================Eval==================== +epoch=14 global_step=5865 loss=1.88836 test_loss_avg=0.83738 acc=0.60156 test_acc_avg=0.78604 test_acc_top5_avg=0.98011 time=239.81it/s +epoch=14 global_step=5865 loss=1.23268 test_loss_avg=0.98402 acc=0.81250 test_acc_avg=0.75020 test_acc_top5_avg=0.97488 time=511.75it/s +curr_acc 0.7502 +BEST_ACC 0.7986 +curr_acc_top5 0.9749 +BEST_ACC_top5 0.9810 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=3.20572 loss_avg=3.56243 acc=0.69531 acc_top1_avg=0.65915 acc_top5_avg=0.91228 lr=0.01000 gn=8.17762 time=56.06it/s +epoch=15 global_step=5950 loss=3.32939 loss_avg=3.52306 acc=0.67969 acc_top1_avg=0.66425 acc_top5_avg=0.90460 lr=0.01000 gn=7.70450 time=54.30it/s +epoch=15 global_step=6000 loss=3.31149 loss_avg=3.51439 acc=0.70312 acc_top1_avg=0.66557 acc_top5_avg=0.90752 lr=0.01000 gn=8.08356 time=60.69it/s +epoch=15 global_step=6050 loss=3.37394 loss_avg=3.50815 acc=0.68750 acc_top1_avg=0.66533 acc_top5_avg=0.90781 lr=0.01000 gn=9.08545 time=59.85it/s +epoch=15 global_step=6100 loss=2.95728 loss_avg=3.52331 acc=0.69531 acc_top1_avg=0.66346 acc_top5_avg=0.90891 lr=0.01000 gn=8.32019 time=55.24it/s +epoch=15 global_step=6150 loss=3.95534 loss_avg=3.54091 acc=0.60938 acc_top1_avg=0.66113 acc_top5_avg=0.90803 lr=0.01000 gn=7.04757 time=57.36it/s +epoch=15 global_step=6200 loss=3.43209 loss_avg=3.55427 acc=0.67969 acc_top1_avg=0.65982 acc_top5_avg=0.90746 lr=0.01000 gn=9.14137 time=48.48it/s +epoch=15 global_step=6250 loss=3.03524 loss_avg=3.55220 acc=0.70312 acc_top1_avg=0.65968 acc_top5_avg=0.90787 lr=0.01000 gn=8.48094 time=57.55it/s +====================Eval==================== +epoch=15 global_step=6256 loss=1.18617 test_loss_avg=1.13880 acc=0.63281 test_acc_avg=0.68021 test_acc_top5_avg=0.98021 time=232.22it/s +epoch=15 global_step=6256 loss=0.51800 test_loss_avg=1.50209 acc=0.89844 test_acc_avg=0.68197 test_acc_top5_avg=0.96358 time=140.15it/s +epoch=15 global_step=6256 loss=1.26696 test_loss_avg=1.37485 acc=0.75000 test_acc_avg=0.70134 test_acc_top5_avg=0.96806 time=830.39it/s +curr_acc 0.7013 +BEST_ACC 0.7986 +curr_acc_top5 0.9681 +BEST_ACC_top5 0.9810 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=2.74971 loss_avg=3.42837 acc=0.75000 acc_top1_avg=0.66992 acc_top5_avg=0.91495 lr=0.01000 gn=8.47904 time=59.48it/s +epoch=16 global_step=6350 loss=3.53523 loss_avg=3.43738 acc=0.65625 acc_top1_avg=0.66847 acc_top5_avg=0.91556 lr=0.01000 gn=8.36950 time=57.35it/s +epoch=16 global_step=6400 loss=3.42508 loss_avg=3.45434 acc=0.66406 acc_top1_avg=0.66743 acc_top5_avg=0.91650 lr=0.01000 gn=10.07973 time=53.81it/s +epoch=16 global_step=6450 loss=3.88992 loss_avg=3.47399 acc=0.60156 acc_top1_avg=0.66575 acc_top5_avg=0.91442 lr=0.01000 gn=8.41187 time=58.93it/s +epoch=16 global_step=6500 loss=3.22570 loss_avg=3.49960 acc=0.70312 acc_top1_avg=0.66307 acc_top5_avg=0.91413 lr=0.01000 gn=10.97334 time=51.85it/s +epoch=16 global_step=6550 loss=3.78038 loss_avg=3.51679 acc=0.64844 acc_top1_avg=0.66202 acc_top5_avg=0.91241 lr=0.01000 gn=10.44536 time=51.82it/s +epoch=16 global_step=6600 loss=4.51821 loss_avg=3.52587 acc=0.57031 acc_top1_avg=0.66118 acc_top5_avg=0.91131 lr=0.01000 gn=8.21321 time=58.59it/s +====================Eval==================== +epoch=16 global_step=6647 loss=1.16371 test_loss_avg=1.57504 acc=0.70312 test_acc_avg=0.64931 test_acc_top5_avg=0.95399 time=103.30it/s +epoch=16 global_step=6647 loss=2.36419 test_loss_avg=1.21061 acc=0.56250 test_acc_avg=0.71371 test_acc_top5_avg=0.96499 time=874.91it/s +curr_acc 0.7137 +BEST_ACC 0.7986 +curr_acc_top5 0.9650 +BEST_ACC_top5 0.9810 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=3.93417 loss_avg=3.69063 acc=0.62500 acc_top1_avg=0.64583 acc_top5_avg=0.89844 lr=0.01000 gn=9.14672 time=59.46it/s +epoch=17 global_step=6700 loss=3.31608 loss_avg=3.58854 acc=0.68750 acc_top1_avg=0.65492 acc_top5_avg=0.90537 lr=0.01000 gn=9.58327 time=56.04it/s +epoch=17 global_step=6750 loss=3.57200 loss_avg=3.55610 acc=0.67188 acc_top1_avg=0.65883 acc_top5_avg=0.90959 lr=0.01000 gn=7.68353 time=60.71it/s +epoch=17 global_step=6800 loss=4.15321 loss_avg=3.56970 acc=0.59375 acc_top1_avg=0.65773 acc_top5_avg=0.90911 lr=0.01000 gn=9.42306 time=56.80it/s +epoch=17 global_step=6850 loss=3.61189 loss_avg=3.54796 acc=0.66406 acc_top1_avg=0.66052 acc_top5_avg=0.90991 lr=0.01000 gn=7.92922 time=53.06it/s +epoch=17 global_step=6900 loss=3.72038 loss_avg=3.54447 acc=0.65625 acc_top1_avg=0.66116 acc_top5_avg=0.91023 lr=0.01000 gn=12.34281 time=55.40it/s +epoch=17 global_step=6950 loss=3.84009 loss_avg=3.54537 acc=0.64062 acc_top1_avg=0.66105 acc_top5_avg=0.91012 lr=0.01000 gn=11.03411 time=53.87it/s +epoch=17 global_step=7000 loss=4.41676 loss_avg=3.53689 acc=0.56250 acc_top1_avg=0.66156 acc_top5_avg=0.91041 lr=0.01000 gn=9.42728 time=60.48it/s +====================Eval==================== +epoch=17 global_step=7038 loss=0.96887 test_loss_avg=1.00437 acc=0.71094 test_acc_avg=0.71317 test_acc_top5_avg=0.98326 time=234.04it/s +epoch=17 global_step=7038 loss=0.79023 test_loss_avg=0.81399 acc=0.78906 test_acc_avg=0.77645 test_acc_top5_avg=0.98177 time=232.29it/s +epoch=17 global_step=7038 loss=0.36633 test_loss_avg=0.73872 acc=0.87500 test_acc_avg=0.79727 test_acc_top5_avg=0.98230 time=787.22it/s +curr_acc 0.7973 +BEST_ACC 0.7986 +curr_acc_top5 0.9823 +BEST_ACC_top5 0.9810 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=4.09845 loss_avg=3.55171 acc=0.59375 acc_top1_avg=0.66406 acc_top5_avg=0.89062 lr=0.01000 gn=8.47077 time=53.71it/s +epoch=18 global_step=7100 loss=3.82218 loss_avg=3.46425 acc=0.62500 acc_top1_avg=0.67200 acc_top5_avg=0.91003 lr=0.01000 gn=10.09502 time=58.98it/s +epoch=18 global_step=7150 loss=3.22702 loss_avg=3.43318 acc=0.69531 acc_top1_avg=0.67341 acc_top5_avg=0.91113 lr=0.01000 gn=11.28933 time=55.18it/s +epoch=18 global_step=7200 loss=4.27238 loss_avg=3.46614 acc=0.57031 acc_top1_avg=0.66922 acc_top5_avg=0.90963 lr=0.01000 gn=8.80898 time=56.51it/s +epoch=18 global_step=7250 loss=3.61558 loss_avg=3.48011 acc=0.64844 acc_top1_avg=0.66745 acc_top5_avg=0.91082 lr=0.01000 gn=9.13339 time=50.25it/s +epoch=18 global_step=7300 loss=3.73935 loss_avg=3.49532 acc=0.62500 acc_top1_avg=0.66564 acc_top5_avg=0.91147 lr=0.01000 gn=11.21453 time=50.20it/s +epoch=18 global_step=7350 loss=3.31399 loss_avg=3.51065 acc=0.67188 acc_top1_avg=0.66384 acc_top5_avg=0.91076 lr=0.01000 gn=8.36823 time=42.45it/s +epoch=18 global_step=7400 loss=3.97694 loss_avg=3.52711 acc=0.61719 acc_top1_avg=0.66227 acc_top5_avg=0.91013 lr=0.01000 gn=9.43588 time=61.15it/s +====================Eval==================== +epoch=18 global_step=7429 loss=1.94143 test_loss_avg=0.83414 acc=0.53906 test_acc_avg=0.79213 test_acc_top5_avg=0.96205 time=246.85it/s +epoch=18 global_step=7429 loss=0.67409 test_loss_avg=1.04447 acc=0.77344 test_acc_avg=0.72556 test_acc_top5_avg=0.95944 time=251.71it/s +epoch=18 global_step=7429 loss=0.72533 test_loss_avg=1.04043 acc=0.87500 test_acc_avg=0.72745 test_acc_top5_avg=0.95995 time=866.23it/s +curr_acc 0.7275 +BEST_ACC 0.7986 +curr_acc_top5 0.9599 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=3.07186 loss_avg=3.37245 acc=0.71875 acc_top1_avg=0.68080 acc_top5_avg=0.91890 lr=0.01000 gn=8.70203 time=59.76it/s +epoch=19 global_step=7500 loss=3.22276 loss_avg=3.46595 acc=0.71875 acc_top1_avg=0.66835 acc_top5_avg=0.91087 lr=0.01000 gn=10.69169 time=59.34it/s +epoch=19 global_step=7550 loss=3.81600 loss_avg=3.48558 acc=0.64844 acc_top1_avg=0.66690 acc_top5_avg=0.91038 lr=0.01000 gn=9.17622 time=61.03it/s +epoch=19 global_step=7600 loss=3.65080 loss_avg=3.49717 acc=0.64844 acc_top1_avg=0.66566 acc_top5_avg=0.91045 lr=0.01000 gn=9.39505 time=58.67it/s +epoch=19 global_step=7650 loss=3.66609 loss_avg=3.48672 acc=0.62500 acc_top1_avg=0.66686 acc_top5_avg=0.91173 lr=0.01000 gn=9.52084 time=58.10it/s +epoch=19 global_step=7700 loss=4.06413 loss_avg=3.50062 acc=0.61719 acc_top1_avg=0.66565 acc_top5_avg=0.91173 lr=0.01000 gn=10.82027 time=54.14it/s +epoch=19 global_step=7750 loss=3.87075 loss_avg=3.50379 acc=0.60156 acc_top1_avg=0.66518 acc_top5_avg=0.91121 lr=0.01000 gn=8.99884 time=49.24it/s +epoch=19 global_step=7800 loss=3.58310 loss_avg=3.50649 acc=0.66406 acc_top1_avg=0.66509 acc_top5_avg=0.91103 lr=0.01000 gn=10.48145 time=56.55it/s +====================Eval==================== +epoch=19 global_step=7820 loss=1.00533 test_loss_avg=0.97863 acc=0.75781 test_acc_avg=0.75893 test_acc_top5_avg=0.97816 time=234.52it/s +epoch=19 global_step=7820 loss=0.79621 test_loss_avg=1.03471 acc=0.68750 test_acc_avg=0.73754 test_acc_top5_avg=0.97261 time=880.60it/s +curr_acc 0.7375 +BEST_ACC 0.7986 +curr_acc_top5 0.9726 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=3.00708 loss_avg=3.41986 acc=0.72656 acc_top1_avg=0.67057 acc_top5_avg=0.91797 lr=0.01000 gn=8.36606 time=55.69it/s +epoch=20 global_step=7900 loss=3.98993 loss_avg=3.51334 acc=0.60938 acc_top1_avg=0.66250 acc_top5_avg=0.91143 lr=0.01000 gn=9.55198 time=59.77it/s +epoch=20 global_step=7950 loss=3.77970 loss_avg=3.51466 acc=0.64062 acc_top1_avg=0.66280 acc_top5_avg=0.90980 lr=0.01000 gn=10.92100 time=51.35it/s +epoch=20 global_step=8000 loss=4.42890 loss_avg=3.49970 acc=0.56250 acc_top1_avg=0.66463 acc_top5_avg=0.90964 lr=0.01000 gn=9.39033 time=55.52it/s +epoch=20 global_step=8050 loss=3.06965 loss_avg=3.49164 acc=0.71875 acc_top1_avg=0.66583 acc_top5_avg=0.90941 lr=0.01000 gn=9.09679 time=53.64it/s +epoch=20 global_step=8100 loss=3.22894 loss_avg=3.48491 acc=0.67969 acc_top1_avg=0.66671 acc_top5_avg=0.91030 lr=0.01000 gn=9.09149 time=48.84it/s +epoch=20 global_step=8150 loss=4.18882 loss_avg=3.49328 acc=0.57031 acc_top1_avg=0.66570 acc_top5_avg=0.91058 lr=0.01000 gn=9.42276 time=57.79it/s +epoch=20 global_step=8200 loss=3.27589 loss_avg=3.50455 acc=0.69531 acc_top1_avg=0.66472 acc_top5_avg=0.91088 lr=0.01000 gn=8.75236 time=57.96it/s +====================Eval==================== +epoch=20 global_step=8211 loss=1.25283 test_loss_avg=0.84781 acc=0.60938 test_acc_avg=0.76875 test_acc_top5_avg=0.98711 time=242.59it/s +epoch=20 global_step=8211 loss=0.01562 test_loss_avg=0.88660 acc=0.99219 test_acc_avg=0.75826 test_acc_top5_avg=0.97589 time=242.59it/s +epoch=20 global_step=8211 loss=0.66527 test_loss_avg=0.89090 acc=0.87500 test_acc_avg=0.75900 test_acc_top5_avg=0.97745 time=870.37it/s +curr_acc 0.7590 +BEST_ACC 0.7986 +curr_acc_top5 0.9775 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=3.60767 loss_avg=3.44247 acc=0.65625 acc_top1_avg=0.67147 acc_top5_avg=0.91266 lr=0.01000 gn=8.54336 time=54.17it/s +epoch=21 global_step=8300 loss=3.81888 loss_avg=3.49090 acc=0.64062 acc_top1_avg=0.66696 acc_top5_avg=0.91160 lr=0.01000 gn=9.78712 time=53.03it/s +epoch=21 global_step=8350 loss=4.04394 loss_avg=3.49436 acc=0.62500 acc_top1_avg=0.66665 acc_top5_avg=0.91092 lr=0.01000 gn=12.47793 time=60.53it/s +epoch=21 global_step=8400 loss=4.28064 loss_avg=3.49171 acc=0.57812 acc_top1_avg=0.66712 acc_top5_avg=0.91129 lr=0.01000 gn=11.03499 time=60.68it/s +epoch=21 global_step=8450 loss=3.49245 loss_avg=3.48989 acc=0.65625 acc_top1_avg=0.66753 acc_top5_avg=0.91060 lr=0.01000 gn=8.90767 time=48.35it/s +epoch=21 global_step=8500 loss=3.89173 loss_avg=3.48713 acc=0.62500 acc_top1_avg=0.66782 acc_top5_avg=0.91120 lr=0.01000 gn=9.18593 time=58.49it/s +epoch=21 global_step=8550 loss=3.52014 loss_avg=3.50335 acc=0.66406 acc_top1_avg=0.66568 acc_top5_avg=0.91097 lr=0.01000 gn=8.59122 time=61.47it/s +epoch=21 global_step=8600 loss=3.35656 loss_avg=3.50619 acc=0.67969 acc_top1_avg=0.66491 acc_top5_avg=0.91035 lr=0.01000 gn=8.87471 time=58.90it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.51683 test_loss_avg=0.99858 acc=0.64844 test_acc_avg=0.74505 test_acc_top5_avg=0.97447 time=236.35it/s +epoch=21 global_step=8602 loss=0.09704 test_loss_avg=0.86001 acc=1.00000 test_acc_avg=0.78273 test_acc_top5_avg=0.98042 time=379.20it/s +curr_acc 0.7827 +BEST_ACC 0.7986 +curr_acc_top5 0.9804 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=3.36703 loss_avg=3.44929 acc=0.66406 acc_top1_avg=0.66911 acc_top5_avg=0.91211 lr=0.01000 gn=8.80642 time=54.93it/s +epoch=22 global_step=8700 loss=3.58861 loss_avg=3.47071 acc=0.63281 acc_top1_avg=0.66637 acc_top5_avg=0.91390 lr=0.01000 gn=8.97874 time=60.46it/s +epoch=22 global_step=8750 loss=3.90759 loss_avg=3.48601 acc=0.62500 acc_top1_avg=0.66580 acc_top5_avg=0.91380 lr=0.01000 gn=8.93607 time=57.50it/s +epoch=22 global_step=8800 loss=3.75426 loss_avg=3.52734 acc=0.64062 acc_top1_avg=0.66189 acc_top5_avg=0.91166 lr=0.01000 gn=8.95972 time=55.04it/s +epoch=22 global_step=8850 loss=4.24074 loss_avg=3.50588 acc=0.58594 acc_top1_avg=0.66365 acc_top5_avg=0.91129 lr=0.01000 gn=8.58657 time=58.05it/s +epoch=22 global_step=8900 loss=3.06088 loss_avg=3.49027 acc=0.70312 acc_top1_avg=0.66498 acc_top5_avg=0.91173 lr=0.01000 gn=9.25092 time=59.50it/s +epoch=22 global_step=8950 loss=3.19259 loss_avg=3.49191 acc=0.69531 acc_top1_avg=0.66485 acc_top5_avg=0.91173 lr=0.01000 gn=10.31049 time=58.82it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.26270 test_loss_avg=0.91623 acc=0.89062 test_acc_avg=0.75130 test_acc_top5_avg=0.97396 time=236.19it/s +epoch=22 global_step=8993 loss=1.64978 test_loss_avg=1.41237 acc=0.59375 test_acc_avg=0.68233 test_acc_top5_avg=0.97303 time=244.00it/s +epoch=22 global_step=8993 loss=1.00970 test_loss_avg=1.29952 acc=0.68750 test_acc_avg=0.70115 test_acc_top5_avg=0.97251 time=545.07it/s +curr_acc 0.7011 +BEST_ACC 0.7986 +curr_acc_top5 0.9725 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=3.72339 loss_avg=3.72017 acc=0.64062 acc_top1_avg=0.63951 acc_top5_avg=0.89509 lr=0.01000 gn=8.91579 time=56.59it/s +epoch=23 global_step=9050 loss=3.63604 loss_avg=3.45371 acc=0.64844 acc_top1_avg=0.67050 acc_top5_avg=0.91420 lr=0.01000 gn=9.80977 time=57.63it/s +epoch=23 global_step=9100 loss=3.80673 loss_avg=3.45317 acc=0.64844 acc_top1_avg=0.66917 acc_top5_avg=0.91691 lr=0.01000 gn=8.35267 time=47.91it/s +epoch=23 global_step=9150 loss=3.19397 loss_avg=3.46106 acc=0.71094 acc_top1_avg=0.66740 acc_top5_avg=0.91650 lr=0.01000 gn=9.77862 time=60.69it/s +epoch=23 global_step=9200 loss=3.15215 loss_avg=3.43635 acc=0.69531 acc_top1_avg=0.67048 acc_top5_avg=0.91648 lr=0.01000 gn=11.36525 time=55.18it/s +epoch=23 global_step=9250 loss=3.95064 loss_avg=3.47000 acc=0.60938 acc_top1_avg=0.66747 acc_top5_avg=0.91583 lr=0.01000 gn=8.32677 time=54.04it/s +epoch=23 global_step=9300 loss=3.61809 loss_avg=3.47384 acc=0.66406 acc_top1_avg=0.66717 acc_top5_avg=0.91505 lr=0.01000 gn=10.24690 time=58.54it/s +epoch=23 global_step=9350 loss=3.33679 loss_avg=3.47350 acc=0.69531 acc_top1_avg=0.66741 acc_top5_avg=0.91459 lr=0.01000 gn=7.56682 time=52.36it/s +====================Eval==================== +epoch=23 global_step=9384 loss=0.45281 test_loss_avg=1.32688 acc=0.86719 test_acc_avg=0.67116 test_acc_top5_avg=0.95241 time=231.74it/s +epoch=23 global_step=9384 loss=0.40870 test_loss_avg=1.03230 acc=0.87500 test_acc_avg=0.74347 test_acc_top5_avg=0.96697 time=722.41it/s +curr_acc 0.7435 +BEST_ACC 0.7986 +curr_acc_top5 0.9670 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=3.89190 loss_avg=3.51187 acc=0.61719 acc_top1_avg=0.66309 acc_top5_avg=0.90576 lr=0.01000 gn=8.75877 time=58.39it/s +epoch=24 global_step=9450 loss=3.34079 loss_avg=3.41302 acc=0.69531 acc_top1_avg=0.67247 acc_top5_avg=0.91264 lr=0.01000 gn=8.45112 time=54.31it/s +epoch=24 global_step=9500 loss=2.88391 loss_avg=3.39438 acc=0.72656 acc_top1_avg=0.67511 acc_top5_avg=0.91312 lr=0.01000 gn=11.67215 time=58.27it/s +epoch=24 global_step=9550 loss=3.64660 loss_avg=3.42602 acc=0.64062 acc_top1_avg=0.67244 acc_top5_avg=0.91237 lr=0.01000 gn=7.39902 time=59.58it/s +epoch=24 global_step=9600 loss=3.62028 loss_avg=3.44420 acc=0.67969 acc_top1_avg=0.67068 acc_top5_avg=0.91316 lr=0.01000 gn=9.95798 time=53.08it/s +epoch=24 global_step=9650 loss=3.48282 loss_avg=3.45065 acc=0.66406 acc_top1_avg=0.66947 acc_top5_avg=0.91368 lr=0.01000 gn=10.38534 time=56.85it/s +epoch=24 global_step=9700 loss=3.66791 loss_avg=3.45850 acc=0.63281 acc_top1_avg=0.66856 acc_top5_avg=0.91295 lr=0.01000 gn=10.92679 time=51.10it/s +epoch=24 global_step=9750 loss=3.62538 loss_avg=3.46765 acc=0.64062 acc_top1_avg=0.66739 acc_top5_avg=0.91319 lr=0.01000 gn=8.49103 time=51.83it/s +====================Eval==================== +epoch=24 global_step=9775 loss=0.54426 test_loss_avg=0.74117 acc=0.84375 test_acc_avg=0.80273 test_acc_top5_avg=0.99219 time=251.07it/s +epoch=24 global_step=9775 loss=1.32295 test_loss_avg=0.91155 acc=0.64062 test_acc_avg=0.75651 test_acc_top5_avg=0.97772 time=243.74it/s +epoch=24 global_step=9775 loss=0.29126 test_loss_avg=0.77144 acc=0.93750 test_acc_avg=0.78857 test_acc_top5_avg=0.98220 time=665.02it/s +curr_acc 0.7886 +BEST_ACC 0.7986 +curr_acc_top5 0.9822 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=3.29497 loss_avg=3.35746 acc=0.67188 acc_top1_avg=0.68375 acc_top5_avg=0.92063 lr=0.01000 gn=8.77562 time=50.86it/s +epoch=25 global_step=9850 loss=4.48113 loss_avg=3.41589 acc=0.56250 acc_top1_avg=0.67563 acc_top5_avg=0.91917 lr=0.01000 gn=9.54892 time=51.71it/s +epoch=25 global_step=9900 loss=3.25830 loss_avg=3.41174 acc=0.69531 acc_top1_avg=0.67531 acc_top5_avg=0.91956 lr=0.01000 gn=9.67411 time=53.18it/s +epoch=25 global_step=9950 loss=3.05137 loss_avg=3.43190 acc=0.71094 acc_top1_avg=0.67183 acc_top5_avg=0.91754 lr=0.01000 gn=9.21286 time=53.00it/s +epoch=25 global_step=10000 loss=4.26059 loss_avg=3.43656 acc=0.58594 acc_top1_avg=0.67080 acc_top5_avg=0.91698 lr=0.01000 gn=9.10806 time=59.68it/s +epoch=25 global_step=10050 loss=3.51371 loss_avg=3.44032 acc=0.67188 acc_top1_avg=0.67043 acc_top5_avg=0.91659 lr=0.01000 gn=8.54703 time=60.04it/s +epoch=25 global_step=10100 loss=3.22013 loss_avg=3.45741 acc=0.68750 acc_top1_avg=0.66877 acc_top5_avg=0.91575 lr=0.01000 gn=10.94674 time=58.78it/s +epoch=25 global_step=10150 loss=3.58514 loss_avg=3.45950 acc=0.64844 acc_top1_avg=0.66877 acc_top5_avg=0.91460 lr=0.01000 gn=9.46207 time=58.07it/s +====================Eval==================== +epoch=25 global_step=10166 loss=1.67559 test_loss_avg=0.70893 acc=0.63281 test_acc_avg=0.81844 test_acc_top5_avg=0.99406 time=234.16it/s +epoch=25 global_step=10166 loss=0.72886 test_loss_avg=1.03749 acc=0.80469 test_acc_avg=0.74938 test_acc_top5_avg=0.98135 time=244.44it/s +epoch=25 global_step=10166 loss=0.70855 test_loss_avg=1.02636 acc=0.87500 test_acc_avg=0.75148 test_acc_top5_avg=0.98180 time=532.81it/s +curr_acc 0.7515 +BEST_ACC 0.7986 +curr_acc_top5 0.9818 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=3.62122 loss_avg=3.28827 acc=0.63281 acc_top1_avg=0.68612 acc_top5_avg=0.92233 lr=0.01000 gn=8.92552 time=60.76it/s +epoch=26 global_step=10250 loss=3.42447 loss_avg=3.36662 acc=0.67188 acc_top1_avg=0.67941 acc_top5_avg=0.91722 lr=0.01000 gn=9.36827 time=50.58it/s +epoch=26 global_step=10300 loss=3.25872 loss_avg=3.37155 acc=0.70312 acc_top1_avg=0.67881 acc_top5_avg=0.91680 lr=0.01000 gn=9.79102 time=52.44it/s +epoch=26 global_step=10350 loss=3.36100 loss_avg=3.40198 acc=0.67188 acc_top1_avg=0.67548 acc_top5_avg=0.91423 lr=0.01000 gn=6.60155 time=59.16it/s +epoch=26 global_step=10400 loss=3.77537 loss_avg=3.44100 acc=0.62500 acc_top1_avg=0.67131 acc_top5_avg=0.91433 lr=0.01000 gn=9.34242 time=55.16it/s +epoch=26 global_step=10450 loss=3.44344 loss_avg=3.47200 acc=0.67969 acc_top1_avg=0.66780 acc_top5_avg=0.91332 lr=0.01000 gn=11.92051 time=58.64it/s +epoch=26 global_step=10500 loss=3.37040 loss_avg=3.47773 acc=0.67188 acc_top1_avg=0.66745 acc_top5_avg=0.91282 lr=0.01000 gn=12.23273 time=54.23it/s +epoch=26 global_step=10550 loss=3.90903 loss_avg=3.47626 acc=0.61719 acc_top1_avg=0.66760 acc_top5_avg=0.91347 lr=0.01000 gn=9.78027 time=58.88it/s +====================Eval==================== +epoch=26 global_step=10557 loss=2.76967 test_loss_avg=1.21928 acc=0.39844 test_acc_avg=0.70839 test_acc_top5_avg=0.97877 time=233.06it/s +epoch=26 global_step=10557 loss=0.51999 test_loss_avg=1.05619 acc=0.87500 test_acc_avg=0.74011 test_acc_top5_avg=0.97834 time=863.74it/s +curr_acc 0.7401 +BEST_ACC 0.7986 +curr_acc_top5 0.9783 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=3.14796 loss_avg=3.54186 acc=0.69531 acc_top1_avg=0.66025 acc_top5_avg=0.90607 lr=0.01000 gn=10.01823 time=51.14it/s +epoch=27 global_step=10650 loss=3.44365 loss_avg=3.41800 acc=0.65625 acc_top1_avg=0.67221 acc_top5_avg=0.91263 lr=0.01000 gn=7.75478 time=55.92it/s +epoch=27 global_step=10700 loss=3.43782 loss_avg=3.41341 acc=0.67188 acc_top1_avg=0.67264 acc_top5_avg=0.91373 lr=0.01000 gn=7.14075 time=59.90it/s +epoch=27 global_step=10750 loss=3.56306 loss_avg=3.43199 acc=0.63281 acc_top1_avg=0.67131 acc_top5_avg=0.91370 lr=0.01000 gn=10.68893 time=59.55it/s +epoch=27 global_step=10800 loss=3.08336 loss_avg=3.42996 acc=0.71875 acc_top1_avg=0.67197 acc_top5_avg=0.91374 lr=0.01000 gn=10.15696 time=49.99it/s +epoch=27 global_step=10850 loss=3.62723 loss_avg=3.44009 acc=0.64062 acc_top1_avg=0.67129 acc_top5_avg=0.91409 lr=0.01000 gn=9.77615 time=53.77it/s +epoch=27 global_step=10900 loss=3.80857 loss_avg=3.45603 acc=0.63281 acc_top1_avg=0.66912 acc_top5_avg=0.91459 lr=0.01000 gn=9.70187 time=52.76it/s +====================Eval==================== +epoch=27 global_step=10948 loss=1.51760 test_loss_avg=0.59104 acc=0.65625 test_acc_avg=0.85202 test_acc_top5_avg=0.98897 time=226.85it/s +epoch=27 global_step=10948 loss=0.38247 test_loss_avg=1.00946 acc=0.88281 test_acc_avg=0.76259 test_acc_top5_avg=0.97691 time=229.54it/s +epoch=27 global_step=10948 loss=0.63778 test_loss_avg=0.96433 acc=0.87500 test_acc_avg=0.77027 test_acc_top5_avg=0.97854 time=885.62it/s +curr_acc 0.7703 +BEST_ACC 0.7986 +curr_acc_top5 0.9785 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=3.23950 loss_avg=3.06416 acc=0.68750 acc_top1_avg=0.71484 acc_top5_avg=0.93359 lr=0.01000 gn=9.19549 time=46.56it/s +epoch=28 global_step=11000 loss=2.73914 loss_avg=3.39322 acc=0.75781 acc_top1_avg=0.67578 acc_top5_avg=0.91541 lr=0.01000 gn=9.17562 time=45.71it/s +epoch=28 global_step=11050 loss=3.50600 loss_avg=3.44233 acc=0.65625 acc_top1_avg=0.67057 acc_top5_avg=0.91559 lr=0.01000 gn=9.89083 time=57.86it/s +epoch=28 global_step=11100 loss=3.46626 loss_avg=3.46474 acc=0.67188 acc_top1_avg=0.66879 acc_top5_avg=0.91375 lr=0.01000 gn=8.74155 time=53.43it/s +epoch=28 global_step=11150 loss=2.82461 loss_avg=3.46342 acc=0.72656 acc_top1_avg=0.66886 acc_top5_avg=0.91449 lr=0.01000 gn=8.89435 time=54.64it/s +epoch=28 global_step=11200 loss=3.81843 loss_avg=3.45360 acc=0.62500 acc_top1_avg=0.67001 acc_top5_avg=0.91474 lr=0.01000 gn=9.88363 time=45.86it/s +epoch=28 global_step=11250 loss=4.04978 loss_avg=3.45035 acc=0.60156 acc_top1_avg=0.67004 acc_top5_avg=0.91373 lr=0.01000 gn=9.60232 time=52.42it/s +epoch=28 global_step=11300 loss=3.70143 loss_avg=3.46007 acc=0.64844 acc_top1_avg=0.66883 acc_top5_avg=0.91435 lr=0.01000 gn=9.43130 time=56.60it/s +====================Eval==================== +epoch=28 global_step=11339 loss=1.01968 test_loss_avg=1.24990 acc=0.71094 test_acc_avg=0.70312 test_acc_top5_avg=0.96978 time=226.34it/s +epoch=28 global_step=11339 loss=0.61534 test_loss_avg=0.96268 acc=0.81250 test_acc_avg=0.76691 test_acc_top5_avg=0.97795 time=547.06it/s +curr_acc 0.7669 +BEST_ACC 0.7986 +curr_acc_top5 0.9779 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=3.71172 loss_avg=3.31552 acc=0.63281 acc_top1_avg=0.68395 acc_top5_avg=0.91335 lr=0.01000 gn=11.20230 time=55.56it/s +epoch=29 global_step=11400 loss=3.34797 loss_avg=3.32710 acc=0.67969 acc_top1_avg=0.68174 acc_top5_avg=0.91778 lr=0.01000 gn=8.25351 time=52.68it/s +epoch=29 global_step=11450 loss=3.53083 loss_avg=3.41540 acc=0.65625 acc_top1_avg=0.67279 acc_top5_avg=0.91512 lr=0.01000 gn=11.42003 time=58.30it/s +epoch=29 global_step=11500 loss=3.41279 loss_avg=3.40105 acc=0.67188 acc_top1_avg=0.67435 acc_top5_avg=0.91537 lr=0.01000 gn=10.39098 time=58.84it/s +epoch=29 global_step=11550 loss=4.15057 loss_avg=3.43273 acc=0.59375 acc_top1_avg=0.67102 acc_top5_avg=0.91414 lr=0.01000 gn=8.24510 time=56.45it/s +epoch=29 global_step=11600 loss=3.83167 loss_avg=3.45048 acc=0.62500 acc_top1_avg=0.66945 acc_top5_avg=0.91304 lr=0.01000 gn=9.68452 time=52.06it/s +epoch=29 global_step=11650 loss=3.07301 loss_avg=3.45848 acc=0.70312 acc_top1_avg=0.66858 acc_top5_avg=0.91288 lr=0.01000 gn=10.60656 time=52.16it/s +epoch=29 global_step=11700 loss=4.25874 loss_avg=3.46124 acc=0.57812 acc_top1_avg=0.66846 acc_top5_avg=0.91352 lr=0.01000 gn=9.32107 time=51.09it/s +====================Eval==================== +epoch=29 global_step=11730 loss=1.12560 test_loss_avg=0.97754 acc=0.75781 test_acc_avg=0.74826 test_acc_top5_avg=0.98524 time=224.19it/s +epoch=29 global_step=11730 loss=0.18330 test_loss_avg=1.29733 acc=0.94531 test_acc_avg=0.66896 test_acc_top5_avg=0.95432 time=232.14it/s +epoch=29 global_step=11730 loss=0.00662 test_loss_avg=1.00803 acc=1.00000 test_acc_avg=0.74061 test_acc_top5_avg=0.96559 time=549.86it/s +curr_acc 0.7406 +BEST_ACC 0.7986 +curr_acc_top5 0.9656 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=3.74018 loss_avg=3.41699 acc=0.64062 acc_top1_avg=0.67383 acc_top5_avg=0.91445 lr=0.01000 gn=8.41549 time=50.24it/s +epoch=30 global_step=11800 loss=4.36654 loss_avg=3.42218 acc=0.57031 acc_top1_avg=0.67310 acc_top5_avg=0.91328 lr=0.01000 gn=8.94167 time=56.98it/s +epoch=30 global_step=11850 loss=3.47304 loss_avg=3.40940 acc=0.68750 acc_top1_avg=0.67428 acc_top5_avg=0.91595 lr=0.01000 gn=9.80574 time=57.41it/s +epoch=30 global_step=11900 loss=3.07161 loss_avg=3.41322 acc=0.72656 acc_top1_avg=0.67371 acc_top5_avg=0.91498 lr=0.01000 gn=9.11960 time=57.42it/s +epoch=30 global_step=11950 loss=3.30769 loss_avg=3.42859 acc=0.69531 acc_top1_avg=0.67170 acc_top5_avg=0.91449 lr=0.01000 gn=10.19805 time=58.65it/s +epoch=30 global_step=12000 loss=3.23095 loss_avg=3.43010 acc=0.69531 acc_top1_avg=0.67124 acc_top5_avg=0.91400 lr=0.01000 gn=11.38696 time=55.84it/s +epoch=30 global_step=12050 loss=3.31698 loss_avg=3.43374 acc=0.65625 acc_top1_avg=0.67092 acc_top5_avg=0.91418 lr=0.01000 gn=11.22212 time=57.15it/s +epoch=30 global_step=12100 loss=3.91574 loss_avg=3.44878 acc=0.61719 acc_top1_avg=0.66902 acc_top5_avg=0.91467 lr=0.01000 gn=7.92543 time=56.70it/s +====================Eval==================== +epoch=30 global_step=12121 loss=0.83398 test_loss_avg=1.02173 acc=0.82031 test_acc_avg=0.75807 test_acc_top5_avg=0.97266 time=234.00it/s +epoch=30 global_step=12121 loss=0.39303 test_loss_avg=0.95617 acc=0.93750 test_acc_avg=0.77205 test_acc_top5_avg=0.97854 time=869.47it/s +curr_acc 0.7721 +BEST_ACC 0.7986 +curr_acc_top5 0.9785 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=3.67931 loss_avg=3.43951 acc=0.65625 acc_top1_avg=0.66999 acc_top5_avg=0.91622 lr=0.01000 gn=9.24016 time=59.83it/s +epoch=31 global_step=12200 loss=2.99220 loss_avg=3.46287 acc=0.71094 acc_top1_avg=0.66673 acc_top5_avg=0.91594 lr=0.01000 gn=9.26051 time=56.93it/s +epoch=31 global_step=12250 loss=3.88132 loss_avg=3.47023 acc=0.60938 acc_top1_avg=0.66715 acc_top5_avg=0.91334 lr=0.01000 gn=7.91516 time=50.43it/s +epoch=31 global_step=12300 loss=2.92119 loss_avg=3.44477 acc=0.71875 acc_top1_avg=0.66947 acc_top5_avg=0.91472 lr=0.01000 gn=10.18084 time=56.56it/s +epoch=31 global_step=12350 loss=2.40080 loss_avg=3.44111 acc=0.76562 acc_top1_avg=0.67051 acc_top5_avg=0.91437 lr=0.01000 gn=8.00240 time=49.85it/s +epoch=31 global_step=12400 loss=3.38110 loss_avg=3.44044 acc=0.67188 acc_top1_avg=0.67075 acc_top5_avg=0.91375 lr=0.01000 gn=7.87088 time=53.25it/s +epoch=31 global_step=12450 loss=3.80459 loss_avg=3.42213 acc=0.61719 acc_top1_avg=0.67275 acc_top5_avg=0.91447 lr=0.01000 gn=9.29307 time=59.31it/s +epoch=31 global_step=12500 loss=3.44550 loss_avg=3.43734 acc=0.67188 acc_top1_avg=0.67109 acc_top5_avg=0.91386 lr=0.01000 gn=8.66192 time=58.23it/s +====================Eval==================== +epoch=31 global_step=12512 loss=1.79775 test_loss_avg=1.79775 acc=0.50781 test_acc_avg=0.50781 test_acc_top5_avg=0.95312 time=197.23it/s +epoch=31 global_step=12512 loss=0.92653 test_loss_avg=0.99748 acc=0.78906 test_acc_avg=0.72580 test_acc_top5_avg=0.97381 time=232.40it/s +epoch=31 global_step=12512 loss=0.53865 test_loss_avg=0.86812 acc=0.87500 test_acc_avg=0.76147 test_acc_top5_avg=0.97785 time=555.68it/s +curr_acc 0.7615 +BEST_ACC 0.7986 +curr_acc_top5 0.9778 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=3.62722 loss_avg=3.31069 acc=0.64844 acc_top1_avg=0.68257 acc_top5_avg=0.91715 lr=0.01000 gn=14.01247 time=57.52it/s +epoch=32 global_step=12600 loss=3.41583 loss_avg=3.37098 acc=0.67969 acc_top1_avg=0.67694 acc_top5_avg=0.91513 lr=0.01000 gn=9.11939 time=59.21it/s +epoch=32 global_step=12650 loss=3.16786 loss_avg=3.37416 acc=0.71875 acc_top1_avg=0.67737 acc_top5_avg=0.91582 lr=0.01000 gn=12.17274 time=54.08it/s +epoch=32 global_step=12700 loss=3.32436 loss_avg=3.38761 acc=0.68750 acc_top1_avg=0.67503 acc_top5_avg=0.91527 lr=0.01000 gn=9.09236 time=59.73it/s +epoch=32 global_step=12750 loss=3.11182 loss_avg=3.42685 acc=0.69531 acc_top1_avg=0.67132 acc_top5_avg=0.91472 lr=0.01000 gn=8.34847 time=55.28it/s +epoch=32 global_step=12800 loss=3.41757 loss_avg=3.43984 acc=0.66406 acc_top1_avg=0.66992 acc_top5_avg=0.91412 lr=0.01000 gn=8.37329 time=44.88it/s +epoch=32 global_step=12850 loss=4.10349 loss_avg=3.45104 acc=0.60938 acc_top1_avg=0.66912 acc_top5_avg=0.91388 lr=0.01000 gn=9.59997 time=55.00it/s +epoch=32 global_step=12900 loss=3.73524 loss_avg=3.44992 acc=0.63281 acc_top1_avg=0.66952 acc_top5_avg=0.91440 lr=0.01000 gn=9.29572 time=55.08it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.56536 test_loss_avg=1.01481 acc=0.60938 test_acc_avg=0.72692 test_acc_top5_avg=0.98651 time=234.33it/s +epoch=32 global_step=12903 loss=0.14940 test_loss_avg=0.90468 acc=0.95312 test_acc_avg=0.75456 test_acc_top5_avg=0.96908 time=229.75it/s +epoch=32 global_step=12903 loss=0.00607 test_loss_avg=0.83513 acc=1.00000 test_acc_avg=0.77324 test_acc_top5_avg=0.97142 time=559.69it/s +curr_acc 0.7732 +BEST_ACC 0.7986 +curr_acc_top5 0.9714 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=3.48772 loss_avg=3.34173 acc=0.66406 acc_top1_avg=0.67769 acc_top5_avg=0.91772 lr=0.01000 gn=8.58720 time=53.03it/s +epoch=33 global_step=13000 loss=3.54383 loss_avg=3.31791 acc=0.66406 acc_top1_avg=0.68291 acc_top5_avg=0.92026 lr=0.01000 gn=10.95647 time=52.23it/s +epoch=33 global_step=13050 loss=3.90566 loss_avg=3.35402 acc=0.61719 acc_top1_avg=0.67852 acc_top5_avg=0.91773 lr=0.01000 gn=10.48361 time=50.95it/s +epoch=33 global_step=13100 loss=3.02426 loss_avg=3.38130 acc=0.71094 acc_top1_avg=0.67628 acc_top5_avg=0.91636 lr=0.01000 gn=7.60268 time=45.22it/s +epoch=33 global_step=13150 loss=3.44823 loss_avg=3.40530 acc=0.66406 acc_top1_avg=0.67396 acc_top5_avg=0.91536 lr=0.01000 gn=9.72430 time=58.14it/s +epoch=33 global_step=13200 loss=4.08464 loss_avg=3.41212 acc=0.58594 acc_top1_avg=0.67274 acc_top5_avg=0.91575 lr=0.01000 gn=11.93956 time=56.60it/s +epoch=33 global_step=13250 loss=3.39273 loss_avg=3.41187 acc=0.67969 acc_top1_avg=0.67289 acc_top5_avg=0.91575 lr=0.01000 gn=6.92341 time=58.72it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.42449 test_loss_avg=1.05770 acc=0.89844 test_acc_avg=0.74600 test_acc_top5_avg=0.98056 time=227.75it/s +epoch=33 global_step=13294 loss=0.90582 test_loss_avg=0.92488 acc=0.81250 test_acc_avg=0.77383 test_acc_top5_avg=0.97834 time=716.85it/s +curr_acc 0.7738 +BEST_ACC 0.7986 +curr_acc_top5 0.9783 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=3.35103 loss_avg=3.42132 acc=0.67969 acc_top1_avg=0.67318 acc_top5_avg=0.91927 lr=0.01000 gn=11.30336 time=51.00it/s +epoch=34 global_step=13350 loss=3.68594 loss_avg=3.43968 acc=0.64844 acc_top1_avg=0.67174 acc_top5_avg=0.90862 lr=0.01000 gn=12.20306 time=59.11it/s +epoch=34 global_step=13400 loss=3.05795 loss_avg=3.41955 acc=0.72656 acc_top1_avg=0.67364 acc_top5_avg=0.91193 lr=0.01000 gn=10.95135 time=49.18it/s +epoch=34 global_step=13450 loss=3.33436 loss_avg=3.42358 acc=0.69531 acc_top1_avg=0.67318 acc_top5_avg=0.91121 lr=0.01000 gn=12.48335 time=51.18it/s +epoch=34 global_step=13500 loss=3.63767 loss_avg=3.40645 acc=0.63281 acc_top1_avg=0.67480 acc_top5_avg=0.91414 lr=0.01000 gn=9.68000 time=52.69it/s +epoch=34 global_step=13550 loss=3.48922 loss_avg=3.40257 acc=0.66406 acc_top1_avg=0.67542 acc_top5_avg=0.91486 lr=0.01000 gn=7.93339 time=51.46it/s +epoch=34 global_step=13600 loss=3.29333 loss_avg=3.42191 acc=0.67969 acc_top1_avg=0.67307 acc_top5_avg=0.91432 lr=0.01000 gn=8.79489 time=58.68it/s +epoch=34 global_step=13650 loss=3.58320 loss_avg=3.42730 acc=0.64062 acc_top1_avg=0.67256 acc_top5_avg=0.91367 lr=0.01000 gn=7.40846 time=58.76it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.34949 test_loss_avg=0.86396 acc=0.89062 test_acc_avg=0.78348 test_acc_top5_avg=0.96931 time=229.70it/s +epoch=34 global_step=13685 loss=0.31973 test_loss_avg=1.05355 acc=0.91406 test_acc_avg=0.75623 test_acc_top5_avg=0.96802 time=218.27it/s +epoch=34 global_step=13685 loss=0.23041 test_loss_avg=0.90855 acc=0.81250 test_acc_avg=0.78540 test_acc_top5_avg=0.97330 time=566.87it/s +curr_acc 0.7854 +BEST_ACC 0.7986 +curr_acc_top5 0.9733 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=3.00696 loss_avg=3.28734 acc=0.71875 acc_top1_avg=0.69219 acc_top5_avg=0.91979 lr=0.01000 gn=9.87081 time=56.70it/s +epoch=35 global_step=13750 loss=3.07708 loss_avg=3.40199 acc=0.71875 acc_top1_avg=0.67656 acc_top5_avg=0.91322 lr=0.01000 gn=9.85357 time=59.44it/s +epoch=35 global_step=13800 loss=3.34951 loss_avg=3.41279 acc=0.67188 acc_top1_avg=0.67439 acc_top5_avg=0.91508 lr=0.01000 gn=9.98171 time=56.94it/s +epoch=35 global_step=13850 loss=2.97157 loss_avg=3.45508 acc=0.71094 acc_top1_avg=0.66993 acc_top5_avg=0.91392 lr=0.01000 gn=8.92864 time=57.27it/s +epoch=35 global_step=13900 loss=3.11105 loss_avg=3.43745 acc=0.68750 acc_top1_avg=0.67151 acc_top5_avg=0.91417 lr=0.01000 gn=9.44557 time=54.78it/s +epoch=35 global_step=13950 loss=3.47386 loss_avg=3.42012 acc=0.64844 acc_top1_avg=0.67358 acc_top5_avg=0.91462 lr=0.01000 gn=9.17609 time=53.49it/s +epoch=35 global_step=14000 loss=4.06983 loss_avg=3.42049 acc=0.58594 acc_top1_avg=0.67324 acc_top5_avg=0.91399 lr=0.01000 gn=8.19659 time=49.79it/s +epoch=35 global_step=14050 loss=4.11179 loss_avg=3.42185 acc=0.60938 acc_top1_avg=0.67339 acc_top5_avg=0.91404 lr=0.01000 gn=9.44212 time=56.81it/s +====================Eval==================== +epoch=35 global_step=14076 loss=1.93964 test_loss_avg=1.35083 acc=0.51562 test_acc_avg=0.70759 test_acc_top5_avg=0.93527 time=237.73it/s +epoch=35 global_step=14076 loss=0.19822 test_loss_avg=1.05492 acc=0.93750 test_acc_avg=0.74703 test_acc_top5_avg=0.96114 time=863.74it/s +curr_acc 0.7470 +BEST_ACC 0.7986 +curr_acc_top5 0.9611 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=3.39713 loss_avg=3.44073 acc=0.70312 acc_top1_avg=0.67741 acc_top5_avg=0.91536 lr=0.01000 gn=9.34682 time=59.13it/s +epoch=36 global_step=14150 loss=3.32175 loss_avg=3.34150 acc=0.71094 acc_top1_avg=0.68528 acc_top5_avg=0.91543 lr=0.01000 gn=9.97825 time=51.62it/s +epoch=36 global_step=14200 loss=3.15649 loss_avg=3.37351 acc=0.70312 acc_top1_avg=0.68208 acc_top5_avg=0.91608 lr=0.01000 gn=8.52331 time=52.54it/s +epoch=36 global_step=14250 loss=2.67223 loss_avg=3.39510 acc=0.74219 acc_top1_avg=0.67955 acc_top5_avg=0.91532 lr=0.01000 gn=8.89175 time=55.17it/s +epoch=36 global_step=14300 loss=2.86849 loss_avg=3.38833 acc=0.72656 acc_top1_avg=0.67976 acc_top5_avg=0.91469 lr=0.01000 gn=9.51390 time=55.28it/s +epoch=36 global_step=14350 loss=2.91576 loss_avg=3.38630 acc=0.74219 acc_top1_avg=0.67986 acc_top5_avg=0.91512 lr=0.01000 gn=10.04965 time=53.97it/s +epoch=36 global_step=14400 loss=3.21933 loss_avg=3.40407 acc=0.69531 acc_top1_avg=0.67720 acc_top5_avg=0.91423 lr=0.01000 gn=10.21108 time=52.16it/s +epoch=36 global_step=14450 loss=3.23048 loss_avg=3.41268 acc=0.69531 acc_top1_avg=0.67595 acc_top5_avg=0.91408 lr=0.01000 gn=10.47547 time=43.71it/s +====================Eval==================== +epoch=36 global_step=14467 loss=0.87840 test_loss_avg=0.67706 acc=0.78125 test_acc_avg=0.81771 test_acc_top5_avg=0.97786 time=235.69it/s +epoch=36 global_step=14467 loss=0.47992 test_loss_avg=0.95934 acc=0.84375 test_acc_avg=0.74386 test_acc_top5_avg=0.98103 time=231.93it/s +epoch=36 global_step=14467 loss=0.33439 test_loss_avg=0.78845 acc=0.87500 test_acc_avg=0.78649 test_acc_top5_avg=0.98418 time=550.29it/s +curr_acc 0.7865 +BEST_ACC 0.7986 +curr_acc_top5 0.9842 +BEST_ACC_top5 0.9823 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=2.64991 loss_avg=3.32305 acc=0.75000 acc_top1_avg=0.68158 acc_top5_avg=0.91572 lr=0.01000 gn=9.38186 time=56.78it/s +epoch=37 global_step=14550 loss=3.22942 loss_avg=3.34949 acc=0.68750 acc_top1_avg=0.68025 acc_top5_avg=0.91943 lr=0.01000 gn=9.22661 time=58.07it/s +epoch=37 global_step=14600 loss=3.53063 loss_avg=3.38828 acc=0.67969 acc_top1_avg=0.67563 acc_top5_avg=0.91665 lr=0.01000 gn=12.44776 time=58.27it/s +epoch=37 global_step=14650 loss=3.62325 loss_avg=3.40214 acc=0.64062 acc_top1_avg=0.67452 acc_top5_avg=0.91539 lr=0.01000 gn=7.89609 time=49.81it/s +epoch=37 global_step=14700 loss=2.88496 loss_avg=3.42191 acc=0.72656 acc_top1_avg=0.67281 acc_top5_avg=0.91540 lr=0.01000 gn=9.90777 time=58.82it/s +epoch=37 global_step=14750 loss=3.02982 loss_avg=3.41197 acc=0.69531 acc_top1_avg=0.67417 acc_top5_avg=0.91613 lr=0.01000 gn=12.52752 time=57.51it/s +epoch=37 global_step=14800 loss=3.84224 loss_avg=3.41918 acc=0.66406 acc_top1_avg=0.67335 acc_top5_avg=0.91589 lr=0.01000 gn=12.96567 time=55.37it/s +epoch=37 global_step=14850 loss=4.17962 loss_avg=3.42863 acc=0.57812 acc_top1_avg=0.67194 acc_top5_avg=0.91543 lr=0.01000 gn=9.19277 time=58.57it/s +====================Eval==================== +epoch=37 global_step=14858 loss=1.39795 test_loss_avg=0.61362 acc=0.60938 test_acc_avg=0.83304 test_acc_top5_avg=0.98524 time=242.80it/s +epoch=37 global_step=14858 loss=1.94119 test_loss_avg=0.86509 acc=0.53906 test_acc_avg=0.76684 test_acc_top5_avg=0.97291 time=248.27it/s +epoch=37 global_step=14858 loss=1.29433 test_loss_avg=0.87854 acc=0.81250 test_acc_avg=0.76543 test_acc_top5_avg=0.97271 time=884.50it/s +curr_acc 0.7654 +BEST_ACC 0.7986 +curr_acc_top5 0.9727 +BEST_ACC_top5 0.9842 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=3.13940 loss_avg=3.30493 acc=0.70312 acc_top1_avg=0.68583 acc_top5_avg=0.91760 lr=0.01000 gn=9.98092 time=58.27it/s +epoch=38 global_step=14950 loss=3.71461 loss_avg=3.35874 acc=0.61719 acc_top1_avg=0.67969 acc_top5_avg=0.91636 lr=0.01000 gn=12.05657 time=54.20it/s +epoch=38 global_step=15000 loss=3.35263 loss_avg=3.37930 acc=0.67188 acc_top1_avg=0.67710 acc_top5_avg=0.91412 lr=0.01000 gn=10.05469 time=52.63it/s +epoch=38 global_step=15050 loss=3.69577 loss_avg=3.39912 acc=0.63281 acc_top1_avg=0.67493 acc_top5_avg=0.91410 lr=0.01000 gn=10.77462 time=52.49it/s +epoch=38 global_step=15100 loss=3.18430 loss_avg=3.42494 acc=0.67969 acc_top1_avg=0.67168 acc_top5_avg=0.91377 lr=0.01000 gn=12.21003 time=54.06it/s +epoch=38 global_step=15150 loss=3.67376 loss_avg=3.41461 acc=0.64062 acc_top1_avg=0.67268 acc_top5_avg=0.91438 lr=0.01000 gn=10.84017 time=50.80it/s +epoch=38 global_step=15200 loss=3.35186 loss_avg=3.41248 acc=0.69531 acc_top1_avg=0.67315 acc_top5_avg=0.91482 lr=0.01000 gn=10.44959 time=56.42it/s +====================Eval==================== +epoch=38 global_step=15249 loss=1.12105 test_loss_avg=1.08801 acc=0.75000 test_acc_avg=0.73730 test_acc_top5_avg=0.97054 time=239.37it/s +epoch=38 global_step=15249 loss=0.82869 test_loss_avg=0.93412 acc=0.75000 test_acc_avg=0.76820 test_acc_top5_avg=0.97468 time=868.21it/s +curr_acc 0.7682 +BEST_ACC 0.7986 +curr_acc_top5 0.9747 +BEST_ACC_top5 0.9842 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=3.75295 loss_avg=3.75295 acc=0.63281 acc_top1_avg=0.63281 acc_top5_avg=0.89844 lr=0.01000 gn=8.13779 time=51.13it/s +epoch=39 global_step=15300 loss=3.10377 loss_avg=3.28938 acc=0.70312 acc_top1_avg=0.68612 acc_top5_avg=0.92004 lr=0.01000 gn=8.95571 time=55.46it/s +epoch=39 global_step=15350 loss=3.25222 loss_avg=3.34627 acc=0.69531 acc_top1_avg=0.68054 acc_top5_avg=0.92033 lr=0.01000 gn=8.66478 time=49.08it/s +epoch=39 global_step=15400 loss=3.13447 loss_avg=3.38734 acc=0.69531 acc_top1_avg=0.67638 acc_top5_avg=0.91779 lr=0.01000 gn=9.52101 time=55.68it/s +epoch=39 global_step=15450 loss=3.31452 loss_avg=3.38745 acc=0.69531 acc_top1_avg=0.67588 acc_top5_avg=0.91628 lr=0.01000 gn=10.23887 time=55.55it/s +epoch=39 global_step=15500 loss=3.48827 loss_avg=3.40060 acc=0.65625 acc_top1_avg=0.67458 acc_top5_avg=0.91487 lr=0.01000 gn=9.56461 time=53.39it/s +epoch=39 global_step=15550 loss=3.16109 loss_avg=3.40480 acc=0.68750 acc_top1_avg=0.67346 acc_top5_avg=0.91593 lr=0.01000 gn=9.20954 time=47.13it/s +epoch=39 global_step=15600 loss=3.34477 loss_avg=3.42594 acc=0.70312 acc_top1_avg=0.67130 acc_top5_avg=0.91544 lr=0.01000 gn=11.12391 time=58.49it/s +====================Eval==================== +epoch=39 global_step=15640 loss=1.78220 test_loss_avg=0.66398 acc=0.53906 test_acc_avg=0.81579 test_acc_top5_avg=0.98561 time=231.36it/s +epoch=39 global_step=15640 loss=0.65494 test_loss_avg=1.04461 acc=0.81250 test_acc_avg=0.74830 test_acc_top5_avg=0.97871 time=212.42it/s +epoch=39 global_step=15640 loss=0.93904 test_loss_avg=1.02329 acc=0.81250 test_acc_avg=0.74901 test_acc_top5_avg=0.97973 time=837.52it/s +curr_acc 0.7490 +BEST_ACC 0.7986 +curr_acc_top5 0.9797 +BEST_ACC_top5 0.9842 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=3.71672 loss_avg=3.31062 acc=0.64844 acc_top1_avg=0.68516 acc_top5_avg=0.91875 lr=0.00100 gn=8.85107 time=49.01it/s +epoch=40 global_step=15700 loss=2.64052 loss_avg=3.24145 acc=0.74219 acc_top1_avg=0.69349 acc_top5_avg=0.91758 lr=0.00100 gn=7.01970 time=51.91it/s +epoch=40 global_step=15750 loss=2.12014 loss_avg=3.17732 acc=0.80469 acc_top1_avg=0.69822 acc_top5_avg=0.92010 lr=0.00100 gn=8.99012 time=50.13it/s +epoch=40 global_step=15800 loss=3.12747 loss_avg=3.13193 acc=0.71094 acc_top1_avg=0.70205 acc_top5_avg=0.92266 lr=0.00100 gn=7.36182 time=57.77it/s +epoch=40 global_step=15850 loss=2.72079 loss_avg=3.10568 acc=0.75781 acc_top1_avg=0.70525 acc_top5_avg=0.92344 lr=0.00100 gn=9.23650 time=49.57it/s +epoch=40 global_step=15900 loss=2.59492 loss_avg=3.08092 acc=0.76562 acc_top1_avg=0.70781 acc_top5_avg=0.92377 lr=0.00100 gn=8.11621 time=59.45it/s +epoch=40 global_step=15950 loss=2.73688 loss_avg=3.05432 acc=0.73438 acc_top1_avg=0.71046 acc_top5_avg=0.92414 lr=0.00100 gn=9.44097 time=59.41it/s +epoch=40 global_step=16000 loss=2.91649 loss_avg=3.03241 acc=0.71875 acc_top1_avg=0.71241 acc_top5_avg=0.92489 lr=0.00100 gn=8.43324 time=50.92it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.89639 test_loss_avg=0.48284 acc=0.75000 test_acc_avg=0.86309 test_acc_top5_avg=0.99062 time=223.60it/s +epoch=40 global_step=16031 loss=0.41280 test_loss_avg=0.42913 acc=0.87500 test_acc_avg=0.87866 test_acc_top5_avg=0.99179 time=540.43it/s +curr_acc 0.8787 +BEST_ACC 0.7986 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9842 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=3.05748 loss_avg=2.94271 acc=0.71094 acc_top1_avg=0.71793 acc_top5_avg=0.93544 lr=0.00100 gn=9.58542 time=52.43it/s +epoch=41 global_step=16100 loss=2.74330 loss_avg=2.92805 acc=0.72656 acc_top1_avg=0.72135 acc_top5_avg=0.92844 lr=0.00100 gn=7.11606 time=56.08it/s +epoch=41 global_step=16150 loss=3.31078 loss_avg=2.91510 acc=0.68750 acc_top1_avg=0.72302 acc_top5_avg=0.92969 lr=0.00100 gn=8.07594 time=55.26it/s +epoch=41 global_step=16200 loss=2.77419 loss_avg=2.87696 acc=0.71875 acc_top1_avg=0.72670 acc_top5_avg=0.93084 lr=0.00100 gn=7.13471 time=57.50it/s +epoch=41 global_step=16250 loss=2.82306 loss_avg=2.87137 acc=0.72656 acc_top1_avg=0.72738 acc_top5_avg=0.93104 lr=0.00100 gn=9.40567 time=55.59it/s +epoch=41 global_step=16300 loss=3.31148 loss_avg=2.86242 acc=0.67188 acc_top1_avg=0.72819 acc_top5_avg=0.93009 lr=0.00100 gn=8.86270 time=56.85it/s +epoch=41 global_step=16350 loss=2.99944 loss_avg=2.87015 acc=0.71094 acc_top1_avg=0.72725 acc_top5_avg=0.92964 lr=0.00100 gn=9.77500 time=58.03it/s +epoch=41 global_step=16400 loss=2.16138 loss_avg=2.85284 acc=0.80469 acc_top1_avg=0.72893 acc_top5_avg=0.93007 lr=0.00100 gn=9.09567 time=50.89it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.15358 test_loss_avg=0.35295 acc=0.94531 test_acc_avg=0.88991 test_acc_top5_avg=0.99503 time=232.73it/s +epoch=41 global_step=16422 loss=0.14325 test_loss_avg=0.47442 acc=0.95312 test_acc_avg=0.86309 test_acc_top5_avg=0.99065 time=224.49it/s +epoch=41 global_step=16422 loss=0.26433 test_loss_avg=0.41222 acc=0.87500 test_acc_avg=0.87905 test_acc_top5_avg=0.99169 time=867.85it/s +curr_acc 0.8791 +BEST_ACC 0.8787 +curr_acc_top5 0.9917 +BEST_ACC_top5 0.9918 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=2.86859 loss_avg=2.75224 acc=0.72656 acc_top1_avg=0.73772 acc_top5_avg=0.93025 lr=0.00100 gn=8.13331 time=53.49it/s +epoch=42 global_step=16500 loss=2.67018 loss_avg=2.79121 acc=0.74219 acc_top1_avg=0.73387 acc_top5_avg=0.93119 lr=0.00100 gn=9.20370 time=52.29it/s +epoch=42 global_step=16550 loss=2.64997 loss_avg=2.79335 acc=0.75000 acc_top1_avg=0.73425 acc_top5_avg=0.93103 lr=0.00100 gn=9.70905 time=58.12it/s +epoch=42 global_step=16600 loss=2.90251 loss_avg=2.81430 acc=0.74219 acc_top1_avg=0.73266 acc_top5_avg=0.93070 lr=0.00100 gn=9.76427 time=52.89it/s +epoch=42 global_step=16650 loss=2.66477 loss_avg=2.81774 acc=0.74219 acc_top1_avg=0.73215 acc_top5_avg=0.93048 lr=0.00100 gn=7.92912 time=52.00it/s +epoch=42 global_step=16700 loss=2.62157 loss_avg=2.80380 acc=0.74219 acc_top1_avg=0.73353 acc_top5_avg=0.93115 lr=0.00100 gn=8.97557 time=48.64it/s +epoch=42 global_step=16750 loss=3.10643 loss_avg=2.79552 acc=0.68750 acc_top1_avg=0.73485 acc_top5_avg=0.93133 lr=0.00100 gn=9.74045 time=58.87it/s +epoch=42 global_step=16800 loss=2.53619 loss_avg=2.78744 acc=0.76562 acc_top1_avg=0.73597 acc_top5_avg=0.93173 lr=0.00100 gn=9.20455 time=51.75it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.53264 test_loss_avg=0.56201 acc=0.83594 test_acc_avg=0.84546 test_acc_top5_avg=0.98633 time=213.67it/s +epoch=42 global_step=16813 loss=0.38760 test_loss_avg=0.39665 acc=0.87500 test_acc_avg=0.88884 test_acc_top5_avg=0.99140 time=865.34it/s +curr_acc 0.8888 +BEST_ACC 0.8791 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9918 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=2.65326 loss_avg=2.70718 acc=0.75000 acc_top1_avg=0.74345 acc_top5_avg=0.93011 lr=0.00100 gn=9.85050 time=53.92it/s +epoch=43 global_step=16900 loss=2.78846 loss_avg=2.71450 acc=0.75000 acc_top1_avg=0.74228 acc_top5_avg=0.93166 lr=0.00100 gn=8.44107 time=53.20it/s +epoch=43 global_step=16950 loss=2.59302 loss_avg=2.75017 acc=0.75000 acc_top1_avg=0.73956 acc_top5_avg=0.93203 lr=0.00100 gn=7.71591 time=45.27it/s +epoch=43 global_step=17000 loss=2.44520 loss_avg=2.75312 acc=0.77344 acc_top1_avg=0.73960 acc_top5_avg=0.93190 lr=0.00100 gn=9.59955 time=58.81it/s +epoch=43 global_step=17050 loss=2.31451 loss_avg=2.75573 acc=0.77344 acc_top1_avg=0.73909 acc_top5_avg=0.93137 lr=0.00100 gn=8.84757 time=57.09it/s +epoch=43 global_step=17100 loss=2.44263 loss_avg=2.74321 acc=0.78906 acc_top1_avg=0.74045 acc_top5_avg=0.93187 lr=0.00100 gn=9.24609 time=53.36it/s +epoch=43 global_step=17150 loss=2.43647 loss_avg=2.73741 acc=0.77344 acc_top1_avg=0.74098 acc_top5_avg=0.93214 lr=0.00100 gn=7.59006 time=56.60it/s +epoch=43 global_step=17200 loss=2.42785 loss_avg=2.74138 acc=0.76562 acc_top1_avg=0.74053 acc_top5_avg=0.93253 lr=0.00100 gn=9.88604 time=48.50it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.22703 test_loss_avg=0.34852 acc=0.92969 test_acc_avg=0.89323 test_acc_top5_avg=0.99479 time=233.11it/s +epoch=43 global_step=17204 loss=0.19462 test_loss_avg=0.44481 acc=0.95312 test_acc_avg=0.87692 test_acc_top5_avg=0.99233 time=231.10it/s +epoch=43 global_step=17204 loss=0.45342 test_loss_avg=0.38506 acc=0.87500 test_acc_avg=0.89122 test_acc_top5_avg=0.99347 time=879.68it/s +curr_acc 0.8912 +BEST_ACC 0.8888 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9918 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=2.76012 loss_avg=2.64401 acc=0.74219 acc_top1_avg=0.75255 acc_top5_avg=0.93257 lr=0.00100 gn=11.68837 time=57.90it/s +epoch=44 global_step=17300 loss=2.51771 loss_avg=2.68834 acc=0.75781 acc_top1_avg=0.74707 acc_top5_avg=0.93465 lr=0.00100 gn=9.33495 time=59.11it/s +epoch=44 global_step=17350 loss=2.26218 loss_avg=2.68165 acc=0.77344 acc_top1_avg=0.74749 acc_top5_avg=0.93327 lr=0.00100 gn=7.84723 time=53.62it/s +epoch=44 global_step=17400 loss=3.41255 loss_avg=2.67433 acc=0.66406 acc_top1_avg=0.74833 acc_top5_avg=0.93431 lr=0.00100 gn=10.82124 time=52.17it/s +epoch=44 global_step=17450 loss=2.43372 loss_avg=2.68934 acc=0.77344 acc_top1_avg=0.74692 acc_top5_avg=0.93353 lr=0.00100 gn=8.26531 time=55.78it/s +epoch=44 global_step=17500 loss=2.37317 loss_avg=2.67981 acc=0.77344 acc_top1_avg=0.74781 acc_top5_avg=0.93320 lr=0.00100 gn=9.87025 time=51.54it/s +epoch=44 global_step=17550 loss=2.51898 loss_avg=2.69470 acc=0.76562 acc_top1_avg=0.74596 acc_top5_avg=0.93262 lr=0.00100 gn=11.58074 time=52.27it/s +====================Eval==================== +epoch=44 global_step=17595 loss=0.88370 test_loss_avg=0.49625 acc=0.75781 test_acc_avg=0.85970 test_acc_top5_avg=0.99056 time=238.91it/s +epoch=44 global_step=17595 loss=0.18950 test_loss_avg=0.40375 acc=0.94531 test_acc_avg=0.88714 test_acc_top5_avg=0.99314 time=239.52it/s +epoch=44 global_step=17595 loss=0.19114 test_loss_avg=0.39213 acc=0.87500 test_acc_avg=0.88914 test_acc_top5_avg=0.99347 time=745.12it/s +curr_acc 0.8891 +BEST_ACC 0.8912 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9935 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=3.07305 loss_avg=2.66202 acc=0.72656 acc_top1_avg=0.75469 acc_top5_avg=0.94219 lr=0.00100 gn=10.77246 time=57.84it/s +epoch=45 global_step=17650 loss=2.41801 loss_avg=2.67865 acc=0.78125 acc_top1_avg=0.74759 acc_top5_avg=0.93210 lr=0.00100 gn=9.20914 time=52.84it/s +epoch=45 global_step=17700 loss=2.68669 loss_avg=2.69644 acc=0.74219 acc_top1_avg=0.74531 acc_top5_avg=0.93199 lr=0.00100 gn=12.36479 time=55.98it/s +epoch=45 global_step=17750 loss=2.90609 loss_avg=2.69325 acc=0.74219 acc_top1_avg=0.74551 acc_top5_avg=0.93251 lr=0.00100 gn=10.53437 time=49.68it/s +epoch=45 global_step=17800 loss=2.32343 loss_avg=2.66733 acc=0.78906 acc_top1_avg=0.74844 acc_top5_avg=0.93300 lr=0.00100 gn=12.35610 time=57.25it/s +epoch=45 global_step=17850 loss=2.27532 loss_avg=2.65448 acc=0.79688 acc_top1_avg=0.74994 acc_top5_avg=0.93343 lr=0.00100 gn=15.71500 time=52.27it/s +epoch=45 global_step=17900 loss=2.29184 loss_avg=2.66763 acc=0.78125 acc_top1_avg=0.74823 acc_top5_avg=0.93312 lr=0.00100 gn=11.55603 time=52.06it/s +epoch=45 global_step=17950 loss=2.28664 loss_avg=2.65551 acc=0.78906 acc_top1_avg=0.74969 acc_top5_avg=0.93376 lr=0.00100 gn=11.55842 time=52.50it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.54203 test_loss_avg=0.52865 acc=0.86719 test_acc_avg=0.85503 test_acc_top5_avg=0.99028 time=199.07it/s +epoch=45 global_step=17986 loss=0.19624 test_loss_avg=0.39314 acc=0.93750 test_acc_avg=0.89191 test_acc_top5_avg=0.99209 time=859.84it/s +curr_acc 0.8919 +BEST_ACC 0.8912 +curr_acc_top5 0.9921 +BEST_ACC_top5 0.9935 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=1.98268 loss_avg=2.69506 acc=0.82812 acc_top1_avg=0.74833 acc_top5_avg=0.93862 lr=0.00100 gn=12.33390 time=58.74it/s +epoch=46 global_step=18050 loss=3.10757 loss_avg=2.66673 acc=0.70312 acc_top1_avg=0.74719 acc_top5_avg=0.92981 lr=0.00100 gn=9.46734 time=56.02it/s +epoch=46 global_step=18100 loss=3.41759 loss_avg=2.62835 acc=0.67969 acc_top1_avg=0.75151 acc_top5_avg=0.93250 lr=0.00100 gn=9.69302 time=53.72it/s +epoch=46 global_step=18150 loss=2.94911 loss_avg=2.62917 acc=0.72656 acc_top1_avg=0.75157 acc_top5_avg=0.93436 lr=0.00100 gn=11.69240 time=51.45it/s +epoch=46 global_step=18200 loss=2.81113 loss_avg=2.63652 acc=0.73438 acc_top1_avg=0.75069 acc_top5_avg=0.93374 lr=0.00100 gn=9.29630 time=51.25it/s +epoch=46 global_step=18250 loss=3.06634 loss_avg=2.65003 acc=0.70312 acc_top1_avg=0.74935 acc_top5_avg=0.93324 lr=0.00100 gn=11.68636 time=52.84it/s +epoch=46 global_step=18300 loss=2.62179 loss_avg=2.64924 acc=0.75781 acc_top1_avg=0.74945 acc_top5_avg=0.93384 lr=0.00100 gn=11.54611 time=58.56it/s +epoch=46 global_step=18350 loss=2.59144 loss_avg=2.63976 acc=0.76562 acc_top1_avg=0.75039 acc_top5_avg=0.93394 lr=0.00100 gn=10.59843 time=46.84it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.45397 test_loss_avg=0.31471 acc=0.89844 test_acc_avg=0.90527 test_acc_top5_avg=0.99805 time=237.65it/s +epoch=46 global_step=18377 loss=0.09567 test_loss_avg=0.38923 acc=0.97656 test_acc_avg=0.89299 test_acc_top5_avg=0.99361 time=222.25it/s +epoch=46 global_step=18377 loss=0.36989 test_loss_avg=0.36269 acc=0.87500 test_acc_avg=0.89953 test_acc_top5_avg=0.99387 time=882.45it/s +curr_acc 0.8995 +BEST_ACC 0.8919 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9935 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=2.74449 loss_avg=2.57345 acc=0.73438 acc_top1_avg=0.75951 acc_top5_avg=0.93750 lr=0.00100 gn=11.08714 time=56.68it/s +epoch=47 global_step=18450 loss=2.71284 loss_avg=2.57414 acc=0.73438 acc_top1_avg=0.75856 acc_top5_avg=0.93429 lr=0.00100 gn=8.34384 time=52.39it/s +epoch=47 global_step=18500 loss=3.01147 loss_avg=2.60294 acc=0.71094 acc_top1_avg=0.75572 acc_top5_avg=0.93566 lr=0.00100 gn=8.37980 time=57.98it/s +epoch=47 global_step=18550 loss=2.38670 loss_avg=2.59713 acc=0.76562 acc_top1_avg=0.75596 acc_top5_avg=0.93605 lr=0.00100 gn=8.98615 time=57.64it/s +epoch=47 global_step=18600 loss=2.30417 loss_avg=2.61003 acc=0.76562 acc_top1_avg=0.75473 acc_top5_avg=0.93592 lr=0.00100 gn=10.49203 time=58.37it/s +epoch=47 global_step=18650 loss=2.75738 loss_avg=2.61238 acc=0.74219 acc_top1_avg=0.75464 acc_top5_avg=0.93530 lr=0.00100 gn=11.70908 time=53.02it/s +epoch=47 global_step=18700 loss=2.01174 loss_avg=2.61697 acc=0.80469 acc_top1_avg=0.75406 acc_top5_avg=0.93479 lr=0.00100 gn=10.62435 time=52.94it/s +epoch=47 global_step=18750 loss=2.63761 loss_avg=2.60648 acc=0.75781 acc_top1_avg=0.75503 acc_top5_avg=0.93513 lr=0.00100 gn=10.78894 time=52.09it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.17369 test_loss_avg=0.45101 acc=0.93750 test_acc_avg=0.87141 test_acc_top5_avg=0.99303 time=228.20it/s +epoch=47 global_step=18768 loss=0.30270 test_loss_avg=0.35832 acc=0.87500 test_acc_avg=0.89686 test_acc_top5_avg=0.99367 time=881.53it/s +curr_acc 0.8969 +BEST_ACC 0.8995 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=2.39303 loss_avg=2.55705 acc=0.78125 acc_top1_avg=0.75879 acc_top5_avg=0.93237 lr=0.00100 gn=12.51748 time=58.03it/s +epoch=48 global_step=18850 loss=2.42640 loss_avg=2.56988 acc=0.77344 acc_top1_avg=0.75734 acc_top5_avg=0.93321 lr=0.00100 gn=10.33467 time=53.01it/s +epoch=48 global_step=18900 loss=2.89307 loss_avg=2.60570 acc=0.72656 acc_top1_avg=0.75343 acc_top5_avg=0.93324 lr=0.00100 gn=12.50741 time=52.52it/s +epoch=48 global_step=18950 loss=2.18794 loss_avg=2.59136 acc=0.78906 acc_top1_avg=0.75528 acc_top5_avg=0.93445 lr=0.00100 gn=9.77275 time=55.66it/s +epoch=48 global_step=19000 loss=2.69769 loss_avg=2.58132 acc=0.74219 acc_top1_avg=0.75613 acc_top5_avg=0.93504 lr=0.00100 gn=12.10620 time=58.89it/s +epoch=48 global_step=19050 loss=2.88174 loss_avg=2.56951 acc=0.73438 acc_top1_avg=0.75709 acc_top5_avg=0.93562 lr=0.00100 gn=9.67743 time=51.86it/s +epoch=48 global_step=19100 loss=2.75433 loss_avg=2.57803 acc=0.75000 acc_top1_avg=0.75668 acc_top5_avg=0.93564 lr=0.00100 gn=13.79363 time=52.62it/s +epoch=48 global_step=19150 loss=2.58014 loss_avg=2.57721 acc=0.75781 acc_top1_avg=0.75699 acc_top5_avg=0.93629 lr=0.00100 gn=14.14495 time=59.13it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.25135 test_loss_avg=0.35368 acc=0.91406 test_acc_avg=0.88184 test_acc_top5_avg=0.99707 time=241.87it/s +epoch=48 global_step=19159 loss=0.24887 test_loss_avg=0.40727 acc=0.91406 test_acc_avg=0.88429 test_acc_top5_avg=0.99300 time=219.83it/s +epoch=48 global_step=19159 loss=0.37084 test_loss_avg=0.35243 acc=0.87500 test_acc_avg=0.89864 test_acc_top5_avg=0.99367 time=716.85it/s +curr_acc 0.8986 +BEST_ACC 0.8995 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=3.44312 loss_avg=2.58414 acc=0.65625 acc_top1_avg=0.75572 acc_top5_avg=0.94112 lr=0.00100 gn=11.47063 time=45.29it/s +epoch=49 global_step=19250 loss=2.92438 loss_avg=2.55326 acc=0.71875 acc_top1_avg=0.75901 acc_top5_avg=0.93776 lr=0.00100 gn=15.79600 time=45.80it/s +epoch=49 global_step=19300 loss=3.01818 loss_avg=2.53606 acc=0.71094 acc_top1_avg=0.76147 acc_top5_avg=0.93678 lr=0.00100 gn=11.13332 time=56.88it/s +epoch=49 global_step=19350 loss=2.55092 loss_avg=2.56488 acc=0.75781 acc_top1_avg=0.75871 acc_top5_avg=0.93611 lr=0.00100 gn=13.94528 time=59.18it/s +epoch=49 global_step=19400 loss=2.47133 loss_avg=2.57758 acc=0.75000 acc_top1_avg=0.75723 acc_top5_avg=0.93588 lr=0.00100 gn=12.27603 time=44.59it/s +epoch=49 global_step=19450 loss=2.42552 loss_avg=2.56828 acc=0.78125 acc_top1_avg=0.75819 acc_top5_avg=0.93643 lr=0.00100 gn=11.40113 time=57.42it/s +epoch=49 global_step=19500 loss=3.11624 loss_avg=2.57123 acc=0.68750 acc_top1_avg=0.75788 acc_top5_avg=0.93590 lr=0.00100 gn=9.45154 time=49.17it/s +epoch=49 global_step=19550 loss=3.27099 loss_avg=2.56672 acc=0.70000 acc_top1_avg=0.75818 acc_top5_avg=0.93563 lr=0.00100 gn=23.21634 time=61.69it/s +====================Eval==================== +epoch=49 global_step=19550 loss=0.71789 test_loss_avg=0.49121 acc=0.78906 test_acc_avg=0.86234 test_acc_top5_avg=0.99165 time=115.17it/s +epoch=49 global_step=19550 loss=0.41512 test_loss_avg=0.36284 acc=0.93750 test_acc_avg=0.89666 test_acc_top5_avg=0.99397 time=544.71it/s +epoch=49 global_step=19550 loss=0.41512 test_loss_avg=0.36284 acc=0.93750 test_acc_avg=0.89666 test_acc_top5_avg=0.99397 time=544.71it/s +curr_acc 0.8967 +BEST_ACC 0.8995 +curr_acc_top5 0.9940 +BEST_ACC_top5 0.9939 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=1.95169 loss_avg=2.57670 acc=0.83594 acc_top1_avg=0.75797 acc_top5_avg=0.93875 lr=0.00100 gn=9.48146 time=57.66it/s +epoch=50 global_step=19650 loss=2.39557 loss_avg=2.53410 acc=0.78125 acc_top1_avg=0.76172 acc_top5_avg=0.93766 lr=0.00100 gn=10.68474 time=50.34it/s +epoch=50 global_step=19700 loss=2.53454 loss_avg=2.52345 acc=0.75781 acc_top1_avg=0.76224 acc_top5_avg=0.93828 lr=0.00100 gn=10.91553 time=51.56it/s +epoch=50 global_step=19750 loss=1.93420 loss_avg=2.52267 acc=0.82812 acc_top1_avg=0.76277 acc_top5_avg=0.93859 lr=0.00100 gn=12.80090 time=56.89it/s +epoch=50 global_step=19800 loss=2.98434 loss_avg=2.53878 acc=0.69531 acc_top1_avg=0.76116 acc_top5_avg=0.93794 lr=0.00100 gn=12.73968 time=52.32it/s +epoch=50 global_step=19850 loss=3.17729 loss_avg=2.54771 acc=0.69531 acc_top1_avg=0.76060 acc_top5_avg=0.93682 lr=0.00100 gn=12.85409 time=57.05it/s +epoch=50 global_step=19900 loss=3.00070 loss_avg=2.54754 acc=0.71875 acc_top1_avg=0.76056 acc_top5_avg=0.93654 lr=0.00100 gn=10.72388 time=58.47it/s +====================Eval==================== +epoch=50 global_step=19941 loss=0.26386 test_loss_avg=0.51068 acc=0.91406 test_acc_avg=0.86234 test_acc_top5_avg=0.99281 time=230.13it/s +epoch=50 global_step=19941 loss=0.27527 test_loss_avg=0.40719 acc=0.87500 test_acc_avg=0.88805 test_acc_top5_avg=0.99397 time=664.71it/s +curr_acc 0.8881 +BEST_ACC 0.8995 +curr_acc_top5 0.9940 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=2.15189 loss_avg=2.44117 acc=0.79688 acc_top1_avg=0.76736 acc_top5_avg=0.93924 lr=0.00100 gn=7.58852 time=58.26it/s +epoch=51 global_step=20000 loss=2.67040 loss_avg=2.48025 acc=0.74219 acc_top1_avg=0.76615 acc_top5_avg=0.93816 lr=0.00100 gn=13.83435 time=50.51it/s +epoch=51 global_step=20050 loss=2.61898 loss_avg=2.49693 acc=0.74219 acc_top1_avg=0.76577 acc_top5_avg=0.93865 lr=0.00100 gn=11.26172 time=50.78it/s +epoch=51 global_step=20100 loss=2.09978 loss_avg=2.50665 acc=0.81250 acc_top1_avg=0.76567 acc_top5_avg=0.93721 lr=0.00100 gn=12.64719 time=50.85it/s +epoch=51 global_step=20150 loss=2.48067 loss_avg=2.50628 acc=0.75781 acc_top1_avg=0.76533 acc_top5_avg=0.93795 lr=0.00100 gn=9.10099 time=50.68it/s +epoch=51 global_step=20200 loss=2.87866 loss_avg=2.50901 acc=0.73438 acc_top1_avg=0.76481 acc_top5_avg=0.93774 lr=0.00100 gn=15.46695 time=56.99it/s +epoch=51 global_step=20250 loss=2.53435 loss_avg=2.51773 acc=0.76562 acc_top1_avg=0.76348 acc_top5_avg=0.93790 lr=0.00100 gn=12.36964 time=57.38it/s +epoch=51 global_step=20300 loss=2.05884 loss_avg=2.52463 acc=0.79688 acc_top1_avg=0.76271 acc_top5_avg=0.93720 lr=0.00100 gn=13.80835 time=57.28it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.62347 test_loss_avg=0.34207 acc=0.83594 test_acc_avg=0.90030 test_acc_top5_avg=0.99591 time=225.19it/s +epoch=51 global_step=20332 loss=0.22401 test_loss_avg=0.36991 acc=0.93750 test_acc_avg=0.89855 test_acc_top5_avg=0.99362 time=233.68it/s +epoch=51 global_step=20332 loss=0.31640 test_loss_avg=0.36088 acc=0.87500 test_acc_avg=0.89943 test_acc_top5_avg=0.99367 time=860.55it/s +curr_acc 0.8994 +BEST_ACC 0.8995 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=2.89390 loss_avg=2.39384 acc=0.73438 acc_top1_avg=0.77734 acc_top5_avg=0.93967 lr=0.00100 gn=14.31725 time=45.72it/s +epoch=52 global_step=20400 loss=2.60945 loss_avg=2.49793 acc=0.75000 acc_top1_avg=0.76482 acc_top5_avg=0.93302 lr=0.00100 gn=8.43502 time=54.37it/s +epoch=52 global_step=20450 loss=2.06641 loss_avg=2.47415 acc=0.80469 acc_top1_avg=0.76788 acc_top5_avg=0.93505 lr=0.00100 gn=15.27830 time=50.79it/s +epoch=52 global_step=20500 loss=2.60420 loss_avg=2.50291 acc=0.75000 acc_top1_avg=0.76493 acc_top5_avg=0.93462 lr=0.00100 gn=16.30086 time=57.55it/s +epoch=52 global_step=20550 loss=2.50575 loss_avg=2.50059 acc=0.78906 acc_top1_avg=0.76516 acc_top5_avg=0.93546 lr=0.00100 gn=12.47156 time=55.26it/s +epoch=52 global_step=20600 loss=2.48782 loss_avg=2.50341 acc=0.78125 acc_top1_avg=0.76504 acc_top5_avg=0.93613 lr=0.00100 gn=12.84599 time=48.96it/s +epoch=52 global_step=20650 loss=2.75920 loss_avg=2.49883 acc=0.75000 acc_top1_avg=0.76521 acc_top5_avg=0.93701 lr=0.00100 gn=12.84025 time=57.17it/s +epoch=52 global_step=20700 loss=2.50735 loss_avg=2.50475 acc=0.75000 acc_top1_avg=0.76439 acc_top5_avg=0.93650 lr=0.00100 gn=10.19334 time=54.37it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.66282 test_loss_avg=0.44317 acc=0.82812 test_acc_avg=0.87500 test_acc_top5_avg=0.99200 time=239.29it/s +epoch=52 global_step=20723 loss=0.25201 test_loss_avg=0.38364 acc=0.87500 test_acc_avg=0.89211 test_acc_top5_avg=0.99268 time=537.46it/s +curr_acc 0.8921 +BEST_ACC 0.8995 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=2.93790 loss_avg=2.62931 acc=0.71875 acc_top1_avg=0.75231 acc_top5_avg=0.93316 lr=0.00100 gn=9.75631 time=56.96it/s +epoch=53 global_step=20800 loss=2.00465 loss_avg=2.56252 acc=0.82031 acc_top1_avg=0.76025 acc_top5_avg=0.93273 lr=0.00100 gn=12.58060 time=55.66it/s +epoch=53 global_step=20850 loss=2.50606 loss_avg=2.53745 acc=0.76562 acc_top1_avg=0.76163 acc_top5_avg=0.93547 lr=0.00100 gn=12.52775 time=58.21it/s +epoch=53 global_step=20900 loss=2.40282 loss_avg=2.51079 acc=0.76562 acc_top1_avg=0.76390 acc_top5_avg=0.93525 lr=0.00100 gn=10.13347 time=50.85it/s +epoch=53 global_step=20950 loss=2.36690 loss_avg=2.48012 acc=0.78125 acc_top1_avg=0.76697 acc_top5_avg=0.93602 lr=0.00100 gn=12.41432 time=51.06it/s +epoch=53 global_step=21000 loss=2.73099 loss_avg=2.48476 acc=0.74219 acc_top1_avg=0.76667 acc_top5_avg=0.93702 lr=0.00100 gn=12.69653 time=55.83it/s +epoch=53 global_step=21050 loss=2.79968 loss_avg=2.48088 acc=0.72656 acc_top1_avg=0.76723 acc_top5_avg=0.93795 lr=0.00100 gn=10.70382 time=54.73it/s +epoch=53 global_step=21100 loss=2.32938 loss_avg=2.47695 acc=0.78125 acc_top1_avg=0.76743 acc_top5_avg=0.93783 lr=0.00100 gn=15.17365 time=56.19it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.09604 test_loss_avg=0.32398 acc=0.96875 test_acc_avg=0.89603 test_acc_top5_avg=0.99820 time=229.70it/s +epoch=53 global_step=21114 loss=0.17467 test_loss_avg=0.45149 acc=0.96094 test_acc_avg=0.87612 test_acc_top5_avg=0.99206 time=228.68it/s +epoch=53 global_step=21114 loss=0.45472 test_loss_avg=0.39975 acc=0.87500 test_acc_avg=0.88934 test_acc_top5_avg=0.99308 time=676.17it/s +curr_acc 0.8893 +BEST_ACC 0.8995 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=2.13338 loss_avg=2.43273 acc=0.79688 acc_top1_avg=0.76888 acc_top5_avg=0.93902 lr=0.00100 gn=11.30087 time=53.75it/s +epoch=54 global_step=21200 loss=2.92118 loss_avg=2.46433 acc=0.72656 acc_top1_avg=0.76708 acc_top5_avg=0.94104 lr=0.00100 gn=12.24665 time=48.82it/s +epoch=54 global_step=21250 loss=2.48759 loss_avg=2.47779 acc=0.75781 acc_top1_avg=0.76637 acc_top5_avg=0.94020 lr=0.00100 gn=9.28409 time=51.56it/s +epoch=54 global_step=21300 loss=2.80834 loss_avg=2.49444 acc=0.72656 acc_top1_avg=0.76457 acc_top5_avg=0.93792 lr=0.00100 gn=11.23192 time=51.82it/s +epoch=54 global_step=21350 loss=2.23562 loss_avg=2.49517 acc=0.78906 acc_top1_avg=0.76523 acc_top5_avg=0.93773 lr=0.00100 gn=10.13222 time=52.42it/s +epoch=54 global_step=21400 loss=2.38257 loss_avg=2.48494 acc=0.77344 acc_top1_avg=0.76614 acc_top5_avg=0.93780 lr=0.00100 gn=14.61573 time=49.99it/s +epoch=54 global_step=21450 loss=2.20906 loss_avg=2.47789 acc=0.79688 acc_top1_avg=0.76711 acc_top5_avg=0.93815 lr=0.00100 gn=14.15950 time=52.75it/s +epoch=54 global_step=21500 loss=2.71707 loss_avg=2.48407 acc=0.74219 acc_top1_avg=0.76625 acc_top5_avg=0.93793 lr=0.00100 gn=12.73202 time=55.90it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.22941 test_loss_avg=0.38154 acc=0.95312 test_acc_avg=0.89062 test_acc_top5_avg=0.99334 time=233.06it/s +epoch=54 global_step=21505 loss=0.79787 test_loss_avg=0.36972 acc=0.81250 test_acc_avg=0.89498 test_acc_top5_avg=0.99387 time=881.71it/s +curr_acc 0.8950 +BEST_ACC 0.8995 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=2.30887 loss_avg=2.49114 acc=0.78906 acc_top1_avg=0.76545 acc_top5_avg=0.93663 lr=0.00100 gn=15.42160 time=57.64it/s +epoch=55 global_step=21600 loss=2.37606 loss_avg=2.46959 acc=0.76562 acc_top1_avg=0.76817 acc_top5_avg=0.93577 lr=0.00100 gn=15.35192 time=55.75it/s +epoch=55 global_step=21650 loss=2.39653 loss_avg=2.46090 acc=0.76562 acc_top1_avg=0.76902 acc_top5_avg=0.93766 lr=0.00100 gn=10.84241 time=57.29it/s +epoch=55 global_step=21700 loss=2.44290 loss_avg=2.46302 acc=0.76562 acc_top1_avg=0.76931 acc_top5_avg=0.93790 lr=0.00100 gn=14.26346 time=57.50it/s +epoch=55 global_step=21750 loss=2.79635 loss_avg=2.45062 acc=0.74219 acc_top1_avg=0.77047 acc_top5_avg=0.93718 lr=0.00100 gn=13.47714 time=51.77it/s +epoch=55 global_step=21800 loss=2.88330 loss_avg=2.45388 acc=0.71875 acc_top1_avg=0.76997 acc_top5_avg=0.93729 lr=0.00100 gn=11.71068 time=51.56it/s +epoch=55 global_step=21850 loss=2.74183 loss_avg=2.46112 acc=0.74219 acc_top1_avg=0.76941 acc_top5_avg=0.93748 lr=0.00100 gn=12.55052 time=49.79it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.26617 test_loss_avg=0.39202 acc=0.89062 test_acc_avg=0.87500 test_acc_top5_avg=0.99375 time=246.67it/s +epoch=55 global_step=21896 loss=0.29618 test_loss_avg=0.42705 acc=0.90625 test_acc_avg=0.88224 test_acc_top5_avg=0.99148 time=208.26it/s +epoch=55 global_step=21896 loss=0.31388 test_loss_avg=0.36521 acc=0.87500 test_acc_avg=0.89844 test_acc_top5_avg=0.99268 time=892.41it/s +curr_acc 0.8984 +BEST_ACC 0.8995 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=2.44904 loss_avg=2.35343 acc=0.75781 acc_top1_avg=0.77344 acc_top5_avg=0.95508 lr=0.00100 gn=5.31660 time=58.05it/s +epoch=56 global_step=21950 loss=1.85846 loss_avg=2.43381 acc=0.82812 acc_top1_avg=0.77040 acc_top5_avg=0.93895 lr=0.00100 gn=12.65487 time=58.87it/s +epoch=56 global_step=22000 loss=2.43796 loss_avg=2.42660 acc=0.77344 acc_top1_avg=0.77178 acc_top5_avg=0.93795 lr=0.00100 gn=12.63283 time=54.96it/s +epoch=56 global_step=22050 loss=1.76660 loss_avg=2.43688 acc=0.83594 acc_top1_avg=0.77161 acc_top5_avg=0.93826 lr=0.00100 gn=14.51010 time=51.88it/s +epoch=56 global_step=22100 loss=2.71849 loss_avg=2.44630 acc=0.75000 acc_top1_avg=0.77057 acc_top5_avg=0.93823 lr=0.00100 gn=13.16119 time=53.40it/s +epoch=56 global_step=22150 loss=2.72759 loss_avg=2.44736 acc=0.75000 acc_top1_avg=0.77073 acc_top5_avg=0.93864 lr=0.00100 gn=18.10756 time=55.55it/s +epoch=56 global_step=22200 loss=2.53293 loss_avg=2.45261 acc=0.75781 acc_top1_avg=0.77015 acc_top5_avg=0.93896 lr=0.00100 gn=12.32503 time=51.49it/s +epoch=56 global_step=22250 loss=1.96275 loss_avg=2.44476 acc=0.81250 acc_top1_avg=0.77094 acc_top5_avg=0.93863 lr=0.00100 gn=15.12511 time=56.68it/s +====================Eval==================== +epoch=56 global_step=22287 loss=0.74436 test_loss_avg=0.50144 acc=0.80469 test_acc_avg=0.85998 test_acc_top5_avg=0.98858 time=240.51it/s +epoch=56 global_step=22287 loss=0.31356 test_loss_avg=0.38706 acc=0.91406 test_acc_avg=0.89258 test_acc_top5_avg=0.99198 time=247.28it/s +epoch=56 global_step=22287 loss=0.42960 test_loss_avg=0.38570 acc=0.87500 test_acc_avg=0.89241 test_acc_top5_avg=0.99219 time=882.45it/s +curr_acc 0.8924 +BEST_ACC 0.8995 +curr_acc_top5 0.9922 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=1.90916 loss_avg=2.54402 acc=0.82031 acc_top1_avg=0.76022 acc_top5_avg=0.91767 lr=0.00100 gn=19.08920 time=48.61it/s +epoch=57 global_step=22350 loss=2.49983 loss_avg=2.44217 acc=0.75781 acc_top1_avg=0.77207 acc_top5_avg=0.93812 lr=0.00100 gn=13.89015 time=51.46it/s +epoch=57 global_step=22400 loss=2.03328 loss_avg=2.41525 acc=0.81250 acc_top1_avg=0.77441 acc_top5_avg=0.93764 lr=0.00100 gn=11.46060 time=58.47it/s +epoch=57 global_step=22450 loss=1.99014 loss_avg=2.41192 acc=0.82031 acc_top1_avg=0.77464 acc_top5_avg=0.93774 lr=0.00100 gn=12.89757 time=58.06it/s +epoch=57 global_step=22500 loss=2.09160 loss_avg=2.41978 acc=0.80469 acc_top1_avg=0.77402 acc_top5_avg=0.93812 lr=0.00100 gn=10.94780 time=52.05it/s +epoch=57 global_step=22550 loss=2.41709 loss_avg=2.42713 acc=0.78906 acc_top1_avg=0.77317 acc_top5_avg=0.93839 lr=0.00100 gn=17.99872 time=51.22it/s +epoch=57 global_step=22600 loss=2.32328 loss_avg=2.42315 acc=0.78125 acc_top1_avg=0.77326 acc_top5_avg=0.93880 lr=0.00100 gn=12.73372 time=57.00it/s +epoch=57 global_step=22650 loss=2.93143 loss_avg=2.42962 acc=0.71094 acc_top1_avg=0.77283 acc_top5_avg=0.93922 lr=0.00100 gn=13.17248 time=52.03it/s +====================Eval==================== +epoch=57 global_step=22678 loss=0.40174 test_loss_avg=0.43841 acc=0.85938 test_acc_avg=0.87550 test_acc_top5_avg=0.99136 time=224.73it/s +epoch=57 global_step=22678 loss=0.32153 test_loss_avg=0.35876 acc=0.87500 test_acc_avg=0.89686 test_acc_top5_avg=0.99328 time=895.07it/s +curr_acc 0.8969 +BEST_ACC 0.8995 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=2.55853 loss_avg=2.41300 acc=0.76562 acc_top1_avg=0.77557 acc_top5_avg=0.93395 lr=0.00100 gn=16.40748 time=57.98it/s +epoch=58 global_step=22750 loss=2.50868 loss_avg=2.43779 acc=0.78125 acc_top1_avg=0.77246 acc_top5_avg=0.93457 lr=0.00100 gn=14.80910 time=56.77it/s +epoch=58 global_step=22800 loss=2.10588 loss_avg=2.42514 acc=0.81250 acc_top1_avg=0.77305 acc_top5_avg=0.93654 lr=0.00100 gn=14.62298 time=52.64it/s +epoch=58 global_step=22850 loss=2.65274 loss_avg=2.42798 acc=0.75000 acc_top1_avg=0.77298 acc_top5_avg=0.93609 lr=0.00100 gn=11.57335 time=51.76it/s +epoch=58 global_step=22900 loss=2.06490 loss_avg=2.42012 acc=0.80469 acc_top1_avg=0.77361 acc_top5_avg=0.93722 lr=0.00100 gn=9.33753 time=55.25it/s +epoch=58 global_step=22950 loss=2.40339 loss_avg=2.41047 acc=0.75781 acc_top1_avg=0.77479 acc_top5_avg=0.93721 lr=0.00100 gn=15.98904 time=56.08it/s +epoch=58 global_step=23000 loss=2.14744 loss_avg=2.40166 acc=0.80469 acc_top1_avg=0.77589 acc_top5_avg=0.93765 lr=0.00100 gn=14.56754 time=55.55it/s +epoch=58 global_step=23050 loss=2.66768 loss_avg=2.41300 acc=0.75781 acc_top1_avg=0.77468 acc_top5_avg=0.93683 lr=0.00100 gn=14.97724 time=55.83it/s +====================Eval==================== +epoch=58 global_step=23069 loss=0.70668 test_loss_avg=0.33209 acc=0.81250 test_acc_avg=0.90148 test_acc_top5_avg=0.99436 time=219.69it/s +epoch=58 global_step=23069 loss=0.25306 test_loss_avg=0.40113 acc=0.94531 test_acc_avg=0.89040 test_acc_top5_avg=0.99322 time=226.30it/s +epoch=58 global_step=23069 loss=0.29871 test_loss_avg=0.37869 acc=0.87500 test_acc_avg=0.89508 test_acc_top5_avg=0.99377 time=888.25it/s +curr_acc 0.8951 +BEST_ACC 0.8995 +curr_acc_top5 0.9938 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=2.34494 loss_avg=2.49906 acc=0.78125 acc_top1_avg=0.76512 acc_top5_avg=0.93095 lr=0.00100 gn=15.42855 time=51.39it/s +epoch=59 global_step=23150 loss=2.74114 loss_avg=2.39497 acc=0.73438 acc_top1_avg=0.77566 acc_top5_avg=0.93769 lr=0.00100 gn=13.74571 time=52.09it/s +epoch=59 global_step=23200 loss=1.97047 loss_avg=2.39490 acc=0.82812 acc_top1_avg=0.77570 acc_top5_avg=0.93917 lr=0.00100 gn=14.92932 time=47.72it/s +epoch=59 global_step=23250 loss=2.19566 loss_avg=2.38879 acc=0.78906 acc_top1_avg=0.77642 acc_top5_avg=0.93810 lr=0.00100 gn=12.76229 time=58.87it/s +epoch=59 global_step=23300 loss=3.30647 loss_avg=2.39782 acc=0.67969 acc_top1_avg=0.77574 acc_top5_avg=0.93784 lr=0.00100 gn=19.21258 time=52.07it/s +epoch=59 global_step=23350 loss=2.45110 loss_avg=2.39916 acc=0.75781 acc_top1_avg=0.77600 acc_top5_avg=0.93872 lr=0.00100 gn=11.97386 time=52.50it/s +epoch=59 global_step=23400 loss=2.43934 loss_avg=2.39403 acc=0.77344 acc_top1_avg=0.77636 acc_top5_avg=0.93887 lr=0.00100 gn=11.87757 time=49.83it/s +epoch=59 global_step=23450 loss=2.38116 loss_avg=2.40622 acc=0.78906 acc_top1_avg=0.77485 acc_top5_avg=0.93848 lr=0.00100 gn=16.61052 time=55.94it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.18741 test_loss_avg=0.42775 acc=0.96094 test_acc_avg=0.87680 test_acc_top5_avg=0.99239 time=227.22it/s +epoch=59 global_step=23460 loss=0.32068 test_loss_avg=0.36459 acc=0.87500 test_acc_avg=0.89557 test_acc_top5_avg=0.99288 time=540.50it/s +curr_acc 0.8956 +BEST_ACC 0.8995 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=2.10186 loss_avg=2.31420 acc=0.79688 acc_top1_avg=0.78418 acc_top5_avg=0.94199 lr=0.00100 gn=13.08180 time=58.70it/s +epoch=60 global_step=23550 loss=2.20439 loss_avg=2.36670 acc=0.78906 acc_top1_avg=0.77856 acc_top5_avg=0.93915 lr=0.00100 gn=11.36435 time=57.72it/s +epoch=60 global_step=23600 loss=2.90918 loss_avg=2.35697 acc=0.73438 acc_top1_avg=0.77963 acc_top5_avg=0.93979 lr=0.00100 gn=15.15151 time=49.11it/s +epoch=60 global_step=23650 loss=2.79027 loss_avg=2.35322 acc=0.74219 acc_top1_avg=0.78047 acc_top5_avg=0.94058 lr=0.00100 gn=19.76640 time=52.66it/s +epoch=60 global_step=23700 loss=1.93361 loss_avg=2.36805 acc=0.82812 acc_top1_avg=0.77923 acc_top5_avg=0.94053 lr=0.00100 gn=17.25713 time=54.64it/s +epoch=60 global_step=23750 loss=3.00069 loss_avg=2.36885 acc=0.70312 acc_top1_avg=0.77891 acc_top5_avg=0.93995 lr=0.00100 gn=11.30194 time=51.44it/s +epoch=60 global_step=23800 loss=2.70556 loss_avg=2.38355 acc=0.75000 acc_top1_avg=0.77723 acc_top5_avg=0.93961 lr=0.00100 gn=18.70154 time=57.30it/s +epoch=60 global_step=23850 loss=2.10409 loss_avg=2.38428 acc=0.82031 acc_top1_avg=0.77752 acc_top5_avg=0.93944 lr=0.00100 gn=18.36291 time=58.32it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.08336 test_loss_avg=0.43662 acc=0.98438 test_acc_avg=0.86719 test_acc_top5_avg=0.99609 time=225.66it/s +epoch=60 global_step=23851 loss=0.17643 test_loss_avg=0.42051 acc=0.94531 test_acc_avg=0.88073 test_acc_top5_avg=0.99258 time=229.71it/s +epoch=60 global_step=23851 loss=0.24178 test_loss_avg=0.37356 acc=0.87500 test_acc_avg=0.89320 test_acc_top5_avg=0.99318 time=872.90it/s +curr_acc 0.8932 +BEST_ACC 0.8995 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=2.39301 loss_avg=2.39071 acc=0.77344 acc_top1_avg=0.77726 acc_top5_avg=0.94149 lr=0.00100 gn=17.81721 time=47.87it/s +epoch=61 global_step=23950 loss=2.14879 loss_avg=2.37059 acc=0.80469 acc_top1_avg=0.77936 acc_top5_avg=0.93932 lr=0.00100 gn=17.11940 time=52.57it/s +epoch=61 global_step=24000 loss=2.15949 loss_avg=2.37986 acc=0.79688 acc_top1_avg=0.77784 acc_top5_avg=0.93965 lr=0.00100 gn=13.32345 time=51.05it/s +epoch=61 global_step=24050 loss=2.49222 loss_avg=2.38672 acc=0.77344 acc_top1_avg=0.77705 acc_top5_avg=0.93809 lr=0.00100 gn=15.33576 time=58.03it/s +epoch=61 global_step=24100 loss=2.48703 loss_avg=2.38752 acc=0.75781 acc_top1_avg=0.77698 acc_top5_avg=0.93816 lr=0.00100 gn=15.37763 time=52.32it/s +epoch=61 global_step=24150 loss=2.70698 loss_avg=2.39572 acc=0.75000 acc_top1_avg=0.77657 acc_top5_avg=0.93899 lr=0.00100 gn=19.13131 time=57.45it/s +epoch=61 global_step=24200 loss=2.33730 loss_avg=2.38044 acc=0.78125 acc_top1_avg=0.77800 acc_top5_avg=0.93927 lr=0.00100 gn=21.43465 time=55.38it/s +====================Eval==================== +epoch=61 global_step=24242 loss=0.83255 test_loss_avg=0.51022 acc=0.81250 test_acc_avg=0.85685 test_acc_top5_avg=0.99017 time=232.20it/s +epoch=61 global_step=24242 loss=0.08703 test_loss_avg=0.36645 acc=1.00000 test_acc_avg=0.89725 test_acc_top5_avg=0.99248 time=889.38it/s +curr_acc 0.8973 +BEST_ACC 0.8995 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=1.94472 loss_avg=2.47459 acc=0.82812 acc_top1_avg=0.76855 acc_top5_avg=0.95020 lr=0.00100 gn=13.14730 time=58.89it/s +epoch=62 global_step=24300 loss=1.94654 loss_avg=2.35291 acc=0.82031 acc_top1_avg=0.78138 acc_top5_avg=0.93871 lr=0.00100 gn=16.60867 time=52.63it/s +epoch=62 global_step=24350 loss=2.42500 loss_avg=2.31356 acc=0.77344 acc_top1_avg=0.78465 acc_top5_avg=0.94076 lr=0.00100 gn=17.08371 time=53.48it/s +epoch=62 global_step=24400 loss=2.21388 loss_avg=2.32508 acc=0.78125 acc_top1_avg=0.78352 acc_top5_avg=0.94066 lr=0.00100 gn=15.54302 time=58.02it/s +epoch=62 global_step=24450 loss=2.04982 loss_avg=2.34413 acc=0.81250 acc_top1_avg=0.78204 acc_top5_avg=0.93972 lr=0.00100 gn=13.17123 time=56.08it/s +epoch=62 global_step=24500 loss=2.75314 loss_avg=2.35394 acc=0.73438 acc_top1_avg=0.78104 acc_top5_avg=0.93889 lr=0.00100 gn=13.33051 time=50.03it/s +epoch=62 global_step=24550 loss=1.98952 loss_avg=2.35447 acc=0.82812 acc_top1_avg=0.78130 acc_top5_avg=0.93877 lr=0.00100 gn=15.40167 time=56.81it/s +epoch=62 global_step=24600 loss=2.65637 loss_avg=2.36127 acc=0.75781 acc_top1_avg=0.78060 acc_top5_avg=0.93918 lr=0.00100 gn=19.08041 time=50.16it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.39946 test_loss_avg=0.37053 acc=0.85156 test_acc_avg=0.87500 test_acc_top5_avg=0.98828 time=240.86it/s +epoch=62 global_step=24633 loss=0.22764 test_loss_avg=0.43510 acc=0.93750 test_acc_avg=0.87861 test_acc_top5_avg=0.99234 time=241.12it/s +epoch=62 global_step=24633 loss=0.17697 test_loss_avg=0.36387 acc=0.93750 test_acc_avg=0.89784 test_acc_top5_avg=0.99337 time=860.02it/s +curr_acc 0.8978 +BEST_ACC 0.8995 +curr_acc_top5 0.9934 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=2.43584 loss_avg=2.44318 acc=0.78125 acc_top1_avg=0.77206 acc_top5_avg=0.93750 lr=0.00100 gn=16.07024 time=52.46it/s +epoch=63 global_step=24700 loss=2.02362 loss_avg=2.37260 acc=0.80469 acc_top1_avg=0.77892 acc_top5_avg=0.93808 lr=0.00100 gn=12.18587 time=58.84it/s +epoch=63 global_step=24750 loss=2.05458 loss_avg=2.34435 acc=0.80469 acc_top1_avg=0.78205 acc_top5_avg=0.93783 lr=0.00100 gn=14.20569 time=52.08it/s +epoch=63 global_step=24800 loss=2.46443 loss_avg=2.34831 acc=0.76562 acc_top1_avg=0.78162 acc_top5_avg=0.93975 lr=0.00100 gn=15.83887 time=53.59it/s +epoch=63 global_step=24850 loss=2.44850 loss_avg=2.35146 acc=0.76562 acc_top1_avg=0.78067 acc_top5_avg=0.93959 lr=0.00100 gn=15.99602 time=57.91it/s +epoch=63 global_step=24900 loss=2.68229 loss_avg=2.35245 acc=0.75000 acc_top1_avg=0.78064 acc_top5_avg=0.93955 lr=0.00100 gn=15.56445 time=48.49it/s +epoch=63 global_step=24950 loss=1.89910 loss_avg=2.34498 acc=0.83594 acc_top1_avg=0.78160 acc_top5_avg=0.94011 lr=0.00100 gn=15.76255 time=57.33it/s +epoch=63 global_step=25000 loss=2.49086 loss_avg=2.34788 acc=0.78125 acc_top1_avg=0.78159 acc_top5_avg=0.93963 lr=0.00100 gn=19.07118 time=57.80it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.76765 test_loss_avg=0.41510 acc=0.81250 test_acc_avg=0.87636 test_acc_top5_avg=0.99117 time=230.81it/s +epoch=63 global_step=25024 loss=0.38463 test_loss_avg=0.38544 acc=0.89062 test_acc_avg=0.88774 test_acc_top5_avg=0.99240 time=224.68it/s +epoch=63 global_step=25024 loss=0.22401 test_loss_avg=0.37472 acc=0.87500 test_acc_avg=0.88973 test_acc_top5_avg=0.99268 time=831.38it/s +curr_acc 0.8897 +BEST_ACC 0.8995 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=2.12223 loss_avg=2.26673 acc=0.80469 acc_top1_avg=0.78966 acc_top5_avg=0.94681 lr=0.00100 gn=15.06429 time=55.07it/s +epoch=64 global_step=25100 loss=2.59700 loss_avg=2.31549 acc=0.74219 acc_top1_avg=0.78423 acc_top5_avg=0.94038 lr=0.00100 gn=13.52067 time=56.45it/s +epoch=64 global_step=25150 loss=2.88937 loss_avg=2.31880 acc=0.73438 acc_top1_avg=0.78491 acc_top5_avg=0.94079 lr=0.00100 gn=18.97137 time=46.31it/s +epoch=64 global_step=25200 loss=2.14560 loss_avg=2.33315 acc=0.81250 acc_top1_avg=0.78378 acc_top5_avg=0.94043 lr=0.00100 gn=19.88978 time=53.28it/s +epoch=64 global_step=25250 loss=2.58724 loss_avg=2.34989 acc=0.76562 acc_top1_avg=0.78163 acc_top5_avg=0.93989 lr=0.00100 gn=19.24023 time=53.34it/s +epoch=64 global_step=25300 loss=2.11882 loss_avg=2.34868 acc=0.80469 acc_top1_avg=0.78159 acc_top5_avg=0.93948 lr=0.00100 gn=20.72541 time=49.69it/s +epoch=64 global_step=25350 loss=2.59616 loss_avg=2.34476 acc=0.75781 acc_top1_avg=0.78187 acc_top5_avg=0.94021 lr=0.00100 gn=21.23446 time=53.47it/s +epoch=64 global_step=25400 loss=2.43655 loss_avg=2.35025 acc=0.77344 acc_top1_avg=0.78150 acc_top5_avg=0.94078 lr=0.00100 gn=16.65472 time=52.08it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.92366 test_loss_avg=0.45007 acc=0.75781 test_acc_avg=0.87553 test_acc_top5_avg=0.99112 time=222.11it/s +epoch=64 global_step=25415 loss=0.12979 test_loss_avg=0.36823 acc=0.93750 test_acc_avg=0.89715 test_acc_top5_avg=0.99298 time=892.79it/s +curr_acc 0.8972 +BEST_ACC 0.8995 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=2.52887 loss_avg=2.36271 acc=0.75781 acc_top1_avg=0.77879 acc_top5_avg=0.94085 lr=0.00100 gn=16.31861 time=54.93it/s +epoch=65 global_step=25500 loss=2.42144 loss_avg=2.31094 acc=0.78125 acc_top1_avg=0.78520 acc_top5_avg=0.93869 lr=0.00100 gn=17.64428 time=47.85it/s +epoch=65 global_step=25550 loss=2.51059 loss_avg=2.31918 acc=0.75000 acc_top1_avg=0.78426 acc_top5_avg=0.93883 lr=0.00100 gn=8.48328 time=53.44it/s +epoch=65 global_step=25600 loss=2.53246 loss_avg=2.31807 acc=0.75781 acc_top1_avg=0.78471 acc_top5_avg=0.93902 lr=0.00100 gn=13.87064 time=58.30it/s +epoch=65 global_step=25650 loss=2.19014 loss_avg=2.31044 acc=0.78906 acc_top1_avg=0.78600 acc_top5_avg=0.93939 lr=0.00100 gn=21.67444 time=56.48it/s +epoch=65 global_step=25700 loss=2.34428 loss_avg=2.32159 acc=0.78125 acc_top1_avg=0.78495 acc_top5_avg=0.93893 lr=0.00100 gn=17.58932 time=58.64it/s +epoch=65 global_step=25750 loss=2.15079 loss_avg=2.32643 acc=0.80469 acc_top1_avg=0.78414 acc_top5_avg=0.93892 lr=0.00100 gn=19.75634 time=49.88it/s +epoch=65 global_step=25800 loss=1.90383 loss_avg=2.33757 acc=0.82812 acc_top1_avg=0.78285 acc_top5_avg=0.93902 lr=0.00100 gn=23.65617 time=56.89it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.22696 test_loss_avg=0.38036 acc=0.92969 test_acc_avg=0.88542 test_acc_top5_avg=0.99687 time=225.13it/s +epoch=65 global_step=25806 loss=0.12091 test_loss_avg=0.40906 acc=0.96094 test_acc_avg=0.88269 test_acc_top5_avg=0.99135 time=224.45it/s +epoch=65 global_step=25806 loss=0.07872 test_loss_avg=0.36341 acc=1.00000 test_acc_avg=0.89567 test_acc_top5_avg=0.99248 time=839.70it/s +curr_acc 0.8957 +BEST_ACC 0.8995 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=2.06901 loss_avg=2.26956 acc=0.82031 acc_top1_avg=0.79048 acc_top5_avg=0.93945 lr=0.00100 gn=15.73005 time=55.94it/s +epoch=66 global_step=25900 loss=2.58193 loss_avg=2.32895 acc=0.75781 acc_top1_avg=0.78383 acc_top5_avg=0.94074 lr=0.00100 gn=16.58734 time=56.98it/s +epoch=66 global_step=25950 loss=3.01536 loss_avg=2.30014 acc=0.70312 acc_top1_avg=0.78684 acc_top5_avg=0.94265 lr=0.00100 gn=21.48807 time=57.56it/s +epoch=66 global_step=26000 loss=2.95962 loss_avg=2.33402 acc=0.71094 acc_top1_avg=0.78326 acc_top5_avg=0.94133 lr=0.00100 gn=15.77882 time=51.56it/s +epoch=66 global_step=26050 loss=1.74517 loss_avg=2.33152 acc=0.83594 acc_top1_avg=0.78372 acc_top5_avg=0.94054 lr=0.00100 gn=21.00025 time=55.01it/s +epoch=66 global_step=26100 loss=2.39756 loss_avg=2.32407 acc=0.78125 acc_top1_avg=0.78425 acc_top5_avg=0.94042 lr=0.00100 gn=21.22469 time=50.09it/s +epoch=66 global_step=26150 loss=2.29722 loss_avg=2.33109 acc=0.78906 acc_top1_avg=0.78327 acc_top5_avg=0.94011 lr=0.00100 gn=15.96375 time=50.95it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.26928 test_loss_avg=0.45887 acc=0.89844 test_acc_avg=0.87196 test_acc_top5_avg=0.99262 time=237.11it/s +epoch=66 global_step=26197 loss=0.21091 test_loss_avg=0.37534 acc=0.93750 test_acc_avg=0.89468 test_acc_top5_avg=0.99288 time=849.74it/s +curr_acc 0.8947 +BEST_ACC 0.8995 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=2.06238 loss_avg=2.19890 acc=0.82031 acc_top1_avg=0.79688 acc_top5_avg=0.94010 lr=0.00100 gn=21.17826 time=34.57it/s +epoch=67 global_step=26250 loss=2.26748 loss_avg=2.31935 acc=0.78906 acc_top1_avg=0.78508 acc_top5_avg=0.93942 lr=0.00100 gn=20.15886 time=51.97it/s +epoch=67 global_step=26300 loss=2.07684 loss_avg=2.32425 acc=0.82031 acc_top1_avg=0.78383 acc_top5_avg=0.94076 lr=0.00100 gn=22.47177 time=54.18it/s +epoch=67 global_step=26350 loss=2.07961 loss_avg=2.29761 acc=0.79688 acc_top1_avg=0.78641 acc_top5_avg=0.94296 lr=0.00100 gn=14.13663 time=55.91it/s +epoch=67 global_step=26400 loss=2.48053 loss_avg=2.30820 acc=0.75781 acc_top1_avg=0.78602 acc_top5_avg=0.94189 lr=0.00100 gn=15.69653 time=52.35it/s +epoch=67 global_step=26450 loss=2.10108 loss_avg=2.30414 acc=0.80469 acc_top1_avg=0.78668 acc_top5_avg=0.94124 lr=0.00100 gn=18.84189 time=49.07it/s +epoch=67 global_step=26500 loss=2.22499 loss_avg=2.31722 acc=0.80469 acc_top1_avg=0.78525 acc_top5_avg=0.94114 lr=0.00100 gn=18.14967 time=48.19it/s +epoch=67 global_step=26550 loss=2.44863 loss_avg=2.31465 acc=0.76562 acc_top1_avg=0.78534 acc_top5_avg=0.94111 lr=0.00100 gn=14.77603 time=49.39it/s +====================Eval==================== +epoch=67 global_step=26588 loss=0.58130 test_loss_avg=0.61620 acc=0.83594 test_acc_avg=0.81920 test_acc_top5_avg=0.99330 time=213.64it/s +epoch=67 global_step=26588 loss=0.18837 test_loss_avg=0.44676 acc=0.94531 test_acc_avg=0.87212 test_acc_top5_avg=0.99191 time=223.15it/s +epoch=67 global_step=26588 loss=0.11878 test_loss_avg=0.36946 acc=0.93750 test_acc_avg=0.89409 test_acc_top5_avg=0.99298 time=765.80it/s +curr_acc 0.8941 +BEST_ACC 0.8995 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=2.20683 loss_avg=2.19673 acc=0.79688 acc_top1_avg=0.79557 acc_top5_avg=0.94271 lr=0.00100 gn=15.91607 time=52.85it/s +epoch=68 global_step=26650 loss=2.52992 loss_avg=2.31538 acc=0.76562 acc_top1_avg=0.78427 acc_top5_avg=0.93851 lr=0.00100 gn=23.47239 time=50.99it/s +epoch=68 global_step=26700 loss=1.99300 loss_avg=2.33327 acc=0.81250 acc_top1_avg=0.78230 acc_top5_avg=0.93785 lr=0.00100 gn=13.58979 time=48.87it/s +epoch=68 global_step=26750 loss=2.21515 loss_avg=2.30588 acc=0.78906 acc_top1_avg=0.78545 acc_top5_avg=0.93904 lr=0.00100 gn=16.70315 time=49.07it/s +epoch=68 global_step=26800 loss=2.34721 loss_avg=2.30491 acc=0.78125 acc_top1_avg=0.78571 acc_top5_avg=0.94004 lr=0.00100 gn=17.15082 time=58.35it/s +epoch=68 global_step=26850 loss=2.58461 loss_avg=2.28685 acc=0.75781 acc_top1_avg=0.78778 acc_top5_avg=0.94078 lr=0.00100 gn=18.38035 time=53.96it/s +epoch=68 global_step=26900 loss=1.84042 loss_avg=2.29504 acc=0.83594 acc_top1_avg=0.78671 acc_top5_avg=0.94033 lr=0.00100 gn=13.13888 time=56.12it/s +epoch=68 global_step=26950 loss=2.55837 loss_avg=2.29551 acc=0.75781 acc_top1_avg=0.78665 acc_top5_avg=0.94063 lr=0.00100 gn=22.99039 time=45.70it/s +====================Eval==================== +epoch=68 global_step=26979 loss=0.47220 test_loss_avg=0.37643 acc=0.86719 test_acc_avg=0.89369 test_acc_top5_avg=0.99330 time=230.90it/s +epoch=68 global_step=26979 loss=0.22408 test_loss_avg=0.36443 acc=0.92969 test_acc_avg=0.89924 test_acc_top5_avg=0.99239 time=241.40it/s +epoch=68 global_step=26979 loss=0.36380 test_loss_avg=0.36442 acc=0.87500 test_acc_avg=0.89893 test_acc_top5_avg=0.99248 time=866.95it/s +curr_acc 0.8989 +BEST_ACC 0.8995 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=2.26050 loss_avg=2.21209 acc=0.78125 acc_top1_avg=0.79390 acc_top5_avg=0.94271 lr=0.00100 gn=16.02992 time=55.27it/s +epoch=69 global_step=27050 loss=1.94073 loss_avg=2.29134 acc=0.82031 acc_top1_avg=0.78697 acc_top5_avg=0.94179 lr=0.00100 gn=17.53227 time=58.76it/s +epoch=69 global_step=27100 loss=2.34283 loss_avg=2.34047 acc=0.78906 acc_top1_avg=0.78196 acc_top5_avg=0.93982 lr=0.00100 gn=18.87888 time=50.86it/s +epoch=69 global_step=27150 loss=1.91716 loss_avg=2.31689 acc=0.83594 acc_top1_avg=0.78486 acc_top5_avg=0.94029 lr=0.00100 gn=19.19847 time=51.72it/s +epoch=69 global_step=27200 loss=2.21694 loss_avg=2.31577 acc=0.79688 acc_top1_avg=0.78521 acc_top5_avg=0.94026 lr=0.00100 gn=16.29894 time=56.77it/s +epoch=69 global_step=27250 loss=2.25967 loss_avg=2.33045 acc=0.78906 acc_top1_avg=0.78370 acc_top5_avg=0.93966 lr=0.00100 gn=20.65254 time=56.24it/s +epoch=69 global_step=27300 loss=2.17073 loss_avg=2.32612 acc=0.80469 acc_top1_avg=0.78422 acc_top5_avg=0.93935 lr=0.00100 gn=21.75018 time=55.36it/s +epoch=69 global_step=27350 loss=2.24013 loss_avg=2.31396 acc=0.79688 acc_top1_avg=0.78548 acc_top5_avg=0.94013 lr=0.00100 gn=19.19589 time=52.38it/s +====================Eval==================== +epoch=69 global_step=27370 loss=0.28732 test_loss_avg=0.42412 acc=0.89844 test_acc_avg=0.87994 test_acc_top5_avg=0.99251 time=219.68it/s +epoch=69 global_step=27370 loss=0.34974 test_loss_avg=0.37108 acc=0.87500 test_acc_avg=0.89330 test_acc_top5_avg=0.99258 time=865.34it/s +curr_acc 0.8933 +BEST_ACC 0.8995 +curr_acc_top5 0.9926 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=1.88909 loss_avg=2.18110 acc=0.81250 acc_top1_avg=0.79974 acc_top5_avg=0.93958 lr=0.00100 gn=11.40960 time=56.82it/s +epoch=70 global_step=27450 loss=2.42646 loss_avg=2.24332 acc=0.77344 acc_top1_avg=0.79199 acc_top5_avg=0.94199 lr=0.00100 gn=16.45424 time=53.51it/s +epoch=70 global_step=27500 loss=2.99628 loss_avg=2.29735 acc=0.71094 acc_top1_avg=0.78558 acc_top5_avg=0.94069 lr=0.00100 gn=21.67980 time=55.19it/s +epoch=70 global_step=27550 loss=2.29849 loss_avg=2.29307 acc=0.78125 acc_top1_avg=0.78633 acc_top5_avg=0.94115 lr=0.00100 gn=19.07953 time=58.01it/s +epoch=70 global_step=27600 loss=2.14949 loss_avg=2.28704 acc=0.79688 acc_top1_avg=0.78767 acc_top5_avg=0.94090 lr=0.00100 gn=18.05137 time=58.34it/s +epoch=70 global_step=27650 loss=2.43564 loss_avg=2.27621 acc=0.77344 acc_top1_avg=0.78909 acc_top5_avg=0.94096 lr=0.00100 gn=14.62479 time=51.90it/s +epoch=70 global_step=27700 loss=2.44005 loss_avg=2.29705 acc=0.77344 acc_top1_avg=0.78684 acc_top5_avg=0.94006 lr=0.00100 gn=20.54247 time=55.77it/s +epoch=70 global_step=27750 loss=2.68312 loss_avg=2.29571 acc=0.75000 acc_top1_avg=0.78707 acc_top5_avg=0.94040 lr=0.00100 gn=24.70563 time=56.56it/s +====================Eval==================== +epoch=70 global_step=27761 loss=0.60277 test_loss_avg=0.29943 acc=0.79688 test_acc_avg=0.90703 test_acc_top5_avg=0.99531 time=225.15it/s +epoch=70 global_step=27761 loss=0.14801 test_loss_avg=0.36987 acc=0.94531 test_acc_avg=0.89208 test_acc_top5_avg=0.99163 time=240.61it/s +epoch=70 global_step=27761 loss=0.30084 test_loss_avg=0.36555 acc=0.87500 test_acc_avg=0.89211 test_acc_top5_avg=0.99189 time=556.42it/s +curr_acc 0.8921 +BEST_ACC 0.8995 +curr_acc_top5 0.9919 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=2.15579 loss_avg=2.33363 acc=0.79688 acc_top1_avg=0.78305 acc_top5_avg=0.93810 lr=0.00100 gn=20.23643 time=56.82it/s +epoch=71 global_step=27850 loss=2.20919 loss_avg=2.33615 acc=0.79688 acc_top1_avg=0.78248 acc_top5_avg=0.93706 lr=0.00100 gn=20.02352 time=48.76it/s +epoch=71 global_step=27900 loss=2.28309 loss_avg=2.30072 acc=0.78125 acc_top1_avg=0.78715 acc_top5_avg=0.93997 lr=0.00100 gn=8.84170 time=52.56it/s +epoch=71 global_step=27950 loss=2.37624 loss_avg=2.30060 acc=0.78906 acc_top1_avg=0.78720 acc_top5_avg=0.94023 lr=0.00100 gn=17.24028 time=58.14it/s +epoch=71 global_step=28000 loss=2.48169 loss_avg=2.29899 acc=0.77344 acc_top1_avg=0.78740 acc_top5_avg=0.94162 lr=0.00100 gn=17.25357 time=50.30it/s +epoch=71 global_step=28050 loss=2.64509 loss_avg=2.28987 acc=0.75781 acc_top1_avg=0.78855 acc_top5_avg=0.94137 lr=0.00100 gn=18.23989 time=52.84it/s +epoch=71 global_step=28100 loss=1.97160 loss_avg=2.28972 acc=0.82812 acc_top1_avg=0.78853 acc_top5_avg=0.94158 lr=0.00100 gn=23.49597 time=58.61it/s +epoch=71 global_step=28150 loss=2.34402 loss_avg=2.28814 acc=0.77344 acc_top1_avg=0.78864 acc_top5_avg=0.94142 lr=0.00100 gn=16.27626 time=58.97it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.72526 test_loss_avg=0.43645 acc=0.83594 test_acc_avg=0.87329 test_acc_top5_avg=0.99162 time=223.36it/s +epoch=71 global_step=28152 loss=0.12524 test_loss_avg=0.35283 acc=0.93750 test_acc_avg=0.89775 test_acc_top5_avg=0.99318 time=540.57it/s +curr_acc 0.8977 +BEST_ACC 0.8995 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=1.41790 loss_avg=2.23614 acc=0.88281 acc_top1_avg=0.79395 acc_top5_avg=0.94352 lr=0.00100 gn=22.34321 time=52.14it/s +epoch=72 global_step=28250 loss=2.36145 loss_avg=2.24150 acc=0.78906 acc_top1_avg=0.79305 acc_top5_avg=0.93925 lr=0.00100 gn=20.50704 time=49.99it/s +epoch=72 global_step=28300 loss=2.16764 loss_avg=2.27109 acc=0.78906 acc_top1_avg=0.79049 acc_top5_avg=0.93993 lr=0.00100 gn=17.48477 time=51.32it/s +epoch=72 global_step=28350 loss=2.32987 loss_avg=2.27343 acc=0.78906 acc_top1_avg=0.79044 acc_top5_avg=0.94010 lr=0.00100 gn=19.48610 time=51.24it/s +epoch=72 global_step=28400 loss=1.83541 loss_avg=2.26867 acc=0.82031 acc_top1_avg=0.79083 acc_top5_avg=0.94068 lr=0.00100 gn=19.49494 time=58.51it/s +epoch=72 global_step=28450 loss=2.32281 loss_avg=2.27644 acc=0.80469 acc_top1_avg=0.78993 acc_top5_avg=0.93999 lr=0.00100 gn=22.66306 time=49.95it/s +epoch=72 global_step=28500 loss=2.61072 loss_avg=2.28614 acc=0.75781 acc_top1_avg=0.78873 acc_top5_avg=0.93959 lr=0.00100 gn=21.68985 time=58.45it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.08194 test_loss_avg=0.34533 acc=0.97656 test_acc_avg=0.89583 test_acc_top5_avg=0.99740 time=224.43it/s +epoch=72 global_step=28543 loss=0.30165 test_loss_avg=0.44283 acc=0.91406 test_acc_avg=0.87601 test_acc_top5_avg=0.98841 time=222.47it/s +epoch=72 global_step=28543 loss=0.19747 test_loss_avg=0.39156 acc=0.93750 test_acc_avg=0.88993 test_acc_top5_avg=0.99011 time=891.46it/s +curr_acc 0.8899 +BEST_ACC 0.8995 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=1.71078 loss_avg=2.14214 acc=0.85156 acc_top1_avg=0.80692 acc_top5_avg=0.94196 lr=0.00100 gn=16.65470 time=48.08it/s +epoch=73 global_step=28600 loss=2.42127 loss_avg=2.20450 acc=0.77344 acc_top1_avg=0.79715 acc_top5_avg=0.94339 lr=0.00100 gn=21.40802 time=57.80it/s +epoch=73 global_step=28650 loss=2.65595 loss_avg=2.22051 acc=0.75000 acc_top1_avg=0.79541 acc_top5_avg=0.94334 lr=0.00100 gn=17.75679 time=53.39it/s +epoch=73 global_step=28700 loss=2.22218 loss_avg=2.21301 acc=0.79688 acc_top1_avg=0.79623 acc_top5_avg=0.94263 lr=0.00100 gn=17.00889 time=52.51it/s +epoch=73 global_step=28750 loss=3.12141 loss_avg=2.23223 acc=0.70312 acc_top1_avg=0.79431 acc_top5_avg=0.94192 lr=0.00100 gn=15.34861 time=50.29it/s +epoch=73 global_step=28800 loss=1.54662 loss_avg=2.24476 acc=0.85156 acc_top1_avg=0.79329 acc_top5_avg=0.94109 lr=0.00100 gn=20.42037 time=58.99it/s +epoch=73 global_step=28850 loss=2.08657 loss_avg=2.24962 acc=0.81250 acc_top1_avg=0.79263 acc_top5_avg=0.94073 lr=0.00100 gn=19.83191 time=57.12it/s +epoch=73 global_step=28900 loss=2.63247 loss_avg=2.26615 acc=0.75000 acc_top1_avg=0.79092 acc_top5_avg=0.94013 lr=0.00100 gn=14.57463 time=51.90it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.19273 test_loss_avg=0.59058 acc=0.94531 test_acc_avg=0.83594 test_acc_top5_avg=0.98911 time=219.44it/s +epoch=73 global_step=28934 loss=0.22571 test_loss_avg=0.40708 acc=0.93750 test_acc_avg=0.88588 test_acc_top5_avg=0.99130 time=894.50it/s +curr_acc 0.8859 +BEST_ACC 0.8995 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=2.55465 loss_avg=2.24263 acc=0.75781 acc_top1_avg=0.79199 acc_top5_avg=0.93408 lr=0.00100 gn=13.16703 time=52.64it/s +epoch=74 global_step=29000 loss=1.91457 loss_avg=2.29215 acc=0.82812 acc_top1_avg=0.78717 acc_top5_avg=0.93821 lr=0.00100 gn=20.73746 time=56.93it/s +epoch=74 global_step=29050 loss=2.10597 loss_avg=2.29578 acc=0.80469 acc_top1_avg=0.78724 acc_top5_avg=0.93898 lr=0.00100 gn=18.54245 time=57.33it/s +epoch=74 global_step=29100 loss=2.25728 loss_avg=2.29197 acc=0.78906 acc_top1_avg=0.78765 acc_top5_avg=0.93882 lr=0.00100 gn=16.79012 time=56.30it/s +epoch=74 global_step=29150 loss=2.13008 loss_avg=2.28297 acc=0.80469 acc_top1_avg=0.78870 acc_top5_avg=0.93992 lr=0.00100 gn=18.48455 time=53.49it/s +epoch=74 global_step=29200 loss=2.56142 loss_avg=2.27543 acc=0.76562 acc_top1_avg=0.78986 acc_top5_avg=0.93967 lr=0.00100 gn=24.03158 time=56.39it/s +epoch=74 global_step=29250 loss=1.79873 loss_avg=2.26802 acc=0.84375 acc_top1_avg=0.79035 acc_top5_avg=0.93973 lr=0.00100 gn=22.25972 time=57.56it/s +epoch=74 global_step=29300 loss=2.13395 loss_avg=2.27068 acc=0.81250 acc_top1_avg=0.79013 acc_top5_avg=0.94013 lr=0.00100 gn=20.31848 time=56.58it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.57132 test_loss_avg=0.60374 acc=0.82031 test_acc_avg=0.82422 test_acc_top5_avg=0.99023 time=233.09it/s +epoch=74 global_step=29325 loss=0.17516 test_loss_avg=0.44160 acc=0.96875 test_acc_avg=0.87109 test_acc_top5_avg=0.98944 time=227.42it/s +epoch=74 global_step=29325 loss=0.50118 test_loss_avg=0.37231 acc=0.87500 test_acc_avg=0.89092 test_acc_top5_avg=0.99159 time=869.47it/s +curr_acc 0.8909 +BEST_ACC 0.8995 +curr_acc_top5 0.9916 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=1.81237 loss_avg=2.21343 acc=0.83594 acc_top1_avg=0.79438 acc_top5_avg=0.94594 lr=0.00100 gn=23.37507 time=28.81it/s +epoch=75 global_step=29400 loss=2.43438 loss_avg=2.31192 acc=0.77344 acc_top1_avg=0.78406 acc_top5_avg=0.93833 lr=0.00100 gn=15.09232 time=55.03it/s +epoch=75 global_step=29450 loss=1.96376 loss_avg=2.31186 acc=0.81250 acc_top1_avg=0.78450 acc_top5_avg=0.93769 lr=0.00100 gn=15.43830 time=45.10it/s +epoch=75 global_step=29500 loss=2.23604 loss_avg=2.28464 acc=0.78906 acc_top1_avg=0.78772 acc_top5_avg=0.93875 lr=0.00100 gn=17.22529 time=54.74it/s +epoch=75 global_step=29550 loss=2.09152 loss_avg=2.25391 acc=0.82031 acc_top1_avg=0.79135 acc_top5_avg=0.93906 lr=0.00100 gn=24.10935 time=52.75it/s +epoch=75 global_step=29600 loss=1.93328 loss_avg=2.24779 acc=0.83594 acc_top1_avg=0.79207 acc_top5_avg=0.93946 lr=0.00100 gn=23.92778 time=58.20it/s +epoch=75 global_step=29650 loss=2.18533 loss_avg=2.25504 acc=0.78906 acc_top1_avg=0.79166 acc_top5_avg=0.93983 lr=0.00100 gn=13.54719 time=53.34it/s +epoch=75 global_step=29700 loss=2.47399 loss_avg=2.25981 acc=0.76562 acc_top1_avg=0.79127 acc_top5_avg=0.94025 lr=0.00100 gn=20.33556 time=57.44it/s +====================Eval==================== +epoch=75 global_step=29716 loss=0.59852 test_loss_avg=0.46106 acc=0.82031 test_acc_avg=0.86969 test_acc_top5_avg=0.98844 time=227.54it/s +epoch=75 global_step=29716 loss=0.24778 test_loss_avg=0.37741 acc=0.92969 test_acc_avg=0.89302 test_acc_top5_avg=0.99146 time=233.69it/s +epoch=75 global_step=29716 loss=0.16687 test_loss_avg=0.37091 acc=0.87500 test_acc_avg=0.89320 test_acc_top5_avg=0.99179 time=819.84it/s +curr_acc 0.8932 +BEST_ACC 0.8995 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=2.50690 loss_avg=2.21473 acc=0.76562 acc_top1_avg=0.79550 acc_top5_avg=0.94646 lr=0.00100 gn=15.02753 time=51.54it/s +epoch=76 global_step=29800 loss=2.50077 loss_avg=2.24711 acc=0.77344 acc_top1_avg=0.79297 acc_top5_avg=0.94327 lr=0.00100 gn=20.94504 time=53.83it/s +epoch=76 global_step=29850 loss=2.58303 loss_avg=2.26759 acc=0.75781 acc_top1_avg=0.79075 acc_top5_avg=0.94146 lr=0.00100 gn=17.23425 time=52.92it/s +epoch=76 global_step=29900 loss=2.60382 loss_avg=2.26429 acc=0.75781 acc_top1_avg=0.79144 acc_top5_avg=0.94119 lr=0.00100 gn=17.22209 time=52.13it/s +epoch=76 global_step=29950 loss=2.23414 loss_avg=2.25510 acc=0.79688 acc_top1_avg=0.79247 acc_top5_avg=0.94241 lr=0.00100 gn=26.37004 time=54.67it/s +epoch=76 global_step=30000 loss=2.20652 loss_avg=2.25442 acc=0.80469 acc_top1_avg=0.79261 acc_top5_avg=0.94165 lr=0.00100 gn=16.56762 time=56.44it/s +epoch=76 global_step=30050 loss=2.54564 loss_avg=2.25974 acc=0.77344 acc_top1_avg=0.79196 acc_top5_avg=0.94157 lr=0.00100 gn=23.35597 time=54.60it/s +epoch=76 global_step=30100 loss=1.92819 loss_avg=2.26336 acc=0.83594 acc_top1_avg=0.79163 acc_top5_avg=0.94194 lr=0.00100 gn=23.84239 time=38.54it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.57257 test_loss_avg=0.43073 acc=0.86719 test_acc_avg=0.87925 test_acc_top5_avg=0.99015 time=205.22it/s +epoch=76 global_step=30107 loss=0.15910 test_loss_avg=0.37044 acc=0.93750 test_acc_avg=0.89399 test_acc_top5_avg=0.99150 time=817.92it/s +curr_acc 0.8940 +BEST_ACC 0.8995 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=2.01322 loss_avg=2.12949 acc=0.84375 acc_top1_avg=0.80541 acc_top5_avg=0.94967 lr=0.00100 gn=24.89617 time=49.63it/s +epoch=77 global_step=30200 loss=2.07600 loss_avg=2.18325 acc=0.82031 acc_top1_avg=0.79956 acc_top5_avg=0.94481 lr=0.00100 gn=24.17028 time=55.37it/s +epoch=77 global_step=30250 loss=2.18201 loss_avg=2.22630 acc=0.78906 acc_top1_avg=0.79502 acc_top5_avg=0.94313 lr=0.00100 gn=26.39844 time=58.17it/s +epoch=77 global_step=30300 loss=2.73775 loss_avg=2.23890 acc=0.75000 acc_top1_avg=0.79416 acc_top5_avg=0.94171 lr=0.00100 gn=21.59852 time=49.55it/s +epoch=77 global_step=30350 loss=2.43392 loss_avg=2.22683 acc=0.78906 acc_top1_avg=0.79524 acc_top5_avg=0.94274 lr=0.00100 gn=25.70934 time=52.80it/s +epoch=77 global_step=30400 loss=2.67004 loss_avg=2.22632 acc=0.75000 acc_top1_avg=0.79544 acc_top5_avg=0.94286 lr=0.00100 gn=22.42220 time=53.61it/s +epoch=77 global_step=30450 loss=1.91594 loss_avg=2.23983 acc=0.84375 acc_top1_avg=0.79416 acc_top5_avg=0.94278 lr=0.00100 gn=29.88681 time=56.93it/s +====================Eval==================== +epoch=77 global_step=30498 loss=1.03486 test_loss_avg=0.39128 acc=0.71094 test_acc_avg=0.89017 test_acc_top5_avg=0.99586 time=226.05it/s +epoch=77 global_step=30498 loss=0.13893 test_loss_avg=0.42726 acc=0.96875 test_acc_avg=0.88153 test_acc_top5_avg=0.99056 time=218.12it/s +epoch=77 global_step=30498 loss=0.18950 test_loss_avg=0.38978 acc=0.87500 test_acc_avg=0.89132 test_acc_top5_avg=0.99140 time=544.71it/s +curr_acc 0.8913 +BEST_ACC 0.8995 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=2.02292 loss_avg=2.19326 acc=0.81250 acc_top1_avg=0.79297 acc_top5_avg=0.92969 lr=0.00100 gn=15.47582 time=50.60it/s +epoch=78 global_step=30550 loss=2.88957 loss_avg=2.24919 acc=0.71875 acc_top1_avg=0.79357 acc_top5_avg=0.94020 lr=0.00100 gn=14.75796 time=54.95it/s +epoch=78 global_step=30600 loss=1.95010 loss_avg=2.22726 acc=0.83594 acc_top1_avg=0.79542 acc_top5_avg=0.94010 lr=0.00100 gn=20.69630 time=57.40it/s +epoch=78 global_step=30650 loss=2.36402 loss_avg=2.21943 acc=0.78125 acc_top1_avg=0.79600 acc_top5_avg=0.94089 lr=0.00100 gn=21.69465 time=51.71it/s +epoch=78 global_step=30700 loss=2.10640 loss_avg=2.21578 acc=0.81250 acc_top1_avg=0.79633 acc_top5_avg=0.94059 lr=0.00100 gn=23.81293 time=54.13it/s +epoch=78 global_step=30750 loss=2.14590 loss_avg=2.22531 acc=0.81250 acc_top1_avg=0.79532 acc_top5_avg=0.94023 lr=0.00100 gn=23.88380 time=50.25it/s +epoch=78 global_step=30800 loss=2.58024 loss_avg=2.23030 acc=0.76562 acc_top1_avg=0.79483 acc_top5_avg=0.94120 lr=0.00100 gn=25.58404 time=53.71it/s +epoch=78 global_step=30850 loss=2.82508 loss_avg=2.23895 acc=0.72656 acc_top1_avg=0.79403 acc_top5_avg=0.94061 lr=0.00100 gn=22.26319 time=57.21it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.39489 test_loss_avg=0.46104 acc=0.85938 test_acc_avg=0.86637 test_acc_top5_avg=0.99013 time=210.66it/s +epoch=78 global_step=30889 loss=0.32384 test_loss_avg=0.40285 acc=0.87500 test_acc_avg=0.88341 test_acc_top5_avg=0.99179 time=851.81it/s +curr_acc 0.8834 +BEST_ACC 0.8995 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=2.72339 loss_avg=2.37951 acc=0.73438 acc_top1_avg=0.77699 acc_top5_avg=0.93466 lr=0.00100 gn=19.10313 time=51.78it/s +epoch=79 global_step=30950 loss=2.16226 loss_avg=2.22371 acc=0.81250 acc_top1_avg=0.79393 acc_top5_avg=0.93981 lr=0.00100 gn=23.24543 time=53.81it/s +epoch=79 global_step=31000 loss=2.53777 loss_avg=2.21993 acc=0.75781 acc_top1_avg=0.79462 acc_top5_avg=0.94116 lr=0.00100 gn=20.45558 time=58.85it/s +epoch=79 global_step=31050 loss=2.50132 loss_avg=2.23435 acc=0.76562 acc_top1_avg=0.79372 acc_top5_avg=0.94090 lr=0.00100 gn=27.90744 time=57.10it/s +epoch=79 global_step=31100 loss=2.08351 loss_avg=2.21527 acc=0.80469 acc_top1_avg=0.79606 acc_top5_avg=0.94165 lr=0.00100 gn=20.07373 time=52.07it/s +epoch=79 global_step=31150 loss=2.47171 loss_avg=2.21206 acc=0.78125 acc_top1_avg=0.79670 acc_top5_avg=0.94190 lr=0.00100 gn=21.97520 time=49.81it/s +epoch=79 global_step=31200 loss=2.22148 loss_avg=2.21920 acc=0.78906 acc_top1_avg=0.79600 acc_top5_avg=0.94142 lr=0.00100 gn=22.03320 time=51.70it/s +epoch=79 global_step=31250 loss=1.98364 loss_avg=2.23170 acc=0.81250 acc_top1_avg=0.79465 acc_top5_avg=0.94120 lr=0.00100 gn=18.77228 time=53.88it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.14674 test_loss_avg=0.55216 acc=0.97656 test_acc_avg=0.83594 test_acc_top5_avg=0.99045 time=226.03it/s +epoch=79 global_step=31280 loss=0.13786 test_loss_avg=0.45113 acc=0.93750 test_acc_avg=0.86984 test_acc_top5_avg=0.99153 time=231.67it/s +epoch=79 global_step=31280 loss=0.26820 test_loss_avg=0.39781 acc=0.87500 test_acc_avg=0.88430 test_acc_top5_avg=0.99179 time=886.18it/s +curr_acc 0.8843 +BEST_ACC 0.8995 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=2.29663 loss_avg=2.18275 acc=0.78906 acc_top1_avg=0.79883 acc_top5_avg=0.94609 lr=0.00010 gn=22.19156 time=57.98it/s +epoch=80 global_step=31350 loss=1.93270 loss_avg=2.22235 acc=0.82812 acc_top1_avg=0.79442 acc_top5_avg=0.94263 lr=0.00010 gn=18.78244 time=57.91it/s +epoch=80 global_step=31400 loss=2.21922 loss_avg=2.18753 acc=0.80469 acc_top1_avg=0.79850 acc_top5_avg=0.94368 lr=0.00010 gn=20.70494 time=51.71it/s +epoch=80 global_step=31450 loss=2.14535 loss_avg=2.18526 acc=0.80469 acc_top1_avg=0.79835 acc_top5_avg=0.94472 lr=0.00010 gn=19.29470 time=57.99it/s +epoch=80 global_step=31500 loss=2.13594 loss_avg=2.16559 acc=0.80469 acc_top1_avg=0.80000 acc_top5_avg=0.94450 lr=0.00010 gn=23.61137 time=52.60it/s +epoch=80 global_step=31550 loss=2.93101 loss_avg=2.16622 acc=0.73438 acc_top1_avg=0.79991 acc_top5_avg=0.94346 lr=0.00010 gn=19.41570 time=58.25it/s +epoch=80 global_step=31600 loss=2.13503 loss_avg=2.15346 acc=0.79688 acc_top1_avg=0.80100 acc_top5_avg=0.94358 lr=0.00010 gn=11.72593 time=59.58it/s +epoch=80 global_step=31650 loss=1.88641 loss_avg=2.15788 acc=0.82031 acc_top1_avg=0.80027 acc_top5_avg=0.94299 lr=0.00010 gn=14.65315 time=51.58it/s +====================Eval==================== +epoch=80 global_step=31671 loss=0.67977 test_loss_avg=0.44548 acc=0.78125 test_acc_avg=0.87292 test_acc_top5_avg=0.99062 time=233.34it/s +epoch=80 global_step=31671 loss=0.18489 test_loss_avg=0.34965 acc=0.87500 test_acc_avg=0.89804 test_acc_top5_avg=0.99268 time=532.47it/s +curr_acc 0.8980 +BEST_ACC 0.8995 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=2.00314 loss_avg=2.10225 acc=0.82031 acc_top1_avg=0.80388 acc_top5_avg=0.94370 lr=0.00010 gn=23.16946 time=55.90it/s +epoch=81 global_step=31750 loss=1.46661 loss_avg=2.11432 acc=0.86719 acc_top1_avg=0.80330 acc_top5_avg=0.94284 lr=0.00010 gn=13.59182 time=58.20it/s +epoch=81 global_step=31800 loss=2.14282 loss_avg=2.13146 acc=0.79688 acc_top1_avg=0.80154 acc_top5_avg=0.94162 lr=0.00010 gn=13.71574 time=56.61it/s +epoch=81 global_step=31850 loss=2.29264 loss_avg=2.11037 acc=0.80469 acc_top1_avg=0.80347 acc_top5_avg=0.94169 lr=0.00010 gn=23.95756 time=58.42it/s +epoch=81 global_step=31900 loss=1.88924 loss_avg=2.12467 acc=0.82031 acc_top1_avg=0.80240 acc_top5_avg=0.94115 lr=0.00010 gn=13.13624 time=57.85it/s +epoch=81 global_step=31950 loss=1.88693 loss_avg=2.12473 acc=0.82812 acc_top1_avg=0.80295 acc_top5_avg=0.94170 lr=0.00010 gn=23.82356 time=42.61it/s +epoch=81 global_step=32000 loss=3.09340 loss_avg=2.12330 acc=0.69531 acc_top1_avg=0.80310 acc_top5_avg=0.94256 lr=0.00010 gn=19.46805 time=49.44it/s +epoch=81 global_step=32050 loss=1.81783 loss_avg=2.12879 acc=0.83594 acc_top1_avg=0.80250 acc_top5_avg=0.94300 lr=0.00010 gn=19.90540 time=58.04it/s +====================Eval==================== +epoch=81 global_step=32062 loss=0.33134 test_loss_avg=0.33134 acc=0.89844 test_acc_avg=0.89844 test_acc_top5_avg=0.98438 time=205.25it/s +epoch=81 global_step=32062 loss=0.13290 test_loss_avg=0.42195 acc=0.96875 test_acc_avg=0.87944 test_acc_top5_avg=0.99020 time=226.95it/s +epoch=81 global_step=32062 loss=0.14668 test_loss_avg=0.34968 acc=0.87500 test_acc_avg=0.89893 test_acc_top5_avg=0.99199 time=877.84it/s +curr_acc 0.8989 +BEST_ACC 0.8995 +curr_acc_top5 0.9920 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=2.51278 loss_avg=2.01475 acc=0.75781 acc_top1_avg=0.81394 acc_top5_avg=0.95107 lr=0.00010 gn=19.90264 time=52.89it/s +epoch=82 global_step=32150 loss=2.08466 loss_avg=2.06952 acc=0.80469 acc_top1_avg=0.80850 acc_top5_avg=0.94407 lr=0.00010 gn=18.97448 time=55.40it/s +epoch=82 global_step=32200 loss=2.19656 loss_avg=2.08156 acc=0.78906 acc_top1_avg=0.80701 acc_top5_avg=0.94509 lr=0.00010 gn=13.79669 time=57.33it/s +epoch=82 global_step=32250 loss=1.83038 loss_avg=2.08855 acc=0.82812 acc_top1_avg=0.80589 acc_top5_avg=0.94390 lr=0.00010 gn=16.04119 time=53.34it/s +epoch=82 global_step=32300 loss=2.16375 loss_avg=2.10050 acc=0.80469 acc_top1_avg=0.80482 acc_top5_avg=0.94397 lr=0.00010 gn=18.34262 time=51.50it/s +epoch=82 global_step=32350 loss=1.88936 loss_avg=2.10673 acc=0.82031 acc_top1_avg=0.80406 acc_top5_avg=0.94301 lr=0.00010 gn=23.67398 time=56.96it/s +epoch=82 global_step=32400 loss=1.98093 loss_avg=2.10363 acc=0.83594 acc_top1_avg=0.80448 acc_top5_avg=0.94291 lr=0.00010 gn=25.46402 time=54.96it/s +epoch=82 global_step=32450 loss=1.84265 loss_avg=2.10641 acc=0.82031 acc_top1_avg=0.80392 acc_top5_avg=0.94316 lr=0.00010 gn=20.70287 time=49.77it/s +====================Eval==================== +epoch=82 global_step=32453 loss=0.68111 test_loss_avg=0.37795 acc=0.81250 test_acc_avg=0.89027 test_acc_top5_avg=0.99290 time=231.19it/s +epoch=82 global_step=32453 loss=0.36830 test_loss_avg=0.35801 acc=0.88281 test_acc_avg=0.89855 test_acc_top5_avg=0.99240 time=225.94it/s +epoch=82 global_step=32453 loss=0.12646 test_loss_avg=0.34642 acc=0.87500 test_acc_avg=0.90051 test_acc_top5_avg=0.99248 time=544.93it/s +curr_acc 0.9005 +BEST_ACC 0.8995 +curr_acc_top5 0.9925 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=2.08445 loss_avg=2.15258 acc=0.80469 acc_top1_avg=0.79870 acc_top5_avg=0.94348 lr=0.00010 gn=18.50114 time=58.42it/s +epoch=83 global_step=32550 loss=2.35689 loss_avg=2.12743 acc=0.77344 acc_top1_avg=0.80163 acc_top5_avg=0.94370 lr=0.00010 gn=18.18508 time=58.96it/s +epoch=83 global_step=32600 loss=1.38557 loss_avg=2.10148 acc=0.87500 acc_top1_avg=0.80442 acc_top5_avg=0.94398 lr=0.00010 gn=15.74297 time=54.38it/s +epoch=83 global_step=32650 loss=2.16128 loss_avg=2.09801 acc=0.80469 acc_top1_avg=0.80457 acc_top5_avg=0.94392 lr=0.00010 gn=19.30532 time=55.70it/s +epoch=83 global_step=32700 loss=1.91387 loss_avg=2.09586 acc=0.81250 acc_top1_avg=0.80488 acc_top5_avg=0.94386 lr=0.00010 gn=18.29245 time=51.27it/s +epoch=83 global_step=32750 loss=2.38446 loss_avg=2.08968 acc=0.77344 acc_top1_avg=0.80561 acc_top5_avg=0.94337 lr=0.00010 gn=18.45205 time=57.53it/s +epoch=83 global_step=32800 loss=2.45616 loss_avg=2.07886 acc=0.76562 acc_top1_avg=0.80660 acc_top5_avg=0.94470 lr=0.00010 gn=16.26125 time=58.84it/s +====================Eval==================== +epoch=83 global_step=32844 loss=0.81244 test_loss_avg=0.44662 acc=0.79688 test_acc_avg=0.87536 test_acc_top5_avg=0.98946 time=228.16it/s +epoch=83 global_step=32844 loss=0.20259 test_loss_avg=0.35823 acc=0.87500 test_acc_avg=0.89784 test_acc_top5_avg=0.99179 time=875.09it/s +curr_acc 0.8978 +BEST_ACC 0.9005 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=1.91156 loss_avg=2.15906 acc=0.82031 acc_top1_avg=0.80339 acc_top5_avg=0.94922 lr=0.00010 gn=18.80403 time=58.32it/s +epoch=84 global_step=32900 loss=1.97364 loss_avg=2.07433 acc=0.81250 acc_top1_avg=0.80664 acc_top5_avg=0.94629 lr=0.00010 gn=16.61001 time=51.32it/s +epoch=84 global_step=32950 loss=2.31006 loss_avg=2.05483 acc=0.78125 acc_top1_avg=0.80881 acc_top5_avg=0.94354 lr=0.00010 gn=17.07873 time=55.60it/s +epoch=84 global_step=33000 loss=1.70755 loss_avg=2.05707 acc=0.84375 acc_top1_avg=0.80824 acc_top5_avg=0.94291 lr=0.00010 gn=18.77971 time=53.63it/s +epoch=84 global_step=33050 loss=2.38642 loss_avg=2.06212 acc=0.75781 acc_top1_avg=0.80761 acc_top5_avg=0.94296 lr=0.00010 gn=13.80227 time=50.40it/s +epoch=84 global_step=33100 loss=1.66235 loss_avg=2.08953 acc=0.84375 acc_top1_avg=0.80499 acc_top5_avg=0.94156 lr=0.00010 gn=12.96580 time=58.78it/s +epoch=84 global_step=33150 loss=2.08610 loss_avg=2.08832 acc=0.79688 acc_top1_avg=0.80530 acc_top5_avg=0.94192 lr=0.00010 gn=21.33152 time=57.16it/s +epoch=84 global_step=33200 loss=1.74811 loss_avg=2.08146 acc=0.84375 acc_top1_avg=0.80607 acc_top5_avg=0.94277 lr=0.00010 gn=24.20188 time=47.34it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.12672 test_loss_avg=0.30640 acc=0.94531 test_acc_avg=0.90904 test_acc_top5_avg=0.99665 time=244.31it/s +epoch=84 global_step=33235 loss=0.26027 test_loss_avg=0.38309 acc=0.93750 test_acc_avg=0.89270 test_acc_top5_avg=0.99146 time=242.14it/s +epoch=84 global_step=33235 loss=0.15469 test_loss_avg=0.35012 acc=0.87500 test_acc_avg=0.90081 test_acc_top5_avg=0.99219 time=867.49it/s +curr_acc 0.9008 +BEST_ACC 0.9005 +curr_acc_top5 0.9922 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=1.76114 loss_avg=2.08451 acc=0.85938 acc_top1_avg=0.80781 acc_top5_avg=0.95312 lr=0.00010 gn=25.76811 time=56.87it/s +epoch=85 global_step=33300 loss=2.19994 loss_avg=2.03424 acc=0.78125 acc_top1_avg=0.81058 acc_top5_avg=0.94639 lr=0.00010 gn=9.59455 time=57.09it/s +epoch=85 global_step=33350 loss=2.02580 loss_avg=2.05808 acc=0.82031 acc_top1_avg=0.80849 acc_top5_avg=0.94579 lr=0.00010 gn=21.15810 time=52.16it/s +epoch=85 global_step=33400 loss=1.90987 loss_avg=2.05207 acc=0.82812 acc_top1_avg=0.80947 acc_top5_avg=0.94489 lr=0.00010 gn=22.21173 time=52.25it/s +epoch=85 global_step=33450 loss=2.10446 loss_avg=2.08184 acc=0.81250 acc_top1_avg=0.80625 acc_top5_avg=0.94350 lr=0.00010 gn=24.20843 time=51.40it/s +epoch=85 global_step=33500 loss=2.43241 loss_avg=2.08622 acc=0.77344 acc_top1_avg=0.80560 acc_top5_avg=0.94284 lr=0.00010 gn=26.24405 time=54.48it/s +epoch=85 global_step=33550 loss=1.74480 loss_avg=2.08722 acc=0.83594 acc_top1_avg=0.80538 acc_top5_avg=0.94303 lr=0.00010 gn=19.61168 time=58.45it/s +epoch=85 global_step=33600 loss=1.78682 loss_avg=2.08186 acc=0.82812 acc_top1_avg=0.80610 acc_top5_avg=0.94270 lr=0.00010 gn=14.44347 time=50.73it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.15691 test_loss_avg=0.48110 acc=0.93750 test_acc_avg=0.86763 test_acc_top5_avg=0.98661 time=231.79it/s +epoch=85 global_step=33626 loss=0.16596 test_loss_avg=0.36047 acc=0.87500 test_acc_avg=0.89725 test_acc_top5_avg=0.99110 time=877.65it/s +curr_acc 0.8973 +BEST_ACC 0.9008 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=2.20721 loss_avg=2.13020 acc=0.79688 acc_top1_avg=0.80013 acc_top5_avg=0.94108 lr=0.00010 gn=22.05818 time=51.99it/s +epoch=86 global_step=33700 loss=2.18073 loss_avg=2.04335 acc=0.79688 acc_top1_avg=0.81007 acc_top5_avg=0.94531 lr=0.00010 gn=18.95889 time=58.96it/s +epoch=86 global_step=33750 loss=2.20680 loss_avg=2.03476 acc=0.79688 acc_top1_avg=0.81061 acc_top5_avg=0.94468 lr=0.00010 gn=18.00217 time=50.57it/s +epoch=86 global_step=33800 loss=2.28880 loss_avg=2.05359 acc=0.78125 acc_top1_avg=0.80873 acc_top5_avg=0.94383 lr=0.00010 gn=21.69204 time=57.42it/s +epoch=86 global_step=33850 loss=2.57980 loss_avg=2.07406 acc=0.76562 acc_top1_avg=0.80657 acc_top5_avg=0.94378 lr=0.00010 gn=22.16396 time=48.24it/s +epoch=86 global_step=33900 loss=2.73204 loss_avg=2.06701 acc=0.73438 acc_top1_avg=0.80705 acc_top5_avg=0.94380 lr=0.00010 gn=19.59344 time=52.74it/s +epoch=86 global_step=33950 loss=2.33571 loss_avg=2.08699 acc=0.77344 acc_top1_avg=0.80524 acc_top5_avg=0.94283 lr=0.00010 gn=15.29602 time=51.02it/s +epoch=86 global_step=34000 loss=2.18789 loss_avg=2.07685 acc=0.80469 acc_top1_avg=0.80646 acc_top5_avg=0.94370 lr=0.00010 gn=19.60345 time=58.27it/s +====================Eval==================== +epoch=86 global_step=34017 loss=0.48565 test_loss_avg=0.42294 acc=0.88281 test_acc_avg=0.87500 test_acc_top5_avg=0.99609 time=245.76it/s +epoch=86 global_step=34017 loss=0.19751 test_loss_avg=0.40622 acc=0.92188 test_acc_avg=0.88630 test_acc_top5_avg=0.98996 time=230.11it/s +epoch=86 global_step=34017 loss=0.10815 test_loss_avg=0.34917 acc=0.93750 test_acc_avg=0.90170 test_acc_top5_avg=0.99130 time=558.72it/s +curr_acc 0.9017 +BEST_ACC 0.9008 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=2.13840 loss_avg=2.16949 acc=0.79688 acc_top1_avg=0.79593 acc_top5_avg=0.94105 lr=0.00010 gn=14.53300 time=56.45it/s +epoch=87 global_step=34100 loss=1.92093 loss_avg=2.05966 acc=0.81250 acc_top1_avg=0.80676 acc_top5_avg=0.94456 lr=0.00010 gn=15.96009 time=55.35it/s +epoch=87 global_step=34150 loss=2.03844 loss_avg=2.04870 acc=0.80469 acc_top1_avg=0.80839 acc_top5_avg=0.94596 lr=0.00010 gn=17.88522 time=54.75it/s +epoch=87 global_step=34200 loss=1.86851 loss_avg=2.06338 acc=0.83594 acc_top1_avg=0.80695 acc_top5_avg=0.94429 lr=0.00010 gn=21.78986 time=51.47it/s +epoch=87 global_step=34250 loss=2.16130 loss_avg=2.08006 acc=0.78906 acc_top1_avg=0.80502 acc_top5_avg=0.94364 lr=0.00010 gn=25.19581 time=51.53it/s +epoch=87 global_step=34300 loss=2.05401 loss_avg=2.07993 acc=0.82031 acc_top1_avg=0.80516 acc_top5_avg=0.94327 lr=0.00010 gn=19.54809 time=57.22it/s +epoch=87 global_step=34350 loss=2.01906 loss_avg=2.07245 acc=0.81250 acc_top1_avg=0.80588 acc_top5_avg=0.94362 lr=0.00010 gn=24.89097 time=57.43it/s +epoch=87 global_step=34400 loss=2.58120 loss_avg=2.07311 acc=0.75781 acc_top1_avg=0.80579 acc_top5_avg=0.94380 lr=0.00010 gn=18.06347 time=56.54it/s +====================Eval==================== +epoch=87 global_step=34408 loss=0.73694 test_loss_avg=0.46712 acc=0.81250 test_acc_avg=0.86719 test_acc_top5_avg=0.98785 time=243.67it/s +epoch=87 global_step=34408 loss=0.30539 test_loss_avg=0.36031 acc=0.91406 test_acc_avg=0.89813 test_acc_top5_avg=0.99006 time=230.10it/s +epoch=87 global_step=34408 loss=0.12998 test_loss_avg=0.35510 acc=0.93750 test_acc_avg=0.89913 test_acc_top5_avg=0.99031 time=556.20it/s +curr_acc 0.8991 +BEST_ACC 0.9017 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=1.85019 loss_avg=2.05745 acc=0.83594 acc_top1_avg=0.80785 acc_top5_avg=0.94513 lr=0.00010 gn=20.43497 time=52.90it/s +epoch=88 global_step=34500 loss=2.42732 loss_avg=2.04534 acc=0.76562 acc_top1_avg=0.80876 acc_top5_avg=0.94412 lr=0.00010 gn=19.41230 time=55.41it/s +epoch=88 global_step=34550 loss=2.26745 loss_avg=2.04608 acc=0.78125 acc_top1_avg=0.80887 acc_top5_avg=0.94284 lr=0.00010 gn=22.86830 time=53.72it/s +epoch=88 global_step=34600 loss=2.35143 loss_avg=2.04648 acc=0.78125 acc_top1_avg=0.80908 acc_top5_avg=0.94324 lr=0.00010 gn=21.90067 time=52.17it/s +epoch=88 global_step=34650 loss=2.12333 loss_avg=2.04303 acc=0.79688 acc_top1_avg=0.80943 acc_top5_avg=0.94350 lr=0.00010 gn=12.87732 time=58.02it/s +epoch=88 global_step=34700 loss=1.69503 loss_avg=2.06001 acc=0.84375 acc_top1_avg=0.80747 acc_top5_avg=0.94315 lr=0.00010 gn=16.27336 time=52.92it/s +epoch=88 global_step=34750 loss=1.79878 loss_avg=2.06634 acc=0.83594 acc_top1_avg=0.80679 acc_top5_avg=0.94291 lr=0.00010 gn=18.90230 time=52.08it/s +====================Eval==================== +epoch=88 global_step=34799 loss=0.32330 test_loss_avg=0.45333 acc=0.92188 test_acc_avg=0.87337 test_acc_top5_avg=0.98844 time=227.36it/s +epoch=88 global_step=34799 loss=0.13766 test_loss_avg=0.35601 acc=0.93750 test_acc_avg=0.89982 test_acc_top5_avg=0.99110 time=540.16it/s +curr_acc 0.8998 +BEST_ACC 0.9017 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=1.60875 loss_avg=1.60875 acc=0.85938 acc_top1_avg=0.85938 acc_top5_avg=0.98438 lr=0.00010 gn=25.20341 time=47.14it/s +epoch=89 global_step=34850 loss=2.46234 loss_avg=2.07370 acc=0.76562 acc_top1_avg=0.80576 acc_top5_avg=0.93964 lr=0.00010 gn=18.94715 time=52.64it/s +epoch=89 global_step=34900 loss=1.86161 loss_avg=2.07884 acc=0.82031 acc_top1_avg=0.80469 acc_top5_avg=0.94083 lr=0.00010 gn=19.68652 time=52.79it/s +epoch=89 global_step=34950 loss=1.94232 loss_avg=2.08756 acc=0.82031 acc_top1_avg=0.80427 acc_top5_avg=0.94179 lr=0.00010 gn=14.34146 time=56.49it/s +epoch=89 global_step=35000 loss=1.89094 loss_avg=2.08473 acc=0.83594 acc_top1_avg=0.80500 acc_top5_avg=0.94255 lr=0.00010 gn=21.71855 time=50.09it/s +epoch=89 global_step=35050 loss=2.64091 loss_avg=2.06823 acc=0.73438 acc_top1_avg=0.80693 acc_top5_avg=0.94397 lr=0.00010 gn=18.73870 time=57.33it/s +epoch=89 global_step=35100 loss=1.74697 loss_avg=2.06641 acc=0.84375 acc_top1_avg=0.80721 acc_top5_avg=0.94394 lr=0.00010 gn=21.95657 time=55.16it/s +epoch=89 global_step=35150 loss=2.10174 loss_avg=2.06896 acc=0.81250 acc_top1_avg=0.80676 acc_top5_avg=0.94355 lr=0.00010 gn=26.63539 time=49.68it/s +====================Eval==================== +epoch=89 global_step=35190 loss=0.63896 test_loss_avg=0.36952 acc=0.83594 test_acc_avg=0.89227 test_acc_top5_avg=0.99137 time=228.47it/s +epoch=89 global_step=35190 loss=0.14790 test_loss_avg=0.37267 acc=0.95312 test_acc_avg=0.89527 test_acc_top5_avg=0.99026 time=227.52it/s +epoch=89 global_step=35190 loss=0.15638 test_loss_avg=0.35709 acc=0.87500 test_acc_avg=0.89844 test_acc_top5_avg=0.99080 time=870.01it/s +curr_acc 0.8984 +BEST_ACC 0.9017 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=2.16144 loss_avg=2.07544 acc=0.78906 acc_top1_avg=0.80391 acc_top5_avg=0.94766 lr=0.00010 gn=21.43809 time=57.14it/s +epoch=90 global_step=35250 loss=1.52643 loss_avg=2.01653 acc=0.86719 acc_top1_avg=0.81289 acc_top5_avg=0.94831 lr=0.00010 gn=19.43517 time=53.27it/s +epoch=90 global_step=35300 loss=2.14417 loss_avg=2.01002 acc=0.79688 acc_top1_avg=0.81321 acc_top5_avg=0.94652 lr=0.00010 gn=12.66698 time=51.92it/s +epoch=90 global_step=35350 loss=1.60678 loss_avg=2.01943 acc=0.85938 acc_top1_avg=0.81226 acc_top5_avg=0.94507 lr=0.00010 gn=18.79200 time=54.34it/s +epoch=90 global_step=35400 loss=1.81502 loss_avg=2.03334 acc=0.84375 acc_top1_avg=0.81071 acc_top5_avg=0.94501 lr=0.00010 gn=30.64620 time=57.83it/s +epoch=90 global_step=35450 loss=2.19664 loss_avg=2.04050 acc=0.78906 acc_top1_avg=0.80992 acc_top5_avg=0.94489 lr=0.00010 gn=18.33495 time=52.04it/s +epoch=90 global_step=35500 loss=1.76980 loss_avg=2.04877 acc=0.85156 acc_top1_avg=0.80907 acc_top5_avg=0.94410 lr=0.00010 gn=20.04704 time=55.75it/s +epoch=90 global_step=35550 loss=1.62547 loss_avg=2.05609 acc=0.85938 acc_top1_avg=0.80822 acc_top5_avg=0.94408 lr=0.00010 gn=29.92604 time=52.61it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.57907 test_loss_avg=0.41838 acc=0.84375 test_acc_avg=0.88145 test_acc_top5_avg=0.98906 time=232.73it/s +epoch=90 global_step=35581 loss=0.10053 test_loss_avg=0.35453 acc=1.00000 test_acc_avg=0.90012 test_acc_top5_avg=0.99080 time=521.36it/s +curr_acc 0.9001 +BEST_ACC 0.9017 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=1.98923 loss_avg=2.07799 acc=0.80469 acc_top1_avg=0.80551 acc_top5_avg=0.94285 lr=0.00010 gn=18.76762 time=53.11it/s +epoch=91 global_step=35650 loss=1.96008 loss_avg=2.07244 acc=0.81250 acc_top1_avg=0.80639 acc_top5_avg=0.94452 lr=0.00010 gn=23.14560 time=52.18it/s +epoch=91 global_step=35700 loss=1.74311 loss_avg=2.06601 acc=0.84375 acc_top1_avg=0.80699 acc_top5_avg=0.94347 lr=0.00010 gn=20.96249 time=53.60it/s +epoch=91 global_step=35750 loss=1.93431 loss_avg=2.04577 acc=0.81250 acc_top1_avg=0.80899 acc_top5_avg=0.94337 lr=0.00010 gn=14.54214 time=56.36it/s +epoch=91 global_step=35800 loss=1.88075 loss_avg=2.05309 acc=0.82812 acc_top1_avg=0.80815 acc_top5_avg=0.94417 lr=0.00010 gn=18.43756 time=57.97it/s +epoch=91 global_step=35850 loss=2.13620 loss_avg=2.05002 acc=0.79688 acc_top1_avg=0.80867 acc_top5_avg=0.94493 lr=0.00010 gn=17.31072 time=50.78it/s +epoch=91 global_step=35900 loss=1.79366 loss_avg=2.04880 acc=0.82812 acc_top1_avg=0.80870 acc_top5_avg=0.94460 lr=0.00010 gn=18.60068 time=58.33it/s +epoch=91 global_step=35950 loss=2.35906 loss_avg=2.05350 acc=0.78125 acc_top1_avg=0.80827 acc_top5_avg=0.94432 lr=0.00010 gn=16.84696 time=53.96it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.15159 test_loss_avg=0.36690 acc=0.96094 test_acc_avg=0.89560 test_acc_top5_avg=0.99432 time=240.72it/s +epoch=91 global_step=35972 loss=0.12337 test_loss_avg=0.39826 acc=0.95312 test_acc_avg=0.88832 test_acc_top5_avg=0.99039 time=221.73it/s +epoch=91 global_step=35972 loss=0.11364 test_loss_avg=0.35795 acc=0.93750 test_acc_avg=0.89933 test_acc_top5_avg=0.99110 time=635.89it/s +curr_acc 0.8993 +BEST_ACC 0.9017 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=2.36325 loss_avg=2.08164 acc=0.78125 acc_top1_avg=0.80580 acc_top5_avg=0.94141 lr=0.00010 gn=22.39473 time=56.58it/s +epoch=92 global_step=36050 loss=1.78973 loss_avg=2.06394 acc=0.83594 acc_top1_avg=0.80649 acc_top5_avg=0.94181 lr=0.00010 gn=23.88512 time=57.08it/s +epoch=92 global_step=36100 loss=1.84096 loss_avg=2.03199 acc=0.83594 acc_top1_avg=0.81061 acc_top5_avg=0.94379 lr=0.00010 gn=24.43353 time=55.75it/s +epoch=92 global_step=36150 loss=2.27056 loss_avg=2.02937 acc=0.79688 acc_top1_avg=0.81118 acc_top5_avg=0.94404 lr=0.00010 gn=24.60792 time=55.13it/s +epoch=92 global_step=36200 loss=2.49627 loss_avg=2.03949 acc=0.78125 acc_top1_avg=0.81027 acc_top5_avg=0.94343 lr=0.00010 gn=28.13998 time=54.25it/s +epoch=92 global_step=36250 loss=2.53779 loss_avg=2.04683 acc=0.75000 acc_top1_avg=0.80955 acc_top5_avg=0.94382 lr=0.00010 gn=20.27498 time=50.97it/s +epoch=92 global_step=36300 loss=1.79984 loss_avg=2.05290 acc=0.82031 acc_top1_avg=0.80888 acc_top5_avg=0.94360 lr=0.00010 gn=19.85142 time=58.22it/s +epoch=92 global_step=36350 loss=1.85272 loss_avg=2.05054 acc=0.82812 acc_top1_avg=0.80911 acc_top5_avg=0.94374 lr=0.00010 gn=19.82640 time=58.62it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.34461 test_loss_avg=0.48392 acc=0.92188 test_acc_avg=0.86694 test_acc_top5_avg=0.98804 time=228.46it/s +epoch=92 global_step=36363 loss=0.10223 test_loss_avg=0.35415 acc=0.93750 test_acc_avg=0.90002 test_acc_top5_avg=0.99120 time=547.13it/s +curr_acc 0.9000 +BEST_ACC 0.9017 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=1.93795 loss_avg=2.09600 acc=0.82812 acc_top1_avg=0.80321 acc_top5_avg=0.93898 lr=0.00010 gn=22.38779 time=56.14it/s +epoch=93 global_step=36450 loss=2.43703 loss_avg=2.06488 acc=0.78125 acc_top1_avg=0.80639 acc_top5_avg=0.94352 lr=0.00010 gn=16.74620 time=52.38it/s +epoch=93 global_step=36500 loss=1.81516 loss_avg=2.06001 acc=0.83594 acc_top1_avg=0.80748 acc_top5_avg=0.94286 lr=0.00010 gn=20.76419 time=58.41it/s +epoch=93 global_step=36550 loss=2.15952 loss_avg=2.05719 acc=0.80469 acc_top1_avg=0.80765 acc_top5_avg=0.94435 lr=0.00010 gn=24.19561 time=57.30it/s +epoch=93 global_step=36600 loss=2.12976 loss_avg=2.06209 acc=0.80469 acc_top1_avg=0.80719 acc_top5_avg=0.94498 lr=0.00010 gn=23.02077 time=57.72it/s +epoch=93 global_step=36650 loss=2.20744 loss_avg=2.05038 acc=0.78906 acc_top1_avg=0.80844 acc_top5_avg=0.94496 lr=0.00010 gn=18.56958 time=51.02it/s +epoch=93 global_step=36700 loss=1.50935 loss_avg=2.04707 acc=0.87500 acc_top1_avg=0.80905 acc_top5_avg=0.94485 lr=0.00010 gn=22.31119 time=54.69it/s +epoch=93 global_step=36750 loss=1.67130 loss_avg=2.04866 acc=0.84375 acc_top1_avg=0.80881 acc_top5_avg=0.94497 lr=0.00010 gn=19.08472 time=59.28it/s +====================Eval==================== +epoch=93 global_step=36754 loss=0.64029 test_loss_avg=0.55960 acc=0.81250 test_acc_avg=0.82812 test_acc_top5_avg=0.98958 time=246.61it/s +epoch=93 global_step=36754 loss=0.12334 test_loss_avg=0.43827 acc=0.96875 test_acc_avg=0.87780 test_acc_top5_avg=0.98747 time=235.01it/s +epoch=93 global_step=36754 loss=0.10464 test_loss_avg=0.36130 acc=0.93750 test_acc_avg=0.89883 test_acc_top5_avg=0.99011 time=538.56it/s +curr_acc 0.8988 +BEST_ACC 0.9017 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=1.75714 loss_avg=2.00123 acc=0.83594 acc_top1_avg=0.81488 acc_top5_avg=0.94633 lr=0.00010 gn=17.16797 time=52.50it/s +epoch=94 global_step=36850 loss=2.21405 loss_avg=2.02539 acc=0.79688 acc_top1_avg=0.81144 acc_top5_avg=0.94425 lr=0.00010 gn=23.22646 time=57.36it/s +epoch=94 global_step=36900 loss=2.10204 loss_avg=2.04358 acc=0.79688 acc_top1_avg=0.80956 acc_top5_avg=0.94317 lr=0.00010 gn=19.16274 time=55.01it/s +epoch=94 global_step=36950 loss=2.12963 loss_avg=2.04435 acc=0.79688 acc_top1_avg=0.80939 acc_top5_avg=0.94336 lr=0.00010 gn=19.34005 time=53.96it/s +epoch=94 global_step=37000 loss=1.96735 loss_avg=2.04190 acc=0.80469 acc_top1_avg=0.80977 acc_top5_avg=0.94401 lr=0.00010 gn=18.23434 time=53.83it/s +epoch=94 global_step=37050 loss=2.31328 loss_avg=2.03934 acc=0.78906 acc_top1_avg=0.81020 acc_top5_avg=0.94386 lr=0.00010 gn=20.57355 time=53.11it/s +epoch=94 global_step=37100 loss=2.25730 loss_avg=2.03949 acc=0.78906 acc_top1_avg=0.80995 acc_top5_avg=0.94378 lr=0.00010 gn=26.13818 time=59.32it/s +====================Eval==================== +epoch=94 global_step=37145 loss=0.82309 test_loss_avg=0.43156 acc=0.78125 test_acc_avg=0.87077 test_acc_top5_avg=0.98828 time=237.56it/s +epoch=94 global_step=37145 loss=0.25300 test_loss_avg=0.37275 acc=0.91406 test_acc_avg=0.89263 test_acc_top5_avg=0.99008 time=241.15it/s +epoch=94 global_step=37145 loss=0.12303 test_loss_avg=0.36494 acc=0.93750 test_acc_avg=0.89438 test_acc_top5_avg=0.99031 time=551.59it/s +curr_acc 0.8944 +BEST_ACC 0.9017 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=2.03088 loss_avg=2.25451 acc=0.80469 acc_top1_avg=0.78438 acc_top5_avg=0.93906 lr=0.00010 gn=16.26024 time=55.54it/s +epoch=95 global_step=37200 loss=1.70565 loss_avg=2.09050 acc=0.85156 acc_top1_avg=0.80412 acc_top5_avg=0.94318 lr=0.00010 gn=19.46477 time=59.02it/s +epoch=95 global_step=37250 loss=1.48509 loss_avg=2.06069 acc=0.86719 acc_top1_avg=0.80766 acc_top5_avg=0.94308 lr=0.00010 gn=18.77342 time=52.36it/s +epoch=95 global_step=37300 loss=2.18003 loss_avg=2.06505 acc=0.78906 acc_top1_avg=0.80731 acc_top5_avg=0.94345 lr=0.00010 gn=13.65631 time=47.81it/s +epoch=95 global_step=37350 loss=1.97171 loss_avg=2.04203 acc=0.82031 acc_top1_avg=0.80987 acc_top5_avg=0.94459 lr=0.00010 gn=22.24638 time=48.23it/s +epoch=95 global_step=37400 loss=2.06440 loss_avg=2.05291 acc=0.80469 acc_top1_avg=0.80843 acc_top5_avg=0.94485 lr=0.00010 gn=13.99759 time=51.00it/s +epoch=95 global_step=37450 loss=2.37549 loss_avg=2.04050 acc=0.78125 acc_top1_avg=0.80963 acc_top5_avg=0.94554 lr=0.00010 gn=22.55890 time=53.58it/s +epoch=95 global_step=37500 loss=2.36305 loss_avg=2.03779 acc=0.78125 acc_top1_avg=0.80993 acc_top5_avg=0.94520 lr=0.00010 gn=25.91821 time=49.96it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.47009 test_loss_avg=0.47241 acc=0.90625 test_acc_avg=0.86545 test_acc_top5_avg=0.98785 time=231.92it/s +epoch=95 global_step=37536 loss=0.09296 test_loss_avg=0.36132 acc=0.93750 test_acc_avg=0.89636 test_acc_top5_avg=0.99061 time=869.47it/s +curr_acc 0.8964 +BEST_ACC 0.9017 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=2.20522 loss_avg=2.13359 acc=0.80469 acc_top1_avg=0.80134 acc_top5_avg=0.94085 lr=0.00010 gn=29.61324 time=52.51it/s +epoch=96 global_step=37600 loss=2.09781 loss_avg=2.00749 acc=0.79688 acc_top1_avg=0.81274 acc_top5_avg=0.94458 lr=0.00010 gn=13.37171 time=54.59it/s +epoch=96 global_step=37650 loss=1.98184 loss_avg=2.01742 acc=0.81250 acc_top1_avg=0.81168 acc_top5_avg=0.94422 lr=0.00010 gn=19.98531 time=53.18it/s +epoch=96 global_step=37700 loss=1.80143 loss_avg=2.01538 acc=0.82812 acc_top1_avg=0.81226 acc_top5_avg=0.94503 lr=0.00010 gn=23.42213 time=52.85it/s +epoch=96 global_step=37750 loss=1.90694 loss_avg=2.01889 acc=0.84375 acc_top1_avg=0.81188 acc_top5_avg=0.94469 lr=0.00010 gn=29.23964 time=51.68it/s +epoch=96 global_step=37800 loss=1.60632 loss_avg=2.02145 acc=0.86719 acc_top1_avg=0.81161 acc_top5_avg=0.94437 lr=0.00010 gn=22.72150 time=53.83it/s +epoch=96 global_step=37850 loss=1.65759 loss_avg=2.03657 acc=0.85156 acc_top1_avg=0.80996 acc_top5_avg=0.94467 lr=0.00010 gn=20.17119 time=51.61it/s +epoch=96 global_step=37900 loss=1.76755 loss_avg=2.04256 acc=0.84375 acc_top1_avg=0.80937 acc_top5_avg=0.94445 lr=0.00010 gn=22.48971 time=50.87it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.39501 test_loss_avg=0.30704 acc=0.86719 test_acc_avg=0.90820 test_acc_top5_avg=0.99512 time=222.73it/s +epoch=96 global_step=37927 loss=0.10374 test_loss_avg=0.38389 acc=0.96094 test_acc_avg=0.89205 test_acc_top5_avg=0.99053 time=217.47it/s +epoch=96 global_step=37927 loss=0.12141 test_loss_avg=0.36151 acc=0.87500 test_acc_avg=0.89656 test_acc_top5_avg=0.99120 time=881.16it/s +curr_acc 0.8966 +BEST_ACC 0.9017 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=1.83875 loss_avg=2.12270 acc=0.83594 acc_top1_avg=0.80163 acc_top5_avg=0.94022 lr=0.00010 gn=23.00894 time=54.00it/s +epoch=97 global_step=38000 loss=2.05337 loss_avg=2.06844 acc=0.82031 acc_top1_avg=0.80736 acc_top5_avg=0.94071 lr=0.00010 gn=25.31428 time=51.40it/s +epoch=97 global_step=38050 loss=2.66606 loss_avg=2.04768 acc=0.74219 acc_top1_avg=0.80888 acc_top5_avg=0.94131 lr=0.00010 gn=17.82266 time=59.23it/s +epoch=97 global_step=38100 loss=1.38216 loss_avg=2.03368 acc=0.87500 acc_top1_avg=0.81002 acc_top5_avg=0.94292 lr=0.00010 gn=17.26295 time=50.92it/s +epoch=97 global_step=38150 loss=2.11396 loss_avg=2.03253 acc=0.79688 acc_top1_avg=0.81026 acc_top5_avg=0.94325 lr=0.00010 gn=12.48482 time=52.01it/s +epoch=97 global_step=38200 loss=1.83609 loss_avg=2.03278 acc=0.82031 acc_top1_avg=0.81050 acc_top5_avg=0.94362 lr=0.00010 gn=17.73207 time=52.66it/s +epoch=97 global_step=38250 loss=2.18377 loss_avg=2.02667 acc=0.78125 acc_top1_avg=0.81086 acc_top5_avg=0.94415 lr=0.00010 gn=15.57587 time=50.80it/s +epoch=97 global_step=38300 loss=1.92578 loss_avg=2.03807 acc=0.82031 acc_top1_avg=0.80971 acc_top5_avg=0.94366 lr=0.00010 gn=18.67746 time=52.81it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.21736 test_loss_avg=0.47543 acc=0.92969 test_acc_avg=0.86761 test_acc_top5_avg=0.98818 time=229.00it/s +epoch=97 global_step=38318 loss=0.11716 test_loss_avg=0.37102 acc=0.93750 test_acc_avg=0.89527 test_acc_top5_avg=0.99061 time=874.91it/s +curr_acc 0.8953 +BEST_ACC 0.9017 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=1.74493 loss_avg=2.05642 acc=0.82812 acc_top1_avg=0.80811 acc_top5_avg=0.94189 lr=0.00010 gn=14.52161 time=58.22it/s +epoch=98 global_step=38400 loss=2.24519 loss_avg=2.03421 acc=0.78906 acc_top1_avg=0.80964 acc_top5_avg=0.94531 lr=0.00010 gn=17.42404 time=53.35it/s +epoch=98 global_step=38450 loss=2.75490 loss_avg=2.07027 acc=0.73438 acc_top1_avg=0.80611 acc_top5_avg=0.94377 lr=0.00010 gn=18.38704 time=50.71it/s +epoch=98 global_step=38500 loss=2.54099 loss_avg=2.05644 acc=0.75000 acc_top1_avg=0.80778 acc_top5_avg=0.94420 lr=0.00010 gn=14.38896 time=53.40it/s +epoch=98 global_step=38550 loss=2.54242 loss_avg=2.05732 acc=0.75781 acc_top1_avg=0.80762 acc_top5_avg=0.94457 lr=0.00010 gn=22.35498 time=57.29it/s +epoch=98 global_step=38600 loss=1.43935 loss_avg=2.04380 acc=0.86719 acc_top1_avg=0.80895 acc_top5_avg=0.94479 lr=0.00010 gn=18.37395 time=57.80it/s +epoch=98 global_step=38650 loss=1.89413 loss_avg=2.03173 acc=0.82812 acc_top1_avg=0.81029 acc_top5_avg=0.94437 lr=0.00010 gn=20.18543 time=49.26it/s +epoch=98 global_step=38700 loss=2.01224 loss_avg=2.03303 acc=0.81250 acc_top1_avg=0.81011 acc_top5_avg=0.94492 lr=0.00010 gn=23.03345 time=55.80it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.24899 test_loss_avg=0.45226 acc=0.92188 test_acc_avg=0.86426 test_acc_top5_avg=0.99121 time=223.30it/s +epoch=98 global_step=38709 loss=0.28305 test_loss_avg=0.41274 acc=0.92969 test_acc_avg=0.88443 test_acc_top5_avg=0.98963 time=225.65it/s +epoch=98 global_step=38709 loss=0.08243 test_loss_avg=0.35766 acc=1.00000 test_acc_avg=0.90012 test_acc_top5_avg=0.99080 time=866.23it/s +curr_acc 0.9001 +BEST_ACC 0.9017 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=1.88829 loss_avg=2.05167 acc=0.82812 acc_top1_avg=0.80812 acc_top5_avg=0.94931 lr=0.00010 gn=20.58745 time=55.50it/s +epoch=99 global_step=38800 loss=1.94639 loss_avg=2.02442 acc=0.82031 acc_top1_avg=0.81121 acc_top5_avg=0.94789 lr=0.00010 gn=20.34200 time=56.99it/s +epoch=99 global_step=38850 loss=1.62833 loss_avg=2.01858 acc=0.85156 acc_top1_avg=0.81195 acc_top5_avg=0.94670 lr=0.00010 gn=18.17611 time=51.79it/s +epoch=99 global_step=38900 loss=2.08761 loss_avg=2.01800 acc=0.80469 acc_top1_avg=0.81205 acc_top5_avg=0.94507 lr=0.00010 gn=15.17505 time=51.75it/s +epoch=99 global_step=38950 loss=2.35707 loss_avg=2.02075 acc=0.76562 acc_top1_avg=0.81172 acc_top5_avg=0.94473 lr=0.00010 gn=14.74060 time=50.57it/s +epoch=99 global_step=39000 loss=2.14202 loss_avg=2.02973 acc=0.80469 acc_top1_avg=0.81081 acc_top5_avg=0.94491 lr=0.00010 gn=21.06008 time=49.36it/s +epoch=99 global_step=39050 loss=1.84198 loss_avg=2.02538 acc=0.84375 acc_top1_avg=0.81156 acc_top5_avg=0.94458 lr=0.00010 gn=23.06733 time=58.40it/s +epoch=99 global_step=39100 loss=1.89801 loss_avg=2.02972 acc=0.81250 acc_top1_avg=0.81126 acc_top5_avg=0.94439 lr=0.00010 gn=19.43436 time=78.03it/s +====================Eval==================== +epoch=99 global_step=39100 loss=0.52097 test_loss_avg=0.47609 acc=0.84375 test_acc_avg=0.86988 test_acc_top5_avg=0.98842 time=243.16it/s +epoch=99 global_step=39100 loss=0.09330 test_loss_avg=0.36561 acc=1.00000 test_acc_avg=0.89903 test_acc_top5_avg=0.99061 time=564.97it/s +epoch=99 global_step=39100 loss=0.09330 test_loss_avg=0.36561 acc=1.00000 test_acc_avg=0.89903 test_acc_top5_avg=0.99061 time=564.97it/s +curr_acc 0.8990 +BEST_ACC 0.9017 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=1.84465 loss_avg=2.01038 acc=0.83594 acc_top1_avg=0.81266 acc_top5_avg=0.94734 lr=0.00010 gn=23.78900 time=55.29it/s +epoch=100 global_step=39200 loss=1.85437 loss_avg=1.98707 acc=0.83594 acc_top1_avg=0.81492 acc_top5_avg=0.94852 lr=0.00010 gn=22.70739 time=56.46it/s +epoch=100 global_step=39250 loss=1.71256 loss_avg=2.01598 acc=0.83594 acc_top1_avg=0.81156 acc_top5_avg=0.94740 lr=0.00010 gn=18.14596 time=54.59it/s +epoch=100 global_step=39300 loss=1.71560 loss_avg=2.01964 acc=0.83594 acc_top1_avg=0.81109 acc_top5_avg=0.94563 lr=0.00010 gn=11.67338 time=52.17it/s +epoch=100 global_step=39350 loss=1.56559 loss_avg=2.02555 acc=0.85938 acc_top1_avg=0.81088 acc_top5_avg=0.94563 lr=0.00010 gn=19.76629 time=58.40it/s +epoch=100 global_step=39400 loss=1.86652 loss_avg=2.02957 acc=0.82812 acc_top1_avg=0.81047 acc_top5_avg=0.94521 lr=0.00010 gn=19.05045 time=50.16it/s +epoch=100 global_step=39450 loss=1.55150 loss_avg=2.01877 acc=0.86719 acc_top1_avg=0.81167 acc_top5_avg=0.94498 lr=0.00010 gn=24.31482 time=51.59it/s +====================Eval==================== +epoch=100 global_step=39491 loss=0.20010 test_loss_avg=0.46691 acc=0.94531 test_acc_avg=0.87281 test_acc_top5_avg=0.98672 time=229.22it/s +epoch=100 global_step=39491 loss=0.08816 test_loss_avg=0.36986 acc=1.00000 test_acc_avg=0.89784 test_acc_top5_avg=0.98991 time=897.95it/s +curr_acc 0.8978 +BEST_ACC 0.9017 +curr_acc_top5 0.9899 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=1.70092 loss_avg=1.96919 acc=0.84375 acc_top1_avg=0.82031 acc_top5_avg=0.94878 lr=0.00010 gn=21.99055 time=54.15it/s +epoch=101 global_step=39550 loss=1.82912 loss_avg=1.98648 acc=0.84375 acc_top1_avg=0.81647 acc_top5_avg=0.94372 lr=0.00010 gn=20.63898 time=52.27it/s +epoch=101 global_step=39600 loss=1.91222 loss_avg=2.00902 acc=0.82812 acc_top1_avg=0.81350 acc_top5_avg=0.94452 lr=0.00010 gn=18.57658 time=51.00it/s +epoch=101 global_step=39650 loss=1.79309 loss_avg=2.02540 acc=0.83594 acc_top1_avg=0.81235 acc_top5_avg=0.94320 lr=0.00010 gn=23.17616 time=48.07it/s +epoch=101 global_step=39700 loss=1.81662 loss_avg=2.01604 acc=0.83594 acc_top1_avg=0.81325 acc_top5_avg=0.94412 lr=0.00010 gn=25.43519 time=57.31it/s +epoch=101 global_step=39750 loss=1.85973 loss_avg=2.01746 acc=0.81250 acc_top1_avg=0.81289 acc_top5_avg=0.94429 lr=0.00010 gn=14.18601 time=53.58it/s +epoch=101 global_step=39800 loss=2.18073 loss_avg=2.01877 acc=0.78906 acc_top1_avg=0.81273 acc_top5_avg=0.94453 lr=0.00010 gn=16.97181 time=51.74it/s +epoch=101 global_step=39850 loss=2.37696 loss_avg=2.02102 acc=0.76562 acc_top1_avg=0.81243 acc_top5_avg=0.94464 lr=0.00010 gn=17.11391 time=56.47it/s +====================Eval==================== +epoch=101 global_step=39882 loss=0.58690 test_loss_avg=0.39864 acc=0.85938 test_acc_avg=0.88653 test_acc_top5_avg=0.99219 time=247.67it/s +epoch=101 global_step=39882 loss=0.24900 test_loss_avg=0.37547 acc=0.92969 test_acc_avg=0.89503 test_acc_top5_avg=0.99021 time=235.90it/s +epoch=101 global_step=39882 loss=0.09832 test_loss_avg=0.36384 acc=1.00000 test_acc_avg=0.89794 test_acc_top5_avg=0.99051 time=870.73it/s +curr_acc 0.8979 +BEST_ACC 0.9017 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=2.09039 loss_avg=2.05506 acc=0.79688 acc_top1_avg=0.80859 acc_top5_avg=0.94010 lr=0.00010 gn=17.77162 time=57.53it/s +epoch=102 global_step=39950 loss=2.03534 loss_avg=2.06692 acc=0.82031 acc_top1_avg=0.80722 acc_top5_avg=0.94451 lr=0.00010 gn=21.15302 time=58.93it/s +epoch=102 global_step=40000 loss=2.21834 loss_avg=2.05538 acc=0.79688 acc_top1_avg=0.80853 acc_top5_avg=0.94386 lr=0.00010 gn=20.35004 time=51.16it/s +epoch=102 global_step=40050 loss=2.52381 loss_avg=2.03864 acc=0.75781 acc_top1_avg=0.81041 acc_top5_avg=0.94327 lr=0.00010 gn=15.27521 time=57.77it/s +epoch=102 global_step=40100 loss=1.98967 loss_avg=2.01714 acc=0.82031 acc_top1_avg=0.81275 acc_top5_avg=0.94510 lr=0.00010 gn=22.24792 time=53.73it/s +epoch=102 global_step=40150 loss=2.05531 loss_avg=2.01110 acc=0.81250 acc_top1_avg=0.81337 acc_top5_avg=0.94590 lr=0.00010 gn=25.90412 time=48.99it/s +epoch=102 global_step=40200 loss=2.58807 loss_avg=2.02802 acc=0.74219 acc_top1_avg=0.81147 acc_top5_avg=0.94497 lr=0.00010 gn=13.76903 time=55.21it/s +epoch=102 global_step=40250 loss=2.15000 loss_avg=2.01458 acc=0.79688 acc_top1_avg=0.81286 acc_top5_avg=0.94546 lr=0.00010 gn=26.67033 time=58.19it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.56708 test_loss_avg=0.46366 acc=0.81250 test_acc_avg=0.87091 test_acc_top5_avg=0.98717 time=230.60it/s +epoch=102 global_step=40273 loss=0.08340 test_loss_avg=0.36609 acc=1.00000 test_acc_avg=0.89844 test_acc_top5_avg=0.99011 time=464.69it/s +curr_acc 0.8984 +BEST_ACC 0.9017 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=1.83592 loss_avg=1.96057 acc=0.82812 acc_top1_avg=0.81539 acc_top5_avg=0.94676 lr=0.00010 gn=15.18354 time=52.60it/s +epoch=103 global_step=40350 loss=1.57660 loss_avg=1.97269 acc=0.85156 acc_top1_avg=0.81504 acc_top5_avg=0.94653 lr=0.00010 gn=16.41081 time=50.04it/s +epoch=103 global_step=40400 loss=1.64036 loss_avg=1.98439 acc=0.85156 acc_top1_avg=0.81391 acc_top5_avg=0.94439 lr=0.00010 gn=21.16416 time=52.18it/s +epoch=103 global_step=40450 loss=1.78308 loss_avg=1.99908 acc=0.84375 acc_top1_avg=0.81294 acc_top5_avg=0.94465 lr=0.00010 gn=28.37574 time=52.63it/s +epoch=103 global_step=40500 loss=2.12386 loss_avg=2.00086 acc=0.79688 acc_top1_avg=0.81288 acc_top5_avg=0.94445 lr=0.00010 gn=18.09108 time=54.53it/s +epoch=103 global_step=40550 loss=1.71793 loss_avg=2.00653 acc=0.83594 acc_top1_avg=0.81239 acc_top5_avg=0.94416 lr=0.00010 gn=15.93982 time=44.43it/s +epoch=103 global_step=40600 loss=2.46576 loss_avg=2.01485 acc=0.76562 acc_top1_avg=0.81176 acc_top5_avg=0.94395 lr=0.00010 gn=19.75946 time=52.69it/s +epoch=103 global_step=40650 loss=1.63490 loss_avg=2.01218 acc=0.85938 acc_top1_avg=0.81209 acc_top5_avg=0.94467 lr=0.00010 gn=22.82704 time=56.64it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.08227 test_loss_avg=0.38193 acc=0.96875 test_acc_avg=0.89243 test_acc_top5_avg=0.99159 time=125.44it/s +epoch=103 global_step=40664 loss=0.19023 test_loss_avg=0.41397 acc=0.93750 test_acc_avg=0.88504 test_acc_top5_avg=0.98859 time=207.71it/s +epoch=103 global_step=40664 loss=0.06543 test_loss_avg=0.37195 acc=1.00000 test_acc_avg=0.89676 test_acc_top5_avg=0.98942 time=606.73it/s +curr_acc 0.8968 +BEST_ACC 0.9017 +curr_acc_top5 0.9894 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=1.67792 loss_avg=1.85250 acc=0.85156 acc_top1_avg=0.82943 acc_top5_avg=0.94596 lr=0.00010 gn=19.79723 time=47.40it/s +epoch=104 global_step=40750 loss=2.67469 loss_avg=1.93595 acc=0.74219 acc_top1_avg=0.81959 acc_top5_avg=0.94395 lr=0.00010 gn=21.30383 time=52.85it/s +epoch=104 global_step=40800 loss=2.10382 loss_avg=1.97643 acc=0.79688 acc_top1_avg=0.81560 acc_top5_avg=0.94324 lr=0.00010 gn=14.21922 time=58.95it/s +epoch=104 global_step=40850 loss=2.74146 loss_avg=1.99655 acc=0.71875 acc_top1_avg=0.81351 acc_top5_avg=0.94267 lr=0.00010 gn=22.25073 time=49.41it/s +epoch=104 global_step=40900 loss=2.21470 loss_avg=1.99811 acc=0.79688 acc_top1_avg=0.81359 acc_top5_avg=0.94283 lr=0.00010 gn=20.02654 time=58.23it/s +epoch=104 global_step=40950 loss=1.99417 loss_avg=2.00699 acc=0.80469 acc_top1_avg=0.81296 acc_top5_avg=0.94332 lr=0.00010 gn=16.65124 time=57.94it/s +epoch=104 global_step=41000 loss=2.15610 loss_avg=2.01551 acc=0.79688 acc_top1_avg=0.81227 acc_top5_avg=0.94387 lr=0.00010 gn=21.13459 time=54.63it/s +epoch=104 global_step=41050 loss=1.41846 loss_avg=2.02024 acc=0.86719 acc_top1_avg=0.81179 acc_top5_avg=0.94355 lr=0.00010 gn=20.73188 time=52.17it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.30375 test_loss_avg=0.50111 acc=0.90625 test_acc_avg=0.85892 test_acc_top5_avg=0.98598 time=225.59it/s +epoch=104 global_step=41055 loss=0.07002 test_loss_avg=0.37246 acc=1.00000 test_acc_avg=0.89567 test_acc_top5_avg=0.98952 time=555.24it/s +curr_acc 0.8957 +BEST_ACC 0.9017 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=2.25202 loss_avg=2.02975 acc=0.78906 acc_top1_avg=0.81111 acc_top5_avg=0.94861 lr=0.00010 gn=19.10884 time=48.05it/s +epoch=105 global_step=41150 loss=1.73023 loss_avg=2.00709 acc=0.82812 acc_top1_avg=0.81168 acc_top5_avg=0.94844 lr=0.00010 gn=17.97239 time=56.62it/s +epoch=105 global_step=41200 loss=2.02653 loss_avg=2.00408 acc=0.80469 acc_top1_avg=0.81239 acc_top5_avg=0.94758 lr=0.00010 gn=17.51875 time=50.06it/s +epoch=105 global_step=41250 loss=1.44448 loss_avg=1.99416 acc=0.87500 acc_top1_avg=0.81382 acc_top5_avg=0.94708 lr=0.00010 gn=20.60304 time=52.19it/s +epoch=105 global_step=41300 loss=1.93758 loss_avg=2.00057 acc=0.82031 acc_top1_avg=0.81291 acc_top5_avg=0.94624 lr=0.00010 gn=22.94226 time=50.07it/s +epoch=105 global_step=41350 loss=1.77690 loss_avg=2.01527 acc=0.82812 acc_top1_avg=0.81131 acc_top5_avg=0.94558 lr=0.00010 gn=18.47229 time=56.47it/s +epoch=105 global_step=41400 loss=1.66306 loss_avg=2.00973 acc=0.84375 acc_top1_avg=0.81202 acc_top5_avg=0.94592 lr=0.00010 gn=17.42708 time=52.54it/s +====================Eval==================== +epoch=105 global_step=41446 loss=0.37957 test_loss_avg=0.51923 acc=0.89062 test_acc_avg=0.84219 test_acc_top5_avg=0.98750 time=230.39it/s +epoch=105 global_step=41446 loss=0.32108 test_loss_avg=0.44810 acc=0.90625 test_acc_avg=0.87429 test_acc_top5_avg=0.98722 time=231.33it/s +epoch=105 global_step=41446 loss=0.10285 test_loss_avg=0.37479 acc=1.00000 test_acc_avg=0.89517 test_acc_top5_avg=0.98942 time=539.67it/s +curr_acc 0.8952 +BEST_ACC 0.9017 +curr_acc_top5 0.9894 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=1.70330 loss_avg=1.96918 acc=0.83594 acc_top1_avg=0.81445 acc_top5_avg=0.93555 lr=0.00010 gn=17.63774 time=51.56it/s +epoch=106 global_step=41500 loss=1.93339 loss_avg=2.01251 acc=0.82031 acc_top1_avg=0.81134 acc_top5_avg=0.94473 lr=0.00010 gn=22.20750 time=48.86it/s +epoch=106 global_step=41550 loss=1.92914 loss_avg=1.99042 acc=0.81250 acc_top1_avg=0.81415 acc_top5_avg=0.94426 lr=0.00010 gn=19.38194 time=55.98it/s +epoch=106 global_step=41600 loss=2.41109 loss_avg=2.01924 acc=0.77344 acc_top1_avg=0.81118 acc_top5_avg=0.94486 lr=0.00010 gn=18.96732 time=54.37it/s +epoch=106 global_step=41650 loss=2.22084 loss_avg=2.01049 acc=0.78125 acc_top1_avg=0.81208 acc_top5_avg=0.94562 lr=0.00010 gn=18.96168 time=57.75it/s +epoch=106 global_step=41700 loss=2.16390 loss_avg=2.00330 acc=0.79688 acc_top1_avg=0.81287 acc_top5_avg=0.94537 lr=0.00010 gn=21.85928 time=53.33it/s +epoch=106 global_step=41750 loss=1.50621 loss_avg=2.00711 acc=0.85938 acc_top1_avg=0.81268 acc_top5_avg=0.94449 lr=0.00010 gn=18.36309 time=54.30it/s +epoch=106 global_step=41800 loss=1.56955 loss_avg=2.00368 acc=0.86719 acc_top1_avg=0.81316 acc_top5_avg=0.94474 lr=0.00010 gn=23.42765 time=51.53it/s +====================Eval==================== +epoch=106 global_step=41837 loss=0.83305 test_loss_avg=0.46000 acc=0.81250 test_acc_avg=0.87350 test_acc_top5_avg=0.98768 time=231.81it/s +epoch=106 global_step=41837 loss=0.29068 test_loss_avg=0.37507 acc=0.92188 test_acc_avg=0.89576 test_acc_top5_avg=0.98982 time=245.19it/s +epoch=106 global_step=41837 loss=0.11756 test_loss_avg=0.36859 acc=1.00000 test_acc_avg=0.89784 test_acc_top5_avg=0.99001 time=883.01it/s +curr_acc 0.8978 +BEST_ACC 0.9017 +curr_acc_top5 0.9900 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=1.56368 loss_avg=1.92418 acc=0.85938 acc_top1_avg=0.81731 acc_top5_avg=0.95012 lr=0.00010 gn=18.93651 time=51.85it/s +epoch=107 global_step=41900 loss=2.55704 loss_avg=1.97549 acc=0.75781 acc_top1_avg=0.81560 acc_top5_avg=0.94606 lr=0.00010 gn=18.05124 time=51.04it/s +epoch=107 global_step=41950 loss=1.92648 loss_avg=1.98025 acc=0.81250 acc_top1_avg=0.81492 acc_top5_avg=0.94545 lr=0.00010 gn=14.15950 time=51.60it/s +epoch=107 global_step=42000 loss=1.78254 loss_avg=2.00998 acc=0.82812 acc_top1_avg=0.81236 acc_top5_avg=0.94550 lr=0.00010 gn=17.20883 time=58.55it/s +epoch=107 global_step=42050 loss=2.26576 loss_avg=2.00707 acc=0.78125 acc_top1_avg=0.81257 acc_top5_avg=0.94605 lr=0.00010 gn=13.10805 time=52.76it/s +epoch=107 global_step=42100 loss=1.97866 loss_avg=2.01607 acc=0.81250 acc_top1_avg=0.81170 acc_top5_avg=0.94484 lr=0.00010 gn=19.17751 time=50.52it/s +epoch=107 global_step=42150 loss=1.87653 loss_avg=2.00826 acc=0.82812 acc_top1_avg=0.81257 acc_top5_avg=0.94469 lr=0.00010 gn=20.40899 time=59.06it/s +epoch=107 global_step=42200 loss=2.09211 loss_avg=2.00671 acc=0.79688 acc_top1_avg=0.81284 acc_top5_avg=0.94484 lr=0.00010 gn=18.58182 time=52.44it/s +====================Eval==================== +epoch=107 global_step=42228 loss=0.56472 test_loss_avg=0.48482 acc=0.85938 test_acc_avg=0.86403 test_acc_top5_avg=0.98687 time=231.59it/s +epoch=107 global_step=42228 loss=0.11688 test_loss_avg=0.37522 acc=1.00000 test_acc_avg=0.89567 test_acc_top5_avg=0.99011 time=547.06it/s +curr_acc 0.8957 +BEST_ACC 0.9017 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=1.41489 loss_avg=1.92330 acc=0.87500 acc_top1_avg=0.82280 acc_top5_avg=0.94886 lr=0.00010 gn=18.69174 time=56.50it/s +epoch=108 global_step=42300 loss=1.78018 loss_avg=1.95152 acc=0.82812 acc_top1_avg=0.81879 acc_top5_avg=0.94705 lr=0.00010 gn=22.53294 time=55.29it/s +epoch=108 global_step=42350 loss=2.11264 loss_avg=1.98553 acc=0.81250 acc_top1_avg=0.81525 acc_top5_avg=0.94518 lr=0.00010 gn=22.39518 time=54.45it/s +epoch=108 global_step=42400 loss=2.26112 loss_avg=1.99505 acc=0.80469 acc_top1_avg=0.81436 acc_top5_avg=0.94545 lr=0.00010 gn=22.63962 time=56.45it/s +epoch=108 global_step=42450 loss=2.48050 loss_avg=1.97564 acc=0.76562 acc_top1_avg=0.81623 acc_top5_avg=0.94626 lr=0.00010 gn=18.97515 time=55.39it/s +epoch=108 global_step=42500 loss=2.44271 loss_avg=1.99194 acc=0.76562 acc_top1_avg=0.81463 acc_top5_avg=0.94557 lr=0.00010 gn=21.75397 time=52.08it/s +epoch=108 global_step=42550 loss=1.43950 loss_avg=1.99581 acc=0.87500 acc_top1_avg=0.81403 acc_top5_avg=0.94505 lr=0.00010 gn=19.71648 time=51.34it/s +epoch=108 global_step=42600 loss=2.70820 loss_avg=2.00311 acc=0.73438 acc_top1_avg=0.81330 acc_top5_avg=0.94462 lr=0.00010 gn=20.31464 time=48.10it/s +====================Eval==================== +epoch=108 global_step=42619 loss=0.68185 test_loss_avg=0.37471 acc=0.82812 test_acc_avg=0.88932 test_acc_top5_avg=0.99219 time=235.79it/s +epoch=108 global_step=42619 loss=0.33338 test_loss_avg=0.39082 acc=0.93750 test_acc_avg=0.88982 test_acc_top5_avg=0.98943 time=235.09it/s +epoch=108 global_step=42619 loss=0.10975 test_loss_avg=0.37084 acc=1.00000 test_acc_avg=0.89527 test_acc_top5_avg=0.99011 time=547.42it/s +curr_acc 0.8953 +BEST_ACC 0.9017 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=1.71507 loss_avg=1.98712 acc=0.83594 acc_top1_avg=0.81628 acc_top5_avg=0.94808 lr=0.00010 gn=21.11648 time=50.31it/s +epoch=109 global_step=42700 loss=1.93453 loss_avg=2.01548 acc=0.82812 acc_top1_avg=0.81125 acc_top5_avg=0.94473 lr=0.00010 gn=22.76561 time=52.06it/s +epoch=109 global_step=42750 loss=1.84197 loss_avg=1.99353 acc=0.85156 acc_top1_avg=0.81399 acc_top5_avg=0.94448 lr=0.00010 gn=30.75917 time=57.07it/s +epoch=109 global_step=42800 loss=2.02288 loss_avg=1.99592 acc=0.80469 acc_top1_avg=0.81418 acc_top5_avg=0.94436 lr=0.00010 gn=26.92730 time=50.55it/s +epoch=109 global_step=42850 loss=2.23035 loss_avg=1.99349 acc=0.78906 acc_top1_avg=0.81439 acc_top5_avg=0.94521 lr=0.00010 gn=26.51417 time=54.45it/s +epoch=109 global_step=42900 loss=1.67237 loss_avg=1.99388 acc=0.83594 acc_top1_avg=0.81422 acc_top5_avg=0.94534 lr=0.00010 gn=17.80975 time=54.42it/s +epoch=109 global_step=42950 loss=2.23084 loss_avg=1.99810 acc=0.78906 acc_top1_avg=0.81387 acc_top5_avg=0.94472 lr=0.00010 gn=20.64891 time=56.74it/s +epoch=109 global_step=43000 loss=1.97354 loss_avg=2.00449 acc=0.81250 acc_top1_avg=0.81320 acc_top5_avg=0.94435 lr=0.00010 gn=16.47975 time=52.25it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.17914 test_loss_avg=0.44859 acc=0.96094 test_acc_avg=0.87280 test_acc_top5_avg=0.98698 time=229.49it/s +epoch=109 global_step=43010 loss=0.11822 test_loss_avg=0.37530 acc=1.00000 test_acc_avg=0.89409 test_acc_top5_avg=0.98942 time=893.93it/s +curr_acc 0.8941 +BEST_ACC 0.9017 +curr_acc_top5 0.9894 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=2.23847 loss_avg=2.05211 acc=0.79688 acc_top1_avg=0.80684 acc_top5_avg=0.94766 lr=0.00010 gn=27.78895 time=57.26it/s +epoch=110 global_step=43100 loss=1.53646 loss_avg=2.04520 acc=0.86719 acc_top1_avg=0.80816 acc_top5_avg=0.94505 lr=0.00010 gn=22.18789 time=58.52it/s +epoch=110 global_step=43150 loss=2.06666 loss_avg=2.04584 acc=0.80469 acc_top1_avg=0.80837 acc_top5_avg=0.94459 lr=0.00010 gn=24.72063 time=56.19it/s +epoch=110 global_step=43200 loss=1.39338 loss_avg=2.02554 acc=0.88281 acc_top1_avg=0.81032 acc_top5_avg=0.94379 lr=0.00010 gn=18.43620 time=57.34it/s +epoch=110 global_step=43250 loss=1.36508 loss_avg=2.02197 acc=0.88281 acc_top1_avg=0.81068 acc_top5_avg=0.94502 lr=0.00010 gn=16.26521 time=54.65it/s +epoch=110 global_step=43300 loss=1.31394 loss_avg=2.01002 acc=0.89062 acc_top1_avg=0.81185 acc_top5_avg=0.94537 lr=0.00010 gn=22.57348 time=59.07it/s +epoch=110 global_step=43350 loss=1.81630 loss_avg=2.00407 acc=0.82812 acc_top1_avg=0.81245 acc_top5_avg=0.94550 lr=0.00010 gn=22.53564 time=49.28it/s +epoch=110 global_step=43400 loss=2.23950 loss_avg=1.99716 acc=0.78125 acc_top1_avg=0.81370 acc_top5_avg=0.94557 lr=0.00010 gn=18.39411 time=59.34it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.11462 test_loss_avg=0.45019 acc=0.96094 test_acc_avg=0.86719 test_acc_top5_avg=0.99219 time=228.62it/s +epoch=110 global_step=43401 loss=0.10140 test_loss_avg=0.43080 acc=0.98438 test_acc_avg=0.87799 test_acc_top5_avg=0.98815 time=208.25it/s +epoch=110 global_step=43401 loss=0.07620 test_loss_avg=0.37894 acc=1.00000 test_acc_avg=0.89300 test_acc_top5_avg=0.98981 time=885.81it/s +curr_acc 0.8930 +BEST_ACC 0.9017 +curr_acc_top5 0.9898 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=2.10984 loss_avg=1.99025 acc=0.80469 acc_top1_avg=0.81489 acc_top5_avg=0.94627 lr=0.00010 gn=20.33651 time=59.08it/s +epoch=111 global_step=43500 loss=2.08345 loss_avg=2.01479 acc=0.80469 acc_top1_avg=0.81242 acc_top5_avg=0.94642 lr=0.00010 gn=26.92694 time=58.67it/s +epoch=111 global_step=43550 loss=2.21701 loss_avg=1.97500 acc=0.79688 acc_top1_avg=0.81654 acc_top5_avg=0.94683 lr=0.00010 gn=17.11792 time=55.49it/s +epoch=111 global_step=43600 loss=1.54325 loss_avg=1.98687 acc=0.86719 acc_top1_avg=0.81529 acc_top5_avg=0.94669 lr=0.00010 gn=19.23445 time=57.42it/s +epoch=111 global_step=43650 loss=1.93780 loss_avg=1.98259 acc=0.82031 acc_top1_avg=0.81583 acc_top5_avg=0.94556 lr=0.00010 gn=19.10294 time=58.67it/s +epoch=111 global_step=43700 loss=1.49522 loss_avg=1.99480 acc=0.86719 acc_top1_avg=0.81412 acc_top5_avg=0.94474 lr=0.00010 gn=24.52664 time=49.99it/s +epoch=111 global_step=43750 loss=1.48808 loss_avg=1.98929 acc=0.86719 acc_top1_avg=0.81472 acc_top5_avg=0.94558 lr=0.00010 gn=20.59043 time=56.07it/s +====================Eval==================== +epoch=111 global_step=43792 loss=0.74423 test_loss_avg=0.50577 acc=0.81250 test_acc_avg=0.85509 test_acc_top5_avg=0.98488 time=246.97it/s +epoch=111 global_step=43792 loss=0.09982 test_loss_avg=0.37843 acc=1.00000 test_acc_avg=0.89250 test_acc_top5_avg=0.98942 time=885.25it/s +curr_acc 0.8925 +BEST_ACC 0.9017 +curr_acc_top5 0.9894 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=2.08904 loss_avg=2.10185 acc=0.80469 acc_top1_avg=0.79980 acc_top5_avg=0.94336 lr=0.00010 gn=20.10175 time=53.45it/s +epoch=112 global_step=43850 loss=2.36672 loss_avg=2.03326 acc=0.78125 acc_top1_avg=0.81075 acc_top5_avg=0.94599 lr=0.00010 gn=25.11228 time=58.50it/s +epoch=112 global_step=43900 loss=2.02130 loss_avg=1.99774 acc=0.79688 acc_top1_avg=0.81366 acc_top5_avg=0.94618 lr=0.00010 gn=15.30047 time=59.03it/s +epoch=112 global_step=43950 loss=2.21202 loss_avg=1.97090 acc=0.78906 acc_top1_avg=0.81675 acc_top5_avg=0.94665 lr=0.00010 gn=13.50599 time=54.92it/s +epoch=112 global_step=44000 loss=2.33415 loss_avg=1.98762 acc=0.77344 acc_top1_avg=0.81520 acc_top5_avg=0.94595 lr=0.00010 gn=16.62814 time=55.90it/s +epoch=112 global_step=44050 loss=2.51711 loss_avg=1.97884 acc=0.75781 acc_top1_avg=0.81622 acc_top5_avg=0.94649 lr=0.00010 gn=21.38295 time=54.88it/s +epoch=112 global_step=44100 loss=2.05827 loss_avg=1.97740 acc=0.80469 acc_top1_avg=0.81641 acc_top5_avg=0.94678 lr=0.00010 gn=22.83211 time=56.10it/s +epoch=112 global_step=44150 loss=1.83964 loss_avg=1.97851 acc=0.83594 acc_top1_avg=0.81632 acc_top5_avg=0.94612 lr=0.00010 gn=20.49318 time=58.08it/s +====================Eval==================== +epoch=112 global_step=44183 loss=0.65761 test_loss_avg=0.56508 acc=0.80469 test_acc_avg=0.82422 test_acc_top5_avg=0.98438 time=242.53it/s +epoch=112 global_step=44183 loss=0.15992 test_loss_avg=0.45760 acc=0.93750 test_acc_avg=0.87049 test_acc_top5_avg=0.98693 time=229.13it/s +epoch=112 global_step=44183 loss=0.10043 test_loss_avg=0.37913 acc=1.00000 test_acc_avg=0.89270 test_acc_top5_avg=0.98952 time=881.34it/s +curr_acc 0.8927 +BEST_ACC 0.9017 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=2.02523 loss_avg=2.09489 acc=0.81250 acc_top1_avg=0.80377 acc_top5_avg=0.94164 lr=0.00010 gn=26.46604 time=50.24it/s +epoch=113 global_step=44250 loss=1.93401 loss_avg=2.02071 acc=0.82812 acc_top1_avg=0.81145 acc_top5_avg=0.94415 lr=0.00010 gn=21.45123 time=55.96it/s +epoch=113 global_step=44300 loss=2.12033 loss_avg=2.00707 acc=0.80469 acc_top1_avg=0.81297 acc_top5_avg=0.94491 lr=0.00010 gn=21.17787 time=56.85it/s +epoch=113 global_step=44350 loss=1.99061 loss_avg=2.01295 acc=0.81250 acc_top1_avg=0.81217 acc_top5_avg=0.94419 lr=0.00010 gn=20.57192 time=56.77it/s +epoch=113 global_step=44400 loss=2.14444 loss_avg=2.01420 acc=0.79688 acc_top1_avg=0.81236 acc_top5_avg=0.94384 lr=0.00010 gn=22.87813 time=55.67it/s +epoch=113 global_step=44450 loss=1.62808 loss_avg=2.00230 acc=0.85156 acc_top1_avg=0.81382 acc_top5_avg=0.94473 lr=0.00010 gn=18.57608 time=51.50it/s +epoch=113 global_step=44500 loss=2.63372 loss_avg=1.99888 acc=0.75000 acc_top1_avg=0.81390 acc_top5_avg=0.94539 lr=0.00010 gn=22.05374 time=55.03it/s +epoch=113 global_step=44550 loss=1.78717 loss_avg=1.99641 acc=0.82031 acc_top1_avg=0.81418 acc_top5_avg=0.94565 lr=0.00010 gn=17.56110 time=57.74it/s +====================Eval==================== +epoch=113 global_step=44574 loss=0.68938 test_loss_avg=0.46623 acc=0.81250 test_acc_avg=0.86413 test_acc_top5_avg=0.98845 time=230.63it/s +epoch=113 global_step=44574 loss=0.32365 test_loss_avg=0.39491 acc=0.92188 test_acc_avg=0.88859 test_acc_top5_avg=0.98951 time=239.56it/s +epoch=113 global_step=44574 loss=0.07549 test_loss_avg=0.38121 acc=1.00000 test_acc_avg=0.89231 test_acc_top5_avg=0.98991 time=898.14it/s +curr_acc 0.8923 +BEST_ACC 0.9017 +curr_acc_top5 0.9899 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=1.67237 loss_avg=2.08232 acc=0.84375 acc_top1_avg=0.80409 acc_top5_avg=0.94020 lr=0.00010 gn=20.15532 time=53.61it/s +epoch=114 global_step=44650 loss=2.44720 loss_avg=2.02863 acc=0.77344 acc_top1_avg=0.81065 acc_top5_avg=0.94171 lr=0.00010 gn=24.17343 time=57.04it/s +epoch=114 global_step=44700 loss=2.24112 loss_avg=2.03169 acc=0.80469 acc_top1_avg=0.81021 acc_top5_avg=0.94184 lr=0.00010 gn=28.36491 time=59.22it/s +epoch=114 global_step=44750 loss=2.34590 loss_avg=2.00614 acc=0.77344 acc_top1_avg=0.81268 acc_top5_avg=0.94136 lr=0.00010 gn=24.45658 time=52.58it/s +epoch=114 global_step=44800 loss=2.01605 loss_avg=2.00212 acc=0.81250 acc_top1_avg=0.81340 acc_top5_avg=0.94310 lr=0.00010 gn=22.97535 time=55.94it/s +epoch=114 global_step=44850 loss=2.20192 loss_avg=1.99054 acc=0.78906 acc_top1_avg=0.81474 acc_top5_avg=0.94339 lr=0.00010 gn=19.26512 time=49.61it/s +epoch=114 global_step=44900 loss=1.79487 loss_avg=1.98246 acc=0.83594 acc_top1_avg=0.81545 acc_top5_avg=0.94378 lr=0.00010 gn=22.63247 time=57.75it/s +epoch=114 global_step=44950 loss=1.94860 loss_avg=1.97601 acc=0.82812 acc_top1_avg=0.81620 acc_top5_avg=0.94454 lr=0.00010 gn=25.26684 time=48.43it/s +====================Eval==================== +epoch=114 global_step=44965 loss=0.70650 test_loss_avg=0.47979 acc=0.80469 test_acc_avg=0.86452 test_acc_top5_avg=0.98704 time=236.03it/s +epoch=114 global_step=44965 loss=0.09289 test_loss_avg=0.37271 acc=1.00000 test_acc_avg=0.89527 test_acc_top5_avg=0.99021 time=887.50it/s +curr_acc 0.8953 +BEST_ACC 0.9017 +curr_acc_top5 0.9902 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=1.44712 loss_avg=1.95735 acc=0.88281 acc_top1_avg=0.82009 acc_top5_avg=0.95022 lr=0.00010 gn=23.97065 time=59.41it/s +epoch=115 global_step=45050 loss=2.15765 loss_avg=1.95931 acc=0.79688 acc_top1_avg=0.81976 acc_top5_avg=0.94761 lr=0.00010 gn=19.77381 time=56.53it/s +epoch=115 global_step=45100 loss=1.69006 loss_avg=1.96933 acc=0.84375 acc_top1_avg=0.81817 acc_top5_avg=0.94867 lr=0.00010 gn=23.08083 time=58.41it/s +epoch=115 global_step=45150 loss=1.84632 loss_avg=1.96734 acc=0.82812 acc_top1_avg=0.81820 acc_top5_avg=0.94679 lr=0.00010 gn=13.61662 time=55.37it/s +epoch=115 global_step=45200 loss=1.65156 loss_avg=1.97845 acc=0.84375 acc_top1_avg=0.81676 acc_top5_avg=0.94584 lr=0.00010 gn=19.86561 time=54.64it/s +epoch=115 global_step=45250 loss=1.54388 loss_avg=1.98777 acc=0.86719 acc_top1_avg=0.81557 acc_top5_avg=0.94578 lr=0.00010 gn=23.02155 time=51.53it/s +epoch=115 global_step=45300 loss=1.92933 loss_avg=1.97768 acc=0.81250 acc_top1_avg=0.81623 acc_top5_avg=0.94569 lr=0.00010 gn=23.18717 time=56.67it/s +epoch=115 global_step=45350 loss=2.29463 loss_avg=1.98120 acc=0.79688 acc_top1_avg=0.81579 acc_top5_avg=0.94621 lr=0.00010 gn=31.68166 time=56.71it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.21410 test_loss_avg=0.34662 acc=0.92969 test_acc_avg=0.89635 test_acc_top5_avg=0.99219 time=241.26it/s +epoch=115 global_step=45356 loss=0.11200 test_loss_avg=0.41148 acc=0.96875 test_acc_avg=0.88522 test_acc_top5_avg=0.98786 time=242.91it/s +epoch=115 global_step=45356 loss=0.11879 test_loss_avg=0.38176 acc=1.00000 test_acc_avg=0.89349 test_acc_top5_avg=0.98892 time=904.72it/s +curr_acc 0.8935 +BEST_ACC 0.9017 +curr_acc_top5 0.9889 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=1.82198 loss_avg=1.94295 acc=0.83594 acc_top1_avg=0.82049 acc_top5_avg=0.95188 lr=0.00010 gn=23.11887 time=56.75it/s +epoch=116 global_step=45450 loss=1.64935 loss_avg=1.95626 acc=0.85156 acc_top1_avg=0.81981 acc_top5_avg=0.94806 lr=0.00010 gn=19.50758 time=56.47it/s +epoch=116 global_step=45500 loss=2.04291 loss_avg=1.94322 acc=0.80469 acc_top1_avg=0.82107 acc_top5_avg=0.94759 lr=0.00010 gn=21.97840 time=56.63it/s +epoch=116 global_step=45550 loss=1.95100 loss_avg=1.97213 acc=0.82812 acc_top1_avg=0.81765 acc_top5_avg=0.94636 lr=0.00010 gn=26.64445 time=56.66it/s +epoch=116 global_step=45600 loss=1.82377 loss_avg=1.96731 acc=0.84375 acc_top1_avg=0.81839 acc_top5_avg=0.94749 lr=0.00010 gn=25.28039 time=54.35it/s +epoch=116 global_step=45650 loss=2.11099 loss_avg=1.96999 acc=0.80469 acc_top1_avg=0.81792 acc_top5_avg=0.94701 lr=0.00010 gn=22.63018 time=56.18it/s +epoch=116 global_step=45700 loss=1.72142 loss_avg=1.97775 acc=0.85156 acc_top1_avg=0.81709 acc_top5_avg=0.94570 lr=0.00010 gn=24.08021 time=56.02it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.20818 test_loss_avg=0.48907 acc=0.91406 test_acc_avg=0.85916 test_acc_top5_avg=0.98546 time=240.02it/s +epoch=116 global_step=45747 loss=0.08363 test_loss_avg=0.38235 acc=1.00000 test_acc_avg=0.89062 test_acc_top5_avg=0.98883 time=872.36it/s +curr_acc 0.8906 +BEST_ACC 0.9017 +curr_acc_top5 0.9888 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=1.77810 loss_avg=2.12400 acc=0.82812 acc_top1_avg=0.79427 acc_top5_avg=0.93229 lr=0.00010 gn=21.16781 time=54.39it/s +epoch=117 global_step=45800 loss=2.57585 loss_avg=2.00215 acc=0.75781 acc_top1_avg=0.81338 acc_top5_avg=0.94546 lr=0.00010 gn=22.27434 time=56.83it/s +epoch=117 global_step=45850 loss=2.67273 loss_avg=1.97278 acc=0.74219 acc_top1_avg=0.81690 acc_top5_avg=0.94675 lr=0.00010 gn=25.12491 time=54.72it/s +epoch=117 global_step=45900 loss=2.19326 loss_avg=1.98416 acc=0.78906 acc_top1_avg=0.81567 acc_top5_avg=0.94567 lr=0.00010 gn=11.92933 time=56.52it/s +epoch=117 global_step=45950 loss=1.85729 loss_avg=1.98488 acc=0.82812 acc_top1_avg=0.81558 acc_top5_avg=0.94581 lr=0.00010 gn=26.35843 time=51.05it/s +epoch=117 global_step=46000 loss=2.14387 loss_avg=1.98030 acc=0.80469 acc_top1_avg=0.81599 acc_top5_avg=0.94556 lr=0.00010 gn=23.20958 time=56.77it/s +epoch=117 global_step=46050 loss=1.64507 loss_avg=1.97641 acc=0.84375 acc_top1_avg=0.81650 acc_top5_avg=0.94573 lr=0.00010 gn=16.66711 time=56.92it/s +epoch=117 global_step=46100 loss=2.14429 loss_avg=1.97104 acc=0.81250 acc_top1_avg=0.81710 acc_top5_avg=0.94600 lr=0.00010 gn=28.58183 time=56.12it/s +====================Eval==================== +epoch=117 global_step=46138 loss=0.48820 test_loss_avg=0.56902 acc=0.85156 test_acc_avg=0.83036 test_acc_top5_avg=0.98772 time=223.66it/s +epoch=117 global_step=46138 loss=0.23833 test_loss_avg=0.45304 acc=0.95312 test_acc_avg=0.87198 test_acc_top5_avg=0.98629 time=242.87it/s +epoch=117 global_step=46138 loss=0.06591 test_loss_avg=0.38490 acc=1.00000 test_acc_avg=0.89161 test_acc_top5_avg=0.98902 time=893.17it/s +curr_acc 0.8916 +BEST_ACC 0.9017 +curr_acc_top5 0.9890 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=2.19305 loss_avg=2.00980 acc=0.79688 acc_top1_avg=0.81706 acc_top5_avg=0.94531 lr=0.00010 gn=27.51946 time=56.22it/s +epoch=118 global_step=46200 loss=1.56287 loss_avg=1.96877 acc=0.85938 acc_top1_avg=0.81817 acc_top5_avg=0.94367 lr=0.00010 gn=23.09384 time=53.97it/s +epoch=118 global_step=46250 loss=1.37108 loss_avg=1.96375 acc=0.87500 acc_top1_avg=0.81822 acc_top5_avg=0.94510 lr=0.00010 gn=24.32085 time=56.47it/s +epoch=118 global_step=46300 loss=1.77974 loss_avg=1.96045 acc=0.84375 acc_top1_avg=0.81848 acc_top5_avg=0.94546 lr=0.00010 gn=26.53147 time=56.42it/s +epoch=118 global_step=46350 loss=1.28188 loss_avg=1.96101 acc=0.88281 acc_top1_avg=0.81799 acc_top5_avg=0.94446 lr=0.00010 gn=17.33956 time=56.49it/s +epoch=118 global_step=46400 loss=1.42537 loss_avg=1.97958 acc=0.87500 acc_top1_avg=0.81614 acc_top5_avg=0.94427 lr=0.00010 gn=21.29661 time=52.65it/s +epoch=118 global_step=46450 loss=1.48513 loss_avg=1.97478 acc=0.85938 acc_top1_avg=0.81653 acc_top5_avg=0.94454 lr=0.00010 gn=25.59229 time=55.94it/s +epoch=118 global_step=46500 loss=1.81220 loss_avg=1.97318 acc=0.84375 acc_top1_avg=0.81703 acc_top5_avg=0.94505 lr=0.00010 gn=26.30465 time=55.58it/s +====================Eval==================== +epoch=118 global_step=46529 loss=0.51733 test_loss_avg=0.51824 acc=0.86719 test_acc_avg=0.85212 test_acc_top5_avg=0.98298 time=230.58it/s +epoch=118 global_step=46529 loss=0.21040 test_loss_avg=0.38907 acc=0.94531 test_acc_avg=0.88942 test_acc_top5_avg=0.98828 time=238.26it/s +epoch=118 global_step=46529 loss=0.07058 test_loss_avg=0.38504 acc=1.00000 test_acc_avg=0.89082 test_acc_top5_avg=0.98843 time=869.83it/s +curr_acc 0.8908 +BEST_ACC 0.9017 +curr_acc_top5 0.9884 +BEST_ACC_top5 0.9940 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=1.28687 loss_avg=2.04375 acc=0.89062 acc_top1_avg=0.80692 acc_top5_avg=0.93192 lr=0.00010 gn=18.68176 time=54.02it/s +epoch=119 global_step=46600 loss=2.19377 loss_avg=1.96463 acc=0.78906 acc_top1_avg=0.81624 acc_top5_avg=0.94036 lr=0.00010 gn=16.29408 time=54.09it/s +epoch=119 global_step=46650 loss=2.25731 loss_avg=1.95386 acc=0.78906 acc_top1_avg=0.81792 acc_top5_avg=0.94338 lr=0.00010 gn=21.22129 time=51.63it/s +epoch=119 global_step=46700 loss=2.00257 loss_avg=1.96370 acc=0.82031 acc_top1_avg=0.81734 acc_top5_avg=0.94449 lr=0.00010 gn=22.18011 time=57.22it/s +epoch=119 global_step=46750 loss=2.13593 loss_avg=1.98133 acc=0.81250 acc_top1_avg=0.81519 acc_top5_avg=0.94326 lr=0.00010 gn=27.44960 time=57.68it/s +epoch=119 global_step=46800 loss=2.19132 loss_avg=1.97753 acc=0.78906 acc_top1_avg=0.81567 acc_top5_avg=0.94373 lr=0.00010 gn=19.15168 time=52.81it/s +epoch=119 global_step=46850 loss=2.09752 loss_avg=1.98017 acc=0.80469 acc_top1_avg=0.81549 acc_top5_avg=0.94346 lr=0.00010 gn=18.28752 time=54.56it/s +epoch=119 global_step=46900 loss=2.02662 loss_avg=1.96607 acc=0.80469 acc_top1_avg=0.81699 acc_top5_avg=0.94436 lr=0.00010 gn=22.29665 time=55.14it/s +====================Eval==================== +epoch=119 global_step=46920 loss=0.12676 test_loss_avg=0.49347 acc=0.95312 test_acc_avg=0.85890 test_acc_top5_avg=0.98533 time=247.00it/s +epoch=119 global_step=46920 loss=0.08803 test_loss_avg=0.39213 acc=1.00000 test_acc_avg=0.88776 test_acc_top5_avg=0.98883 time=921.22it/s +curr_acc 0.8878 +BEST_ACC 0.9017 +curr_acc_top5 0.9888 +BEST_ACC_top5 0.9940 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_4_4.log b/other_methods/sceloss/sceloss_results/out_4_4.log new file mode 100644 index 0000000..bdb09f8 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_4_4.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.4__noise_amount__0.4.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.36307 loss_avg=7.35034 acc=0.25781 acc_top1_avg=0.25688 acc_top5_avg=0.69766 lr=0.01000 gn=6.28846 time=58.01it/s +epoch=0 global_step=100 loss=7.35833 loss_avg=7.13337 acc=0.23438 acc_top1_avg=0.27984 acc_top5_avg=0.72164 lr=0.01000 gn=5.55079 time=52.09it/s +epoch=0 global_step=150 loss=6.18043 loss_avg=7.01823 acc=0.39844 acc_top1_avg=0.29167 acc_top5_avg=0.73375 lr=0.01000 gn=4.85109 time=56.83it/s +epoch=0 global_step=200 loss=6.89550 loss_avg=6.93384 acc=0.31250 acc_top1_avg=0.30066 acc_top5_avg=0.74465 lr=0.01000 gn=4.06699 time=57.36it/s +epoch=0 global_step=250 loss=6.27402 loss_avg=6.85714 acc=0.35938 acc_top1_avg=0.30850 acc_top5_avg=0.75216 lr=0.01000 gn=4.88751 time=61.57it/s +epoch=0 global_step=300 loss=6.90660 loss_avg=6.80188 acc=0.28906 acc_top1_avg=0.31401 acc_top5_avg=0.75703 lr=0.01000 gn=4.32451 time=54.33it/s +epoch=0 global_step=350 loss=6.46476 loss_avg=6.75517 acc=0.34375 acc_top1_avg=0.31884 acc_top5_avg=0.76080 lr=0.01000 gn=4.93247 time=60.32it/s +====================Eval==================== +epoch=0 global_step=391 loss=4.51841 test_loss_avg=3.51425 acc=0.00000 test_acc_avg=0.28453 test_acc_top5_avg=0.75125 time=235.33it/s +epoch=0 global_step=391 loss=4.95165 test_loss_avg=3.29376 acc=0.31250 test_acc_avg=0.33089 test_acc_top5_avg=0.78778 time=33.52it/s +curr_acc 0.3309 +BEST_ACC 0.0000 +curr_acc_top5 0.7878 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.17326 loss_avg=6.14984 acc=0.36719 acc_top1_avg=0.37413 acc_top5_avg=0.80208 lr=0.01000 gn=3.88440 time=57.32it/s +epoch=1 global_step=450 loss=6.30294 loss_avg=6.25894 acc=0.35938 acc_top1_avg=0.36878 acc_top5_avg=0.79767 lr=0.01000 gn=4.71955 time=58.88it/s +epoch=1 global_step=500 loss=6.38161 loss_avg=6.24790 acc=0.33594 acc_top1_avg=0.37070 acc_top5_avg=0.79831 lr=0.01000 gn=3.26946 time=55.53it/s +epoch=1 global_step=550 loss=6.92147 loss_avg=6.25330 acc=0.28125 acc_top1_avg=0.36969 acc_top5_avg=0.79884 lr=0.01000 gn=3.69155 time=63.64it/s +epoch=1 global_step=600 loss=5.62771 loss_avg=6.22778 acc=0.42969 acc_top1_avg=0.37257 acc_top5_avg=0.80155 lr=0.01000 gn=5.39647 time=53.57it/s +epoch=1 global_step=650 loss=5.98499 loss_avg=6.19686 acc=0.42188 acc_top1_avg=0.37663 acc_top5_avg=0.80369 lr=0.01000 gn=4.44258 time=57.29it/s +epoch=1 global_step=700 loss=5.89652 loss_avg=6.17215 acc=0.41406 acc_top1_avg=0.37874 acc_top5_avg=0.80497 lr=0.01000 gn=4.36612 time=54.93it/s +epoch=1 global_step=750 loss=6.40165 loss_avg=6.14951 acc=0.33594 acc_top1_avg=0.38125 acc_top5_avg=0.80641 lr=0.01000 gn=3.93362 time=57.64it/s +====================Eval==================== +epoch=1 global_step=782 loss=4.79714 test_loss_avg=3.32797 acc=0.00000 test_acc_avg=0.30208 test_acc_top5_avg=0.89323 time=226.40it/s +epoch=1 global_step=782 loss=0.39628 test_loss_avg=2.79637 acc=0.92969 test_acc_avg=0.41329 test_acc_top5_avg=0.88006 time=239.13it/s +epoch=1 global_step=782 loss=0.76218 test_loss_avg=2.57570 acc=0.87500 test_acc_avg=0.45807 test_acc_top5_avg=0.89152 time=393.17it/s +curr_acc 0.4581 +BEST_ACC 0.3309 +curr_acc_top5 0.8915 +BEST_ACC_top5 0.7878 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=6.66264 loss_avg=6.09833 acc=0.32031 acc_top1_avg=0.38108 acc_top5_avg=0.81424 lr=0.01000 gn=4.37599 time=62.38it/s +epoch=2 global_step=850 loss=4.76067 loss_avg=5.93644 acc=0.54688 acc_top1_avg=0.40188 acc_top5_avg=0.81997 lr=0.01000 gn=4.38788 time=55.16it/s +epoch=2 global_step=900 loss=6.43068 loss_avg=5.93892 acc=0.35156 acc_top1_avg=0.40301 acc_top5_avg=0.82018 lr=0.01000 gn=4.17086 time=56.75it/s +epoch=2 global_step=950 loss=5.87253 loss_avg=5.93595 acc=0.41406 acc_top1_avg=0.40272 acc_top5_avg=0.82259 lr=0.01000 gn=4.17694 time=55.34it/s +epoch=2 global_step=1000 loss=5.80802 loss_avg=5.91874 acc=0.41406 acc_top1_avg=0.40457 acc_top5_avg=0.82196 lr=0.01000 gn=4.51077 time=59.97it/s +epoch=2 global_step=1050 loss=5.82794 loss_avg=5.90129 acc=0.41406 acc_top1_avg=0.40608 acc_top5_avg=0.82256 lr=0.01000 gn=4.39107 time=57.65it/s +epoch=2 global_step=1100 loss=5.45522 loss_avg=5.89382 acc=0.45312 acc_top1_avg=0.40662 acc_top5_avg=0.82329 lr=0.01000 gn=4.02178 time=54.11it/s +epoch=2 global_step=1150 loss=5.14726 loss_avg=5.86577 acc=0.49219 acc_top1_avg=0.41024 acc_top5_avg=0.82456 lr=0.01000 gn=3.87169 time=56.10it/s +====================Eval==================== +epoch=2 global_step=1173 loss=2.51856 test_loss_avg=2.30186 acc=0.37500 test_acc_avg=0.46540 test_acc_top5_avg=0.83761 time=243.18it/s +epoch=2 global_step=1173 loss=0.57903 test_loss_avg=2.22682 acc=0.87500 test_acc_avg=0.47389 test_acc_top5_avg=0.86946 time=432.45it/s +curr_acc 0.4739 +BEST_ACC 0.4581 +curr_acc_top5 0.8695 +BEST_ACC_top5 0.8915 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=5.93170 loss_avg=5.77686 acc=0.41406 acc_top1_avg=0.42535 acc_top5_avg=0.82986 lr=0.01000 gn=4.18410 time=59.01it/s +epoch=3 global_step=1250 loss=5.77774 loss_avg=5.75898 acc=0.43750 acc_top1_avg=0.42543 acc_top5_avg=0.83026 lr=0.01000 gn=4.37814 time=53.97it/s +epoch=3 global_step=1300 loss=5.49480 loss_avg=5.76109 acc=0.43750 acc_top1_avg=0.42347 acc_top5_avg=0.82911 lr=0.01000 gn=4.53922 time=57.43it/s +epoch=3 global_step=1350 loss=5.33853 loss_avg=5.74807 acc=0.46875 acc_top1_avg=0.42457 acc_top5_avg=0.82980 lr=0.01000 gn=4.29116 time=57.44it/s +epoch=3 global_step=1400 loss=5.81935 loss_avg=5.72282 acc=0.42188 acc_top1_avg=0.42755 acc_top5_avg=0.82912 lr=0.01000 gn=4.42021 time=56.54it/s +epoch=3 global_step=1450 loss=5.67483 loss_avg=5.70708 acc=0.44531 acc_top1_avg=0.42910 acc_top5_avg=0.83123 lr=0.01000 gn=4.86323 time=61.17it/s +epoch=3 global_step=1500 loss=5.02584 loss_avg=5.67726 acc=0.50000 acc_top1_avg=0.43205 acc_top5_avg=0.83211 lr=0.01000 gn=4.89694 time=57.21it/s +epoch=3 global_step=1550 loss=5.66083 loss_avg=5.67028 acc=0.42188 acc_top1_avg=0.43307 acc_top5_avg=0.83293 lr=0.01000 gn=4.91424 time=65.14it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.90821 test_loss_avg=1.62452 acc=0.77344 test_acc_avg=0.59375 test_acc_top5_avg=0.93690 time=244.61it/s +epoch=3 global_step=1564 loss=0.65278 test_loss_avg=2.06000 acc=0.81250 test_acc_avg=0.47830 test_acc_top5_avg=0.91456 time=239.18it/s +epoch=3 global_step=1564 loss=0.18783 test_loss_avg=1.73496 acc=0.93750 test_acc_avg=0.55815 test_acc_top5_avg=0.92613 time=479.62it/s +curr_acc 0.5581 +BEST_ACC 0.4739 +curr_acc_top5 0.9261 +BEST_ACC_top5 0.8915 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=5.69697 loss_avg=5.57843 acc=0.42969 acc_top1_avg=0.44119 acc_top5_avg=0.83832 lr=0.01000 gn=3.86795 time=54.80it/s +epoch=4 global_step=1650 loss=5.55111 loss_avg=5.59815 acc=0.43750 acc_top1_avg=0.43805 acc_top5_avg=0.83348 lr=0.01000 gn=4.18716 time=57.52it/s +epoch=4 global_step=1700 loss=5.73383 loss_avg=5.57595 acc=0.43750 acc_top1_avg=0.44112 acc_top5_avg=0.83852 lr=0.01000 gn=5.15052 time=55.15it/s +epoch=4 global_step=1750 loss=5.91358 loss_avg=5.55897 acc=0.39062 acc_top1_avg=0.44359 acc_top5_avg=0.83951 lr=0.01000 gn=4.15790 time=56.34it/s +epoch=4 global_step=1800 loss=5.08421 loss_avg=5.56120 acc=0.49219 acc_top1_avg=0.44316 acc_top5_avg=0.84011 lr=0.01000 gn=4.45840 time=57.47it/s +epoch=4 global_step=1850 loss=5.47210 loss_avg=5.55446 acc=0.46094 acc_top1_avg=0.44403 acc_top5_avg=0.84173 lr=0.01000 gn=3.93354 time=54.15it/s +epoch=4 global_step=1900 loss=5.56165 loss_avg=5.56102 acc=0.44531 acc_top1_avg=0.44343 acc_top5_avg=0.84022 lr=0.01000 gn=3.90558 time=56.63it/s +epoch=4 global_step=1950 loss=5.55848 loss_avg=5.55701 acc=0.45312 acc_top1_avg=0.44410 acc_top5_avg=0.84082 lr=0.01000 gn=4.64454 time=62.99it/s +====================Eval==================== +epoch=4 global_step=1955 loss=0.42995 test_loss_avg=2.06296 acc=0.85156 test_acc_avg=0.50322 test_acc_top5_avg=0.85409 time=198.96it/s +epoch=4 global_step=1955 loss=0.30814 test_loss_avg=1.64309 acc=0.87500 test_acc_avg=0.57842 test_acc_top5_avg=0.91357 time=857.91it/s +curr_acc 0.5784 +BEST_ACC 0.5581 +curr_acc_top5 0.9136 +BEST_ACC_top5 0.9261 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=4.98258 loss_avg=5.44874 acc=0.50781 acc_top1_avg=0.45642 acc_top5_avg=0.83229 lr=0.01000 gn=4.98264 time=61.10it/s +epoch=5 global_step=2050 loss=5.39809 loss_avg=5.43045 acc=0.44531 acc_top1_avg=0.45658 acc_top5_avg=0.83725 lr=0.01000 gn=3.50006 time=60.88it/s +epoch=5 global_step=2100 loss=5.60057 loss_avg=5.46284 acc=0.42188 acc_top1_avg=0.45318 acc_top5_avg=0.83992 lr=0.01000 gn=3.82728 time=55.50it/s +epoch=5 global_step=2150 loss=5.38017 loss_avg=5.42054 acc=0.46094 acc_top1_avg=0.45777 acc_top5_avg=0.84287 lr=0.01000 gn=4.74441 time=57.66it/s +epoch=5 global_step=2200 loss=5.85294 loss_avg=5.43115 acc=0.39844 acc_top1_avg=0.45644 acc_top5_avg=0.84359 lr=0.01000 gn=4.02683 time=56.94it/s +epoch=5 global_step=2250 loss=5.33233 loss_avg=5.43869 acc=0.49219 acc_top1_avg=0.45564 acc_top5_avg=0.84309 lr=0.01000 gn=4.77268 time=56.72it/s +epoch=5 global_step=2300 loss=4.82884 loss_avg=5.44588 acc=0.52344 acc_top1_avg=0.45485 acc_top5_avg=0.84438 lr=0.01000 gn=3.58040 time=61.21it/s +====================Eval==================== +epoch=5 global_step=2346 loss=2.16799 test_loss_avg=1.96521 acc=0.54688 test_acc_avg=0.54375 test_acc_top5_avg=0.98438 time=244.85it/s +epoch=5 global_step=2346 loss=2.86402 test_loss_avg=2.50348 acc=0.22656 test_acc_avg=0.41207 test_acc_top5_avg=0.85313 time=230.60it/s +epoch=5 global_step=2346 loss=2.25777 test_loss_avg=2.10830 acc=0.56250 test_acc_avg=0.50821 test_acc_top5_avg=0.87896 time=683.00it/s +curr_acc 0.5082 +BEST_ACC 0.5784 +curr_acc_top5 0.8790 +BEST_ACC_top5 0.9261 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=5.46667 loss_avg=5.68158 acc=0.42969 acc_top1_avg=0.42188 acc_top5_avg=0.83594 lr=0.01000 gn=4.92907 time=26.84it/s +epoch=6 global_step=2400 loss=5.91671 loss_avg=5.44309 acc=0.39844 acc_top1_avg=0.45530 acc_top5_avg=0.84635 lr=0.01000 gn=3.95157 time=38.45it/s +epoch=6 global_step=2450 loss=5.57804 loss_avg=5.45955 acc=0.44531 acc_top1_avg=0.45328 acc_top5_avg=0.84367 lr=0.01000 gn=4.72302 time=56.47it/s +epoch=6 global_step=2500 loss=5.74967 loss_avg=5.45280 acc=0.42969 acc_top1_avg=0.45399 acc_top5_avg=0.84705 lr=0.01000 gn=5.36422 time=56.35it/s +epoch=6 global_step=2550 loss=5.91094 loss_avg=5.45316 acc=0.41406 acc_top1_avg=0.45416 acc_top5_avg=0.84658 lr=0.01000 gn=4.44160 time=61.71it/s +epoch=6 global_step=2600 loss=5.07664 loss_avg=5.45710 acc=0.51562 acc_top1_avg=0.45414 acc_top5_avg=0.84646 lr=0.01000 gn=4.57972 time=56.26it/s +epoch=6 global_step=2650 loss=5.58716 loss_avg=5.44333 acc=0.44531 acc_top1_avg=0.45575 acc_top5_avg=0.84673 lr=0.01000 gn=4.55692 time=54.27it/s +epoch=6 global_step=2700 loss=5.25594 loss_avg=5.43896 acc=0.48438 acc_top1_avg=0.45630 acc_top5_avg=0.84611 lr=0.01000 gn=5.05773 time=53.11it/s +====================Eval==================== +epoch=6 global_step=2737 loss=4.61546 test_loss_avg=1.41381 acc=0.00000 test_acc_avg=0.65054 test_acc_top5_avg=0.92969 time=242.80it/s +epoch=6 global_step=2737 loss=0.37260 test_loss_avg=1.62315 acc=0.87500 test_acc_avg=0.59766 test_acc_top5_avg=0.91869 time=239.63it/s +epoch=6 global_step=2737 loss=0.37153 test_loss_avg=1.57578 acc=0.87500 test_acc_avg=0.60848 test_acc_top5_avg=0.92178 time=453.59it/s +curr_acc 0.6085 +BEST_ACC 0.5784 +curr_acc_top5 0.9218 +BEST_ACC_top5 0.9261 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=5.72117 loss_avg=5.55253 acc=0.42969 acc_top1_avg=0.44351 acc_top5_avg=0.83954 lr=0.01000 gn=5.10041 time=49.14it/s +epoch=7 global_step=2800 loss=5.81955 loss_avg=5.33913 acc=0.41406 acc_top1_avg=0.46478 acc_top5_avg=0.84772 lr=0.01000 gn=4.79952 time=52.21it/s +epoch=7 global_step=2850 loss=5.20627 loss_avg=5.36854 acc=0.47656 acc_top1_avg=0.46108 acc_top5_avg=0.84472 lr=0.01000 gn=4.46966 time=56.80it/s +epoch=7 global_step=2900 loss=5.49705 loss_avg=5.34525 acc=0.47656 acc_top1_avg=0.46472 acc_top5_avg=0.84730 lr=0.01000 gn=5.05476 time=60.16it/s +epoch=7 global_step=2950 loss=5.41892 loss_avg=5.37460 acc=0.46094 acc_top1_avg=0.46163 acc_top5_avg=0.84599 lr=0.01000 gn=5.32315 time=56.67it/s +epoch=7 global_step=3000 loss=5.09850 loss_avg=5.36108 acc=0.49219 acc_top1_avg=0.46352 acc_top5_avg=0.84601 lr=0.01000 gn=4.39846 time=52.66it/s +epoch=7 global_step=3050 loss=5.56686 loss_avg=5.35698 acc=0.44531 acc_top1_avg=0.46436 acc_top5_avg=0.84724 lr=0.01000 gn=5.18004 time=58.17it/s +epoch=7 global_step=3100 loss=5.37963 loss_avg=5.36160 acc=0.46875 acc_top1_avg=0.46380 acc_top5_avg=0.84691 lr=0.01000 gn=5.27366 time=55.06it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.98519 test_loss_avg=3.06875 acc=0.50000 test_acc_avg=0.42636 test_acc_top5_avg=0.90293 time=237.99it/s +epoch=7 global_step=3128 loss=2.87615 test_loss_avg=2.70737 acc=0.43750 test_acc_avg=0.47330 test_acc_top5_avg=0.91129 time=846.65it/s +curr_acc 0.4733 +BEST_ACC 0.6085 +curr_acc_top5 0.9113 +BEST_ACC_top5 0.9261 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=5.34450 loss_avg=5.26156 acc=0.46094 acc_top1_avg=0.47124 acc_top5_avg=0.85334 lr=0.01000 gn=4.18410 time=53.76it/s +epoch=8 global_step=3200 loss=5.04489 loss_avg=5.30061 acc=0.48438 acc_top1_avg=0.46918 acc_top5_avg=0.85156 lr=0.01000 gn=5.47475 time=57.17it/s +epoch=8 global_step=3250 loss=5.21130 loss_avg=5.35187 acc=0.46875 acc_top1_avg=0.46382 acc_top5_avg=0.85073 lr=0.01000 gn=5.43944 time=63.14it/s +epoch=8 global_step=3300 loss=4.93969 loss_avg=5.33707 acc=0.53125 acc_top1_avg=0.46652 acc_top5_avg=0.85170 lr=0.01000 gn=5.70746 time=51.62it/s +epoch=8 global_step=3350 loss=5.53441 loss_avg=5.34574 acc=0.45312 acc_top1_avg=0.46548 acc_top5_avg=0.85037 lr=0.01000 gn=6.11232 time=55.94it/s +epoch=8 global_step=3400 loss=5.59145 loss_avg=5.34196 acc=0.42969 acc_top1_avg=0.46573 acc_top5_avg=0.85004 lr=0.01000 gn=5.51908 time=55.53it/s +epoch=8 global_step=3450 loss=4.82918 loss_avg=5.34347 acc=0.52344 acc_top1_avg=0.46581 acc_top5_avg=0.84999 lr=0.01000 gn=5.79223 time=53.50it/s +epoch=8 global_step=3500 loss=5.24567 loss_avg=5.34276 acc=0.50781 acc_top1_avg=0.46579 acc_top5_avg=0.85020 lr=0.01000 gn=5.36140 time=59.27it/s +====================Eval==================== +epoch=8 global_step=3519 loss=2.80824 test_loss_avg=1.23828 acc=0.42188 test_acc_avg=0.68533 test_acc_top5_avg=0.93533 time=234.58it/s +epoch=8 global_step=3519 loss=0.65483 test_loss_avg=1.80889 acc=0.84375 test_acc_avg=0.55974 test_acc_top5_avg=0.92544 time=236.13it/s +epoch=8 global_step=3519 loss=0.21452 test_loss_avg=1.62211 acc=0.93750 test_acc_avg=0.60394 test_acc_top5_avg=0.93354 time=832.20it/s +curr_acc 0.6039 +BEST_ACC 0.6085 +curr_acc_top5 0.9335 +BEST_ACC_top5 0.9261 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=5.85790 loss_avg=5.38271 acc=0.40625 acc_top1_avg=0.46043 acc_top5_avg=0.84652 lr=0.01000 gn=4.02885 time=49.99it/s +epoch=9 global_step=3600 loss=5.65959 loss_avg=5.35645 acc=0.42969 acc_top1_avg=0.46460 acc_top5_avg=0.85002 lr=0.01000 gn=4.77953 time=60.22it/s +epoch=9 global_step=3650 loss=4.98334 loss_avg=5.36711 acc=0.50781 acc_top1_avg=0.46326 acc_top5_avg=0.85168 lr=0.01000 gn=4.86108 time=54.25it/s +epoch=9 global_step=3700 loss=5.08643 loss_avg=5.35338 acc=0.50781 acc_top1_avg=0.46569 acc_top5_avg=0.85350 lr=0.01000 gn=4.87624 time=63.94it/s +epoch=9 global_step=3750 loss=5.40467 loss_avg=5.33323 acc=0.45312 acc_top1_avg=0.46851 acc_top5_avg=0.85308 lr=0.01000 gn=5.16481 time=50.47it/s +epoch=9 global_step=3800 loss=5.03022 loss_avg=5.33072 acc=0.51562 acc_top1_avg=0.46844 acc_top5_avg=0.85226 lr=0.01000 gn=4.92874 time=60.34it/s +epoch=9 global_step=3850 loss=5.77010 loss_avg=5.32805 acc=0.40625 acc_top1_avg=0.46835 acc_top5_avg=0.85319 lr=0.01000 gn=4.97829 time=56.56it/s +epoch=9 global_step=3900 loss=5.55174 loss_avg=5.32515 acc=0.43750 acc_top1_avg=0.46811 acc_top5_avg=0.85220 lr=0.01000 gn=4.11748 time=56.80it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.98031 test_loss_avg=1.64075 acc=0.71094 test_acc_avg=0.57732 test_acc_top5_avg=0.90465 time=195.40it/s +epoch=9 global_step=3910 loss=0.61198 test_loss_avg=1.47847 acc=0.87500 test_acc_avg=0.60216 test_acc_top5_avg=0.93483 time=788.85it/s +curr_acc 0.6022 +BEST_ACC 0.6085 +curr_acc_top5 0.9348 +BEST_ACC_top5 0.9335 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=5.54881 loss_avg=5.38917 acc=0.43750 acc_top1_avg=0.45586 acc_top5_avg=0.85527 lr=0.01000 gn=4.06418 time=54.75it/s +epoch=10 global_step=4000 loss=4.83986 loss_avg=5.34053 acc=0.53125 acc_top1_avg=0.46380 acc_top5_avg=0.85148 lr=0.01000 gn=3.68902 time=58.06it/s +epoch=10 global_step=4050 loss=4.83593 loss_avg=5.31048 acc=0.50781 acc_top1_avg=0.46864 acc_top5_avg=0.85223 lr=0.01000 gn=4.66759 time=55.82it/s +epoch=10 global_step=4100 loss=5.02489 loss_avg=5.30874 acc=0.51562 acc_top1_avg=0.46937 acc_top5_avg=0.85082 lr=0.01000 gn=5.78514 time=52.96it/s +epoch=10 global_step=4150 loss=5.32433 loss_avg=5.32418 acc=0.46094 acc_top1_avg=0.46758 acc_top5_avg=0.85042 lr=0.01000 gn=4.37715 time=58.48it/s +epoch=10 global_step=4200 loss=4.81512 loss_avg=5.32089 acc=0.51562 acc_top1_avg=0.46835 acc_top5_avg=0.85140 lr=0.01000 gn=3.98504 time=63.44it/s +epoch=10 global_step=4250 loss=5.08670 loss_avg=5.32243 acc=0.47656 acc_top1_avg=0.46776 acc_top5_avg=0.85136 lr=0.01000 gn=5.35374 time=45.86it/s +epoch=10 global_step=4300 loss=4.90762 loss_avg=5.30929 acc=0.51562 acc_top1_avg=0.46943 acc_top5_avg=0.85158 lr=0.01000 gn=5.01462 time=62.63it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.66328 test_loss_avg=1.77391 acc=0.85938 test_acc_avg=0.58672 test_acc_top5_avg=0.95859 time=240.57it/s +epoch=10 global_step=4301 loss=0.71115 test_loss_avg=2.05706 acc=0.76562 test_acc_avg=0.51914 test_acc_top5_avg=0.90859 time=221.66it/s +epoch=10 global_step=4301 loss=0.59258 test_loss_avg=1.66184 acc=0.87500 test_acc_avg=0.60947 test_acc_top5_avg=0.92890 time=506.01it/s +curr_acc 0.6095 +BEST_ACC 0.6085 +curr_acc_top5 0.9289 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=5.61609 loss_avg=5.24341 acc=0.42969 acc_top1_avg=0.47688 acc_top5_avg=0.85698 lr=0.01000 gn=5.82272 time=47.97it/s +epoch=11 global_step=4400 loss=5.10204 loss_avg=5.26752 acc=0.48438 acc_top1_avg=0.47309 acc_top5_avg=0.85938 lr=0.01000 gn=3.69687 time=55.35it/s +epoch=11 global_step=4450 loss=5.68295 loss_avg=5.25895 acc=0.45312 acc_top1_avg=0.47462 acc_top5_avg=0.85681 lr=0.01000 gn=5.31043 time=58.60it/s +epoch=11 global_step=4500 loss=4.63400 loss_avg=5.25221 acc=0.56250 acc_top1_avg=0.47515 acc_top5_avg=0.85706 lr=0.01000 gn=5.54112 time=57.13it/s +epoch=11 global_step=4550 loss=6.00118 loss_avg=5.26650 acc=0.39062 acc_top1_avg=0.47386 acc_top5_avg=0.85501 lr=0.01000 gn=5.25200 time=53.73it/s +epoch=11 global_step=4600 loss=5.08342 loss_avg=5.26819 acc=0.50000 acc_top1_avg=0.47348 acc_top5_avg=0.85412 lr=0.01000 gn=5.41664 time=55.29it/s +epoch=11 global_step=4650 loss=5.62502 loss_avg=5.26832 acc=0.42188 acc_top1_avg=0.47363 acc_top5_avg=0.85376 lr=0.01000 gn=4.44570 time=56.80it/s +====================Eval==================== +epoch=11 global_step=4692 loss=5.08077 test_loss_avg=2.06051 acc=0.00000 test_acc_avg=0.55343 test_acc_top5_avg=0.78453 time=239.39it/s +epoch=11 global_step=4692 loss=0.23612 test_loss_avg=1.89116 acc=0.93750 test_acc_avg=0.56290 test_acc_top5_avg=0.87599 time=363.87it/s +curr_acc 0.5629 +BEST_ACC 0.6095 +curr_acc_top5 0.8760 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=5.59620 loss_avg=5.54218 acc=0.42969 acc_top1_avg=0.43457 acc_top5_avg=0.82031 lr=0.01000 gn=3.11117 time=57.36it/s +epoch=12 global_step=4750 loss=5.64130 loss_avg=5.26405 acc=0.40625 acc_top1_avg=0.47198 acc_top5_avg=0.85102 lr=0.01000 gn=4.92352 time=56.17it/s +epoch=12 global_step=4800 loss=5.72141 loss_avg=5.24753 acc=0.42188 acc_top1_avg=0.47627 acc_top5_avg=0.85200 lr=0.01000 gn=4.98981 time=61.39it/s +epoch=12 global_step=4850 loss=5.12987 loss_avg=5.23544 acc=0.48438 acc_top1_avg=0.47711 acc_top5_avg=0.85458 lr=0.01000 gn=4.94395 time=62.30it/s +epoch=12 global_step=4900 loss=5.08300 loss_avg=5.26121 acc=0.50781 acc_top1_avg=0.47450 acc_top5_avg=0.85472 lr=0.01000 gn=5.62967 time=55.28it/s +epoch=12 global_step=4950 loss=4.76227 loss_avg=5.24952 acc=0.52344 acc_top1_avg=0.47568 acc_top5_avg=0.85462 lr=0.01000 gn=5.49196 time=55.74it/s +epoch=12 global_step=5000 loss=5.01458 loss_avg=5.26207 acc=0.50781 acc_top1_avg=0.47436 acc_top5_avg=0.85395 lr=0.01000 gn=7.79689 time=54.65it/s +epoch=12 global_step=5050 loss=5.80678 loss_avg=5.25631 acc=0.43750 acc_top1_avg=0.47530 acc_top5_avg=0.85364 lr=0.01000 gn=6.68369 time=55.28it/s +====================Eval==================== +epoch=12 global_step=5083 loss=2.45791 test_loss_avg=2.58101 acc=0.39062 test_acc_avg=0.37109 test_acc_top5_avg=0.82422 time=234.90it/s +epoch=12 global_step=5083 loss=3.62764 test_loss_avg=2.20911 acc=0.00000 test_acc_avg=0.48498 test_acc_top5_avg=0.91046 time=238.95it/s +epoch=12 global_step=5083 loss=0.37469 test_loss_avg=1.94141 acc=0.87500 test_acc_avg=0.53847 test_acc_top5_avg=0.92108 time=811.28it/s +curr_acc 0.5385 +BEST_ACC 0.6095 +curr_acc_top5 0.9211 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=5.10944 loss_avg=5.17198 acc=0.50000 acc_top1_avg=0.48438 acc_top5_avg=0.85662 lr=0.01000 gn=7.26148 time=58.48it/s +epoch=13 global_step=5150 loss=5.80167 loss_avg=5.30741 acc=0.43750 acc_top1_avg=0.47003 acc_top5_avg=0.85436 lr=0.01000 gn=4.76132 time=57.24it/s +epoch=13 global_step=5200 loss=5.27821 loss_avg=5.24311 acc=0.46875 acc_top1_avg=0.47603 acc_top5_avg=0.85604 lr=0.01000 gn=5.58239 time=59.46it/s +epoch=13 global_step=5250 loss=5.68149 loss_avg=5.29339 acc=0.42969 acc_top1_avg=0.47100 acc_top5_avg=0.85666 lr=0.01000 gn=5.42267 time=56.43it/s +epoch=13 global_step=5300 loss=5.42542 loss_avg=5.26181 acc=0.43750 acc_top1_avg=0.47462 acc_top5_avg=0.85714 lr=0.01000 gn=4.61708 time=54.67it/s +epoch=13 global_step=5350 loss=5.05626 loss_avg=5.24707 acc=0.49219 acc_top1_avg=0.47650 acc_top5_avg=0.85779 lr=0.01000 gn=4.83861 time=54.97it/s +epoch=13 global_step=5400 loss=5.48179 loss_avg=5.24194 acc=0.43750 acc_top1_avg=0.47703 acc_top5_avg=0.85580 lr=0.01000 gn=6.28541 time=54.80it/s +epoch=13 global_step=5450 loss=5.29218 loss_avg=5.25137 acc=0.48438 acc_top1_avg=0.47601 acc_top5_avg=0.85535 lr=0.01000 gn=5.77590 time=52.80it/s +====================Eval==================== +epoch=13 global_step=5474 loss=1.52187 test_loss_avg=1.41742 acc=0.59375 test_acc_avg=0.64368 test_acc_top5_avg=0.94905 time=229.45it/s +epoch=13 global_step=5474 loss=0.23653 test_loss_avg=1.68437 acc=0.93750 test_acc_avg=0.58091 test_acc_top5_avg=0.91824 time=238.45it/s +epoch=13 global_step=5474 loss=0.05043 test_loss_avg=1.56732 acc=0.93750 test_acc_avg=0.60938 test_acc_top5_avg=0.92445 time=501.83it/s +curr_acc 0.6094 +BEST_ACC 0.6095 +curr_acc_top5 0.9244 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=5.00762 loss_avg=5.28320 acc=0.49219 acc_top1_avg=0.46875 acc_top5_avg=0.86208 lr=0.01000 gn=5.51206 time=61.66it/s +epoch=14 global_step=5550 loss=4.87498 loss_avg=5.20786 acc=0.51562 acc_top1_avg=0.47800 acc_top5_avg=0.85794 lr=0.01000 gn=5.41517 time=54.37it/s +epoch=14 global_step=5600 loss=4.95083 loss_avg=5.24804 acc=0.48438 acc_top1_avg=0.47346 acc_top5_avg=0.85807 lr=0.01000 gn=4.91791 time=60.20it/s +epoch=14 global_step=5650 loss=4.89680 loss_avg=5.24850 acc=0.52344 acc_top1_avg=0.47377 acc_top5_avg=0.85733 lr=0.01000 gn=8.02432 time=54.42it/s +epoch=14 global_step=5700 loss=4.62026 loss_avg=5.24594 acc=0.55469 acc_top1_avg=0.47452 acc_top5_avg=0.85664 lr=0.01000 gn=5.37846 time=53.87it/s +epoch=14 global_step=5750 loss=5.59353 loss_avg=5.23900 acc=0.43750 acc_top1_avg=0.47552 acc_top5_avg=0.85666 lr=0.01000 gn=5.15261 time=52.91it/s +epoch=14 global_step=5800 loss=5.36229 loss_avg=5.24287 acc=0.45312 acc_top1_avg=0.47512 acc_top5_avg=0.85580 lr=0.01000 gn=5.47108 time=61.90it/s +epoch=14 global_step=5850 loss=5.16947 loss_avg=5.24403 acc=0.50781 acc_top1_avg=0.47554 acc_top5_avg=0.85624 lr=0.01000 gn=6.60215 time=55.63it/s +====================Eval==================== +epoch=14 global_step=5865 loss=1.42681 test_loss_avg=1.84103 acc=0.59375 test_acc_avg=0.57369 test_acc_top5_avg=0.87287 time=233.34it/s +epoch=14 global_step=5865 loss=0.65488 test_loss_avg=1.75412 acc=0.81250 test_acc_avg=0.56408 test_acc_top5_avg=0.90991 time=489.47it/s +curr_acc 0.5641 +BEST_ACC 0.6095 +curr_acc_top5 0.9099 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=5.31439 loss_avg=5.16004 acc=0.47656 acc_top1_avg=0.48460 acc_top5_avg=0.85781 lr=0.01000 gn=4.61132 time=63.83it/s +epoch=15 global_step=5950 loss=5.71479 loss_avg=5.17554 acc=0.41406 acc_top1_avg=0.48226 acc_top5_avg=0.85340 lr=0.01000 gn=5.40372 time=52.96it/s +epoch=15 global_step=6000 loss=5.25614 loss_avg=5.18710 acc=0.46875 acc_top1_avg=0.48125 acc_top5_avg=0.85330 lr=0.01000 gn=4.32881 time=47.62it/s +epoch=15 global_step=6050 loss=5.05403 loss_avg=5.23497 acc=0.50000 acc_top1_avg=0.47610 acc_top5_avg=0.85304 lr=0.01000 gn=5.84269 time=60.56it/s +epoch=15 global_step=6100 loss=4.89490 loss_avg=5.22017 acc=0.52344 acc_top1_avg=0.47819 acc_top5_avg=0.85512 lr=0.01000 gn=5.96684 time=54.67it/s +epoch=15 global_step=6150 loss=5.36472 loss_avg=5.21191 acc=0.46875 acc_top1_avg=0.47928 acc_top5_avg=0.85630 lr=0.01000 gn=6.25197 time=54.34it/s +epoch=15 global_step=6200 loss=5.56492 loss_avg=5.21638 acc=0.44531 acc_top1_avg=0.47892 acc_top5_avg=0.85592 lr=0.01000 gn=5.29342 time=59.23it/s +epoch=15 global_step=6250 loss=4.36249 loss_avg=5.20373 acc=0.57812 acc_top1_avg=0.48062 acc_top5_avg=0.85712 lr=0.01000 gn=6.67283 time=55.59it/s +====================Eval==================== +epoch=15 global_step=6256 loss=1.61562 test_loss_avg=1.97487 acc=0.60938 test_acc_avg=0.57865 test_acc_top5_avg=0.96667 time=213.98it/s +epoch=15 global_step=6256 loss=0.26678 test_loss_avg=2.29738 acc=0.92188 test_acc_avg=0.49363 test_acc_top5_avg=0.87957 time=241.43it/s +epoch=15 global_step=6256 loss=0.01282 test_loss_avg=1.92225 acc=1.00000 test_acc_avg=0.57437 test_acc_top5_avg=0.90042 time=545.28it/s +curr_acc 0.5744 +BEST_ACC 0.6095 +curr_acc_top5 0.9004 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=5.30278 loss_avg=5.24628 acc=0.46875 acc_top1_avg=0.47603 acc_top5_avg=0.85653 lr=0.01000 gn=4.67809 time=55.95it/s +epoch=16 global_step=6350 loss=4.91790 loss_avg=5.16911 acc=0.50781 acc_top1_avg=0.48562 acc_top5_avg=0.86062 lr=0.01000 gn=6.17639 time=52.76it/s +epoch=16 global_step=6400 loss=4.63671 loss_avg=5.23327 acc=0.54688 acc_top1_avg=0.47765 acc_top5_avg=0.86013 lr=0.01000 gn=6.26138 time=44.57it/s +epoch=16 global_step=6450 loss=5.12917 loss_avg=5.23296 acc=0.50000 acc_top1_avg=0.47761 acc_top5_avg=0.85857 lr=0.01000 gn=5.92300 time=49.64it/s +epoch=16 global_step=6500 loss=5.43971 loss_avg=5.22217 acc=0.44531 acc_top1_avg=0.47832 acc_top5_avg=0.85938 lr=0.01000 gn=5.06873 time=58.00it/s +epoch=16 global_step=6550 loss=5.01317 loss_avg=5.21236 acc=0.49219 acc_top1_avg=0.47962 acc_top5_avg=0.85985 lr=0.01000 gn=6.16875 time=55.60it/s +epoch=16 global_step=6600 loss=5.99831 loss_avg=5.20704 acc=0.39844 acc_top1_avg=0.48051 acc_top5_avg=0.85972 lr=0.01000 gn=6.29989 time=55.17it/s +====================Eval==================== +epoch=16 global_step=6647 loss=0.66299 test_loss_avg=2.00489 acc=0.86719 test_acc_avg=0.55816 test_acc_top5_avg=0.82357 time=250.75it/s +epoch=16 global_step=6647 loss=0.28329 test_loss_avg=1.57901 acc=0.87500 test_acc_avg=0.62915 test_acc_top5_avg=0.90170 time=534.10it/s +curr_acc 0.6292 +BEST_ACC 0.6095 +curr_acc_top5 0.9017 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=5.29447 loss_avg=5.35232 acc=0.46094 acc_top1_avg=0.46094 acc_top5_avg=0.86979 lr=0.01000 gn=5.12812 time=51.04it/s +epoch=17 global_step=6700 loss=5.51438 loss_avg=5.14988 acc=0.43750 acc_top1_avg=0.48408 acc_top5_avg=0.86380 lr=0.01000 gn=5.82444 time=54.04it/s +epoch=17 global_step=6750 loss=5.39200 loss_avg=5.19341 acc=0.45312 acc_top1_avg=0.47975 acc_top5_avg=0.86021 lr=0.01000 gn=5.63257 time=53.80it/s +epoch=17 global_step=6800 loss=4.48170 loss_avg=5.18207 acc=0.56250 acc_top1_avg=0.48192 acc_top5_avg=0.86162 lr=0.01000 gn=5.55404 time=54.12it/s +epoch=17 global_step=6850 loss=5.05181 loss_avg=5.19879 acc=0.51562 acc_top1_avg=0.47999 acc_top5_avg=0.86207 lr=0.01000 gn=6.08166 time=47.73it/s +epoch=17 global_step=6900 loss=5.28817 loss_avg=5.22278 acc=0.46094 acc_top1_avg=0.47730 acc_top5_avg=0.86042 lr=0.01000 gn=4.95799 time=52.77it/s +epoch=17 global_step=6950 loss=4.48794 loss_avg=5.21134 acc=0.56250 acc_top1_avg=0.47863 acc_top5_avg=0.86061 lr=0.01000 gn=6.32673 time=55.16it/s +epoch=17 global_step=7000 loss=5.41876 loss_avg=5.20432 acc=0.46875 acc_top1_avg=0.47964 acc_top5_avg=0.86154 lr=0.01000 gn=7.25842 time=54.71it/s +====================Eval==================== +epoch=17 global_step=7038 loss=1.65562 test_loss_avg=1.49594 acc=0.56250 test_acc_avg=0.59487 test_acc_top5_avg=0.97210 time=207.23it/s +epoch=17 global_step=7038 loss=1.05588 test_loss_avg=1.87860 acc=0.66406 test_acc_avg=0.53797 test_acc_top5_avg=0.90885 time=190.68it/s +epoch=17 global_step=7038 loss=0.40895 test_loss_avg=1.52366 acc=0.87500 test_acc_avg=0.62045 test_acc_top5_avg=0.92850 time=497.13it/s +curr_acc 0.6205 +BEST_ACC 0.6292 +curr_acc_top5 0.9285 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=5.58756 loss_avg=5.24978 acc=0.43750 acc_top1_avg=0.48177 acc_top5_avg=0.86198 lr=0.01000 gn=6.68927 time=48.08it/s +epoch=18 global_step=7100 loss=4.45443 loss_avg=5.21765 acc=0.57812 acc_top1_avg=0.48223 acc_top5_avg=0.86227 lr=0.01000 gn=6.69429 time=53.12it/s +epoch=18 global_step=7150 loss=4.63111 loss_avg=5.18806 acc=0.56250 acc_top1_avg=0.48361 acc_top5_avg=0.86363 lr=0.01000 gn=5.98136 time=54.88it/s +epoch=18 global_step=7200 loss=5.47361 loss_avg=5.17666 acc=0.46094 acc_top1_avg=0.48413 acc_top5_avg=0.86019 lr=0.01000 gn=5.84710 time=50.91it/s +epoch=18 global_step=7250 loss=5.88762 loss_avg=5.16579 acc=0.40625 acc_top1_avg=0.48526 acc_top5_avg=0.86008 lr=0.01000 gn=6.01387 time=57.59it/s +epoch=18 global_step=7300 loss=5.70184 loss_avg=5.19823 acc=0.42188 acc_top1_avg=0.48130 acc_top5_avg=0.85938 lr=0.01000 gn=4.82517 time=56.57it/s +epoch=18 global_step=7350 loss=5.43405 loss_avg=5.19124 acc=0.46875 acc_top1_avg=0.48217 acc_top5_avg=0.86050 lr=0.01000 gn=7.38508 time=55.10it/s +epoch=18 global_step=7400 loss=5.03719 loss_avg=5.19881 acc=0.50000 acc_top1_avg=0.48114 acc_top5_avg=0.85966 lr=0.01000 gn=5.90693 time=55.11it/s +====================Eval==================== +epoch=18 global_step=7429 loss=4.76605 test_loss_avg=1.42833 acc=0.00000 test_acc_avg=0.64007 test_acc_top5_avg=0.87472 time=234.53it/s +epoch=18 global_step=7429 loss=0.97194 test_loss_avg=1.50682 acc=0.65625 test_acc_avg=0.60817 test_acc_top5_avg=0.90525 time=257.75it/s +epoch=18 global_step=7429 loss=0.91676 test_loss_avg=1.49935 acc=0.68750 test_acc_avg=0.60918 test_acc_top5_avg=0.90645 time=826.46it/s +curr_acc 0.6092 +BEST_ACC 0.6292 +curr_acc_top5 0.9064 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=5.87854 loss_avg=5.12075 acc=0.40625 acc_top1_avg=0.48921 acc_top5_avg=0.86124 lr=0.01000 gn=6.54752 time=56.90it/s +epoch=19 global_step=7500 loss=5.37430 loss_avg=5.20053 acc=0.44531 acc_top1_avg=0.48085 acc_top5_avg=0.85486 lr=0.01000 gn=4.17361 time=59.56it/s +epoch=19 global_step=7550 loss=4.36350 loss_avg=5.16522 acc=0.57812 acc_top1_avg=0.48502 acc_top5_avg=0.85996 lr=0.01000 gn=6.65528 time=57.99it/s +epoch=19 global_step=7600 loss=5.73314 loss_avg=5.18082 acc=0.41406 acc_top1_avg=0.48323 acc_top5_avg=0.85924 lr=0.01000 gn=5.77847 time=52.67it/s +epoch=19 global_step=7650 loss=5.31254 loss_avg=5.18200 acc=0.47656 acc_top1_avg=0.48356 acc_top5_avg=0.86029 lr=0.01000 gn=6.02223 time=54.67it/s +epoch=19 global_step=7700 loss=4.67646 loss_avg=5.20285 acc=0.54688 acc_top1_avg=0.48138 acc_top5_avg=0.85909 lr=0.01000 gn=5.68571 time=55.06it/s +epoch=19 global_step=7750 loss=5.54438 loss_avg=5.19058 acc=0.44531 acc_top1_avg=0.48260 acc_top5_avg=0.86003 lr=0.01000 gn=6.29295 time=56.37it/s +epoch=19 global_step=7800 loss=4.77694 loss_avg=5.18690 acc=0.53906 acc_top1_avg=0.48271 acc_top5_avg=0.85950 lr=0.01000 gn=5.00290 time=52.72it/s +====================Eval==================== +epoch=19 global_step=7820 loss=4.34333 test_loss_avg=1.83485 acc=0.00000 test_acc_avg=0.57398 test_acc_top5_avg=0.87612 time=235.38it/s +epoch=19 global_step=7820 loss=0.57475 test_loss_avg=1.61727 acc=0.81250 test_acc_avg=0.62144 test_acc_top5_avg=0.91060 time=845.11it/s +curr_acc 0.6214 +BEST_ACC 0.6292 +curr_acc_top5 0.9106 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=4.77204 loss_avg=5.20964 acc=0.51562 acc_top1_avg=0.48255 acc_top5_avg=0.85911 lr=0.01000 gn=5.73973 time=63.83it/s +epoch=20 global_step=7900 loss=5.31514 loss_avg=5.24783 acc=0.45312 acc_top1_avg=0.47842 acc_top5_avg=0.85732 lr=0.01000 gn=6.08167 time=55.84it/s +epoch=20 global_step=7950 loss=5.06065 loss_avg=5.19065 acc=0.49219 acc_top1_avg=0.48203 acc_top5_avg=0.86064 lr=0.01000 gn=6.06051 time=55.39it/s +epoch=20 global_step=8000 loss=5.47777 loss_avg=5.18961 acc=0.43750 acc_top1_avg=0.48212 acc_top5_avg=0.85946 lr=0.01000 gn=6.81596 time=52.97it/s +epoch=20 global_step=8050 loss=5.26327 loss_avg=5.17512 acc=0.46094 acc_top1_avg=0.48387 acc_top5_avg=0.85938 lr=0.01000 gn=7.77857 time=55.45it/s +epoch=20 global_step=8100 loss=4.56770 loss_avg=5.16330 acc=0.54688 acc_top1_avg=0.48465 acc_top5_avg=0.85868 lr=0.01000 gn=8.35586 time=59.74it/s +epoch=20 global_step=8150 loss=5.13787 loss_avg=5.16200 acc=0.50000 acc_top1_avg=0.48490 acc_top5_avg=0.85949 lr=0.01000 gn=6.66642 time=54.56it/s +epoch=20 global_step=8200 loss=5.57555 loss_avg=5.18957 acc=0.45312 acc_top1_avg=0.48185 acc_top5_avg=0.85896 lr=0.01000 gn=6.29174 time=56.63it/s +====================Eval==================== +epoch=20 global_step=8211 loss=2.30179 test_loss_avg=1.20460 acc=0.43750 test_acc_avg=0.68750 test_acc_top5_avg=0.95234 time=234.08it/s +epoch=20 global_step=8211 loss=0.22874 test_loss_avg=1.66204 acc=0.91406 test_acc_avg=0.57388 test_acc_top5_avg=0.90904 time=238.71it/s +epoch=20 global_step=8211 loss=0.30330 test_loss_avg=1.52236 acc=0.81250 test_acc_avg=0.60631 test_acc_top5_avg=0.91772 time=566.34it/s +curr_acc 0.6063 +BEST_ACC 0.6292 +curr_acc_top5 0.9177 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=5.35587 loss_avg=5.19020 acc=0.46094 acc_top1_avg=0.48037 acc_top5_avg=0.86458 lr=0.01000 gn=6.82622 time=61.40it/s +epoch=21 global_step=8300 loss=4.79626 loss_avg=5.16358 acc=0.54688 acc_top1_avg=0.48464 acc_top5_avg=0.86113 lr=0.01000 gn=6.31964 time=62.88it/s +epoch=21 global_step=8350 loss=5.40921 loss_avg=5.13547 acc=0.43750 acc_top1_avg=0.48775 acc_top5_avg=0.86337 lr=0.01000 gn=7.31119 time=60.19it/s +epoch=21 global_step=8400 loss=5.32244 loss_avg=5.13787 acc=0.47656 acc_top1_avg=0.48723 acc_top5_avg=0.86314 lr=0.01000 gn=6.74483 time=58.53it/s +epoch=21 global_step=8450 loss=4.96733 loss_avg=5.16137 acc=0.50781 acc_top1_avg=0.48477 acc_top5_avg=0.86124 lr=0.01000 gn=5.19175 time=55.59it/s +epoch=21 global_step=8500 loss=5.17583 loss_avg=5.16778 acc=0.48438 acc_top1_avg=0.48389 acc_top5_avg=0.86027 lr=0.01000 gn=5.87517 time=55.50it/s +epoch=21 global_step=8550 loss=4.96945 loss_avg=5.16594 acc=0.50000 acc_top1_avg=0.48412 acc_top5_avg=0.86055 lr=0.01000 gn=5.95544 time=54.31it/s +epoch=21 global_step=8600 loss=4.84468 loss_avg=5.17825 acc=0.52344 acc_top1_avg=0.48317 acc_top5_avg=0.85980 lr=0.01000 gn=5.46172 time=58.26it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.07296 test_loss_avg=1.94277 acc=0.72656 test_acc_avg=0.55697 test_acc_top5_avg=0.87309 time=135.29it/s +epoch=21 global_step=8602 loss=0.02567 test_loss_avg=1.63958 acc=1.00000 test_acc_avg=0.59059 test_acc_top5_avg=0.91901 time=531.60it/s +curr_acc 0.5906 +BEST_ACC 0.6292 +curr_acc_top5 0.9190 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=5.03120 loss_avg=5.16536 acc=0.50000 acc_top1_avg=0.48470 acc_top5_avg=0.85921 lr=0.01000 gn=4.71193 time=52.34it/s +epoch=22 global_step=8700 loss=5.89697 loss_avg=5.16416 acc=0.41406 acc_top1_avg=0.48589 acc_top5_avg=0.85698 lr=0.01000 gn=7.25153 time=54.17it/s +epoch=22 global_step=8750 loss=5.42330 loss_avg=5.18402 acc=0.47656 acc_top1_avg=0.48374 acc_top5_avg=0.85980 lr=0.01000 gn=7.03931 time=63.47it/s +epoch=22 global_step=8800 loss=5.55346 loss_avg=5.19981 acc=0.44531 acc_top1_avg=0.48122 acc_top5_avg=0.85997 lr=0.01000 gn=5.23396 time=57.57it/s +epoch=22 global_step=8850 loss=5.38708 loss_avg=5.20991 acc=0.45312 acc_top1_avg=0.47956 acc_top5_avg=0.85963 lr=0.01000 gn=4.56941 time=61.87it/s +epoch=22 global_step=8900 loss=4.95897 loss_avg=5.19520 acc=0.50781 acc_top1_avg=0.48094 acc_top5_avg=0.85951 lr=0.01000 gn=4.71113 time=56.73it/s +epoch=22 global_step=8950 loss=5.26699 loss_avg=5.19391 acc=0.47656 acc_top1_avg=0.48107 acc_top5_avg=0.86065 lr=0.01000 gn=6.03878 time=53.78it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.64521 test_loss_avg=1.45595 acc=0.83594 test_acc_avg=0.63997 test_acc_top5_avg=0.96354 time=235.54it/s +epoch=22 global_step=8993 loss=0.77327 test_loss_avg=1.93419 acc=0.79688 test_acc_avg=0.53238 test_acc_top5_avg=0.91205 time=234.61it/s +epoch=22 global_step=8993 loss=0.00634 test_loss_avg=1.56238 acc=1.00000 test_acc_avg=0.62065 test_acc_top5_avg=0.93018 time=809.09it/s +curr_acc 0.6206 +BEST_ACC 0.6292 +curr_acc_top5 0.9302 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=5.44524 loss_avg=4.91582 acc=0.44531 acc_top1_avg=0.50670 acc_top5_avg=0.86384 lr=0.01000 gn=5.66005 time=62.16it/s +epoch=23 global_step=9050 loss=5.63305 loss_avg=5.16371 acc=0.41406 acc_top1_avg=0.48561 acc_top5_avg=0.85814 lr=0.01000 gn=6.33767 time=54.16it/s +epoch=23 global_step=9100 loss=5.68650 loss_avg=5.15076 acc=0.40625 acc_top1_avg=0.48744 acc_top5_avg=0.86376 lr=0.01000 gn=4.76990 time=53.43it/s +epoch=23 global_step=9150 loss=4.70113 loss_avg=5.17219 acc=0.53125 acc_top1_avg=0.48388 acc_top5_avg=0.86256 lr=0.01000 gn=6.33844 time=62.22it/s +epoch=23 global_step=9200 loss=5.78182 loss_avg=5.15922 acc=0.40625 acc_top1_avg=0.48539 acc_top5_avg=0.86258 lr=0.01000 gn=6.25353 time=52.23it/s +epoch=23 global_step=9250 loss=4.98644 loss_avg=5.17364 acc=0.50781 acc_top1_avg=0.48383 acc_top5_avg=0.86108 lr=0.01000 gn=6.38101 time=55.21it/s +epoch=23 global_step=9300 loss=4.83540 loss_avg=5.16095 acc=0.50781 acc_top1_avg=0.48529 acc_top5_avg=0.86205 lr=0.01000 gn=5.88843 time=55.37it/s +epoch=23 global_step=9350 loss=4.19180 loss_avg=5.15561 acc=0.57812 acc_top1_avg=0.48569 acc_top5_avg=0.86331 lr=0.01000 gn=5.10303 time=54.25it/s +====================Eval==================== +epoch=23 global_step=9384 loss=0.97990 test_loss_avg=1.66714 acc=0.71875 test_acc_avg=0.60156 test_acc_top5_avg=0.85795 time=238.98it/s +epoch=23 global_step=9384 loss=0.54719 test_loss_avg=1.45982 acc=0.87500 test_acc_avg=0.63627 test_acc_top5_avg=0.92079 time=825.16it/s +curr_acc 0.6363 +BEST_ACC 0.6292 +curr_acc_top5 0.9208 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=5.18283 loss_avg=5.29241 acc=0.49219 acc_top1_avg=0.47168 acc_top5_avg=0.85352 lr=0.01000 gn=5.50107 time=63.59it/s +epoch=24 global_step=9450 loss=4.62740 loss_avg=5.18858 acc=0.55469 acc_top1_avg=0.48059 acc_top5_avg=0.86423 lr=0.01000 gn=6.54422 time=54.87it/s +epoch=24 global_step=9500 loss=4.95176 loss_avg=5.18750 acc=0.52344 acc_top1_avg=0.48283 acc_top5_avg=0.86389 lr=0.01000 gn=6.07866 time=62.94it/s +epoch=24 global_step=9550 loss=5.06965 loss_avg=5.18885 acc=0.50000 acc_top1_avg=0.48179 acc_top5_avg=0.86112 lr=0.01000 gn=6.71205 time=54.05it/s +epoch=24 global_step=9600 loss=5.35171 loss_avg=5.18032 acc=0.46094 acc_top1_avg=0.48213 acc_top5_avg=0.86013 lr=0.01000 gn=5.75646 time=55.39it/s +epoch=24 global_step=9650 loss=5.40499 loss_avg=5.17525 acc=0.48438 acc_top1_avg=0.48291 acc_top5_avg=0.86102 lr=0.01000 gn=8.07825 time=61.63it/s +epoch=24 global_step=9700 loss=5.09079 loss_avg=5.17349 acc=0.50000 acc_top1_avg=0.48274 acc_top5_avg=0.86145 lr=0.01000 gn=7.22626 time=61.92it/s +epoch=24 global_step=9750 loss=5.60726 loss_avg=5.17463 acc=0.43750 acc_top1_avg=0.48275 acc_top5_avg=0.86136 lr=0.01000 gn=5.59782 time=53.17it/s +====================Eval==================== +epoch=24 global_step=9775 loss=1.14210 test_loss_avg=1.27138 acc=0.70312 test_acc_avg=0.69336 test_acc_top5_avg=0.98633 time=241.70it/s +epoch=24 global_step=9775 loss=3.59002 test_loss_avg=1.96950 acc=0.00000 test_acc_avg=0.49537 test_acc_top5_avg=0.86285 time=239.56it/s +epoch=24 global_step=9775 loss=0.63413 test_loss_avg=1.52316 acc=0.81250 test_acc_avg=0.60483 test_acc_top5_avg=0.90002 time=823.70it/s +curr_acc 0.6048 +BEST_ACC 0.6363 +curr_acc_top5 0.9000 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=5.42496 loss_avg=5.20443 acc=0.45312 acc_top1_avg=0.48094 acc_top5_avg=0.85906 lr=0.01000 gn=6.78621 time=61.31it/s +epoch=25 global_step=9850 loss=4.84308 loss_avg=5.19009 acc=0.53125 acc_top1_avg=0.48187 acc_top5_avg=0.86271 lr=0.01000 gn=6.91372 time=54.37it/s +epoch=25 global_step=9900 loss=5.19070 loss_avg=5.15428 acc=0.46875 acc_top1_avg=0.48469 acc_top5_avg=0.86200 lr=0.01000 gn=4.86303 time=61.63it/s +epoch=25 global_step=9950 loss=4.57753 loss_avg=5.15333 acc=0.55469 acc_top1_avg=0.48540 acc_top5_avg=0.86223 lr=0.01000 gn=6.52340 time=41.79it/s +epoch=25 global_step=10000 loss=4.60275 loss_avg=5.12984 acc=0.55469 acc_top1_avg=0.48840 acc_top5_avg=0.86344 lr=0.01000 gn=6.99927 time=51.74it/s +epoch=25 global_step=10050 loss=5.27134 loss_avg=5.13271 acc=0.46094 acc_top1_avg=0.48804 acc_top5_avg=0.86386 lr=0.01000 gn=6.01400 time=55.68it/s +epoch=25 global_step=10100 loss=5.29425 loss_avg=5.14579 acc=0.46875 acc_top1_avg=0.48671 acc_top5_avg=0.86365 lr=0.01000 gn=7.59595 time=52.10it/s +epoch=25 global_step=10150 loss=5.08826 loss_avg=5.15977 acc=0.49219 acc_top1_avg=0.48529 acc_top5_avg=0.86300 lr=0.01000 gn=5.61519 time=58.19it/s +====================Eval==================== +epoch=25 global_step=10166 loss=4.76707 test_loss_avg=1.40875 acc=0.00000 test_acc_avg=0.66063 test_acc_top5_avg=0.89563 time=237.91it/s +epoch=25 global_step=10166 loss=0.59928 test_loss_avg=1.69424 acc=0.82031 test_acc_avg=0.59771 test_acc_top5_avg=0.88719 time=253.51it/s +epoch=25 global_step=10166 loss=0.86116 test_loss_avg=1.65134 acc=0.81250 test_acc_avg=0.60740 test_acc_top5_avg=0.89161 time=515.78it/s +curr_acc 0.6074 +BEST_ACC 0.6363 +curr_acc_top5 0.8916 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=5.48361 loss_avg=5.13218 acc=0.46094 acc_top1_avg=0.48920 acc_top5_avg=0.85938 lr=0.01000 gn=7.06789 time=58.43it/s +epoch=26 global_step=10250 loss=5.39096 loss_avg=5.14532 acc=0.43750 acc_top1_avg=0.48596 acc_top5_avg=0.86523 lr=0.01000 gn=7.43224 time=59.61it/s +epoch=26 global_step=10300 loss=4.88816 loss_avg=5.12262 acc=0.50000 acc_top1_avg=0.48840 acc_top5_avg=0.86526 lr=0.01000 gn=3.55253 time=61.99it/s +epoch=26 global_step=10350 loss=5.78075 loss_avg=5.13908 acc=0.42969 acc_top1_avg=0.48709 acc_top5_avg=0.86481 lr=0.01000 gn=6.40443 time=51.00it/s +epoch=26 global_step=10400 loss=5.37773 loss_avg=5.15983 acc=0.45312 acc_top1_avg=0.48454 acc_top5_avg=0.86325 lr=0.01000 gn=3.44180 time=55.75it/s +epoch=26 global_step=10450 loss=4.53803 loss_avg=5.14511 acc=0.54688 acc_top1_avg=0.48636 acc_top5_avg=0.86391 lr=0.01000 gn=5.86403 time=55.20it/s +epoch=26 global_step=10500 loss=5.81851 loss_avg=5.15195 acc=0.42188 acc_top1_avg=0.48585 acc_top5_avg=0.86342 lr=0.01000 gn=5.96717 time=57.46it/s +epoch=26 global_step=10550 loss=4.96570 loss_avg=5.14726 acc=0.51562 acc_top1_avg=0.48647 acc_top5_avg=0.86239 lr=0.01000 gn=6.95387 time=62.23it/s +====================Eval==================== +epoch=26 global_step=10557 loss=2.97776 test_loss_avg=1.91690 acc=0.26562 test_acc_avg=0.54755 test_acc_top5_avg=0.81607 time=174.37it/s +epoch=26 global_step=10557 loss=0.08385 test_loss_avg=1.64523 acc=0.93750 test_acc_avg=0.60166 test_acc_top5_avg=0.85918 time=498.85it/s +curr_acc 0.6017 +BEST_ACC 0.6363 +curr_acc_top5 0.8592 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=4.08455 loss_avg=5.17898 acc=0.60938 acc_top1_avg=0.48274 acc_top5_avg=0.86010 lr=0.01000 gn=5.79324 time=62.82it/s +epoch=27 global_step=10650 loss=5.46464 loss_avg=5.13569 acc=0.45312 acc_top1_avg=0.48816 acc_top5_avg=0.86307 lr=0.01000 gn=5.61786 time=60.07it/s +epoch=27 global_step=10700 loss=5.60236 loss_avg=5.15199 acc=0.41406 acc_top1_avg=0.48743 acc_top5_avg=0.86347 lr=0.01000 gn=6.73603 time=62.73it/s +epoch=27 global_step=10750 loss=4.85645 loss_avg=5.14803 acc=0.52344 acc_top1_avg=0.48749 acc_top5_avg=0.86427 lr=0.01000 gn=7.51220 time=42.90it/s +epoch=27 global_step=10800 loss=5.16217 loss_avg=5.15736 acc=0.50781 acc_top1_avg=0.48685 acc_top5_avg=0.86407 lr=0.01000 gn=5.41379 time=63.34it/s +epoch=27 global_step=10850 loss=5.11508 loss_avg=5.16097 acc=0.47656 acc_top1_avg=0.48624 acc_top5_avg=0.86257 lr=0.01000 gn=6.16550 time=57.84it/s +epoch=27 global_step=10900 loss=5.08141 loss_avg=5.15287 acc=0.47656 acc_top1_avg=0.48679 acc_top5_avg=0.86190 lr=0.01000 gn=6.76058 time=57.53it/s +====================Eval==================== +epoch=27 global_step=10948 loss=2.91980 test_loss_avg=0.86272 acc=0.37500 test_acc_avg=0.77941 test_acc_top5_avg=0.97243 time=238.42it/s +epoch=27 global_step=10948 loss=0.26490 test_loss_avg=1.76255 acc=0.92188 test_acc_avg=0.58477 test_acc_top5_avg=0.90951 time=238.68it/s +epoch=27 global_step=10948 loss=0.12152 test_loss_avg=1.55457 acc=0.93750 test_acc_avg=0.63281 test_acc_top5_avg=0.92227 time=597.39it/s +curr_acc 0.6328 +BEST_ACC 0.6363 +curr_acc_top5 0.9223 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=5.18364 loss_avg=5.39682 acc=0.49219 acc_top1_avg=0.45312 acc_top5_avg=0.88281 lr=0.01000 gn=5.33179 time=38.22it/s +epoch=28 global_step=11000 loss=5.19689 loss_avg=5.09079 acc=0.46875 acc_top1_avg=0.49279 acc_top5_avg=0.86328 lr=0.01000 gn=6.06430 time=62.98it/s +epoch=28 global_step=11050 loss=4.98665 loss_avg=5.08532 acc=0.51562 acc_top1_avg=0.49303 acc_top5_avg=0.86382 lr=0.01000 gn=5.70262 time=63.14it/s +epoch=28 global_step=11100 loss=4.90096 loss_avg=5.10589 acc=0.50000 acc_top1_avg=0.49152 acc_top5_avg=0.86544 lr=0.01000 gn=6.94685 time=57.39it/s +epoch=28 global_step=11150 loss=6.27535 loss_avg=5.12133 acc=0.36719 acc_top1_avg=0.48933 acc_top5_avg=0.86533 lr=0.01000 gn=5.69170 time=50.86it/s +epoch=28 global_step=11200 loss=5.45820 loss_avg=5.11687 acc=0.43750 acc_top1_avg=0.48961 acc_top5_avg=0.86604 lr=0.01000 gn=4.87203 time=63.77it/s +epoch=28 global_step=11250 loss=5.74355 loss_avg=5.12934 acc=0.42969 acc_top1_avg=0.48859 acc_top5_avg=0.86589 lr=0.01000 gn=5.21197 time=57.92it/s +epoch=28 global_step=11300 loss=4.99522 loss_avg=5.13572 acc=0.52344 acc_top1_avg=0.48815 acc_top5_avg=0.86594 lr=0.01000 gn=7.81592 time=54.35it/s +====================Eval==================== +epoch=28 global_step=11339 loss=0.34457 test_loss_avg=1.77348 acc=0.89062 test_acc_avg=0.59416 test_acc_top5_avg=0.86554 time=236.59it/s +epoch=28 global_step=11339 loss=0.99686 test_loss_avg=1.58679 acc=0.75000 test_acc_avg=0.60384 test_acc_top5_avg=0.91990 time=531.06it/s +curr_acc 0.6038 +BEST_ACC 0.6363 +curr_acc_top5 0.9199 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=5.21882 loss_avg=5.14231 acc=0.48438 acc_top1_avg=0.48722 acc_top5_avg=0.85866 lr=0.01000 gn=6.27900 time=54.85it/s +epoch=29 global_step=11400 loss=4.48276 loss_avg=5.08518 acc=0.54688 acc_top1_avg=0.49014 acc_top5_avg=0.86373 lr=0.01000 gn=5.30718 time=63.11it/s +epoch=29 global_step=11450 loss=5.26431 loss_avg=5.10401 acc=0.46875 acc_top1_avg=0.49078 acc_top5_avg=0.86367 lr=0.01000 gn=5.41760 time=53.32it/s +epoch=29 global_step=11500 loss=4.65118 loss_avg=5.11092 acc=0.53906 acc_top1_avg=0.49010 acc_top5_avg=0.86389 lr=0.01000 gn=6.09157 time=60.89it/s +epoch=29 global_step=11550 loss=5.39378 loss_avg=5.10679 acc=0.46875 acc_top1_avg=0.49071 acc_top5_avg=0.86448 lr=0.01000 gn=7.29682 time=50.48it/s +epoch=29 global_step=11600 loss=5.05100 loss_avg=5.13149 acc=0.50000 acc_top1_avg=0.48845 acc_top5_avg=0.86345 lr=0.01000 gn=6.97708 time=54.91it/s +epoch=29 global_step=11650 loss=4.98809 loss_avg=5.13763 acc=0.48438 acc_top1_avg=0.48817 acc_top5_avg=0.86264 lr=0.01000 gn=6.85599 time=52.39it/s +epoch=29 global_step=11700 loss=5.21613 loss_avg=5.13922 acc=0.47656 acc_top1_avg=0.48814 acc_top5_avg=0.86227 lr=0.01000 gn=5.77273 time=56.46it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.53425 test_loss_avg=1.80187 acc=0.87500 test_acc_avg=0.52083 test_acc_top5_avg=0.95312 time=242.04it/s +epoch=29 global_step=11730 loss=1.59166 test_loss_avg=2.36934 acc=0.61719 test_acc_avg=0.49047 test_acc_top5_avg=0.91314 time=134.29it/s +epoch=29 global_step=11730 loss=0.05811 test_loss_avg=1.88743 acc=1.00000 test_acc_avg=0.58722 test_acc_top5_avg=0.93176 time=494.61it/s +curr_acc 0.5872 +BEST_ACC 0.6363 +curr_acc_top5 0.9318 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=5.26096 loss_avg=5.04170 acc=0.47656 acc_top1_avg=0.49648 acc_top5_avg=0.87187 lr=0.01000 gn=5.69223 time=56.73it/s +epoch=30 global_step=11800 loss=5.75868 loss_avg=5.14686 acc=0.42969 acc_top1_avg=0.48348 acc_top5_avg=0.86473 lr=0.01000 gn=7.42269 time=61.34it/s +epoch=30 global_step=11850 loss=5.14939 loss_avg=5.17730 acc=0.47656 acc_top1_avg=0.48132 acc_top5_avg=0.86211 lr=0.01000 gn=7.02177 time=55.94it/s +epoch=30 global_step=11900 loss=5.23055 loss_avg=5.15686 acc=0.48438 acc_top1_avg=0.48456 acc_top5_avg=0.86241 lr=0.01000 gn=6.76638 time=53.84it/s +epoch=30 global_step=11950 loss=5.02690 loss_avg=5.13477 acc=0.50781 acc_top1_avg=0.48679 acc_top5_avg=0.86364 lr=0.01000 gn=4.86867 time=52.33it/s +epoch=30 global_step=12000 loss=5.51776 loss_avg=5.15438 acc=0.42188 acc_top1_avg=0.48414 acc_top5_avg=0.86343 lr=0.01000 gn=6.68013 time=31.19it/s +epoch=30 global_step=12050 loss=5.20927 loss_avg=5.14253 acc=0.50000 acc_top1_avg=0.48564 acc_top5_avg=0.86348 lr=0.01000 gn=5.67129 time=54.88it/s +epoch=30 global_step=12100 loss=5.13370 loss_avg=5.14627 acc=0.50781 acc_top1_avg=0.48558 acc_top5_avg=0.86362 lr=0.01000 gn=5.46839 time=59.42it/s +====================Eval==================== +epoch=30 global_step=12121 loss=4.85416 test_loss_avg=1.58495 acc=0.00000 test_acc_avg=0.62526 test_acc_top5_avg=0.82734 time=236.89it/s +epoch=30 global_step=12121 loss=0.32757 test_loss_avg=1.56988 acc=0.87500 test_acc_avg=0.60868 test_acc_top5_avg=0.88845 time=505.58it/s +curr_acc 0.6087 +BEST_ACC 0.6363 +curr_acc_top5 0.8884 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=5.50023 loss_avg=5.15864 acc=0.44531 acc_top1_avg=0.48572 acc_top5_avg=0.85075 lr=0.01000 gn=6.13740 time=54.29it/s +epoch=31 global_step=12200 loss=5.30153 loss_avg=5.09132 acc=0.47656 acc_top1_avg=0.49080 acc_top5_avg=0.86195 lr=0.01000 gn=7.55324 time=59.94it/s +epoch=31 global_step=12250 loss=5.62580 loss_avg=5.14817 acc=0.42969 acc_top1_avg=0.48456 acc_top5_avg=0.85828 lr=0.01000 gn=5.81403 time=55.28it/s +epoch=31 global_step=12300 loss=5.22526 loss_avg=5.13399 acc=0.50000 acc_top1_avg=0.48691 acc_top5_avg=0.85859 lr=0.01000 gn=7.18516 time=52.29it/s +epoch=31 global_step=12350 loss=5.00811 loss_avg=5.13673 acc=0.49219 acc_top1_avg=0.48683 acc_top5_avg=0.86040 lr=0.01000 gn=6.68344 time=30.63it/s +epoch=31 global_step=12400 loss=5.64203 loss_avg=5.15036 acc=0.42188 acc_top1_avg=0.48541 acc_top5_avg=0.85946 lr=0.01000 gn=7.52470 time=56.88it/s +epoch=31 global_step=12450 loss=5.24939 loss_avg=5.14006 acc=0.48438 acc_top1_avg=0.48684 acc_top5_avg=0.86123 lr=0.01000 gn=7.23427 time=57.04it/s +epoch=31 global_step=12500 loss=4.82723 loss_avg=5.13694 acc=0.53125 acc_top1_avg=0.48714 acc_top5_avg=0.86158 lr=0.01000 gn=5.48592 time=55.05it/s +====================Eval==================== +epoch=31 global_step=12512 loss=1.35555 test_loss_avg=1.35555 acc=0.60938 test_acc_avg=0.60938 test_acc_top5_avg=0.97656 time=198.91it/s +epoch=31 global_step=12512 loss=3.77844 test_loss_avg=1.77792 acc=0.00000 test_acc_avg=0.57430 test_acc_top5_avg=0.84972 time=238.46it/s +epoch=31 global_step=12512 loss=0.55571 test_loss_avg=1.49130 acc=0.87500 test_acc_avg=0.63439 test_acc_top5_avg=0.89330 time=492.64it/s +curr_acc 0.6344 +BEST_ACC 0.6363 +curr_acc_top5 0.8933 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=5.39080 loss_avg=5.16469 acc=0.46094 acc_top1_avg=0.48520 acc_top5_avg=0.86266 lr=0.01000 gn=6.74991 time=60.33it/s +epoch=32 global_step=12600 loss=5.50981 loss_avg=5.15460 acc=0.44531 acc_top1_avg=0.48606 acc_top5_avg=0.86142 lr=0.01000 gn=6.86023 time=61.95it/s +epoch=32 global_step=12650 loss=4.91733 loss_avg=5.16439 acc=0.50000 acc_top1_avg=0.48460 acc_top5_avg=0.85779 lr=0.01000 gn=5.76696 time=58.41it/s +epoch=32 global_step=12700 loss=4.78406 loss_avg=5.14682 acc=0.51562 acc_top1_avg=0.48645 acc_top5_avg=0.85992 lr=0.01000 gn=6.51737 time=59.61it/s +epoch=32 global_step=12750 loss=4.87058 loss_avg=5.14304 acc=0.51562 acc_top1_avg=0.48717 acc_top5_avg=0.86111 lr=0.01000 gn=7.25584 time=51.24it/s +epoch=32 global_step=12800 loss=4.37521 loss_avg=5.14842 acc=0.57031 acc_top1_avg=0.48665 acc_top5_avg=0.86263 lr=0.01000 gn=5.46379 time=51.22it/s +epoch=32 global_step=12850 loss=4.90307 loss_avg=5.14484 acc=0.51562 acc_top1_avg=0.48703 acc_top5_avg=0.86409 lr=0.01000 gn=6.44829 time=54.86it/s +epoch=32 global_step=12900 loss=4.53704 loss_avg=5.14178 acc=0.54688 acc_top1_avg=0.48748 acc_top5_avg=0.86376 lr=0.01000 gn=5.26055 time=55.89it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.06133 test_loss_avg=1.13954 acc=0.76562 test_acc_avg=0.70987 test_acc_top5_avg=0.96058 time=232.63it/s +epoch=32 global_step=12903 loss=0.63089 test_loss_avg=1.62068 acc=0.85156 test_acc_avg=0.58724 test_acc_top5_avg=0.92535 time=241.23it/s +epoch=32 global_step=12903 loss=0.11668 test_loss_avg=1.51127 acc=0.93750 test_acc_avg=0.61481 test_acc_top5_avg=0.93107 time=577.01it/s +curr_acc 0.6148 +BEST_ACC 0.6363 +curr_acc_top5 0.9311 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=5.40268 loss_avg=5.08397 acc=0.46875 acc_top1_avg=0.49186 acc_top5_avg=0.87783 lr=0.01000 gn=6.61715 time=59.16it/s +epoch=33 global_step=13000 loss=4.92121 loss_avg=5.11598 acc=0.51562 acc_top1_avg=0.49017 acc_top5_avg=0.86985 lr=0.01000 gn=6.94966 time=59.11it/s +epoch=33 global_step=13050 loss=6.30063 loss_avg=5.14607 acc=0.37500 acc_top1_avg=0.48629 acc_top5_avg=0.86682 lr=0.01000 gn=5.25779 time=28.11it/s +epoch=33 global_step=13100 loss=5.30207 loss_avg=5.12909 acc=0.48438 acc_top1_avg=0.48878 acc_top5_avg=0.86409 lr=0.01000 gn=5.49244 time=55.27it/s +epoch=33 global_step=13150 loss=5.47492 loss_avg=5.12560 acc=0.44531 acc_top1_avg=0.48937 acc_top5_avg=0.86330 lr=0.01000 gn=8.22682 time=50.52it/s +epoch=33 global_step=13200 loss=5.53948 loss_avg=5.12962 acc=0.45312 acc_top1_avg=0.48914 acc_top5_avg=0.86264 lr=0.01000 gn=8.95716 time=56.72it/s +epoch=33 global_step=13250 loss=5.18854 loss_avg=5.13555 acc=0.47656 acc_top1_avg=0.48838 acc_top5_avg=0.86302 lr=0.01000 gn=6.46026 time=61.33it/s +====================Eval==================== +epoch=33 global_step=13294 loss=1.45027 test_loss_avg=1.90374 acc=0.60938 test_acc_avg=0.55378 test_acc_top5_avg=0.87718 time=236.90it/s +epoch=33 global_step=13294 loss=0.06609 test_loss_avg=1.59958 acc=1.00000 test_acc_avg=0.60492 test_acc_top5_avg=0.91228 time=809.55it/s +curr_acc 0.6049 +BEST_ACC 0.6363 +curr_acc_top5 0.9123 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=5.02434 loss_avg=5.14020 acc=0.49219 acc_top1_avg=0.49089 acc_top5_avg=0.85286 lr=0.01000 gn=6.79872 time=62.12it/s +epoch=34 global_step=13350 loss=5.51147 loss_avg=5.10798 acc=0.45312 acc_top1_avg=0.49065 acc_top5_avg=0.85896 lr=0.01000 gn=7.27125 time=63.46it/s +epoch=34 global_step=13400 loss=5.28534 loss_avg=5.11302 acc=0.46875 acc_top1_avg=0.48983 acc_top5_avg=0.86055 lr=0.01000 gn=5.86050 time=56.07it/s +epoch=34 global_step=13450 loss=4.72754 loss_avg=5.12360 acc=0.53906 acc_top1_avg=0.48868 acc_top5_avg=0.86338 lr=0.01000 gn=8.00323 time=54.11it/s +epoch=34 global_step=13500 loss=4.39888 loss_avg=5.13305 acc=0.57812 acc_top1_avg=0.48813 acc_top5_avg=0.86400 lr=0.01000 gn=7.76621 time=51.25it/s +epoch=34 global_step=13550 loss=5.73503 loss_avg=5.13661 acc=0.42188 acc_top1_avg=0.48752 acc_top5_avg=0.86557 lr=0.01000 gn=5.99802 time=61.54it/s +epoch=34 global_step=13600 loss=5.47301 loss_avg=5.15922 acc=0.46094 acc_top1_avg=0.48550 acc_top5_avg=0.86354 lr=0.01000 gn=6.08468 time=56.19it/s +epoch=34 global_step=13650 loss=4.74840 loss_avg=5.15342 acc=0.53906 acc_top1_avg=0.48569 acc_top5_avg=0.86350 lr=0.01000 gn=6.21602 time=53.31it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.05839 test_loss_avg=0.36049 acc=0.98438 test_acc_avg=0.87835 test_acc_top5_avg=1.00000 time=234.82it/s +epoch=34 global_step=13685 loss=0.86391 test_loss_avg=2.04303 acc=0.78906 test_acc_avg=0.51917 test_acc_top5_avg=0.81580 time=230.53it/s +epoch=34 global_step=13685 loss=1.44058 test_loss_avg=1.86354 acc=0.75000 test_acc_avg=0.56151 test_acc_top5_avg=0.84721 time=504.79it/s +curr_acc 0.5615 +BEST_ACC 0.6363 +curr_acc_top5 0.8472 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=5.67195 loss_avg=5.03737 acc=0.39844 acc_top1_avg=0.49740 acc_top5_avg=0.86510 lr=0.01000 gn=7.31582 time=52.33it/s +epoch=35 global_step=13750 loss=4.73528 loss_avg=5.05479 acc=0.51562 acc_top1_avg=0.49435 acc_top5_avg=0.86671 lr=0.01000 gn=6.38370 time=60.52it/s +epoch=35 global_step=13800 loss=4.86427 loss_avg=5.05245 acc=0.50781 acc_top1_avg=0.49504 acc_top5_avg=0.86834 lr=0.01000 gn=5.98268 time=58.05it/s +epoch=35 global_step=13850 loss=4.71893 loss_avg=5.09151 acc=0.51562 acc_top1_avg=0.49129 acc_top5_avg=0.86581 lr=0.01000 gn=6.09189 time=60.51it/s +epoch=35 global_step=13900 loss=5.15072 loss_avg=5.10353 acc=0.46875 acc_top1_avg=0.49084 acc_top5_avg=0.86704 lr=0.01000 gn=4.75187 time=56.53it/s +epoch=35 global_step=13950 loss=5.16898 loss_avg=5.09568 acc=0.49219 acc_top1_avg=0.49183 acc_top5_avg=0.86728 lr=0.01000 gn=6.85515 time=59.30it/s +epoch=35 global_step=14000 loss=5.54595 loss_avg=5.11820 acc=0.45312 acc_top1_avg=0.48909 acc_top5_avg=0.86582 lr=0.01000 gn=6.01278 time=61.21it/s +epoch=35 global_step=14050 loss=5.33963 loss_avg=5.12872 acc=0.47656 acc_top1_avg=0.48795 acc_top5_avg=0.86468 lr=0.01000 gn=8.71855 time=54.48it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.07281 test_loss_avg=2.35461 acc=0.98438 test_acc_avg=0.50357 test_acc_top5_avg=0.88036 time=237.07it/s +epoch=35 global_step=14076 loss=0.03470 test_loss_avg=1.86629 acc=1.00000 test_acc_avg=0.57269 test_acc_top5_avg=0.92989 time=858.43it/s +curr_acc 0.5727 +BEST_ACC 0.6363 +curr_acc_top5 0.9299 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=5.03555 loss_avg=5.12420 acc=0.51562 acc_top1_avg=0.48893 acc_top5_avg=0.86393 lr=0.01000 gn=7.55666 time=56.39it/s +epoch=36 global_step=14150 loss=5.68783 loss_avg=5.12974 acc=0.43750 acc_top1_avg=0.48807 acc_top5_avg=0.86497 lr=0.01000 gn=6.11603 time=63.40it/s +epoch=36 global_step=14200 loss=4.85333 loss_avg=5.11814 acc=0.51562 acc_top1_avg=0.48834 acc_top5_avg=0.86429 lr=0.01000 gn=7.49793 time=53.36it/s +epoch=36 global_step=14250 loss=5.07276 loss_avg=5.15369 acc=0.52344 acc_top1_avg=0.48608 acc_top5_avg=0.86422 lr=0.01000 gn=7.01626 time=54.37it/s +epoch=36 global_step=14300 loss=4.89856 loss_avg=5.15229 acc=0.51562 acc_top1_avg=0.48647 acc_top5_avg=0.86283 lr=0.01000 gn=7.25761 time=56.38it/s +epoch=36 global_step=14350 loss=5.22871 loss_avg=5.16386 acc=0.46094 acc_top1_avg=0.48526 acc_top5_avg=0.86377 lr=0.01000 gn=6.50710 time=58.76it/s +epoch=36 global_step=14400 loss=5.07378 loss_avg=5.14123 acc=0.50000 acc_top1_avg=0.48741 acc_top5_avg=0.86441 lr=0.01000 gn=5.64446 time=59.17it/s +epoch=36 global_step=14450 loss=5.15171 loss_avg=5.13403 acc=0.49219 acc_top1_avg=0.48826 acc_top5_avg=0.86495 lr=0.01000 gn=5.72855 time=56.14it/s +====================Eval==================== +epoch=36 global_step=14467 loss=0.55770 test_loss_avg=0.55750 acc=0.85156 test_acc_avg=0.84766 test_acc_top5_avg=0.98568 time=241.62it/s +epoch=36 global_step=14467 loss=0.75719 test_loss_avg=1.76165 acc=0.78906 test_acc_avg=0.55287 test_acc_top5_avg=0.87500 time=237.17it/s +epoch=36 global_step=14467 loss=0.59628 test_loss_avg=1.43705 acc=0.75000 test_acc_avg=0.62777 test_acc_top5_avg=0.90674 time=500.39it/s +curr_acc 0.6278 +BEST_ACC 0.6363 +curr_acc_top5 0.9067 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=4.85670 loss_avg=5.13035 acc=0.51562 acc_top1_avg=0.48651 acc_top5_avg=0.87027 lr=0.01000 gn=6.24765 time=59.99it/s +epoch=37 global_step=14550 loss=5.06957 loss_avg=5.11798 acc=0.49219 acc_top1_avg=0.49106 acc_top5_avg=0.86803 lr=0.01000 gn=6.08091 time=55.07it/s +epoch=37 global_step=14600 loss=5.08624 loss_avg=5.12533 acc=0.50000 acc_top1_avg=0.49037 acc_top5_avg=0.86977 lr=0.01000 gn=6.81330 time=61.48it/s +epoch=37 global_step=14650 loss=4.29690 loss_avg=5.10665 acc=0.55469 acc_top1_avg=0.49108 acc_top5_avg=0.86877 lr=0.01000 gn=7.59916 time=62.63it/s +epoch=37 global_step=14700 loss=5.22865 loss_avg=5.12888 acc=0.47656 acc_top1_avg=0.48914 acc_top5_avg=0.86571 lr=0.01000 gn=5.23471 time=62.66it/s +epoch=37 global_step=14750 loss=4.38919 loss_avg=5.10991 acc=0.57812 acc_top1_avg=0.49056 acc_top5_avg=0.86622 lr=0.01000 gn=5.97155 time=57.12it/s +epoch=37 global_step=14800 loss=5.87314 loss_avg=5.11895 acc=0.39844 acc_top1_avg=0.48940 acc_top5_avg=0.86501 lr=0.01000 gn=7.18780 time=54.22it/s +epoch=37 global_step=14850 loss=5.04533 loss_avg=5.13520 acc=0.49219 acc_top1_avg=0.48807 acc_top5_avg=0.86421 lr=0.01000 gn=6.38679 time=58.84it/s +====================Eval==================== +epoch=37 global_step=14858 loss=5.10877 test_loss_avg=2.32431 acc=0.00000 test_acc_avg=0.51042 test_acc_top5_avg=0.87182 time=147.70it/s +epoch=37 global_step=14858 loss=0.04255 test_loss_avg=1.89977 acc=0.99219 test_acc_avg=0.57539 test_acc_top5_avg=0.92188 time=240.50it/s +epoch=37 global_step=14858 loss=0.00137 test_loss_avg=1.85237 acc=1.00000 test_acc_avg=0.58594 test_acc_top5_avg=0.92385 time=527.32it/s +curr_acc 0.5859 +BEST_ACC 0.6363 +curr_acc_top5 0.9239 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=4.91582 loss_avg=5.10495 acc=0.50781 acc_top1_avg=0.49070 acc_top5_avg=0.86607 lr=0.01000 gn=5.36826 time=55.01it/s +epoch=38 global_step=14950 loss=4.70550 loss_avg=5.08716 acc=0.52344 acc_top1_avg=0.49253 acc_top5_avg=0.86574 lr=0.01000 gn=4.97387 time=52.83it/s +epoch=38 global_step=15000 loss=4.89805 loss_avg=5.05692 acc=0.50000 acc_top1_avg=0.49609 acc_top5_avg=0.86807 lr=0.01000 gn=6.24302 time=58.22it/s +epoch=38 global_step=15050 loss=5.10060 loss_avg=5.08783 acc=0.47656 acc_top1_avg=0.49276 acc_top5_avg=0.86560 lr=0.01000 gn=5.11671 time=60.91it/s +epoch=38 global_step=15100 loss=3.81098 loss_avg=5.08516 acc=0.62500 acc_top1_avg=0.49293 acc_top5_avg=0.86451 lr=0.01000 gn=5.56052 time=61.36it/s +epoch=38 global_step=15150 loss=5.54500 loss_avg=5.11246 acc=0.43750 acc_top1_avg=0.48965 acc_top5_avg=0.86398 lr=0.01000 gn=6.65445 time=62.65it/s +epoch=38 global_step=15200 loss=4.96956 loss_avg=5.12127 acc=0.51562 acc_top1_avg=0.48922 acc_top5_avg=0.86385 lr=0.01000 gn=6.88790 time=56.78it/s +====================Eval==================== +epoch=38 global_step=15249 loss=4.04083 test_loss_avg=2.20147 acc=0.00000 test_acc_avg=0.54281 test_acc_top5_avg=0.88851 time=223.10it/s +epoch=38 global_step=15249 loss=0.00146 test_loss_avg=1.90847 acc=1.00000 test_acc_avg=0.58475 test_acc_top5_avg=0.92286 time=843.25it/s +curr_acc 0.5848 +BEST_ACC 0.6363 +curr_acc_top5 0.9229 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=4.54681 loss_avg=4.54681 acc=0.56250 acc_top1_avg=0.56250 acc_top5_avg=0.89062 lr=0.01000 gn=5.59344 time=30.31it/s +epoch=39 global_step=15300 loss=5.61577 loss_avg=5.15025 acc=0.46094 acc_top1_avg=0.48729 acc_top5_avg=0.86949 lr=0.01000 gn=6.19667 time=55.87it/s +epoch=39 global_step=15350 loss=5.09511 loss_avg=5.10259 acc=0.51562 acc_top1_avg=0.49234 acc_top5_avg=0.86943 lr=0.01000 gn=7.52910 time=62.96it/s +epoch=39 global_step=15400 loss=5.34744 loss_avg=5.11184 acc=0.46875 acc_top1_avg=0.49115 acc_top5_avg=0.86807 lr=0.01000 gn=6.07223 time=60.69it/s +epoch=39 global_step=15450 loss=5.58409 loss_avg=5.11311 acc=0.45312 acc_top1_avg=0.49122 acc_top5_avg=0.86715 lr=0.01000 gn=6.22039 time=56.27it/s +epoch=39 global_step=15500 loss=5.02689 loss_avg=5.11182 acc=0.50000 acc_top1_avg=0.49097 acc_top5_avg=0.86719 lr=0.01000 gn=7.35613 time=59.96it/s +epoch=39 global_step=15550 loss=4.64094 loss_avg=5.11188 acc=0.53906 acc_top1_avg=0.49105 acc_top5_avg=0.86675 lr=0.01000 gn=6.49476 time=59.86it/s +epoch=39 global_step=15600 loss=5.85092 loss_avg=5.12084 acc=0.41406 acc_top1_avg=0.48989 acc_top5_avg=0.86632 lr=0.01000 gn=7.04864 time=53.35it/s +====================Eval==================== +epoch=39 global_step=15640 loss=2.86171 test_loss_avg=1.39240 acc=0.37500 test_acc_avg=0.64556 test_acc_top5_avg=0.91694 time=247.20it/s +epoch=39 global_step=15640 loss=0.51649 test_loss_avg=1.79604 acc=0.86719 test_acc_avg=0.56069 test_acc_top5_avg=0.92154 time=239.66it/s +epoch=39 global_step=15640 loss=0.62978 test_loss_avg=1.64063 acc=0.87500 test_acc_avg=0.59820 test_acc_top5_avg=0.93028 time=631.77it/s +curr_acc 0.5982 +BEST_ACC 0.6363 +curr_acc_top5 0.9303 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=5.09691 loss_avg=4.99102 acc=0.50781 acc_top1_avg=0.51250 acc_top5_avg=0.87500 lr=0.00100 gn=6.78206 time=56.48it/s +epoch=40 global_step=15700 loss=5.14409 loss_avg=4.98703 acc=0.48438 acc_top1_avg=0.50273 acc_top5_avg=0.87135 lr=0.00100 gn=6.97372 time=62.85it/s +epoch=40 global_step=15750 loss=4.71718 loss_avg=4.91565 acc=0.53125 acc_top1_avg=0.50874 acc_top5_avg=0.87436 lr=0.00100 gn=6.09291 time=50.58it/s +epoch=40 global_step=15800 loss=5.10158 loss_avg=4.90654 acc=0.50781 acc_top1_avg=0.51045 acc_top5_avg=0.87451 lr=0.00100 gn=6.99519 time=54.80it/s +epoch=40 global_step=15850 loss=4.98246 loss_avg=4.88072 acc=0.48438 acc_top1_avg=0.51399 acc_top5_avg=0.87388 lr=0.00100 gn=5.10951 time=55.51it/s +epoch=40 global_step=15900 loss=4.83302 loss_avg=4.86544 acc=0.50781 acc_top1_avg=0.51526 acc_top5_avg=0.87398 lr=0.00100 gn=6.37439 time=51.47it/s +epoch=40 global_step=15950 loss=4.54621 loss_avg=4.86527 acc=0.55469 acc_top1_avg=0.51527 acc_top5_avg=0.87490 lr=0.00100 gn=6.03874 time=45.29it/s +epoch=40 global_step=16000 loss=5.27509 loss_avg=4.86367 acc=0.47656 acc_top1_avg=0.51545 acc_top5_avg=0.87661 lr=0.00100 gn=5.75036 time=51.29it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.71471 test_loss_avg=1.40052 acc=0.80469 test_acc_avg=0.66621 test_acc_top5_avg=0.89473 time=234.06it/s +epoch=40 global_step=16031 loss=0.00396 test_loss_avg=1.17431 acc=1.00000 test_acc_avg=0.69254 test_acc_top5_avg=0.93908 time=488.33it/s +curr_acc 0.6925 +BEST_ACC 0.6363 +curr_acc_top5 0.9391 +BEST_ACC_top5 0.9348 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=4.75487 loss_avg=4.60657 acc=0.54688 acc_top1_avg=0.54071 acc_top5_avg=0.88035 lr=0.00100 gn=4.66072 time=56.69it/s +epoch=41 global_step=16100 loss=5.14160 loss_avg=4.76028 acc=0.48438 acc_top1_avg=0.52683 acc_top5_avg=0.87432 lr=0.00100 gn=5.90212 time=55.14it/s +epoch=41 global_step=16150 loss=4.52895 loss_avg=4.72166 acc=0.56250 acc_top1_avg=0.53125 acc_top5_avg=0.87848 lr=0.00100 gn=7.35466 time=48.13it/s +epoch=41 global_step=16200 loss=4.04658 loss_avg=4.70986 acc=0.58594 acc_top1_avg=0.53199 acc_top5_avg=0.87907 lr=0.00100 gn=5.99903 time=54.59it/s +epoch=41 global_step=16250 loss=5.24897 loss_avg=4.71978 acc=0.49219 acc_top1_avg=0.53107 acc_top5_avg=0.87835 lr=0.00100 gn=6.43121 time=62.72it/s +epoch=41 global_step=16300 loss=4.44458 loss_avg=4.72664 acc=0.56250 acc_top1_avg=0.52957 acc_top5_avg=0.88040 lr=0.00100 gn=5.47977 time=51.02it/s +epoch=41 global_step=16350 loss=4.71008 loss_avg=4.74023 acc=0.52344 acc_top1_avg=0.52829 acc_top5_avg=0.88098 lr=0.00100 gn=7.27962 time=59.28it/s +epoch=41 global_step=16400 loss=4.34411 loss_avg=4.73575 acc=0.56250 acc_top1_avg=0.52858 acc_top5_avg=0.88182 lr=0.00100 gn=5.76056 time=55.44it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.16264 test_loss_avg=0.59297 acc=0.95312 test_acc_avg=0.83594 test_acc_top5_avg=0.98438 time=240.28it/s +epoch=41 global_step=16422 loss=0.27184 test_loss_avg=1.40454 acc=0.92969 test_acc_avg=0.62526 test_acc_top5_avg=0.93058 time=238.38it/s +epoch=41 global_step=16422 loss=0.00726 test_loss_avg=1.12557 acc=1.00000 test_acc_avg=0.69897 test_acc_top5_avg=0.94541 time=490.85it/s +curr_acc 0.6990 +BEST_ACC 0.6925 +curr_acc_top5 0.9454 +BEST_ACC_top5 0.9391 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=4.67144 loss_avg=4.66493 acc=0.54688 acc_top1_avg=0.53767 acc_top5_avg=0.88170 lr=0.00100 gn=5.95950 time=53.43it/s +epoch=42 global_step=16500 loss=4.58184 loss_avg=4.65852 acc=0.53125 acc_top1_avg=0.53656 acc_top5_avg=0.88081 lr=0.00100 gn=7.07120 time=55.40it/s +epoch=42 global_step=16550 loss=4.55043 loss_avg=4.62982 acc=0.54688 acc_top1_avg=0.53955 acc_top5_avg=0.88287 lr=0.00100 gn=6.35325 time=53.69it/s +epoch=42 global_step=16600 loss=4.46469 loss_avg=4.64997 acc=0.54688 acc_top1_avg=0.53718 acc_top5_avg=0.88430 lr=0.00100 gn=5.67925 time=62.05it/s +epoch=42 global_step=16650 loss=4.50535 loss_avg=4.67416 acc=0.54688 acc_top1_avg=0.53416 acc_top5_avg=0.88374 lr=0.00100 gn=6.52544 time=58.93it/s +epoch=42 global_step=16700 loss=4.61437 loss_avg=4.69399 acc=0.53906 acc_top1_avg=0.53277 acc_top5_avg=0.88228 lr=0.00100 gn=7.33761 time=56.93it/s +epoch=42 global_step=16750 loss=4.67935 loss_avg=4.69859 acc=0.53906 acc_top1_avg=0.53225 acc_top5_avg=0.88200 lr=0.00100 gn=6.81201 time=59.70it/s +epoch=42 global_step=16800 loss=4.84655 loss_avg=4.69140 acc=0.50781 acc_top1_avg=0.53288 acc_top5_avg=0.88329 lr=0.00100 gn=7.22521 time=59.35it/s +====================Eval==================== +epoch=42 global_step=16813 loss=1.40106 test_loss_avg=1.57782 acc=0.67188 test_acc_avg=0.62012 test_acc_top5_avg=0.88354 time=217.58it/s +epoch=42 global_step=16813 loss=0.00515 test_loss_avg=1.10796 acc=1.00000 test_acc_avg=0.69828 test_acc_top5_avg=0.94600 time=647.07it/s +curr_acc 0.6983 +BEST_ACC 0.6990 +curr_acc_top5 0.9460 +BEST_ACC_top5 0.9454 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=4.52921 loss_avg=4.71062 acc=0.55469 acc_top1_avg=0.53526 acc_top5_avg=0.88450 lr=0.00100 gn=6.36718 time=55.66it/s +epoch=43 global_step=16900 loss=4.99005 loss_avg=4.68190 acc=0.50000 acc_top1_avg=0.53718 acc_top5_avg=0.88587 lr=0.00100 gn=6.14273 time=52.77it/s +epoch=43 global_step=16950 loss=4.66447 loss_avg=4.64642 acc=0.53906 acc_top1_avg=0.53929 acc_top5_avg=0.88509 lr=0.00100 gn=6.28213 time=42.58it/s +epoch=43 global_step=17000 loss=4.85893 loss_avg=4.66136 acc=0.53125 acc_top1_avg=0.53727 acc_top5_avg=0.88478 lr=0.00100 gn=6.84614 time=53.94it/s +epoch=43 global_step=17050 loss=4.56058 loss_avg=4.66672 acc=0.55469 acc_top1_avg=0.53639 acc_top5_avg=0.88538 lr=0.00100 gn=6.82256 time=57.03it/s +epoch=43 global_step=17100 loss=5.21350 loss_avg=4.67111 acc=0.48438 acc_top1_avg=0.53601 acc_top5_avg=0.88559 lr=0.00100 gn=7.24780 time=54.12it/s +epoch=43 global_step=17150 loss=5.27537 loss_avg=4.65801 acc=0.47656 acc_top1_avg=0.53681 acc_top5_avg=0.88580 lr=0.00100 gn=6.56360 time=52.37it/s +epoch=43 global_step=17200 loss=4.19958 loss_avg=4.65395 acc=0.58594 acc_top1_avg=0.53737 acc_top5_avg=0.88630 lr=0.00100 gn=8.39725 time=55.11it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.77352 test_loss_avg=0.80441 acc=0.76562 test_acc_avg=0.77604 test_acc_top5_avg=0.97396 time=236.38it/s +epoch=43 global_step=17204 loss=3.47394 test_loss_avg=1.53594 acc=0.00000 test_acc_avg=0.60761 test_acc_top5_avg=0.92394 time=235.99it/s +epoch=43 global_step=17204 loss=0.00090 test_loss_avg=1.16523 acc=1.00000 test_acc_avg=0.69709 test_acc_top5_avg=0.94610 time=828.26it/s +curr_acc 0.6971 +BEST_ACC 0.6990 +curr_acc_top5 0.9461 +BEST_ACC_top5 0.9460 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=5.03721 loss_avg=4.63929 acc=0.50781 acc_top1_avg=0.53923 acc_top5_avg=0.88944 lr=0.00100 gn=6.76891 time=59.47it/s +epoch=44 global_step=17300 loss=4.18475 loss_avg=4.67203 acc=0.58594 acc_top1_avg=0.53670 acc_top5_avg=0.89030 lr=0.00100 gn=6.31739 time=58.83it/s +epoch=44 global_step=17350 loss=4.52171 loss_avg=4.65004 acc=0.55469 acc_top1_avg=0.53853 acc_top5_avg=0.88945 lr=0.00100 gn=6.72540 time=54.52it/s +epoch=44 global_step=17400 loss=3.96815 loss_avg=4.64152 acc=0.60938 acc_top1_avg=0.53946 acc_top5_avg=0.88807 lr=0.00100 gn=6.25815 time=50.55it/s +epoch=44 global_step=17450 loss=4.49881 loss_avg=4.64824 acc=0.56250 acc_top1_avg=0.53843 acc_top5_avg=0.88704 lr=0.00100 gn=7.80601 time=59.86it/s +epoch=44 global_step=17500 loss=4.81408 loss_avg=4.62646 acc=0.53125 acc_top1_avg=0.54067 acc_top5_avg=0.88735 lr=0.00100 gn=8.37273 time=53.79it/s +epoch=44 global_step=17550 loss=4.86948 loss_avg=4.62001 acc=0.51562 acc_top1_avg=0.54118 acc_top5_avg=0.88785 lr=0.00100 gn=7.51857 time=54.35it/s +====================Eval==================== +epoch=44 global_step=17595 loss=3.13471 test_loss_avg=0.76655 acc=0.27344 test_acc_avg=0.79980 test_acc_top5_avg=0.96712 time=47.83it/s +epoch=44 global_step=17595 loss=0.18971 test_loss_avg=1.18418 acc=0.93750 test_acc_avg=0.68243 test_acc_top5_avg=0.94975 time=248.60it/s +epoch=44 global_step=17595 loss=0.00494 test_loss_avg=1.11605 acc=1.00000 test_acc_avg=0.70016 test_acc_top5_avg=0.95293 time=819.84it/s +curr_acc 0.7002 +BEST_ACC 0.6990 +curr_acc_top5 0.9529 +BEST_ACC_top5 0.9461 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=3.98042 loss_avg=4.52276 acc=0.60938 acc_top1_avg=0.55000 acc_top5_avg=0.90312 lr=0.00100 gn=6.82790 time=55.10it/s +epoch=45 global_step=17650 loss=4.33621 loss_avg=4.58023 acc=0.57031 acc_top1_avg=0.54446 acc_top5_avg=0.88778 lr=0.00100 gn=6.44514 time=54.57it/s +epoch=45 global_step=17700 loss=4.08729 loss_avg=4.59431 acc=0.60156 acc_top1_avg=0.54368 acc_top5_avg=0.88780 lr=0.00100 gn=6.65761 time=55.46it/s +epoch=45 global_step=17750 loss=5.05988 loss_avg=4.61302 acc=0.49219 acc_top1_avg=0.54173 acc_top5_avg=0.88755 lr=0.00100 gn=6.24101 time=63.42it/s +epoch=45 global_step=17800 loss=4.79273 loss_avg=4.61680 acc=0.50000 acc_top1_avg=0.54074 acc_top5_avg=0.88811 lr=0.00100 gn=10.78672 time=53.56it/s +epoch=45 global_step=17850 loss=4.57985 loss_avg=4.59719 acc=0.56250 acc_top1_avg=0.54338 acc_top5_avg=0.88925 lr=0.00100 gn=8.41192 time=55.13it/s +epoch=45 global_step=17900 loss=4.52978 loss_avg=4.60768 acc=0.55469 acc_top1_avg=0.54193 acc_top5_avg=0.88778 lr=0.00100 gn=7.42710 time=55.85it/s +epoch=45 global_step=17950 loss=4.64657 loss_avg=4.59555 acc=0.54688 acc_top1_avg=0.54364 acc_top5_avg=0.88776 lr=0.00100 gn=7.14724 time=54.85it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.82102 test_loss_avg=1.30434 acc=0.78125 test_acc_avg=0.68455 test_acc_top5_avg=0.91771 time=116.74it/s +epoch=45 global_step=17986 loss=0.00624 test_loss_avg=1.14730 acc=1.00000 test_acc_avg=0.69264 test_acc_top5_avg=0.94848 time=524.42it/s +curr_acc 0.6926 +BEST_ACC 0.7002 +curr_acc_top5 0.9485 +BEST_ACC_top5 0.9529 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=4.16002 loss_avg=4.58615 acc=0.58594 acc_top1_avg=0.54353 acc_top5_avg=0.88895 lr=0.00100 gn=7.19125 time=50.12it/s +epoch=46 global_step=18050 loss=4.42749 loss_avg=4.56853 acc=0.57031 acc_top1_avg=0.54553 acc_top5_avg=0.88916 lr=0.00100 gn=7.78364 time=58.42it/s +epoch=46 global_step=18100 loss=4.23259 loss_avg=4.55762 acc=0.57031 acc_top1_avg=0.54735 acc_top5_avg=0.89186 lr=0.00100 gn=8.30327 time=52.68it/s +epoch=46 global_step=18150 loss=4.90378 loss_avg=4.55211 acc=0.51562 acc_top1_avg=0.54797 acc_top5_avg=0.89110 lr=0.00100 gn=8.85659 time=54.96it/s +epoch=46 global_step=18200 loss=5.08260 loss_avg=4.56609 acc=0.49219 acc_top1_avg=0.54625 acc_top5_avg=0.89070 lr=0.00100 gn=8.32720 time=61.66it/s +epoch=46 global_step=18250 loss=4.39869 loss_avg=4.55992 acc=0.56250 acc_top1_avg=0.54679 acc_top5_avg=0.89071 lr=0.00100 gn=8.68125 time=51.10it/s +epoch=46 global_step=18300 loss=4.29262 loss_avg=4.57460 acc=0.57812 acc_top1_avg=0.54471 acc_top5_avg=0.89045 lr=0.00100 gn=9.02105 time=50.29it/s +epoch=46 global_step=18350 loss=4.47432 loss_avg=4.57568 acc=0.57812 acc_top1_avg=0.54417 acc_top5_avg=0.89015 lr=0.00100 gn=9.30494 time=55.62it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.64884 test_loss_avg=0.42572 acc=0.83594 test_acc_avg=0.87891 test_acc_top5_avg=0.99023 time=219.93it/s +epoch=46 global_step=18377 loss=0.16583 test_loss_avg=1.31216 acc=0.95312 test_acc_avg=0.65459 test_acc_top5_avg=0.93939 time=221.09it/s +epoch=46 global_step=18377 loss=0.02833 test_loss_avg=1.12133 acc=1.00000 test_acc_avg=0.70342 test_acc_top5_avg=0.94927 time=500.10it/s +curr_acc 0.7034 +BEST_ACC 0.7002 +curr_acc_top5 0.9493 +BEST_ACC_top5 0.9529 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=4.38429 loss_avg=4.56942 acc=0.56250 acc_top1_avg=0.54348 acc_top5_avg=0.88519 lr=0.00100 gn=9.16837 time=59.31it/s +epoch=47 global_step=18450 loss=4.42128 loss_avg=4.54127 acc=0.57031 acc_top1_avg=0.54827 acc_top5_avg=0.88741 lr=0.00100 gn=10.12649 time=50.45it/s +epoch=47 global_step=18500 loss=4.86349 loss_avg=4.56607 acc=0.50000 acc_top1_avg=0.54580 acc_top5_avg=0.88599 lr=0.00100 gn=7.85039 time=53.83it/s +epoch=47 global_step=18550 loss=4.56723 loss_avg=4.57184 acc=0.56250 acc_top1_avg=0.54543 acc_top5_avg=0.88764 lr=0.00100 gn=9.76709 time=44.83it/s +epoch=47 global_step=18600 loss=3.66805 loss_avg=4.54339 acc=0.67188 acc_top1_avg=0.54919 acc_top5_avg=0.88877 lr=0.00100 gn=9.52360 time=55.47it/s +epoch=47 global_step=18650 loss=4.39436 loss_avg=4.52017 acc=0.57031 acc_top1_avg=0.55183 acc_top5_avg=0.88948 lr=0.00100 gn=9.14126 time=55.64it/s +epoch=47 global_step=18700 loss=4.53847 loss_avg=4.52180 acc=0.54688 acc_top1_avg=0.55130 acc_top5_avg=0.89041 lr=0.00100 gn=8.35340 time=61.42it/s +epoch=47 global_step=18750 loss=5.20965 loss_avg=4.54681 acc=0.44531 acc_top1_avg=0.54863 acc_top5_avg=0.88979 lr=0.00100 gn=9.81562 time=56.38it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.23979 test_loss_avg=1.30567 acc=0.90625 test_acc_avg=0.67673 test_acc_top5_avg=0.90625 time=232.69it/s +epoch=47 global_step=18768 loss=0.00778 test_loss_avg=1.08973 acc=1.00000 test_acc_avg=0.70560 test_acc_top5_avg=0.94670 time=809.55it/s +curr_acc 0.7056 +BEST_ACC 0.7034 +curr_acc_top5 0.9467 +BEST_ACC_top5 0.9529 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=4.49933 loss_avg=4.48575 acc=0.53125 acc_top1_avg=0.55493 acc_top5_avg=0.88647 lr=0.00100 gn=8.04816 time=52.27it/s +epoch=48 global_step=18850 loss=4.78296 loss_avg=4.48350 acc=0.51562 acc_top1_avg=0.55688 acc_top5_avg=0.88977 lr=0.00100 gn=7.81187 time=63.59it/s +epoch=48 global_step=18900 loss=5.09085 loss_avg=4.46961 acc=0.48438 acc_top1_avg=0.55741 acc_top5_avg=0.89015 lr=0.00100 gn=11.20553 time=55.56it/s +epoch=48 global_step=18950 loss=5.00038 loss_avg=4.49096 acc=0.50000 acc_top1_avg=0.55520 acc_top5_avg=0.89084 lr=0.00100 gn=8.91015 time=53.62it/s +epoch=48 global_step=19000 loss=4.73349 loss_avg=4.50862 acc=0.53125 acc_top1_avg=0.55331 acc_top5_avg=0.89032 lr=0.00100 gn=7.76338 time=60.67it/s +epoch=48 global_step=19050 loss=4.26716 loss_avg=4.50155 acc=0.57812 acc_top1_avg=0.55399 acc_top5_avg=0.89157 lr=0.00100 gn=8.50370 time=57.29it/s +epoch=48 global_step=19100 loss=4.61896 loss_avg=4.51178 acc=0.53906 acc_top1_avg=0.55304 acc_top5_avg=0.89138 lr=0.00100 gn=10.08387 time=50.55it/s +epoch=48 global_step=19150 loss=4.44219 loss_avg=4.51328 acc=0.57031 acc_top1_avg=0.55289 acc_top5_avg=0.89140 lr=0.00100 gn=11.94951 time=58.44it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.44518 test_loss_avg=0.68107 acc=0.89844 test_acc_avg=0.81250 test_acc_top5_avg=0.98730 time=239.69it/s +epoch=48 global_step=19159 loss=0.32363 test_loss_avg=1.51859 acc=0.90625 test_acc_avg=0.60816 test_acc_top5_avg=0.93777 time=231.28it/s +epoch=48 global_step=19159 loss=0.00289 test_loss_avg=1.16508 acc=1.00000 test_acc_avg=0.69699 test_acc_top5_avg=0.95303 time=476.57it/s +curr_acc 0.6970 +BEST_ACC 0.7056 +curr_acc_top5 0.9530 +BEST_ACC_top5 0.9529 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=4.60725 loss_avg=4.50334 acc=0.54688 acc_top1_avg=0.55354 acc_top5_avg=0.88910 lr=0.00100 gn=10.50456 time=56.10it/s +epoch=49 global_step=19250 loss=4.18349 loss_avg=4.45206 acc=0.58594 acc_top1_avg=0.56001 acc_top5_avg=0.89011 lr=0.00100 gn=9.71193 time=51.87it/s +epoch=49 global_step=19300 loss=3.98071 loss_avg=4.42872 acc=0.59375 acc_top1_avg=0.56294 acc_top5_avg=0.89406 lr=0.00100 gn=8.03972 time=51.65it/s +epoch=49 global_step=19350 loss=4.75598 loss_avg=4.46182 acc=0.51562 acc_top1_avg=0.55915 acc_top5_avg=0.89377 lr=0.00100 gn=10.86678 time=54.28it/s +epoch=49 global_step=19400 loss=4.50589 loss_avg=4.47187 acc=0.55469 acc_top1_avg=0.55806 acc_top5_avg=0.89270 lr=0.00100 gn=9.38712 time=58.83it/s +epoch=49 global_step=19450 loss=4.35110 loss_avg=4.47957 acc=0.57812 acc_top1_avg=0.55675 acc_top5_avg=0.89288 lr=0.00100 gn=10.95390 time=54.16it/s +epoch=49 global_step=19500 loss=4.77594 loss_avg=4.48535 acc=0.52344 acc_top1_avg=0.55592 acc_top5_avg=0.89363 lr=0.00100 gn=8.13518 time=55.61it/s +epoch=49 global_step=19550 loss=3.99174 loss_avg=4.48504 acc=0.57500 acc_top1_avg=0.55608 acc_top5_avg=0.89291 lr=0.00100 gn=12.25721 time=73.90it/s +====================Eval==================== +epoch=49 global_step=19550 loss=4.18870 test_loss_avg=1.35216 acc=0.00000 test_acc_avg=0.66056 test_acc_top5_avg=0.91649 time=234.49it/s +epoch=49 global_step=19550 loss=0.00741 test_loss_avg=1.10585 acc=1.00000 test_acc_avg=0.69788 test_acc_top5_avg=0.95273 time=505.83it/s +epoch=49 global_step=19550 loss=0.00741 test_loss_avg=1.10585 acc=1.00000 test_acc_avg=0.69788 test_acc_top5_avg=0.95273 time=505.83it/s +curr_acc 0.6979 +BEST_ACC 0.7056 +curr_acc_top5 0.9527 +BEST_ACC_top5 0.9530 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=4.02617 loss_avg=4.51377 acc=0.60156 acc_top1_avg=0.55078 acc_top5_avg=0.89078 lr=0.00100 gn=11.12310 time=55.62it/s +epoch=50 global_step=19650 loss=4.85907 loss_avg=4.51273 acc=0.51562 acc_top1_avg=0.55289 acc_top5_avg=0.88891 lr=0.00100 gn=9.67504 time=63.19it/s +epoch=50 global_step=19700 loss=4.33435 loss_avg=4.48657 acc=0.57812 acc_top1_avg=0.55604 acc_top5_avg=0.89125 lr=0.00100 gn=9.09615 time=61.89it/s +epoch=50 global_step=19750 loss=4.63227 loss_avg=4.47372 acc=0.53906 acc_top1_avg=0.55758 acc_top5_avg=0.89211 lr=0.00100 gn=10.23459 time=55.75it/s +epoch=50 global_step=19800 loss=4.62510 loss_avg=4.47470 acc=0.54688 acc_top1_avg=0.55703 acc_top5_avg=0.89153 lr=0.00100 gn=11.79365 time=54.60it/s +epoch=50 global_step=19850 loss=4.65600 loss_avg=4.46292 acc=0.53125 acc_top1_avg=0.55823 acc_top5_avg=0.89187 lr=0.00100 gn=10.34587 time=51.78it/s +epoch=50 global_step=19900 loss=4.88943 loss_avg=4.46743 acc=0.50000 acc_top1_avg=0.55721 acc_top5_avg=0.89172 lr=0.00100 gn=11.95173 time=60.51it/s +====================Eval==================== +epoch=50 global_step=19941 loss=3.86393 test_loss_avg=1.42340 acc=0.00000 test_acc_avg=0.64984 test_acc_top5_avg=0.93266 time=225.72it/s +epoch=50 global_step=19941 loss=0.00176 test_loss_avg=1.18132 acc=1.00000 test_acc_avg=0.70283 test_acc_top5_avg=0.95421 time=831.21it/s +curr_acc 0.7028 +BEST_ACC 0.7056 +curr_acc_top5 0.9542 +BEST_ACC_top5 0.9530 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=4.55848 loss_avg=4.52021 acc=0.53906 acc_top1_avg=0.55990 acc_top5_avg=0.89410 lr=0.00100 gn=9.71675 time=61.23it/s +epoch=51 global_step=20000 loss=4.91895 loss_avg=4.46542 acc=0.50000 acc_top1_avg=0.55734 acc_top5_avg=0.88890 lr=0.00100 gn=10.52929 time=53.79it/s +epoch=51 global_step=20050 loss=4.31873 loss_avg=4.46413 acc=0.57812 acc_top1_avg=0.55741 acc_top5_avg=0.88998 lr=0.00100 gn=11.73753 time=61.50it/s +epoch=51 global_step=20100 loss=4.19371 loss_avg=4.44283 acc=0.57031 acc_top1_avg=0.55950 acc_top5_avg=0.89254 lr=0.00100 gn=10.62287 time=55.26it/s +epoch=51 global_step=20150 loss=4.56385 loss_avg=4.43890 acc=0.54688 acc_top1_avg=0.56089 acc_top5_avg=0.89107 lr=0.00100 gn=10.91074 time=54.38it/s +epoch=51 global_step=20200 loss=5.06797 loss_avg=4.43988 acc=0.50000 acc_top1_avg=0.56093 acc_top5_avg=0.89168 lr=0.00100 gn=9.91345 time=55.18it/s +epoch=51 global_step=20250 loss=3.81464 loss_avg=4.43477 acc=0.63281 acc_top1_avg=0.56162 acc_top5_avg=0.89262 lr=0.00100 gn=10.50621 time=47.26it/s +epoch=51 global_step=20300 loss=4.80329 loss_avg=4.44371 acc=0.52344 acc_top1_avg=0.56048 acc_top5_avg=0.89217 lr=0.00100 gn=10.99547 time=54.66it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.92509 test_loss_avg=0.64959 acc=0.75781 test_acc_avg=0.82478 test_acc_top5_avg=0.97098 time=237.99it/s +epoch=51 global_step=20332 loss=0.19959 test_loss_avg=1.27722 acc=0.95312 test_acc_avg=0.67210 test_acc_top5_avg=0.95158 time=235.95it/s +epoch=51 global_step=20332 loss=0.03701 test_loss_avg=1.16611 acc=1.00000 test_acc_avg=0.69966 test_acc_top5_avg=0.95629 time=839.87it/s +curr_acc 0.6997 +BEST_ACC 0.7056 +curr_acc_top5 0.9563 +BEST_ACC_top5 0.9542 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=4.52428 loss_avg=4.49788 acc=0.56250 acc_top1_avg=0.54948 acc_top5_avg=0.88108 lr=0.00100 gn=11.02092 time=52.53it/s +epoch=52 global_step=20400 loss=4.46254 loss_avg=4.41982 acc=0.55469 acc_top1_avg=0.56307 acc_top5_avg=0.88902 lr=0.00100 gn=12.77201 time=60.48it/s +epoch=52 global_step=20450 loss=3.99256 loss_avg=4.43229 acc=0.60938 acc_top1_avg=0.56224 acc_top5_avg=0.89043 lr=0.00100 gn=13.51951 time=54.81it/s +epoch=52 global_step=20500 loss=4.98899 loss_avg=4.44136 acc=0.51562 acc_top1_avg=0.56078 acc_top5_avg=0.89169 lr=0.00100 gn=13.03141 time=59.17it/s +epoch=52 global_step=20550 loss=3.90129 loss_avg=4.44317 acc=0.60938 acc_top1_avg=0.56046 acc_top5_avg=0.89131 lr=0.00100 gn=13.31522 time=55.42it/s +epoch=52 global_step=20600 loss=4.11294 loss_avg=4.44077 acc=0.60156 acc_top1_avg=0.56052 acc_top5_avg=0.89115 lr=0.00100 gn=11.01049 time=55.38it/s +epoch=52 global_step=20650 loss=4.57972 loss_avg=4.43493 acc=0.54688 acc_top1_avg=0.56154 acc_top5_avg=0.89173 lr=0.00100 gn=11.43278 time=55.63it/s +epoch=52 global_step=20700 loss=4.33311 loss_avg=4.43538 acc=0.57812 acc_top1_avg=0.56167 acc_top5_avg=0.89245 lr=0.00100 gn=10.09062 time=55.51it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.68959 test_loss_avg=1.24593 acc=0.78125 test_acc_avg=0.68564 test_acc_top5_avg=0.93676 time=239.55it/s +epoch=52 global_step=20723 loss=0.00420 test_loss_avg=1.12225 acc=1.00000 test_acc_avg=0.69798 test_acc_top5_avg=0.95975 time=824.03it/s +curr_acc 0.6980 +BEST_ACC 0.7056 +curr_acc_top5 0.9598 +BEST_ACC_top5 0.9563 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=3.91181 loss_avg=4.37487 acc=0.61719 acc_top1_avg=0.56771 acc_top5_avg=0.90017 lr=0.00100 gn=11.41567 time=53.47it/s +epoch=53 global_step=20800 loss=4.28758 loss_avg=4.40835 acc=0.57031 acc_top1_avg=0.56351 acc_top5_avg=0.89813 lr=0.00100 gn=11.64821 time=63.17it/s +epoch=53 global_step=20850 loss=4.50161 loss_avg=4.42399 acc=0.54688 acc_top1_avg=0.56195 acc_top5_avg=0.89530 lr=0.00100 gn=14.46766 time=58.57it/s +epoch=53 global_step=20900 loss=4.51212 loss_avg=4.42422 acc=0.56250 acc_top1_avg=0.56228 acc_top5_avg=0.89499 lr=0.00100 gn=10.77067 time=53.02it/s +epoch=53 global_step=20950 loss=4.18198 loss_avg=4.40329 acc=0.60156 acc_top1_avg=0.56453 acc_top5_avg=0.89544 lr=0.00100 gn=12.87217 time=61.24it/s +epoch=53 global_step=21000 loss=5.02730 loss_avg=4.41419 acc=0.47656 acc_top1_avg=0.56391 acc_top5_avg=0.89421 lr=0.00100 gn=11.90161 time=47.33it/s +epoch=53 global_step=21050 loss=3.99507 loss_avg=4.42568 acc=0.61719 acc_top1_avg=0.56298 acc_top5_avg=0.89383 lr=0.00100 gn=10.87347 time=59.95it/s +epoch=53 global_step=21100 loss=4.52273 loss_avg=4.42052 acc=0.53906 acc_top1_avg=0.56358 acc_top5_avg=0.89377 lr=0.00100 gn=12.20034 time=60.91it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.20848 test_loss_avg=0.60690 acc=0.96094 test_acc_avg=0.83353 test_acc_top5_avg=0.98858 time=81.77it/s +epoch=53 global_step=21114 loss=0.09404 test_loss_avg=1.38811 acc=0.97656 test_acc_avg=0.62996 test_acc_top5_avg=0.94432 time=202.63it/s +epoch=53 global_step=21114 loss=0.01296 test_loss_avg=1.14532 acc=1.00000 test_acc_avg=0.69324 test_acc_top5_avg=0.95441 time=529.58it/s +curr_acc 0.6932 +BEST_ACC 0.7056 +curr_acc_top5 0.9544 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=4.58143 loss_avg=4.37956 acc=0.54688 acc_top1_avg=0.56944 acc_top5_avg=0.89410 lr=0.00100 gn=12.97724 time=54.47it/s +epoch=54 global_step=21200 loss=4.07084 loss_avg=4.36438 acc=0.60938 acc_top1_avg=0.57077 acc_top5_avg=0.89553 lr=0.00100 gn=10.43799 time=57.18it/s +epoch=54 global_step=21250 loss=4.22749 loss_avg=4.33691 acc=0.57812 acc_top1_avg=0.57462 acc_top5_avg=0.89476 lr=0.00100 gn=10.18190 time=59.03it/s +epoch=54 global_step=21300 loss=4.35273 loss_avg=4.33305 acc=0.56250 acc_top1_avg=0.57426 acc_top5_avg=0.89554 lr=0.00100 gn=10.67014 time=55.89it/s +epoch=54 global_step=21350 loss=4.85139 loss_avg=4.35303 acc=0.50781 acc_top1_avg=0.57174 acc_top5_avg=0.89509 lr=0.00100 gn=13.23440 time=54.14it/s +epoch=54 global_step=21400 loss=4.93556 loss_avg=4.37185 acc=0.50000 acc_top1_avg=0.56971 acc_top5_avg=0.89521 lr=0.00100 gn=13.93204 time=53.64it/s +epoch=54 global_step=21450 loss=3.89808 loss_avg=4.36843 acc=0.60938 acc_top1_avg=0.57008 acc_top5_avg=0.89481 lr=0.00100 gn=11.49467 time=55.18it/s +epoch=54 global_step=21500 loss=4.50660 loss_avg=4.38529 acc=0.53906 acc_top1_avg=0.56815 acc_top5_avg=0.89315 lr=0.00100 gn=11.96666 time=61.15it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.13846 test_loss_avg=1.50796 acc=0.94531 test_acc_avg=0.63235 test_acc_top5_avg=0.92371 time=233.03it/s +epoch=54 global_step=21505 loss=0.14702 test_loss_avg=1.15407 acc=0.93750 test_acc_avg=0.69947 test_acc_top5_avg=0.95985 time=492.17it/s +curr_acc 0.6995 +BEST_ACC 0.7056 +curr_acc_top5 0.9598 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=4.27632 loss_avg=4.23631 acc=0.57812 acc_top1_avg=0.58299 acc_top5_avg=0.89705 lr=0.00100 gn=12.94767 time=56.06it/s +epoch=55 global_step=21600 loss=4.05102 loss_avg=4.31022 acc=0.59375 acc_top1_avg=0.57508 acc_top5_avg=0.89276 lr=0.00100 gn=12.81847 time=55.98it/s +epoch=55 global_step=21650 loss=4.44715 loss_avg=4.32853 acc=0.55469 acc_top1_avg=0.57365 acc_top5_avg=0.89472 lr=0.00100 gn=10.92599 time=61.76it/s +epoch=55 global_step=21700 loss=4.51141 loss_avg=4.34039 acc=0.54688 acc_top1_avg=0.57244 acc_top5_avg=0.89623 lr=0.00100 gn=14.04422 time=61.02it/s +epoch=55 global_step=21750 loss=4.24455 loss_avg=4.34428 acc=0.60156 acc_top1_avg=0.57235 acc_top5_avg=0.89570 lr=0.00100 gn=13.42129 time=61.39it/s +epoch=55 global_step=21800 loss=4.73193 loss_avg=4.35857 acc=0.52344 acc_top1_avg=0.57090 acc_top5_avg=0.89550 lr=0.00100 gn=13.49048 time=43.56it/s +epoch=55 global_step=21850 loss=4.10998 loss_avg=4.35817 acc=0.58594 acc_top1_avg=0.57083 acc_top5_avg=0.89545 lr=0.00100 gn=9.30256 time=59.23it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.54179 test_loss_avg=0.70889 acc=0.83594 test_acc_avg=0.80000 test_acc_top5_avg=0.98594 time=234.48it/s +epoch=55 global_step=21896 loss=2.78525 test_loss_avg=1.59945 acc=0.22656 test_acc_avg=0.59077 test_acc_top5_avg=0.94034 time=236.75it/s +epoch=55 global_step=21896 loss=0.05938 test_loss_avg=1.18163 acc=0.93750 test_acc_avg=0.69442 test_acc_top5_avg=0.95728 time=839.03it/s +curr_acc 0.6944 +BEST_ACC 0.7056 +curr_acc_top5 0.9573 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=4.23779 loss_avg=3.91421 acc=0.59375 acc_top1_avg=0.62305 acc_top5_avg=0.91797 lr=0.00100 gn=13.77865 time=52.32it/s +epoch=56 global_step=21950 loss=4.84260 loss_avg=4.28523 acc=0.52344 acc_top1_avg=0.57972 acc_top5_avg=0.89453 lr=0.00100 gn=15.69743 time=58.55it/s +epoch=56 global_step=22000 loss=4.11276 loss_avg=4.32017 acc=0.60938 acc_top1_avg=0.57565 acc_top5_avg=0.89250 lr=0.00100 gn=13.12570 time=56.73it/s +epoch=56 global_step=22050 loss=4.46101 loss_avg=4.32588 acc=0.55469 acc_top1_avg=0.57539 acc_top5_avg=0.89184 lr=0.00100 gn=15.09343 time=62.45it/s +epoch=56 global_step=22100 loss=3.98938 loss_avg=4.33832 acc=0.61719 acc_top1_avg=0.57414 acc_top5_avg=0.89216 lr=0.00100 gn=15.26713 time=56.46it/s +epoch=56 global_step=22150 loss=4.69848 loss_avg=4.35216 acc=0.51562 acc_top1_avg=0.57240 acc_top5_avg=0.89293 lr=0.00100 gn=15.21098 time=57.45it/s +epoch=56 global_step=22200 loss=4.07733 loss_avg=4.34671 acc=0.61719 acc_top1_avg=0.57340 acc_top5_avg=0.89358 lr=0.00100 gn=16.23724 time=49.44it/s +epoch=56 global_step=22250 loss=4.33986 loss_avg=4.34853 acc=0.57031 acc_top1_avg=0.57298 acc_top5_avg=0.89369 lr=0.00100 gn=10.27759 time=55.18it/s +====================Eval==================== +epoch=56 global_step=22287 loss=4.37644 test_loss_avg=1.22678 acc=0.00000 test_acc_avg=0.68600 test_acc_top5_avg=0.94020 time=237.76it/s +epoch=56 global_step=22287 loss=0.08325 test_loss_avg=1.29667 acc=0.96875 test_acc_avg=0.66992 test_acc_top5_avg=0.95117 time=247.01it/s +epoch=56 global_step=22287 loss=0.03645 test_loss_avg=1.24945 acc=1.00000 test_acc_avg=0.68186 test_acc_top5_avg=0.95303 time=563.07it/s +curr_acc 0.6819 +BEST_ACC 0.7056 +curr_acc_top5 0.9530 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=4.12538 loss_avg=4.35925 acc=0.57812 acc_top1_avg=0.56851 acc_top5_avg=0.89543 lr=0.00100 gn=15.67274 time=58.25it/s +epoch=57 global_step=22350 loss=4.57199 loss_avg=4.36327 acc=0.55469 acc_top1_avg=0.56957 acc_top5_avg=0.89559 lr=0.00100 gn=15.00134 time=57.24it/s +epoch=57 global_step=22400 loss=3.96216 loss_avg=4.34615 acc=0.63281 acc_top1_avg=0.57183 acc_top5_avg=0.89574 lr=0.00100 gn=14.66343 time=53.36it/s +epoch=57 global_step=22450 loss=4.64622 loss_avg=4.33453 acc=0.52344 acc_top1_avg=0.57324 acc_top5_avg=0.89542 lr=0.00100 gn=17.40380 time=21.10it/s +epoch=57 global_step=22500 loss=4.25733 loss_avg=4.34262 acc=0.56250 acc_top1_avg=0.57248 acc_top5_avg=0.89385 lr=0.00100 gn=14.26549 time=53.31it/s +epoch=57 global_step=22550 loss=3.10257 loss_avg=4.34689 acc=0.71875 acc_top1_avg=0.57204 acc_top5_avg=0.89327 lr=0.00100 gn=15.03569 time=52.08it/s +epoch=57 global_step=22600 loss=4.45942 loss_avg=4.34276 acc=0.57031 acc_top1_avg=0.57251 acc_top5_avg=0.89367 lr=0.00100 gn=11.14352 time=54.24it/s +epoch=57 global_step=22650 loss=4.75873 loss_avg=4.34037 acc=0.51562 acc_top1_avg=0.57251 acc_top5_avg=0.89394 lr=0.00100 gn=16.30255 time=54.25it/s +====================Eval==================== +epoch=57 global_step=22678 loss=1.03154 test_loss_avg=1.21349 acc=0.70312 test_acc_avg=0.69232 test_acc_top5_avg=0.93368 time=154.13it/s +epoch=57 global_step=22678 loss=0.24245 test_loss_avg=1.16437 acc=0.93750 test_acc_avg=0.69956 test_acc_top5_avg=0.95570 time=681.45it/s +curr_acc 0.6996 +BEST_ACC 0.7056 +curr_acc_top5 0.9557 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=4.51526 loss_avg=4.35879 acc=0.55469 acc_top1_avg=0.57102 acc_top5_avg=0.89489 lr=0.00100 gn=15.05926 time=63.10it/s +epoch=58 global_step=22750 loss=4.70974 loss_avg=4.27509 acc=0.53125 acc_top1_avg=0.58008 acc_top5_avg=0.89757 lr=0.00100 gn=13.98221 time=58.65it/s +epoch=58 global_step=22800 loss=4.51476 loss_avg=4.28116 acc=0.55469 acc_top1_avg=0.57819 acc_top5_avg=0.89793 lr=0.00100 gn=16.46252 time=58.87it/s +epoch=58 global_step=22850 loss=3.87594 loss_avg=4.30438 acc=0.62500 acc_top1_avg=0.57699 acc_top5_avg=0.89585 lr=0.00100 gn=12.98399 time=53.55it/s +epoch=58 global_step=22900 loss=3.88685 loss_avg=4.30541 acc=0.61719 acc_top1_avg=0.57658 acc_top5_avg=0.89538 lr=0.00100 gn=16.01046 time=53.88it/s +epoch=58 global_step=22950 loss=4.37101 loss_avg=4.30078 acc=0.55469 acc_top1_avg=0.57686 acc_top5_avg=0.89628 lr=0.00100 gn=17.09355 time=56.45it/s +epoch=58 global_step=23000 loss=4.27199 loss_avg=4.29595 acc=0.58594 acc_top1_avg=0.57762 acc_top5_avg=0.89623 lr=0.00100 gn=16.91812 time=51.76it/s +epoch=58 global_step=23050 loss=3.97081 loss_avg=4.29894 acc=0.61719 acc_top1_avg=0.57756 acc_top5_avg=0.89615 lr=0.00100 gn=15.15811 time=51.99it/s +====================Eval==================== +epoch=58 global_step=23069 loss=1.21309 test_loss_avg=0.50634 acc=0.68750 test_acc_avg=0.85547 test_acc_top5_avg=0.98220 time=81.03it/s +epoch=58 global_step=23069 loss=0.24369 test_loss_avg=1.33845 acc=0.91406 test_acc_avg=0.65924 test_acc_top5_avg=0.94405 time=235.79it/s +epoch=58 global_step=23069 loss=0.03214 test_loss_avg=1.18083 acc=1.00000 test_acc_avg=0.69769 test_acc_top5_avg=0.95164 time=483.38it/s +curr_acc 0.6977 +BEST_ACC 0.7056 +curr_acc_top5 0.9516 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=4.34457 loss_avg=4.30338 acc=0.56250 acc_top1_avg=0.57409 acc_top5_avg=0.88155 lr=0.00100 gn=14.30891 time=59.96it/s +epoch=59 global_step=23150 loss=4.66572 loss_avg=4.33334 acc=0.53906 acc_top1_avg=0.57215 acc_top5_avg=0.89024 lr=0.00100 gn=16.01522 time=62.97it/s +epoch=59 global_step=23200 loss=4.48306 loss_avg=4.33287 acc=0.56250 acc_top1_avg=0.57216 acc_top5_avg=0.89188 lr=0.00100 gn=16.44212 time=53.90it/s +epoch=59 global_step=23250 loss=4.02372 loss_avg=4.33515 acc=0.61719 acc_top1_avg=0.57221 acc_top5_avg=0.89386 lr=0.00100 gn=18.16497 time=55.96it/s +epoch=59 global_step=23300 loss=5.28869 loss_avg=4.33961 acc=0.49219 acc_top1_avg=0.57238 acc_top5_avg=0.89221 lr=0.00100 gn=19.10029 time=30.48it/s +epoch=59 global_step=23350 loss=4.00032 loss_avg=4.32331 acc=0.60156 acc_top1_avg=0.57404 acc_top5_avg=0.89463 lr=0.00100 gn=13.52785 time=56.57it/s +epoch=59 global_step=23400 loss=4.12114 loss_avg=4.30511 acc=0.58594 acc_top1_avg=0.57626 acc_top5_avg=0.89527 lr=0.00100 gn=15.78374 time=62.57it/s +epoch=59 global_step=23450 loss=4.62881 loss_avg=4.29508 acc=0.54688 acc_top1_avg=0.57757 acc_top5_avg=0.89598 lr=0.00100 gn=14.64282 time=58.12it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.19385 test_loss_avg=1.36190 acc=0.94531 test_acc_avg=0.66847 test_acc_top5_avg=0.93570 time=238.90it/s +epoch=59 global_step=23460 loss=0.01935 test_loss_avg=1.18071 acc=1.00000 test_acc_avg=0.69215 test_acc_top5_avg=0.96094 time=818.08it/s +curr_acc 0.6921 +BEST_ACC 0.7056 +curr_acc_top5 0.9609 +BEST_ACC_top5 0.9598 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=4.56234 loss_avg=4.21977 acc=0.53125 acc_top1_avg=0.58926 acc_top5_avg=0.89648 lr=0.00100 gn=12.38728 time=61.99it/s +epoch=60 global_step=23550 loss=5.09434 loss_avg=4.23785 acc=0.50000 acc_top1_avg=0.58533 acc_top5_avg=0.89809 lr=0.00100 gn=18.05529 time=63.41it/s +epoch=60 global_step=23600 loss=3.91418 loss_avg=4.24172 acc=0.61719 acc_top1_avg=0.58477 acc_top5_avg=0.89794 lr=0.00100 gn=16.96873 time=59.08it/s +epoch=60 global_step=23650 loss=4.15645 loss_avg=4.25179 acc=0.59375 acc_top1_avg=0.58359 acc_top5_avg=0.89696 lr=0.00100 gn=17.50149 time=56.62it/s +epoch=60 global_step=23700 loss=4.36001 loss_avg=4.25890 acc=0.57031 acc_top1_avg=0.58265 acc_top5_avg=0.89590 lr=0.00100 gn=18.09184 time=52.36it/s +epoch=60 global_step=23750 loss=3.98503 loss_avg=4.26143 acc=0.60156 acc_top1_avg=0.58217 acc_top5_avg=0.89529 lr=0.00100 gn=14.30224 time=58.46it/s +epoch=60 global_step=23800 loss=4.37555 loss_avg=4.27730 acc=0.57031 acc_top1_avg=0.58012 acc_top5_avg=0.89354 lr=0.00100 gn=17.85310 time=60.71it/s +epoch=60 global_step=23850 loss=4.25272 loss_avg=4.28482 acc=0.57812 acc_top1_avg=0.57945 acc_top5_avg=0.89387 lr=0.00100 gn=18.18490 time=56.37it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.22988 test_loss_avg=0.71435 acc=0.94531 test_acc_avg=0.77422 test_acc_top5_avg=0.98438 time=236.53it/s +epoch=60 global_step=23851 loss=0.22719 test_loss_avg=1.54702 acc=0.93750 test_acc_avg=0.60065 test_acc_top5_avg=0.93672 time=237.97it/s +epoch=60 global_step=23851 loss=0.01674 test_loss_avg=1.21213 acc=1.00000 test_acc_avg=0.68503 test_acc_top5_avg=0.95125 time=507.05it/s +curr_acc 0.6850 +BEST_ACC 0.7056 +curr_acc_top5 0.9512 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=4.16670 loss_avg=4.18657 acc=0.57031 acc_top1_avg=0.59056 acc_top5_avg=0.89094 lr=0.00100 gn=12.19799 time=59.05it/s +epoch=61 global_step=23950 loss=4.64175 loss_avg=4.21845 acc=0.53906 acc_top1_avg=0.58688 acc_top5_avg=0.89268 lr=0.00100 gn=14.86665 time=53.95it/s +epoch=61 global_step=24000 loss=4.74535 loss_avg=4.19841 acc=0.52344 acc_top1_avg=0.58919 acc_top5_avg=0.89451 lr=0.00100 gn=16.73484 time=58.62it/s +epoch=61 global_step=24050 loss=3.39060 loss_avg=4.22279 acc=0.66406 acc_top1_avg=0.58660 acc_top5_avg=0.89314 lr=0.00100 gn=19.37742 time=61.79it/s +epoch=61 global_step=24100 loss=4.71976 loss_avg=4.22128 acc=0.53125 acc_top1_avg=0.58672 acc_top5_avg=0.89342 lr=0.00100 gn=17.96769 time=54.47it/s +epoch=61 global_step=24150 loss=3.80009 loss_avg=4.23590 acc=0.63281 acc_top1_avg=0.58492 acc_top5_avg=0.89347 lr=0.00100 gn=18.98350 time=62.22it/s +epoch=61 global_step=24200 loss=4.53655 loss_avg=4.25461 acc=0.54688 acc_top1_avg=0.58314 acc_top5_avg=0.89284 lr=0.00100 gn=17.31696 time=54.83it/s +====================Eval==================== +epoch=61 global_step=24242 loss=4.74592 test_loss_avg=1.59698 acc=0.00000 test_acc_avg=0.61290 test_acc_top5_avg=0.91482 time=91.89it/s +epoch=61 global_step=24242 loss=0.05369 test_loss_avg=1.24410 acc=1.00000 test_acc_avg=0.68968 test_acc_top5_avg=0.95580 time=805.36it/s +curr_acc 0.6897 +BEST_ACC 0.7056 +curr_acc_top5 0.9558 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=4.06053 loss_avg=4.24127 acc=0.59375 acc_top1_avg=0.58203 acc_top5_avg=0.88672 lr=0.00100 gn=12.87553 time=54.39it/s +epoch=62 global_step=24300 loss=4.35557 loss_avg=4.22926 acc=0.54688 acc_top1_avg=0.58284 acc_top5_avg=0.89871 lr=0.00100 gn=12.05089 time=46.67it/s +epoch=62 global_step=24350 loss=5.12638 loss_avg=4.22624 acc=0.49219 acc_top1_avg=0.58550 acc_top5_avg=0.89815 lr=0.00100 gn=15.32889 time=63.03it/s +epoch=62 global_step=24400 loss=3.40179 loss_avg=4.21846 acc=0.67969 acc_top1_avg=0.58614 acc_top5_avg=0.89686 lr=0.00100 gn=17.95160 time=50.65it/s +epoch=62 global_step=24450 loss=4.67732 loss_avg=4.20447 acc=0.53906 acc_top1_avg=0.58823 acc_top5_avg=0.89701 lr=0.00100 gn=17.82906 time=60.82it/s +epoch=62 global_step=24500 loss=4.50429 loss_avg=4.21449 acc=0.56250 acc_top1_avg=0.58721 acc_top5_avg=0.89456 lr=0.00100 gn=15.65843 time=52.28it/s +epoch=62 global_step=24550 loss=4.27275 loss_avg=4.22727 acc=0.57812 acc_top1_avg=0.58596 acc_top5_avg=0.89428 lr=0.00100 gn=13.09515 time=50.13it/s +epoch=62 global_step=24600 loss=4.10403 loss_avg=4.23497 acc=0.60938 acc_top1_avg=0.58533 acc_top5_avg=0.89453 lr=0.00100 gn=15.16702 time=55.54it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.92948 test_loss_avg=0.88049 acc=0.76562 test_acc_avg=0.76172 test_acc_top5_avg=0.97266 time=243.05it/s +epoch=62 global_step=24633 loss=3.85391 test_loss_avg=1.56509 acc=0.00000 test_acc_avg=0.60457 test_acc_top5_avg=0.93480 time=228.70it/s +epoch=62 global_step=24633 loss=0.11506 test_loss_avg=1.21627 acc=0.93750 test_acc_avg=0.68938 test_acc_top5_avg=0.95402 time=815.70it/s +curr_acc 0.6894 +BEST_ACC 0.7056 +curr_acc_top5 0.9540 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=4.03521 loss_avg=4.18640 acc=0.59375 acc_top1_avg=0.58686 acc_top5_avg=0.89430 lr=0.00100 gn=20.28569 time=57.14it/s +epoch=63 global_step=24700 loss=4.51868 loss_avg=4.18723 acc=0.52344 acc_top1_avg=0.58944 acc_top5_avg=0.89086 lr=0.00100 gn=19.08712 time=57.02it/s +epoch=63 global_step=24750 loss=3.83125 loss_avg=4.19142 acc=0.63281 acc_top1_avg=0.58881 acc_top5_avg=0.89570 lr=0.00100 gn=22.02962 time=57.22it/s +epoch=63 global_step=24800 loss=4.38480 loss_avg=4.18215 acc=0.57031 acc_top1_avg=0.59029 acc_top5_avg=0.89493 lr=0.00100 gn=18.06882 time=48.69it/s +epoch=63 global_step=24850 loss=3.64408 loss_avg=4.16983 acc=0.64062 acc_top1_avg=0.59188 acc_top5_avg=0.89541 lr=0.00100 gn=17.82087 time=34.78it/s +epoch=63 global_step=24900 loss=4.54586 loss_avg=4.19522 acc=0.54688 acc_top1_avg=0.58904 acc_top5_avg=0.89522 lr=0.00100 gn=20.53741 time=62.40it/s +epoch=63 global_step=24950 loss=4.51646 loss_avg=4.19226 acc=0.55469 acc_top1_avg=0.58929 acc_top5_avg=0.89518 lr=0.00100 gn=16.61023 time=52.53it/s +epoch=63 global_step=25000 loss=4.54310 loss_avg=4.21609 acc=0.52344 acc_top1_avg=0.58696 acc_top5_avg=0.89480 lr=0.00100 gn=18.62726 time=51.23it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.95764 test_loss_avg=0.72347 acc=0.73438 test_acc_avg=0.79959 test_acc_top5_avg=0.96943 time=235.03it/s +epoch=63 global_step=25024 loss=0.29030 test_loss_avg=1.25058 acc=0.92188 test_acc_avg=0.67177 test_acc_top5_avg=0.95494 time=233.17it/s +epoch=63 global_step=25024 loss=0.18955 test_loss_avg=1.16592 acc=0.93750 test_acc_avg=0.69373 test_acc_top5_avg=0.95817 time=718.57it/s +curr_acc 0.6937 +BEST_ACC 0.7056 +curr_acc_top5 0.9582 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=4.56019 loss_avg=4.22510 acc=0.56250 acc_top1_avg=0.58744 acc_top5_avg=0.90204 lr=0.00100 gn=19.74514 time=52.39it/s +epoch=64 global_step=25100 loss=4.30803 loss_avg=4.20331 acc=0.56250 acc_top1_avg=0.58964 acc_top5_avg=0.89905 lr=0.00100 gn=20.93934 time=63.25it/s +epoch=64 global_step=25150 loss=4.27728 loss_avg=4.15205 acc=0.59375 acc_top1_avg=0.59487 acc_top5_avg=0.89875 lr=0.00100 gn=16.29754 time=56.46it/s +epoch=64 global_step=25200 loss=3.71551 loss_avg=4.14927 acc=0.63281 acc_top1_avg=0.59486 acc_top5_avg=0.89808 lr=0.00100 gn=19.21476 time=42.32it/s +epoch=64 global_step=25250 loss=4.50248 loss_avg=4.17649 acc=0.57812 acc_top1_avg=0.59219 acc_top5_avg=0.89705 lr=0.00100 gn=23.73131 time=60.84it/s +epoch=64 global_step=25300 loss=4.08032 loss_avg=4.16831 acc=0.60938 acc_top1_avg=0.59321 acc_top5_avg=0.89764 lr=0.00100 gn=19.92504 time=55.10it/s +epoch=64 global_step=25350 loss=4.25543 loss_avg=4.18842 acc=0.57812 acc_top1_avg=0.59102 acc_top5_avg=0.89748 lr=0.00100 gn=20.98463 time=56.75it/s +epoch=64 global_step=25400 loss=4.31169 loss_avg=4.19924 acc=0.57812 acc_top1_avg=0.59013 acc_top5_avg=0.89665 lr=0.00100 gn=20.84158 time=54.58it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.86370 test_loss_avg=1.29868 acc=0.75781 test_acc_avg=0.67649 test_acc_top5_avg=0.91779 time=231.65it/s +epoch=64 global_step=25415 loss=0.24318 test_loss_avg=1.23090 acc=0.93750 test_acc_avg=0.68938 test_acc_top5_avg=0.94749 time=524.94it/s +curr_acc 0.6894 +BEST_ACC 0.7056 +curr_acc_top5 0.9475 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=4.05572 loss_avg=4.10825 acc=0.60938 acc_top1_avg=0.59911 acc_top5_avg=0.89085 lr=0.00100 gn=19.89438 time=52.47it/s +epoch=65 global_step=25500 loss=5.08355 loss_avg=4.15402 acc=0.48438 acc_top1_avg=0.59586 acc_top5_avg=0.89329 lr=0.00100 gn=20.83002 time=46.37it/s +epoch=65 global_step=25550 loss=4.62151 loss_avg=4.17939 acc=0.53906 acc_top1_avg=0.59259 acc_top5_avg=0.89462 lr=0.00100 gn=21.77863 time=63.53it/s +epoch=65 global_step=25600 loss=3.32756 loss_avg=4.17267 acc=0.68750 acc_top1_avg=0.59316 acc_top5_avg=0.89535 lr=0.00100 gn=23.75413 time=62.99it/s +epoch=65 global_step=25650 loss=3.94489 loss_avg=4.16588 acc=0.60938 acc_top1_avg=0.59348 acc_top5_avg=0.89531 lr=0.00100 gn=19.29009 time=57.54it/s +epoch=65 global_step=25700 loss=4.62714 loss_avg=4.18299 acc=0.55469 acc_top1_avg=0.59200 acc_top5_avg=0.89463 lr=0.00100 gn=23.51367 time=46.37it/s +epoch=65 global_step=25750 loss=4.53436 loss_avg=4.18618 acc=0.54688 acc_top1_avg=0.59163 acc_top5_avg=0.89415 lr=0.00100 gn=16.87240 time=55.68it/s +epoch=65 global_step=25800 loss=4.01869 loss_avg=4.17770 acc=0.60156 acc_top1_avg=0.59282 acc_top5_avg=0.89428 lr=0.00100 gn=18.42000 time=63.74it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.19559 test_loss_avg=0.50102 acc=0.94531 test_acc_avg=0.84688 test_acc_top5_avg=0.99323 time=237.62it/s +epoch=65 global_step=25806 loss=0.11219 test_loss_avg=1.39229 acc=0.96094 test_acc_avg=0.64255 test_acc_top5_avg=0.94784 time=233.37it/s +epoch=65 global_step=25806 loss=0.30951 test_loss_avg=1.17986 acc=0.93750 test_acc_avg=0.69600 test_acc_top5_avg=0.95688 time=502.13it/s +curr_acc 0.6960 +BEST_ACC 0.7056 +curr_acc_top5 0.9569 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=3.70686 loss_avg=4.16806 acc=0.66406 acc_top1_avg=0.59375 acc_top5_avg=0.88920 lr=0.00100 gn=19.65667 time=57.73it/s +epoch=66 global_step=25900 loss=4.43804 loss_avg=4.14052 acc=0.56250 acc_top1_avg=0.59583 acc_top5_avg=0.89403 lr=0.00100 gn=23.02171 time=51.85it/s +epoch=66 global_step=25950 loss=4.29228 loss_avg=4.15633 acc=0.57031 acc_top1_avg=0.59451 acc_top5_avg=0.89480 lr=0.00100 gn=20.13605 time=54.83it/s +epoch=66 global_step=26000 loss=4.35439 loss_avg=4.16200 acc=0.57812 acc_top1_avg=0.59496 acc_top5_avg=0.89493 lr=0.00100 gn=19.83112 time=52.79it/s +epoch=66 global_step=26050 loss=4.54051 loss_avg=4.17450 acc=0.53906 acc_top1_avg=0.59333 acc_top5_avg=0.89469 lr=0.00100 gn=23.25319 time=57.63it/s +epoch=66 global_step=26100 loss=3.98624 loss_avg=4.15654 acc=0.62500 acc_top1_avg=0.59545 acc_top5_avg=0.89559 lr=0.00100 gn=22.59543 time=50.28it/s +epoch=66 global_step=26150 loss=4.04750 loss_avg=4.16349 acc=0.60156 acc_top1_avg=0.59507 acc_top5_avg=0.89505 lr=0.00100 gn=14.74903 time=52.94it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.39295 test_loss_avg=1.55075 acc=0.88281 test_acc_avg=0.61827 test_acc_top5_avg=0.91102 time=96.75it/s +epoch=66 global_step=26197 loss=0.06964 test_loss_avg=1.26015 acc=0.93750 test_acc_avg=0.68503 test_acc_top5_avg=0.95026 time=474.15it/s +curr_acc 0.6850 +BEST_ACC 0.7056 +curr_acc_top5 0.9503 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=4.04500 loss_avg=3.95912 acc=0.60156 acc_top1_avg=0.61458 acc_top5_avg=0.89323 lr=0.00100 gn=19.07177 time=50.91it/s +epoch=67 global_step=26250 loss=3.63182 loss_avg=4.05319 acc=0.64844 acc_top1_avg=0.60259 acc_top5_avg=0.89637 lr=0.00100 gn=14.40851 time=56.84it/s +epoch=67 global_step=26300 loss=4.31852 loss_avg=4.11028 acc=0.57031 acc_top1_avg=0.59830 acc_top5_avg=0.89411 lr=0.00100 gn=22.80889 time=50.51it/s +epoch=67 global_step=26350 loss=4.31543 loss_avg=4.11733 acc=0.57812 acc_top1_avg=0.59840 acc_top5_avg=0.89267 lr=0.00100 gn=25.45065 time=53.99it/s +epoch=67 global_step=26400 loss=4.02575 loss_avg=4.12321 acc=0.60938 acc_top1_avg=0.59771 acc_top5_avg=0.89282 lr=0.00100 gn=21.88080 time=55.15it/s +epoch=67 global_step=26450 loss=4.83106 loss_avg=4.13867 acc=0.53125 acc_top1_avg=0.59650 acc_top5_avg=0.89381 lr=0.00100 gn=23.33525 time=55.88it/s +epoch=67 global_step=26500 loss=3.85118 loss_avg=4.13768 acc=0.63281 acc_top1_avg=0.59669 acc_top5_avg=0.89542 lr=0.00100 gn=25.83229 time=50.29it/s +epoch=67 global_step=26550 loss=4.74950 loss_avg=4.15763 acc=0.53125 acc_top1_avg=0.59446 acc_top5_avg=0.89580 lr=0.00100 gn=17.66974 time=54.23it/s +====================Eval==================== +epoch=67 global_step=26588 loss=1.08542 test_loss_avg=1.06358 acc=0.60938 test_acc_avg=0.66741 test_acc_top5_avg=0.98438 time=183.75it/s +epoch=67 global_step=26588 loss=0.40481 test_loss_avg=1.67558 acc=0.89062 test_acc_avg=0.57771 test_acc_top5_avg=0.94038 time=234.62it/s +epoch=67 global_step=26588 loss=0.14849 test_loss_avg=1.27308 acc=0.93750 test_acc_avg=0.67603 test_acc_top5_avg=0.95560 time=819.20it/s +curr_acc 0.6760 +BEST_ACC 0.7056 +curr_acc_top5 0.9556 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=5.06423 loss_avg=4.22424 acc=0.49219 acc_top1_avg=0.58724 acc_top5_avg=0.89909 lr=0.00100 gn=23.14314 time=63.15it/s +epoch=68 global_step=26650 loss=4.15074 loss_avg=4.13658 acc=0.58594 acc_top1_avg=0.59728 acc_top5_avg=0.89882 lr=0.00100 gn=26.39833 time=55.03it/s +epoch=68 global_step=26700 loss=4.26734 loss_avg=4.15595 acc=0.58594 acc_top1_avg=0.59452 acc_top5_avg=0.89593 lr=0.00100 gn=19.60398 time=49.83it/s +epoch=68 global_step=26750 loss=4.11847 loss_avg=4.14286 acc=0.59375 acc_top1_avg=0.59597 acc_top5_avg=0.89622 lr=0.00100 gn=22.18327 time=54.04it/s +epoch=68 global_step=26800 loss=4.17034 loss_avg=4.14624 acc=0.61719 acc_top1_avg=0.59618 acc_top5_avg=0.89667 lr=0.00100 gn=27.56032 time=55.74it/s +epoch=68 global_step=26850 loss=3.83388 loss_avg=4.14471 acc=0.64062 acc_top1_avg=0.59599 acc_top5_avg=0.89593 lr=0.00100 gn=18.01947 time=59.09it/s +epoch=68 global_step=26900 loss=4.27432 loss_avg=4.14641 acc=0.57031 acc_top1_avg=0.59580 acc_top5_avg=0.89498 lr=0.00100 gn=24.08792 time=51.28it/s +epoch=68 global_step=26950 loss=4.23233 loss_avg=4.14881 acc=0.57812 acc_top1_avg=0.59558 acc_top5_avg=0.89516 lr=0.00100 gn=25.79159 time=49.89it/s +====================Eval==================== +epoch=68 global_step=26979 loss=4.38518 test_loss_avg=1.32872 acc=0.00000 test_acc_avg=0.65737 test_acc_top5_avg=0.93331 time=240.29it/s +epoch=68 global_step=26979 loss=0.04550 test_loss_avg=1.30360 acc=0.99219 test_acc_avg=0.67839 test_acc_top5_avg=0.95212 time=253.88it/s +epoch=68 global_step=26979 loss=0.04906 test_loss_avg=1.28772 acc=1.00000 test_acc_avg=0.68246 test_acc_top5_avg=0.95273 time=781.21it/s +curr_acc 0.6825 +BEST_ACC 0.7056 +curr_acc_top5 0.9527 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=3.82986 loss_avg=4.03380 acc=0.62500 acc_top1_avg=0.60900 acc_top5_avg=0.89249 lr=0.00100 gn=24.12901 time=54.73it/s +epoch=69 global_step=27050 loss=4.12419 loss_avg=4.04745 acc=0.59375 acc_top1_avg=0.60706 acc_top5_avg=0.89195 lr=0.00100 gn=24.53891 time=54.19it/s +epoch=69 global_step=27100 loss=3.50716 loss_avg=4.10259 acc=0.67188 acc_top1_avg=0.60169 acc_top5_avg=0.89282 lr=0.00100 gn=18.06708 time=42.32it/s +epoch=69 global_step=27150 loss=4.19283 loss_avg=4.09822 acc=0.57031 acc_top1_avg=0.60248 acc_top5_avg=0.89465 lr=0.00100 gn=22.25479 time=51.84it/s +epoch=69 global_step=27200 loss=4.17443 loss_avg=4.10932 acc=0.59375 acc_top1_avg=0.60135 acc_top5_avg=0.89508 lr=0.00100 gn=26.83607 time=57.98it/s +epoch=69 global_step=27250 loss=4.56771 loss_avg=4.11405 acc=0.55469 acc_top1_avg=0.60044 acc_top5_avg=0.89429 lr=0.00100 gn=27.47941 time=62.28it/s +epoch=69 global_step=27300 loss=4.00253 loss_avg=4.13611 acc=0.63281 acc_top1_avg=0.59825 acc_top5_avg=0.89340 lr=0.00100 gn=21.79471 time=52.05it/s +epoch=69 global_step=27350 loss=3.69871 loss_avg=4.12289 acc=0.64062 acc_top1_avg=0.59956 acc_top5_avg=0.89460 lr=0.00100 gn=23.06790 time=49.98it/s +====================Eval==================== +epoch=69 global_step=27370 loss=4.24678 test_loss_avg=1.43220 acc=0.00000 test_acc_avg=0.64174 test_acc_top5_avg=0.92347 time=239.29it/s +epoch=69 global_step=27370 loss=0.16015 test_loss_avg=1.25976 acc=0.93750 test_acc_avg=0.68325 test_acc_top5_avg=0.94769 time=597.22it/s +curr_acc 0.6832 +BEST_ACC 0.7056 +curr_acc_top5 0.9477 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=5.12887 loss_avg=4.15320 acc=0.50000 acc_top1_avg=0.59479 acc_top5_avg=0.89687 lr=0.00100 gn=23.50824 time=59.72it/s +epoch=70 global_step=27450 loss=4.08495 loss_avg=4.05267 acc=0.60938 acc_top1_avg=0.60674 acc_top5_avg=0.90098 lr=0.00100 gn=29.53299 time=54.39it/s +epoch=70 global_step=27500 loss=3.91896 loss_avg=4.07428 acc=0.60938 acc_top1_avg=0.60493 acc_top5_avg=0.89850 lr=0.00100 gn=20.57992 time=48.76it/s +epoch=70 global_step=27550 loss=5.02576 loss_avg=4.12141 acc=0.50781 acc_top1_avg=0.59961 acc_top5_avg=0.89622 lr=0.00100 gn=29.88885 time=32.63it/s +epoch=70 global_step=27600 loss=3.99259 loss_avg=4.12286 acc=0.59375 acc_top1_avg=0.59959 acc_top5_avg=0.89664 lr=0.00100 gn=17.59412 time=56.23it/s +epoch=70 global_step=27650 loss=3.42135 loss_avg=4.11651 acc=0.67969 acc_top1_avg=0.60020 acc_top5_avg=0.89704 lr=0.00100 gn=22.00363 time=59.26it/s +epoch=70 global_step=27700 loss=4.82186 loss_avg=4.12182 acc=0.52344 acc_top1_avg=0.59995 acc_top5_avg=0.89714 lr=0.00100 gn=27.21000 time=61.84it/s +epoch=70 global_step=27750 loss=3.76200 loss_avg=4.11286 acc=0.61719 acc_top1_avg=0.60074 acc_top5_avg=0.89741 lr=0.00100 gn=19.05157 time=61.99it/s +====================Eval==================== +epoch=70 global_step=27761 loss=0.74583 test_loss_avg=0.70162 acc=0.78906 test_acc_avg=0.79414 test_acc_top5_avg=0.98242 time=223.73it/s +epoch=70 global_step=27761 loss=0.08607 test_loss_avg=1.38420 acc=0.96875 test_acc_avg=0.64241 test_acc_top5_avg=0.93783 time=237.42it/s +epoch=70 global_step=27761 loss=0.05232 test_loss_avg=1.24534 acc=1.00000 test_acc_avg=0.67870 test_acc_top5_avg=0.94442 time=839.03it/s +curr_acc 0.6787 +BEST_ACC 0.7056 +curr_acc_top5 0.9444 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=4.65376 loss_avg=4.15528 acc=0.53906 acc_top1_avg=0.59595 acc_top5_avg=0.89924 lr=0.00100 gn=28.76722 time=62.89it/s +epoch=71 global_step=27850 loss=3.20136 loss_avg=4.11086 acc=0.72656 acc_top1_avg=0.60130 acc_top5_avg=0.89484 lr=0.00100 gn=31.65010 time=56.92it/s +epoch=71 global_step=27900 loss=3.79559 loss_avg=4.07921 acc=0.64844 acc_top1_avg=0.60465 acc_top5_avg=0.89405 lr=0.00100 gn=25.55125 time=52.56it/s +epoch=71 global_step=27950 loss=4.56696 loss_avg=4.08946 acc=0.53906 acc_top1_avg=0.60351 acc_top5_avg=0.89484 lr=0.00100 gn=28.20794 time=59.08it/s +epoch=71 global_step=28000 loss=4.83646 loss_avg=4.11452 acc=0.50781 acc_top1_avg=0.60068 acc_top5_avg=0.89523 lr=0.00100 gn=24.46298 time=50.86it/s +epoch=71 global_step=28050 loss=4.39982 loss_avg=4.10136 acc=0.57031 acc_top1_avg=0.60221 acc_top5_avg=0.89444 lr=0.00100 gn=19.77923 time=53.68it/s +epoch=71 global_step=28100 loss=4.13329 loss_avg=4.08643 acc=0.60938 acc_top1_avg=0.60407 acc_top5_avg=0.89505 lr=0.00100 gn=27.69207 time=57.39it/s +epoch=71 global_step=28150 loss=4.33407 loss_avg=4.09096 acc=0.58594 acc_top1_avg=0.60335 acc_top5_avg=0.89492 lr=0.00100 gn=25.51319 time=54.43it/s +====================Eval==================== +epoch=71 global_step=28152 loss=1.66061 test_loss_avg=1.40365 acc=0.54688 test_acc_avg=0.65663 test_acc_top5_avg=0.90682 time=242.49it/s +epoch=71 global_step=28152 loss=0.06173 test_loss_avg=1.34961 acc=1.00000 test_acc_avg=0.66782 test_acc_top5_avg=0.94076 time=465.93it/s +curr_acc 0.6678 +BEST_ACC 0.7056 +curr_acc_top5 0.9408 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=4.13457 loss_avg=4.12957 acc=0.57812 acc_top1_avg=0.59538 acc_top5_avg=0.89502 lr=0.00100 gn=18.05379 time=61.33it/s +epoch=72 global_step=28250 loss=4.22917 loss_avg=4.07188 acc=0.57812 acc_top1_avg=0.60340 acc_top5_avg=0.89310 lr=0.00100 gn=27.40464 time=50.30it/s +epoch=72 global_step=28300 loss=4.18279 loss_avg=4.07066 acc=0.58594 acc_top1_avg=0.60452 acc_top5_avg=0.89553 lr=0.00100 gn=30.52892 time=54.51it/s +epoch=72 global_step=28350 loss=3.73975 loss_avg=4.04998 acc=0.63281 acc_top1_avg=0.60724 acc_top5_avg=0.89512 lr=0.00100 gn=31.01076 time=57.22it/s +epoch=72 global_step=28400 loss=4.67461 loss_avg=4.08826 acc=0.51562 acc_top1_avg=0.60367 acc_top5_avg=0.89475 lr=0.00100 gn=24.56120 time=56.56it/s +epoch=72 global_step=28450 loss=4.32299 loss_avg=4.07251 acc=0.58594 acc_top1_avg=0.60542 acc_top5_avg=0.89490 lr=0.00100 gn=28.57907 time=57.96it/s +epoch=72 global_step=28500 loss=4.21137 loss_avg=4.07653 acc=0.58594 acc_top1_avg=0.60538 acc_top5_avg=0.89503 lr=0.00100 gn=23.08896 time=58.01it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.18635 test_loss_avg=0.79949 acc=0.94531 test_acc_avg=0.75195 test_acc_top5_avg=0.98438 time=237.33it/s +epoch=72 global_step=28543 loss=0.37403 test_loss_avg=1.54393 acc=0.87500 test_acc_avg=0.60181 test_acc_top5_avg=0.93674 time=236.71it/s +epoch=72 global_step=28543 loss=0.17862 test_loss_avg=1.24990 acc=0.87500 test_acc_avg=0.67573 test_acc_top5_avg=0.94956 time=820.80it/s +curr_acc 0.6757 +BEST_ACC 0.7056 +curr_acc_top5 0.9496 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=3.82898 loss_avg=4.19029 acc=0.61719 acc_top1_avg=0.59040 acc_top5_avg=0.87500 lr=0.00100 gn=24.04643 time=54.43it/s +epoch=73 global_step=28600 loss=3.67224 loss_avg=4.02026 acc=0.64062 acc_top1_avg=0.61280 acc_top5_avg=0.89090 lr=0.00100 gn=20.06502 time=54.41it/s +epoch=73 global_step=28650 loss=4.01451 loss_avg=4.02898 acc=0.61719 acc_top1_avg=0.61098 acc_top5_avg=0.89355 lr=0.00100 gn=29.46961 time=56.32it/s +epoch=73 global_step=28700 loss=4.19583 loss_avg=4.03499 acc=0.60156 acc_top1_avg=0.61037 acc_top5_avg=0.89500 lr=0.00100 gn=26.63730 time=58.53it/s +epoch=73 global_step=28750 loss=3.91768 loss_avg=4.03884 acc=0.61719 acc_top1_avg=0.61054 acc_top5_avg=0.89587 lr=0.00100 gn=28.02245 time=58.02it/s +epoch=73 global_step=28800 loss=3.86979 loss_avg=4.05639 acc=0.62500 acc_top1_avg=0.60825 acc_top5_avg=0.89613 lr=0.00100 gn=23.55529 time=54.29it/s +epoch=73 global_step=28850 loss=3.22515 loss_avg=4.04476 acc=0.68750 acc_top1_avg=0.60917 acc_top5_avg=0.89605 lr=0.00100 gn=24.69370 time=61.34it/s +epoch=73 global_step=28900 loss=3.89843 loss_avg=4.06141 acc=0.62500 acc_top1_avg=0.60727 acc_top5_avg=0.89599 lr=0.00100 gn=20.67878 time=54.30it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.29058 test_loss_avg=1.69489 acc=0.90625 test_acc_avg=0.56960 test_acc_top5_avg=0.90128 time=56.06it/s +epoch=73 global_step=28934 loss=0.09966 test_loss_avg=1.24945 acc=0.93750 test_acc_avg=0.67425 test_acc_top5_avg=0.95075 time=823.54it/s +curr_acc 0.6742 +BEST_ACC 0.7056 +curr_acc_top5 0.9508 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=3.94245 loss_avg=3.92009 acc=0.61719 acc_top1_avg=0.61914 acc_top5_avg=0.90039 lr=0.00100 gn=32.01601 time=57.68it/s +epoch=74 global_step=29000 loss=4.25211 loss_avg=3.96929 acc=0.58594 acc_top1_avg=0.61648 acc_top5_avg=0.89643 lr=0.00100 gn=26.72449 time=52.64it/s +epoch=74 global_step=29050 loss=3.69940 loss_avg=4.00773 acc=0.65625 acc_top1_avg=0.61247 acc_top5_avg=0.89500 lr=0.00100 gn=29.52912 time=49.04it/s +epoch=74 global_step=29100 loss=4.35047 loss_avg=4.02094 acc=0.57031 acc_top1_avg=0.61079 acc_top5_avg=0.89575 lr=0.00100 gn=29.98076 time=54.04it/s +epoch=74 global_step=29150 loss=3.95720 loss_avg=4.03202 acc=0.60156 acc_top1_avg=0.60952 acc_top5_avg=0.89620 lr=0.00100 gn=28.47133 time=57.48it/s +epoch=74 global_step=29200 loss=4.42920 loss_avg=4.04458 acc=0.56250 acc_top1_avg=0.60791 acc_top5_avg=0.89670 lr=0.00100 gn=27.04162 time=55.63it/s +epoch=74 global_step=29250 loss=3.99224 loss_avg=4.05004 acc=0.61719 acc_top1_avg=0.60722 acc_top5_avg=0.89705 lr=0.00100 gn=28.11878 time=50.64it/s +epoch=74 global_step=29300 loss=4.09121 loss_avg=4.05287 acc=0.59375 acc_top1_avg=0.60696 acc_top5_avg=0.89722 lr=0.00100 gn=23.29017 time=58.56it/s +====================Eval==================== +epoch=74 global_step=29325 loss=0.76683 test_loss_avg=0.99384 acc=0.75781 test_acc_avg=0.70312 test_acc_top5_avg=0.96875 time=235.44it/s +epoch=74 global_step=29325 loss=4.25974 test_loss_avg=1.77617 acc=0.00000 test_acc_avg=0.56901 test_acc_top5_avg=0.91247 time=229.01it/s +epoch=74 global_step=29325 loss=0.24425 test_loss_avg=1.34868 acc=0.93750 test_acc_avg=0.66831 test_acc_top5_avg=0.93740 time=811.43it/s +curr_acc 0.6683 +BEST_ACC 0.7056 +curr_acc_top5 0.9374 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=4.64247 loss_avg=4.10614 acc=0.54688 acc_top1_avg=0.60156 acc_top5_avg=0.88625 lr=0.00100 gn=26.31756 time=62.85it/s +epoch=75 global_step=29400 loss=4.10054 loss_avg=4.02607 acc=0.62500 acc_top1_avg=0.61104 acc_top5_avg=0.89333 lr=0.00100 gn=32.68413 time=54.89it/s +epoch=75 global_step=29450 loss=4.41161 loss_avg=3.99961 acc=0.55469 acc_top1_avg=0.61394 acc_top5_avg=0.89381 lr=0.00100 gn=20.70070 time=58.37it/s +epoch=75 global_step=29500 loss=4.60849 loss_avg=4.01681 acc=0.53906 acc_top1_avg=0.61259 acc_top5_avg=0.89469 lr=0.00100 gn=27.36199 time=57.26it/s +epoch=75 global_step=29550 loss=3.75030 loss_avg=4.01398 acc=0.62500 acc_top1_avg=0.61281 acc_top5_avg=0.89535 lr=0.00100 gn=26.71407 time=53.72it/s +epoch=75 global_step=29600 loss=4.11396 loss_avg=4.02472 acc=0.59375 acc_top1_avg=0.61176 acc_top5_avg=0.89619 lr=0.00100 gn=29.22000 time=63.24it/s +epoch=75 global_step=29650 loss=4.06325 loss_avg=4.03702 acc=0.60938 acc_top1_avg=0.61043 acc_top5_avg=0.89596 lr=0.00100 gn=33.77943 time=51.26it/s +epoch=75 global_step=29700 loss=3.95904 loss_avg=4.04153 acc=0.62500 acc_top1_avg=0.61015 acc_top5_avg=0.89583 lr=0.00100 gn=25.98979 time=49.38it/s +====================Eval==================== +epoch=75 global_step=29716 loss=4.42306 test_loss_avg=1.10767 acc=0.00000 test_acc_avg=0.69812 test_acc_top5_avg=0.95094 time=213.43it/s +epoch=75 global_step=29716 loss=0.13387 test_loss_avg=1.37059 acc=0.96094 test_acc_avg=0.65563 test_acc_top5_avg=0.93458 time=228.73it/s +epoch=75 global_step=29716 loss=0.07790 test_loss_avg=1.30695 acc=1.00000 test_acc_avg=0.67148 test_acc_top5_avg=0.93780 time=435.09it/s +curr_acc 0.6715 +BEST_ACC 0.7056 +curr_acc_top5 0.9378 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=4.66710 loss_avg=3.97976 acc=0.56250 acc_top1_avg=0.61742 acc_top5_avg=0.89476 lr=0.00100 gn=32.47095 time=62.36it/s +epoch=76 global_step=29800 loss=4.49141 loss_avg=3.99049 acc=0.56250 acc_top1_avg=0.61589 acc_top5_avg=0.89481 lr=0.00100 gn=30.47283 time=63.19it/s +epoch=76 global_step=29850 loss=3.96437 loss_avg=4.02056 acc=0.62500 acc_top1_avg=0.61112 acc_top5_avg=0.89500 lr=0.00100 gn=25.90768 time=49.12it/s +epoch=76 global_step=29900 loss=4.26482 loss_avg=4.04042 acc=0.57031 acc_top1_avg=0.60912 acc_top5_avg=0.89436 lr=0.00100 gn=25.79914 time=55.19it/s +epoch=76 global_step=29950 loss=3.73502 loss_avg=4.02243 acc=0.64844 acc_top1_avg=0.61104 acc_top5_avg=0.89523 lr=0.00100 gn=30.56389 time=52.81it/s +epoch=76 global_step=30000 loss=4.41837 loss_avg=4.00592 acc=0.57812 acc_top1_avg=0.61325 acc_top5_avg=0.89643 lr=0.00100 gn=34.50382 time=62.06it/s +epoch=76 global_step=30050 loss=4.57878 loss_avg=4.02051 acc=0.54688 acc_top1_avg=0.61176 acc_top5_avg=0.89624 lr=0.00100 gn=24.20992 time=52.42it/s +epoch=76 global_step=30100 loss=3.95506 loss_avg=4.02703 acc=0.62500 acc_top1_avg=0.61129 acc_top5_avg=0.89567 lr=0.00100 gn=28.54956 time=51.43it/s +====================Eval==================== +epoch=76 global_step=30107 loss=1.43216 test_loss_avg=1.44210 acc=0.67969 test_acc_avg=0.63196 test_acc_top5_avg=0.91559 time=238.39it/s +epoch=76 global_step=30107 loss=0.03936 test_loss_avg=1.33296 acc=1.00000 test_acc_avg=0.66653 test_acc_top5_avg=0.94531 time=530.19it/s +curr_acc 0.6665 +BEST_ACC 0.7056 +curr_acc_top5 0.9453 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=3.87539 loss_avg=3.96326 acc=0.63281 acc_top1_avg=0.61737 acc_top5_avg=0.88935 lr=0.00100 gn=27.05269 time=52.61it/s +epoch=77 global_step=30200 loss=4.26662 loss_avg=3.96073 acc=0.60156 acc_top1_avg=0.61895 acc_top5_avg=0.89222 lr=0.00100 gn=27.26992 time=59.69it/s +epoch=77 global_step=30250 loss=3.65152 loss_avg=3.93801 acc=0.65625 acc_top1_avg=0.62189 acc_top5_avg=0.89368 lr=0.00100 gn=31.65942 time=62.30it/s +epoch=77 global_step=30300 loss=4.01011 loss_avg=3.94711 acc=0.61719 acc_top1_avg=0.62124 acc_top5_avg=0.89427 lr=0.00100 gn=22.43061 time=56.04it/s +epoch=77 global_step=30350 loss=4.02544 loss_avg=3.97293 acc=0.62500 acc_top1_avg=0.61831 acc_top5_avg=0.89452 lr=0.00100 gn=23.62008 time=57.13it/s +epoch=77 global_step=30400 loss=3.83558 loss_avg=3.98622 acc=0.64062 acc_top1_avg=0.61644 acc_top5_avg=0.89414 lr=0.00100 gn=32.06327 time=56.36it/s +epoch=77 global_step=30450 loss=3.81419 loss_avg=3.99840 acc=0.64062 acc_top1_avg=0.61521 acc_top5_avg=0.89466 lr=0.00100 gn=34.62957 time=52.10it/s +====================Eval==================== +epoch=77 global_step=30498 loss=1.19627 test_loss_avg=0.60138 acc=0.65625 test_acc_avg=0.81250 test_acc_top5_avg=0.98208 time=232.87it/s +epoch=77 global_step=30498 loss=0.16198 test_loss_avg=1.40414 acc=0.96094 test_acc_avg=0.63876 test_acc_top5_avg=0.93214 time=233.46it/s +epoch=77 global_step=30498 loss=0.20791 test_loss_avg=1.22515 acc=0.93750 test_acc_avg=0.68354 test_acc_top5_avg=0.94185 time=811.43it/s +curr_acc 0.6835 +BEST_ACC 0.7056 +curr_acc_top5 0.9419 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=4.38515 loss_avg=4.28981 acc=0.56250 acc_top1_avg=0.57031 acc_top5_avg=0.88672 lr=0.00100 gn=31.25550 time=31.83it/s +epoch=78 global_step=30550 loss=3.73243 loss_avg=4.04690 acc=0.62500 acc_top1_avg=0.60742 acc_top5_avg=0.88912 lr=0.00100 gn=29.23843 time=51.18it/s +epoch=78 global_step=30600 loss=3.83926 loss_avg=3.99079 acc=0.63281 acc_top1_avg=0.61389 acc_top5_avg=0.89514 lr=0.00100 gn=26.63466 time=54.55it/s +epoch=78 global_step=30650 loss=4.04009 loss_avg=3.97323 acc=0.63281 acc_top1_avg=0.61672 acc_top5_avg=0.89664 lr=0.00100 gn=35.54877 time=63.06it/s +epoch=78 global_step=30700 loss=4.25209 loss_avg=3.96986 acc=0.57031 acc_top1_avg=0.61757 acc_top5_avg=0.89612 lr=0.00100 gn=29.26956 time=52.79it/s +epoch=78 global_step=30750 loss=4.32071 loss_avg=3.98550 acc=0.57031 acc_top1_avg=0.61589 acc_top5_avg=0.89500 lr=0.00100 gn=23.99468 time=59.81it/s +epoch=78 global_step=30800 loss=4.47423 loss_avg=3.99209 acc=0.55469 acc_top1_avg=0.61509 acc_top5_avg=0.89546 lr=0.00100 gn=21.82070 time=61.10it/s +epoch=78 global_step=30850 loss=4.03532 loss_avg=4.00359 acc=0.60938 acc_top1_avg=0.61424 acc_top5_avg=0.89520 lr=0.00100 gn=21.51982 time=57.04it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.32099 test_loss_avg=1.52681 acc=0.87500 test_acc_avg=0.61513 test_acc_top5_avg=0.92167 time=237.50it/s +epoch=78 global_step=30889 loss=0.09182 test_loss_avg=1.28096 acc=0.93750 test_acc_avg=0.67108 test_acc_top5_avg=0.94818 time=550.29it/s +curr_acc 0.6711 +BEST_ACC 0.7056 +curr_acc_top5 0.9482 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=3.78632 loss_avg=3.81739 acc=0.64844 acc_top1_avg=0.63565 acc_top5_avg=0.90767 lr=0.00100 gn=33.01626 time=61.76it/s +epoch=79 global_step=30950 loss=3.88443 loss_avg=3.94725 acc=0.60938 acc_top1_avg=0.62039 acc_top5_avg=0.89831 lr=0.00100 gn=25.57924 time=58.71it/s +epoch=79 global_step=31000 loss=3.49557 loss_avg=3.94692 acc=0.67969 acc_top1_avg=0.62000 acc_top5_avg=0.90048 lr=0.00100 gn=31.68389 time=33.75it/s +epoch=79 global_step=31050 loss=3.88057 loss_avg=3.95871 acc=0.62500 acc_top1_avg=0.61845 acc_top5_avg=0.89878 lr=0.00100 gn=29.73078 time=54.15it/s +epoch=79 global_step=31100 loss=3.49246 loss_avg=3.96119 acc=0.66406 acc_top1_avg=0.61830 acc_top5_avg=0.89803 lr=0.00100 gn=26.66077 time=61.85it/s +epoch=79 global_step=31150 loss=3.74641 loss_avg=3.98661 acc=0.64844 acc_top1_avg=0.61539 acc_top5_avg=0.89715 lr=0.00100 gn=26.05931 time=55.18it/s +epoch=79 global_step=31200 loss=4.57742 loss_avg=3.98671 acc=0.55469 acc_top1_avg=0.61578 acc_top5_avg=0.89693 lr=0.00100 gn=33.78909 time=52.91it/s +epoch=79 global_step=31250 loss=3.67621 loss_avg=3.98320 acc=0.65625 acc_top1_avg=0.61645 acc_top5_avg=0.89697 lr=0.00100 gn=26.07600 time=54.42it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.41503 test_loss_avg=1.05874 acc=0.91406 test_acc_avg=0.68229 test_acc_top5_avg=0.97569 time=239.61it/s +epoch=79 global_step=31280 loss=0.33772 test_loss_avg=1.69555 acc=0.86719 test_acc_avg=0.57667 test_acc_top5_avg=0.93684 time=49.28it/s +epoch=79 global_step=31280 loss=0.00684 test_loss_avg=1.30975 acc=1.00000 test_acc_avg=0.67148 test_acc_top5_avg=0.95194 time=816.01it/s +curr_acc 0.6715 +BEST_ACC 0.7056 +curr_acc_top5 0.9519 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=3.74950 loss_avg=3.87550 acc=0.62500 acc_top1_avg=0.62617 acc_top5_avg=0.89961 lr=0.00010 gn=29.56897 time=61.69it/s +epoch=80 global_step=31350 loss=4.06330 loss_avg=3.88669 acc=0.59375 acc_top1_avg=0.62444 acc_top5_avg=0.89810 lr=0.00010 gn=29.92574 time=58.62it/s +epoch=80 global_step=31400 loss=3.36144 loss_avg=3.85074 acc=0.67969 acc_top1_avg=0.62793 acc_top5_avg=0.89655 lr=0.00010 gn=24.00051 time=50.16it/s +epoch=80 global_step=31450 loss=4.38756 loss_avg=3.85741 acc=0.54688 acc_top1_avg=0.62767 acc_top5_avg=0.89802 lr=0.00010 gn=22.45476 time=52.62it/s +epoch=80 global_step=31500 loss=4.34541 loss_avg=3.85426 acc=0.57812 acc_top1_avg=0.62852 acc_top5_avg=0.89641 lr=0.00010 gn=23.87397 time=60.37it/s +epoch=80 global_step=31550 loss=4.20210 loss_avg=3.86368 acc=0.58594 acc_top1_avg=0.62737 acc_top5_avg=0.89728 lr=0.00010 gn=22.99565 time=52.45it/s +epoch=80 global_step=31600 loss=3.90048 loss_avg=3.85443 acc=0.63281 acc_top1_avg=0.62861 acc_top5_avg=0.89839 lr=0.00010 gn=34.13865 time=49.64it/s +epoch=80 global_step=31650 loss=3.62261 loss_avg=3.86756 acc=0.65625 acc_top1_avg=0.62722 acc_top5_avg=0.89766 lr=0.00010 gn=28.04410 time=62.40it/s +====================Eval==================== +epoch=80 global_step=31671 loss=4.19173 test_loss_avg=1.62631 acc=0.00000 test_acc_avg=0.58594 test_acc_top5_avg=0.88932 time=250.29it/s +epoch=80 global_step=31671 loss=0.09160 test_loss_avg=1.31578 acc=0.93750 test_acc_avg=0.66831 test_acc_top5_avg=0.94057 time=821.29it/s +curr_acc 0.6683 +BEST_ACC 0.7056 +curr_acc_top5 0.9406 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=4.07586 loss_avg=3.80750 acc=0.60938 acc_top1_avg=0.63227 acc_top5_avg=0.89359 lr=0.00010 gn=24.61130 time=46.90it/s +epoch=81 global_step=31750 loss=4.50451 loss_avg=3.80404 acc=0.57031 acc_top1_avg=0.63311 acc_top5_avg=0.89330 lr=0.00010 gn=29.52148 time=44.72it/s +epoch=81 global_step=31800 loss=3.41643 loss_avg=3.81298 acc=0.67969 acc_top1_avg=0.63239 acc_top5_avg=0.89680 lr=0.00010 gn=33.83738 time=59.70it/s +epoch=81 global_step=31850 loss=4.03846 loss_avg=3.79921 acc=0.61719 acc_top1_avg=0.63417 acc_top5_avg=0.89887 lr=0.00010 gn=25.92238 time=54.87it/s +epoch=81 global_step=31900 loss=3.76803 loss_avg=3.78883 acc=0.63281 acc_top1_avg=0.63503 acc_top5_avg=0.89888 lr=0.00010 gn=27.29675 time=59.19it/s +epoch=81 global_step=31950 loss=3.73431 loss_avg=3.80022 acc=0.64062 acc_top1_avg=0.63385 acc_top5_avg=0.89863 lr=0.00010 gn=31.21131 time=54.58it/s +epoch=81 global_step=32000 loss=3.92223 loss_avg=3.80007 acc=0.62500 acc_top1_avg=0.63414 acc_top5_avg=0.89832 lr=0.00010 gn=28.88172 time=54.09it/s +epoch=81 global_step=32050 loss=4.16278 loss_avg=3.81549 acc=0.60156 acc_top1_avg=0.63267 acc_top5_avg=0.89776 lr=0.00010 gn=24.71921 time=62.80it/s +====================Eval==================== +epoch=81 global_step=32062 loss=1.15544 test_loss_avg=1.15544 acc=0.65625 test_acc_avg=0.65625 test_acc_top5_avg=0.96094 time=209.61it/s +epoch=81 global_step=32062 loss=4.38377 test_loss_avg=1.60432 acc=0.00000 test_acc_avg=0.59314 test_acc_top5_avg=0.92050 time=235.56it/s +epoch=81 global_step=32062 loss=0.06847 test_loss_avg=1.29278 acc=0.93750 test_acc_avg=0.67197 test_acc_top5_avg=0.94521 time=836.52it/s +curr_acc 0.6720 +BEST_ACC 0.7056 +curr_acc_top5 0.9452 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=3.41711 loss_avg=3.78974 acc=0.68750 acc_top1_avg=0.63281 acc_top5_avg=0.89659 lr=0.00010 gn=30.88960 time=61.85it/s +epoch=82 global_step=32150 loss=3.34522 loss_avg=3.81515 acc=0.68750 acc_top1_avg=0.63228 acc_top5_avg=0.89728 lr=0.00010 gn=34.34495 time=54.34it/s +epoch=82 global_step=32200 loss=3.12726 loss_avg=3.76797 acc=0.71094 acc_top1_avg=0.63672 acc_top5_avg=0.89906 lr=0.00010 gn=33.57539 time=55.26it/s +epoch=82 global_step=32250 loss=4.12918 loss_avg=3.77031 acc=0.57812 acc_top1_avg=0.63684 acc_top5_avg=0.89802 lr=0.00010 gn=22.71858 time=59.17it/s +epoch=82 global_step=32300 loss=3.62829 loss_avg=3.77050 acc=0.67188 acc_top1_avg=0.63662 acc_top5_avg=0.89870 lr=0.00010 gn=36.74963 time=54.62it/s +epoch=82 global_step=32350 loss=4.10768 loss_avg=3.77510 acc=0.58594 acc_top1_avg=0.63637 acc_top5_avg=0.89941 lr=0.00010 gn=31.62242 time=57.42it/s +epoch=82 global_step=32400 loss=3.87704 loss_avg=3.78237 acc=0.62500 acc_top1_avg=0.63577 acc_top5_avg=0.89837 lr=0.00010 gn=29.94038 time=51.47it/s +epoch=82 global_step=32450 loss=3.47994 loss_avg=3.79335 acc=0.66406 acc_top1_avg=0.63464 acc_top5_avg=0.89844 lr=0.00010 gn=26.28648 time=42.04it/s +====================Eval==================== +epoch=82 global_step=32453 loss=1.49147 test_loss_avg=0.86965 acc=0.61719 test_acc_avg=0.75355 test_acc_top5_avg=0.97337 time=236.99it/s +epoch=82 global_step=32453 loss=0.19940 test_loss_avg=1.42576 acc=0.93750 test_acc_avg=0.64301 test_acc_top5_avg=0.93641 time=240.61it/s +epoch=82 global_step=32453 loss=0.07963 test_loss_avg=1.31293 acc=0.93750 test_acc_avg=0.66990 test_acc_top5_avg=0.94195 time=837.19it/s +curr_acc 0.6699 +BEST_ACC 0.7056 +curr_acc_top5 0.9420 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=2.73427 loss_avg=3.82709 acc=0.76562 acc_top1_avg=0.63215 acc_top5_avg=0.90010 lr=0.00010 gn=32.00423 time=60.78it/s +epoch=83 global_step=32550 loss=3.75204 loss_avg=3.77990 acc=0.64062 acc_top1_avg=0.63660 acc_top5_avg=0.89965 lr=0.00010 gn=24.66168 time=52.71it/s +epoch=83 global_step=32600 loss=4.37917 loss_avg=3.76299 acc=0.57031 acc_top1_avg=0.63733 acc_top5_avg=0.90088 lr=0.00010 gn=30.16133 time=41.88it/s +epoch=83 global_step=32650 loss=4.07458 loss_avg=3.75424 acc=0.60156 acc_top1_avg=0.63785 acc_top5_avg=0.89816 lr=0.00010 gn=33.80107 time=54.13it/s +epoch=83 global_step=32700 loss=4.42737 loss_avg=3.76606 acc=0.57812 acc_top1_avg=0.63661 acc_top5_avg=0.89853 lr=0.00010 gn=31.87818 time=56.65it/s +epoch=83 global_step=32750 loss=3.22901 loss_avg=3.77668 acc=0.70312 acc_top1_avg=0.63589 acc_top5_avg=0.89799 lr=0.00010 gn=27.88623 time=52.92it/s +epoch=83 global_step=32800 loss=3.73183 loss_avg=3.78327 acc=0.64062 acc_top1_avg=0.63531 acc_top5_avg=0.89763 lr=0.00010 gn=36.40571 time=49.90it/s +====================Eval==================== +epoch=83 global_step=32844 loss=1.36203 test_loss_avg=1.41428 acc=0.63281 test_acc_avg=0.63826 test_acc_top5_avg=0.90716 time=227.47it/s +epoch=83 global_step=32844 loss=0.09589 test_loss_avg=1.31423 acc=0.93750 test_acc_avg=0.66772 test_acc_top5_avg=0.94175 time=535.06it/s +curr_acc 0.6677 +BEST_ACC 0.7056 +curr_acc_top5 0.9418 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=3.80424 loss_avg=3.92571 acc=0.64062 acc_top1_avg=0.61458 acc_top5_avg=0.89062 lr=0.00010 gn=29.09026 time=57.88it/s +epoch=84 global_step=32900 loss=3.83475 loss_avg=3.78996 acc=0.63281 acc_top1_avg=0.63421 acc_top5_avg=0.89648 lr=0.00010 gn=34.78496 time=58.39it/s +epoch=84 global_step=32950 loss=4.47608 loss_avg=3.77827 acc=0.56250 acc_top1_avg=0.63569 acc_top5_avg=0.89807 lr=0.00010 gn=31.80601 time=59.08it/s +epoch=84 global_step=33000 loss=3.95823 loss_avg=3.78186 acc=0.61719 acc_top1_avg=0.63537 acc_top5_avg=0.89613 lr=0.00010 gn=32.41586 time=60.55it/s +epoch=84 global_step=33050 loss=4.04593 loss_avg=3.76376 acc=0.60156 acc_top1_avg=0.63759 acc_top5_avg=0.89791 lr=0.00010 gn=27.21537 time=62.37it/s +epoch=84 global_step=33100 loss=4.36689 loss_avg=3.75056 acc=0.58594 acc_top1_avg=0.63885 acc_top5_avg=0.89874 lr=0.00010 gn=27.15896 time=54.22it/s +epoch=84 global_step=33150 loss=4.46481 loss_avg=3.77195 acc=0.56250 acc_top1_avg=0.63634 acc_top5_avg=0.89790 lr=0.00010 gn=32.36436 time=52.19it/s +epoch=84 global_step=33200 loss=3.34343 loss_avg=3.76150 acc=0.69531 acc_top1_avg=0.63782 acc_top5_avg=0.89890 lr=0.00010 gn=32.81984 time=53.65it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.18922 test_loss_avg=0.74164 acc=0.96094 test_acc_avg=0.77902 test_acc_top5_avg=0.98326 time=182.97it/s +epoch=84 global_step=33235 loss=0.26492 test_loss_avg=1.56447 acc=0.92969 test_acc_avg=0.60510 test_acc_top5_avg=0.93494 time=249.35it/s +epoch=84 global_step=33235 loss=0.10727 test_loss_avg=1.29918 acc=0.93750 test_acc_avg=0.67019 test_acc_top5_avg=0.94689 time=853.72it/s +curr_acc 0.6702 +BEST_ACC 0.7056 +curr_acc_top5 0.9469 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=4.20244 loss_avg=3.83287 acc=0.60156 acc_top1_avg=0.63490 acc_top5_avg=0.90312 lr=0.00010 gn=24.83134 time=54.23it/s +epoch=85 global_step=33300 loss=3.04224 loss_avg=3.77796 acc=0.72656 acc_top1_avg=0.63642 acc_top5_avg=0.90445 lr=0.00010 gn=31.98785 time=60.53it/s +epoch=85 global_step=33350 loss=4.57448 loss_avg=3.76166 acc=0.55469 acc_top1_avg=0.63811 acc_top5_avg=0.90177 lr=0.00010 gn=34.55927 time=53.24it/s +epoch=85 global_step=33400 loss=3.64603 loss_avg=3.77403 acc=0.64062 acc_top1_avg=0.63660 acc_top5_avg=0.89891 lr=0.00010 gn=27.83180 time=53.84it/s +epoch=85 global_step=33450 loss=3.63218 loss_avg=3.77835 acc=0.65625 acc_top1_avg=0.63659 acc_top5_avg=0.89782 lr=0.00010 gn=35.65999 time=54.87it/s +epoch=85 global_step=33500 loss=4.05901 loss_avg=3.75284 acc=0.60938 acc_top1_avg=0.63936 acc_top5_avg=0.89912 lr=0.00010 gn=32.07180 time=60.06it/s +epoch=85 global_step=33550 loss=3.50085 loss_avg=3.75509 acc=0.67188 acc_top1_avg=0.63896 acc_top5_avg=0.89938 lr=0.00010 gn=32.85558 time=52.99it/s +epoch=85 global_step=33600 loss=3.76038 loss_avg=3.76356 acc=0.64062 acc_top1_avg=0.63793 acc_top5_avg=0.89869 lr=0.00010 gn=26.99053 time=55.35it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.18671 test_loss_avg=1.64183 acc=0.95312 test_acc_avg=0.58259 test_acc_top5_avg=0.88951 time=229.81it/s +epoch=85 global_step=33626 loss=0.13312 test_loss_avg=1.34429 acc=0.93750 test_acc_avg=0.65991 test_acc_top5_avg=0.94146 time=509.57it/s +curr_acc 0.6599 +BEST_ACC 0.7056 +curr_acc_top5 0.9415 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=4.60200 loss_avg=3.65151 acc=0.53906 acc_top1_avg=0.65267 acc_top5_avg=0.89583 lr=0.00010 gn=37.05615 time=49.48it/s +epoch=86 global_step=33700 loss=3.20647 loss_avg=3.74446 acc=0.71094 acc_top1_avg=0.64020 acc_top5_avg=0.89580 lr=0.00010 gn=32.12542 time=54.91it/s +epoch=86 global_step=33750 loss=4.42261 loss_avg=3.76693 acc=0.58594 acc_top1_avg=0.63817 acc_top5_avg=0.89491 lr=0.00010 gn=37.94298 time=55.43it/s +epoch=86 global_step=33800 loss=3.16247 loss_avg=3.75185 acc=0.71094 acc_top1_avg=0.63964 acc_top5_avg=0.89723 lr=0.00010 gn=30.67093 time=52.42it/s +epoch=86 global_step=33850 loss=3.50351 loss_avg=3.75148 acc=0.66406 acc_top1_avg=0.63972 acc_top5_avg=0.89784 lr=0.00010 gn=29.79552 time=62.03it/s +epoch=86 global_step=33900 loss=3.23005 loss_avg=3.72973 acc=0.69531 acc_top1_avg=0.64205 acc_top5_avg=0.89944 lr=0.00010 gn=30.46738 time=57.57it/s +epoch=86 global_step=33950 loss=2.90507 loss_avg=3.73015 acc=0.73438 acc_top1_avg=0.64200 acc_top5_avg=0.89950 lr=0.00010 gn=26.10761 time=52.54it/s +epoch=86 global_step=34000 loss=4.37620 loss_avg=3.74017 acc=0.56250 acc_top1_avg=0.64088 acc_top5_avg=0.89988 lr=0.00010 gn=21.73290 time=62.93it/s +====================Eval==================== +epoch=86 global_step=34017 loss=1.43726 test_loss_avg=1.32014 acc=0.57031 test_acc_avg=0.60026 test_acc_top5_avg=0.97656 time=240.36it/s +epoch=86 global_step=34017 loss=0.32379 test_loss_avg=1.76420 acc=0.89844 test_acc_avg=0.55483 test_acc_top5_avg=0.92062 time=232.65it/s +epoch=86 global_step=34017 loss=0.13641 test_loss_avg=1.31223 acc=0.93750 test_acc_avg=0.66555 test_acc_top5_avg=0.94235 time=542.32it/s +curr_acc 0.6655 +BEST_ACC 0.7056 +curr_acc_top5 0.9423 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=3.82365 loss_avg=3.68977 acc=0.61719 acc_top1_avg=0.64465 acc_top5_avg=0.90199 lr=0.00010 gn=26.61433 time=61.52it/s +epoch=87 global_step=34100 loss=3.53367 loss_avg=3.75165 acc=0.66406 acc_top1_avg=0.63893 acc_top5_avg=0.89985 lr=0.00010 gn=26.03056 time=51.65it/s +epoch=87 global_step=34150 loss=3.76354 loss_avg=3.72704 acc=0.64062 acc_top1_avg=0.64133 acc_top5_avg=0.89961 lr=0.00010 gn=30.43498 time=52.88it/s +epoch=87 global_step=34200 loss=4.36572 loss_avg=3.73080 acc=0.58594 acc_top1_avg=0.64225 acc_top5_avg=0.90015 lr=0.00010 gn=29.04204 time=61.26it/s +epoch=87 global_step=34250 loss=4.09697 loss_avg=3.71215 acc=0.60156 acc_top1_avg=0.64384 acc_top5_avg=0.90062 lr=0.00010 gn=24.44887 time=58.90it/s +epoch=87 global_step=34300 loss=3.75292 loss_avg=3.72135 acc=0.62500 acc_top1_avg=0.64261 acc_top5_avg=0.89951 lr=0.00010 gn=33.78512 time=49.51it/s +epoch=87 global_step=34350 loss=4.40857 loss_avg=3.71971 acc=0.55469 acc_top1_avg=0.64297 acc_top5_avg=0.89886 lr=0.00010 gn=22.51485 time=53.83it/s +epoch=87 global_step=34400 loss=3.95140 loss_avg=3.73494 acc=0.63281 acc_top1_avg=0.64152 acc_top5_avg=0.89848 lr=0.00010 gn=31.91629 time=63.51it/s +====================Eval==================== +epoch=87 global_step=34408 loss=4.40253 test_loss_avg=1.35231 acc=0.00000 test_acc_avg=0.64815 test_acc_top5_avg=0.92245 time=238.79it/s +epoch=87 global_step=34408 loss=0.18288 test_loss_avg=1.35172 acc=0.92969 test_acc_avg=0.66031 test_acc_top5_avg=0.93943 time=243.53it/s +epoch=87 global_step=34408 loss=0.10126 test_loss_avg=1.32032 acc=0.93750 test_acc_avg=0.66742 test_acc_top5_avg=0.94096 time=748.72it/s +curr_acc 0.6674 +BEST_ACC 0.7056 +curr_acc_top5 0.9410 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=3.36619 loss_avg=3.74402 acc=0.67969 acc_top1_avg=0.64100 acc_top5_avg=0.89639 lr=0.00010 gn=27.55659 time=61.69it/s +epoch=88 global_step=34500 loss=3.28230 loss_avg=3.74441 acc=0.68750 acc_top1_avg=0.64029 acc_top5_avg=0.89920 lr=0.00010 gn=28.52921 time=52.06it/s +epoch=88 global_step=34550 loss=4.11608 loss_avg=3.74488 acc=0.59375 acc_top1_avg=0.63958 acc_top5_avg=0.89860 lr=0.00010 gn=33.75274 time=62.07it/s +epoch=88 global_step=34600 loss=3.67011 loss_avg=3.75957 acc=0.64062 acc_top1_avg=0.63843 acc_top5_avg=0.89836 lr=0.00010 gn=27.81157 time=54.56it/s +epoch=88 global_step=34650 loss=3.56184 loss_avg=3.74975 acc=0.66406 acc_top1_avg=0.63988 acc_top5_avg=0.89776 lr=0.00010 gn=33.68712 time=61.75it/s +epoch=88 global_step=34700 loss=3.68846 loss_avg=3.73869 acc=0.64844 acc_top1_avg=0.64057 acc_top5_avg=0.89852 lr=0.00010 gn=31.31078 time=59.56it/s +epoch=88 global_step=34750 loss=3.53576 loss_avg=3.73445 acc=0.66406 acc_top1_avg=0.64090 acc_top5_avg=0.89841 lr=0.00010 gn=27.98551 time=38.97it/s +====================Eval==================== +epoch=88 global_step=34799 loss=4.32667 test_loss_avg=1.45291 acc=0.00000 test_acc_avg=0.63102 test_acc_top5_avg=0.91504 time=237.40it/s +epoch=88 global_step=34799 loss=0.13067 test_loss_avg=1.30995 acc=0.93750 test_acc_avg=0.67049 test_acc_top5_avg=0.94304 time=508.77it/s +curr_acc 0.6705 +BEST_ACC 0.7056 +curr_acc_top5 0.9430 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=3.34235 loss_avg=3.34235 acc=0.67188 acc_top1_avg=0.67188 acc_top5_avg=0.88281 lr=0.00010 gn=30.16173 time=39.27it/s +epoch=89 global_step=34850 loss=3.70872 loss_avg=3.67513 acc=0.63281 acc_top1_avg=0.64553 acc_top5_avg=0.89170 lr=0.00010 gn=24.86925 time=63.49it/s +epoch=89 global_step=34900 loss=3.24806 loss_avg=3.66675 acc=0.68750 acc_top1_avg=0.64712 acc_top5_avg=0.89372 lr=0.00010 gn=19.73062 time=50.31it/s +epoch=89 global_step=34950 loss=3.79213 loss_avg=3.67959 acc=0.63281 acc_top1_avg=0.64657 acc_top5_avg=0.89730 lr=0.00010 gn=26.36165 time=54.86it/s +epoch=89 global_step=35000 loss=3.48696 loss_avg=3.69569 acc=0.64844 acc_top1_avg=0.64513 acc_top5_avg=0.89871 lr=0.00010 gn=23.52767 time=56.91it/s +epoch=89 global_step=35050 loss=3.96632 loss_avg=3.71253 acc=0.60938 acc_top1_avg=0.64392 acc_top5_avg=0.89810 lr=0.00010 gn=21.83291 time=52.12it/s +epoch=89 global_step=35100 loss=4.22472 loss_avg=3.69967 acc=0.57812 acc_top1_avg=0.64506 acc_top5_avg=0.89732 lr=0.00010 gn=26.53919 time=43.78it/s +epoch=89 global_step=35150 loss=3.55892 loss_avg=3.70969 acc=0.65625 acc_top1_avg=0.64399 acc_top5_avg=0.89759 lr=0.00010 gn=29.50351 time=48.79it/s +====================Eval==================== +epoch=89 global_step=35190 loss=1.10533 test_loss_avg=0.86649 acc=0.70312 test_acc_avg=0.74794 test_acc_top5_avg=0.97862 time=238.83it/s +epoch=89 global_step=35190 loss=0.19298 test_loss_avg=1.48634 acc=0.95312 test_acc_avg=0.62251 test_acc_top5_avg=0.93342 time=226.83it/s +epoch=89 global_step=35190 loss=0.10131 test_loss_avg=1.31723 acc=0.93750 test_acc_avg=0.66386 test_acc_top5_avg=0.94165 time=840.54it/s +curr_acc 0.6639 +BEST_ACC 0.7056 +curr_acc_top5 0.9417 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=3.67608 loss_avg=3.57026 acc=0.64844 acc_top1_avg=0.65938 acc_top5_avg=0.90625 lr=0.00010 gn=27.69491 time=46.10it/s +epoch=90 global_step=35250 loss=3.73343 loss_avg=3.74250 acc=0.63281 acc_top1_avg=0.64141 acc_top5_avg=0.89740 lr=0.00010 gn=24.06992 time=54.23it/s +epoch=90 global_step=35300 loss=4.42897 loss_avg=3.72395 acc=0.57812 acc_top1_avg=0.64276 acc_top5_avg=0.89901 lr=0.00010 gn=29.23985 time=62.86it/s +epoch=90 global_step=35350 loss=4.08694 loss_avg=3.70116 acc=0.60156 acc_top1_avg=0.64492 acc_top5_avg=0.89917 lr=0.00010 gn=32.65789 time=55.73it/s +epoch=90 global_step=35400 loss=3.45623 loss_avg=3.72427 acc=0.67188 acc_top1_avg=0.64271 acc_top5_avg=0.89814 lr=0.00010 gn=33.13992 time=55.23it/s +epoch=90 global_step=35450 loss=3.79183 loss_avg=3.72622 acc=0.63281 acc_top1_avg=0.64222 acc_top5_avg=0.89727 lr=0.00010 gn=28.84779 time=55.08it/s +epoch=90 global_step=35500 loss=2.72381 loss_avg=3.72539 acc=0.75781 acc_top1_avg=0.64234 acc_top5_avg=0.89753 lr=0.00010 gn=26.96454 time=59.72it/s +epoch=90 global_step=35550 loss=4.08058 loss_avg=3.71995 acc=0.60938 acc_top1_avg=0.64273 acc_top5_avg=0.89816 lr=0.00010 gn=38.46754 time=53.24it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.97345 test_loss_avg=1.47347 acc=0.74219 test_acc_avg=0.62617 test_acc_top5_avg=0.90156 time=239.36it/s +epoch=90 global_step=35581 loss=0.16654 test_loss_avg=1.33234 acc=0.93750 test_acc_avg=0.66288 test_acc_top5_avg=0.94076 time=545.49it/s +curr_acc 0.6629 +BEST_ACC 0.7056 +curr_acc_top5 0.9408 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=3.04316 loss_avg=3.66472 acc=0.71094 acc_top1_avg=0.64803 acc_top5_avg=0.89474 lr=0.00010 gn=27.40106 time=57.71it/s +epoch=91 global_step=35650 loss=4.11088 loss_avg=3.71240 acc=0.60938 acc_top1_avg=0.64278 acc_top5_avg=0.89697 lr=0.00010 gn=30.42108 time=54.74it/s +epoch=91 global_step=35700 loss=2.89306 loss_avg=3.69395 acc=0.72656 acc_top1_avg=0.64542 acc_top5_avg=0.89883 lr=0.00010 gn=25.38867 time=60.24it/s +epoch=91 global_step=35750 loss=4.00863 loss_avg=3.67747 acc=0.60938 acc_top1_avg=0.64705 acc_top5_avg=0.89955 lr=0.00010 gn=30.71063 time=54.18it/s +epoch=91 global_step=35800 loss=3.50005 loss_avg=3.69888 acc=0.67188 acc_top1_avg=0.64451 acc_top5_avg=0.89865 lr=0.00010 gn=34.63991 time=54.27it/s +epoch=91 global_step=35850 loss=3.65961 loss_avg=3.70055 acc=0.64844 acc_top1_avg=0.64440 acc_top5_avg=0.89864 lr=0.00010 gn=35.83510 time=55.83it/s +epoch=91 global_step=35900 loss=3.91471 loss_avg=3.70592 acc=0.61719 acc_top1_avg=0.64371 acc_top5_avg=0.89885 lr=0.00010 gn=29.52655 time=49.76it/s +epoch=91 global_step=35950 loss=3.94363 loss_avg=3.70647 acc=0.61719 acc_top1_avg=0.64363 acc_top5_avg=0.89867 lr=0.00010 gn=30.30159 time=55.43it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.20572 test_loss_avg=0.98317 acc=0.92969 test_acc_avg=0.70810 test_acc_top5_avg=0.98509 time=233.54it/s +epoch=91 global_step=35972 loss=0.24439 test_loss_avg=1.66901 acc=0.92188 test_acc_avg=0.57838 test_acc_top5_avg=0.92328 time=234.69it/s +epoch=91 global_step=35972 loss=0.15569 test_loss_avg=1.33174 acc=0.93750 test_acc_avg=0.66129 test_acc_top5_avg=0.93977 time=737.78it/s +curr_acc 0.6613 +BEST_ACC 0.7056 +curr_acc_top5 0.9398 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=4.07303 loss_avg=3.76166 acc=0.60156 acc_top1_avg=0.63867 acc_top5_avg=0.89230 lr=0.00010 gn=26.85298 time=62.00it/s +epoch=92 global_step=36050 loss=3.97442 loss_avg=3.73766 acc=0.62500 acc_top1_avg=0.63942 acc_top5_avg=0.89704 lr=0.00010 gn=32.25292 time=56.78it/s +epoch=92 global_step=36100 loss=3.93972 loss_avg=3.66837 acc=0.64062 acc_top1_avg=0.64771 acc_top5_avg=0.89691 lr=0.00010 gn=35.23682 time=55.22it/s +epoch=92 global_step=36150 loss=3.51676 loss_avg=3.68388 acc=0.65625 acc_top1_avg=0.64646 acc_top5_avg=0.89633 lr=0.00010 gn=25.33122 time=55.18it/s +epoch=92 global_step=36200 loss=3.58268 loss_avg=3.70808 acc=0.64062 acc_top1_avg=0.64429 acc_top5_avg=0.89703 lr=0.00010 gn=34.28776 time=52.35it/s +epoch=92 global_step=36250 loss=3.32294 loss_avg=3.70657 acc=0.70312 acc_top1_avg=0.64419 acc_top5_avg=0.89782 lr=0.00010 gn=32.38431 time=54.87it/s +epoch=92 global_step=36300 loss=3.55490 loss_avg=3.71042 acc=0.65625 acc_top1_avg=0.64360 acc_top5_avg=0.89837 lr=0.00010 gn=27.29659 time=53.98it/s +epoch=92 global_step=36350 loss=2.88837 loss_avg=3.70045 acc=0.72656 acc_top1_avg=0.64451 acc_top5_avg=0.89862 lr=0.00010 gn=26.65344 time=57.47it/s +====================Eval==================== +epoch=92 global_step=36363 loss=1.36829 test_loss_avg=1.72623 acc=0.66406 test_acc_avg=0.56689 test_acc_top5_avg=0.88550 time=238.46it/s +epoch=92 global_step=36363 loss=0.11681 test_loss_avg=1.33519 acc=0.93750 test_acc_avg=0.66129 test_acc_top5_avg=0.94244 time=692.82it/s +curr_acc 0.6613 +BEST_ACC 0.7056 +curr_acc_top5 0.9424 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=3.26406 loss_avg=3.64754 acc=0.70312 acc_top1_avg=0.65097 acc_top5_avg=0.90013 lr=0.00010 gn=36.39613 time=62.24it/s +epoch=93 global_step=36450 loss=4.29014 loss_avg=3.70157 acc=0.57812 acc_top1_avg=0.64404 acc_top5_avg=0.90122 lr=0.00010 gn=28.24806 time=54.78it/s +epoch=93 global_step=36500 loss=3.64917 loss_avg=3.69743 acc=0.65625 acc_top1_avg=0.64467 acc_top5_avg=0.90209 lr=0.00010 gn=33.71611 time=54.60it/s +epoch=93 global_step=36550 loss=3.73952 loss_avg=3.69601 acc=0.63281 acc_top1_avg=0.64472 acc_top5_avg=0.90078 lr=0.00010 gn=26.41391 time=53.25it/s +epoch=93 global_step=36600 loss=3.31213 loss_avg=3.68817 acc=0.67188 acc_top1_avg=0.64573 acc_top5_avg=0.90009 lr=0.00010 gn=27.21459 time=59.87it/s +epoch=93 global_step=36650 loss=3.71339 loss_avg=3.69435 acc=0.64844 acc_top1_avg=0.64501 acc_top5_avg=0.89934 lr=0.00010 gn=25.03607 time=57.32it/s +epoch=93 global_step=36700 loss=3.80001 loss_avg=3.68778 acc=0.64062 acc_top1_avg=0.64600 acc_top5_avg=0.89957 lr=0.00010 gn=31.60908 time=57.43it/s +epoch=93 global_step=36750 loss=4.03249 loss_avg=3.68675 acc=0.60156 acc_top1_avg=0.64616 acc_top5_avg=0.89941 lr=0.00010 gn=32.22812 time=54.68it/s +====================Eval==================== +epoch=93 global_step=36754 loss=1.57899 test_loss_avg=1.45252 acc=0.53906 test_acc_avg=0.57812 test_acc_top5_avg=0.96615 time=237.87it/s +epoch=93 global_step=36754 loss=4.11416 test_loss_avg=1.76140 acc=0.00000 test_acc_avg=0.55719 test_acc_top5_avg=0.91627 time=141.83it/s +epoch=93 global_step=36754 loss=0.10306 test_loss_avg=1.33837 acc=0.93750 test_acc_avg=0.66040 test_acc_top5_avg=0.94096 time=829.90it/s +curr_acc 0.6604 +BEST_ACC 0.7056 +curr_acc_top5 0.9410 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=2.99836 loss_avg=3.72754 acc=0.71094 acc_top1_avg=0.64164 acc_top5_avg=0.89147 lr=0.00010 gn=27.63288 time=54.45it/s +epoch=94 global_step=36850 loss=3.65195 loss_avg=3.68152 acc=0.64844 acc_top1_avg=0.64591 acc_top5_avg=0.89746 lr=0.00010 gn=35.03570 time=42.31it/s +epoch=94 global_step=36900 loss=3.42295 loss_avg=3.68270 acc=0.67969 acc_top1_avg=0.64646 acc_top5_avg=0.89881 lr=0.00010 gn=30.46518 time=57.33it/s +epoch=94 global_step=36950 loss=3.92278 loss_avg=3.67796 acc=0.63281 acc_top1_avg=0.64700 acc_top5_avg=0.89772 lr=0.00010 gn=30.63140 time=59.75it/s +epoch=94 global_step=37000 loss=3.70434 loss_avg=3.68903 acc=0.63281 acc_top1_avg=0.64612 acc_top5_avg=0.89761 lr=0.00010 gn=24.35048 time=53.74it/s +epoch=94 global_step=37050 loss=3.34213 loss_avg=3.69025 acc=0.67969 acc_top1_avg=0.64575 acc_top5_avg=0.89862 lr=0.00010 gn=32.11769 time=53.44it/s +epoch=94 global_step=37100 loss=4.08708 loss_avg=3.69817 acc=0.60156 acc_top1_avg=0.64482 acc_top5_avg=0.89918 lr=0.00010 gn=28.71601 time=54.90it/s +====================Eval==================== +epoch=94 global_step=37145 loss=3.19622 test_loss_avg=1.06124 acc=0.23438 test_acc_avg=0.69954 test_acc_top5_avg=0.96159 time=209.10it/s +epoch=94 global_step=37145 loss=0.21396 test_loss_avg=1.43573 acc=0.93750 test_acc_avg=0.63429 test_acc_top5_avg=0.93687 time=250.77it/s +epoch=94 global_step=37145 loss=0.12907 test_loss_avg=1.35437 acc=0.93750 test_acc_avg=0.65417 test_acc_top5_avg=0.94066 time=805.98it/s +curr_acc 0.6542 +BEST_ACC 0.7056 +curr_acc_top5 0.9407 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=3.79422 loss_avg=3.65943 acc=0.63281 acc_top1_avg=0.64375 acc_top5_avg=0.91875 lr=0.00010 gn=23.80358 time=63.09it/s +epoch=95 global_step=37200 loss=3.85157 loss_avg=3.66689 acc=0.62500 acc_top1_avg=0.64744 acc_top5_avg=0.89986 lr=0.00010 gn=24.40117 time=57.92it/s +epoch=95 global_step=37250 loss=3.60022 loss_avg=3.70821 acc=0.64844 acc_top1_avg=0.64457 acc_top5_avg=0.89658 lr=0.00010 gn=25.79685 time=60.24it/s +epoch=95 global_step=37300 loss=3.52396 loss_avg=3.71861 acc=0.67188 acc_top1_avg=0.64340 acc_top5_avg=0.89824 lr=0.00010 gn=32.88162 time=61.64it/s +epoch=95 global_step=37350 loss=3.92118 loss_avg=3.69155 acc=0.61719 acc_top1_avg=0.64668 acc_top5_avg=0.89783 lr=0.00010 gn=30.12210 time=59.76it/s +epoch=95 global_step=37400 loss=3.74973 loss_avg=3.69251 acc=0.64062 acc_top1_avg=0.64660 acc_top5_avg=0.89874 lr=0.00010 gn=33.21826 time=56.50it/s +epoch=95 global_step=37450 loss=3.54726 loss_avg=3.66815 acc=0.65625 acc_top1_avg=0.64915 acc_top5_avg=0.89967 lr=0.00010 gn=26.14939 time=54.98it/s +epoch=95 global_step=37500 loss=4.23508 loss_avg=3.67061 acc=0.59375 acc_top1_avg=0.64910 acc_top5_avg=0.90000 lr=0.00010 gn=25.97033 time=55.88it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.94427 test_loss_avg=1.44684 acc=0.72656 test_acc_avg=0.62604 test_acc_top5_avg=0.91111 time=229.61it/s +epoch=95 global_step=37536 loss=0.13964 test_loss_avg=1.34458 acc=0.93750 test_acc_avg=0.65862 test_acc_top5_avg=0.94146 time=493.45it/s +curr_acc 0.6586 +BEST_ACC 0.7056 +curr_acc_top5 0.9415 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=3.71438 loss_avg=3.67241 acc=0.65625 acc_top1_avg=0.64788 acc_top5_avg=0.90234 lr=0.00010 gn=30.69896 time=56.56it/s +epoch=96 global_step=37600 loss=3.61028 loss_avg=3.71468 acc=0.64844 acc_top1_avg=0.64294 acc_top5_avg=0.89648 lr=0.00010 gn=30.13755 time=63.25it/s +epoch=96 global_step=37650 loss=3.86015 loss_avg=3.72950 acc=0.62500 acc_top1_avg=0.64165 acc_top5_avg=0.89748 lr=0.00010 gn=29.43777 time=56.03it/s +epoch=96 global_step=37700 loss=3.95723 loss_avg=3.72460 acc=0.62500 acc_top1_avg=0.64229 acc_top5_avg=0.89682 lr=0.00010 gn=31.73796 time=56.20it/s +epoch=96 global_step=37750 loss=3.94762 loss_avg=3.70010 acc=0.60938 acc_top1_avg=0.64566 acc_top5_avg=0.89822 lr=0.00010 gn=29.92816 time=50.17it/s +epoch=96 global_step=37800 loss=4.08140 loss_avg=3.70281 acc=0.60938 acc_top1_avg=0.64536 acc_top5_avg=0.89758 lr=0.00010 gn=31.31557 time=62.75it/s +epoch=96 global_step=37850 loss=3.60973 loss_avg=3.68790 acc=0.64844 acc_top1_avg=0.64682 acc_top5_avg=0.89769 lr=0.00010 gn=26.79052 time=59.70it/s +epoch=96 global_step=37900 loss=3.66295 loss_avg=3.68515 acc=0.64844 acc_top1_avg=0.64715 acc_top5_avg=0.89809 lr=0.00010 gn=29.09241 time=52.39it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.61582 test_loss_avg=0.73832 acc=0.81250 test_acc_avg=0.78662 test_acc_top5_avg=0.98486 time=240.40it/s +epoch=96 global_step=37927 loss=0.14897 test_loss_avg=1.56256 acc=0.95312 test_acc_avg=0.60606 test_acc_top5_avg=0.93111 time=227.49it/s +epoch=96 global_step=37927 loss=0.16818 test_loss_avg=1.33330 acc=0.93750 test_acc_avg=0.66208 test_acc_top5_avg=0.94205 time=657.00it/s +curr_acc 0.6621 +BEST_ACC 0.7056 +curr_acc_top5 0.9420 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=3.77162 loss_avg=3.63689 acc=0.64062 acc_top1_avg=0.65353 acc_top5_avg=0.90251 lr=0.00010 gn=35.18127 time=54.52it/s +epoch=97 global_step=38000 loss=3.47266 loss_avg=3.63346 acc=0.66406 acc_top1_avg=0.65208 acc_top5_avg=0.90379 lr=0.00010 gn=27.50658 time=62.98it/s +epoch=97 global_step=38050 loss=3.67189 loss_avg=3.64866 acc=0.64844 acc_top1_avg=0.65047 acc_top5_avg=0.90282 lr=0.00010 gn=26.72362 time=60.62it/s +epoch=97 global_step=38100 loss=3.85605 loss_avg=3.66109 acc=0.62500 acc_top1_avg=0.64916 acc_top5_avg=0.90336 lr=0.00010 gn=31.64208 time=53.62it/s +epoch=97 global_step=38150 loss=3.71966 loss_avg=3.65331 acc=0.64844 acc_top1_avg=0.64998 acc_top5_avg=0.90166 lr=0.00010 gn=33.28322 time=53.84it/s +epoch=97 global_step=38200 loss=4.05954 loss_avg=3.66669 acc=0.60156 acc_top1_avg=0.64809 acc_top5_avg=0.90104 lr=0.00010 gn=30.22265 time=54.74it/s +epoch=97 global_step=38250 loss=4.12567 loss_avg=3.67707 acc=0.60156 acc_top1_avg=0.64699 acc_top5_avg=0.89991 lr=0.00010 gn=39.27766 time=56.22it/s +epoch=97 global_step=38300 loss=4.04479 loss_avg=3.68465 acc=0.60156 acc_top1_avg=0.64632 acc_top5_avg=0.89940 lr=0.00010 gn=32.67810 time=62.03it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.30852 test_loss_avg=1.57593 acc=0.89844 test_acc_avg=0.60177 test_acc_top5_avg=0.89611 time=235.33it/s +epoch=97 global_step=38318 loss=0.10900 test_loss_avg=1.34316 acc=0.93750 test_acc_avg=0.65951 test_acc_top5_avg=0.94047 time=644.68it/s +curr_acc 0.6595 +BEST_ACC 0.7056 +curr_acc_top5 0.9405 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=3.61866 loss_avg=3.75048 acc=0.64844 acc_top1_avg=0.63892 acc_top5_avg=0.89551 lr=0.00010 gn=24.39436 time=53.79it/s +epoch=98 global_step=38400 loss=3.55361 loss_avg=3.67461 acc=0.65625 acc_top1_avg=0.64768 acc_top5_avg=0.89958 lr=0.00010 gn=35.38874 time=55.58it/s +epoch=98 global_step=38450 loss=3.58480 loss_avg=3.63869 acc=0.65625 acc_top1_avg=0.65193 acc_top5_avg=0.90021 lr=0.00010 gn=35.38837 time=63.20it/s +epoch=98 global_step=38500 loss=3.04677 loss_avg=3.64294 acc=0.71094 acc_top1_avg=0.65174 acc_top5_avg=0.89998 lr=0.00010 gn=23.43180 time=61.59it/s +epoch=98 global_step=38550 loss=3.33268 loss_avg=3.64923 acc=0.67969 acc_top1_avg=0.65079 acc_top5_avg=0.89911 lr=0.00010 gn=33.80387 time=51.95it/s +epoch=98 global_step=38600 loss=3.85450 loss_avg=3.66631 acc=0.63281 acc_top1_avg=0.64907 acc_top5_avg=0.89916 lr=0.00010 gn=37.56013 time=58.15it/s +epoch=98 global_step=38650 loss=3.33234 loss_avg=3.66268 acc=0.67188 acc_top1_avg=0.64910 acc_top5_avg=0.89961 lr=0.00010 gn=31.93425 time=54.24it/s +epoch=98 global_step=38700 loss=4.50680 loss_avg=3.66855 acc=0.55469 acc_top1_avg=0.64860 acc_top5_avg=0.89911 lr=0.00010 gn=35.00886 time=55.94it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.97269 test_loss_avg=1.28297 acc=0.71094 test_acc_avg=0.62793 test_acc_top5_avg=0.97168 time=243.61it/s +epoch=98 global_step=38709 loss=0.41480 test_loss_avg=1.73220 acc=0.89062 test_acc_avg=0.56600 test_acc_top5_avg=0.92821 time=217.19it/s +epoch=98 global_step=38709 loss=0.08118 test_loss_avg=1.32534 acc=0.93750 test_acc_avg=0.66475 test_acc_top5_avg=0.94610 time=520.58it/s +curr_acc 0.6648 +BEST_ACC 0.7056 +curr_acc_top5 0.9461 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=2.66236 loss_avg=3.63991 acc=0.75000 acc_top1_avg=0.65111 acc_top5_avg=0.90415 lr=0.00010 gn=30.92725 time=55.82it/s +epoch=99 global_step=38800 loss=3.75626 loss_avg=3.70975 acc=0.62500 acc_top1_avg=0.64269 acc_top5_avg=0.89749 lr=0.00010 gn=28.03033 time=56.21it/s +epoch=99 global_step=38850 loss=3.46844 loss_avg=3.70198 acc=0.67969 acc_top1_avg=0.64423 acc_top5_avg=0.89783 lr=0.00010 gn=30.59635 time=33.86it/s +epoch=99 global_step=38900 loss=3.01193 loss_avg=3.68194 acc=0.71875 acc_top1_avg=0.64664 acc_top5_avg=0.89885 lr=0.00010 gn=37.50821 time=56.33it/s +epoch=99 global_step=38950 loss=3.79720 loss_avg=3.65842 acc=0.64844 acc_top1_avg=0.64905 acc_top5_avg=0.89853 lr=0.00010 gn=38.29250 time=53.36it/s +epoch=99 global_step=39000 loss=3.19131 loss_avg=3.66220 acc=0.69531 acc_top1_avg=0.64860 acc_top5_avg=0.89860 lr=0.00010 gn=28.58373 time=59.11it/s +epoch=99 global_step=39050 loss=4.36136 loss_avg=3.66778 acc=0.57031 acc_top1_avg=0.64812 acc_top5_avg=0.89832 lr=0.00010 gn=38.11647 time=52.28it/s +epoch=99 global_step=39100 loss=3.68954 loss_avg=3.66419 acc=0.65000 acc_top1_avg=0.64844 acc_top5_avg=0.89912 lr=0.00010 gn=46.37743 time=76.17it/s +====================Eval==================== +epoch=99 global_step=39100 loss=4.36578 test_loss_avg=1.56686 acc=0.00000 test_acc_avg=0.59995 test_acc_top5_avg=0.89547 time=240.93it/s +epoch=99 global_step=39100 loss=0.12796 test_loss_avg=1.32926 acc=0.93750 test_acc_avg=0.66307 test_acc_top5_avg=0.94047 time=507.05it/s +epoch=99 global_step=39100 loss=0.12796 test_loss_avg=1.32926 acc=0.93750 test_acc_avg=0.66307 test_acc_top5_avg=0.94047 time=507.05it/s +curr_acc 0.6631 +BEST_ACC 0.7056 +curr_acc_top5 0.9405 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=3.27434 loss_avg=3.63518 acc=0.68750 acc_top1_avg=0.65484 acc_top5_avg=0.90219 lr=0.00010 gn=28.05232 time=30.52it/s +epoch=100 global_step=39200 loss=3.72494 loss_avg=3.66983 acc=0.65625 acc_top1_avg=0.65102 acc_top5_avg=0.89961 lr=0.00010 gn=33.52611 time=58.91it/s +epoch=100 global_step=39250 loss=3.90101 loss_avg=3.67171 acc=0.61719 acc_top1_avg=0.64943 acc_top5_avg=0.89984 lr=0.00010 gn=32.47647 time=59.91it/s +epoch=100 global_step=39300 loss=3.72904 loss_avg=3.66134 acc=0.64062 acc_top1_avg=0.65055 acc_top5_avg=0.89980 lr=0.00010 gn=35.88527 time=56.28it/s +epoch=100 global_step=39350 loss=3.21988 loss_avg=3.66008 acc=0.71094 acc_top1_avg=0.65059 acc_top5_avg=0.89938 lr=0.00010 gn=35.23425 time=55.50it/s +epoch=100 global_step=39400 loss=3.69264 loss_avg=3.66185 acc=0.64062 acc_top1_avg=0.65057 acc_top5_avg=0.89930 lr=0.00010 gn=28.12689 time=58.91it/s +epoch=100 global_step=39450 loss=4.37561 loss_avg=3.66304 acc=0.59375 acc_top1_avg=0.65031 acc_top5_avg=0.89888 lr=0.00010 gn=32.45334 time=58.07it/s +====================Eval==================== +epoch=100 global_step=39491 loss=4.11280 test_loss_avg=1.65392 acc=0.00000 test_acc_avg=0.58078 test_acc_top5_avg=0.91266 time=237.91it/s +epoch=100 global_step=39491 loss=0.11932 test_loss_avg=1.37003 acc=0.93750 test_acc_avg=0.65299 test_acc_top5_avg=0.94047 time=496.90it/s +curr_acc 0.6530 +BEST_ACC 0.7056 +curr_acc_top5 0.9405 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=3.64876 loss_avg=3.73196 acc=0.63281 acc_top1_avg=0.63715 acc_top5_avg=0.90451 lr=0.00010 gn=34.63529 time=55.91it/s +epoch=101 global_step=39550 loss=3.58065 loss_avg=3.64737 acc=0.65625 acc_top1_avg=0.64963 acc_top5_avg=0.89897 lr=0.00010 gn=26.32701 time=48.98it/s +epoch=101 global_step=39600 loss=3.63172 loss_avg=3.66493 acc=0.64844 acc_top1_avg=0.64822 acc_top5_avg=0.89822 lr=0.00010 gn=30.68136 time=61.70it/s +epoch=101 global_step=39650 loss=3.56140 loss_avg=3.66294 acc=0.66406 acc_top1_avg=0.64824 acc_top5_avg=0.89780 lr=0.00010 gn=44.45026 time=51.73it/s +epoch=101 global_step=39700 loss=3.37086 loss_avg=3.66531 acc=0.68750 acc_top1_avg=0.64806 acc_top5_avg=0.89713 lr=0.00010 gn=35.71043 time=61.39it/s +epoch=101 global_step=39750 loss=3.56877 loss_avg=3.66228 acc=0.67188 acc_top1_avg=0.64865 acc_top5_avg=0.89814 lr=0.00010 gn=35.84337 time=62.00it/s +epoch=101 global_step=39800 loss=3.47485 loss_avg=3.65206 acc=0.67969 acc_top1_avg=0.65033 acc_top5_avg=0.89813 lr=0.00010 gn=35.25184 time=47.90it/s +epoch=101 global_step=39850 loss=3.62084 loss_avg=3.64723 acc=0.65625 acc_top1_avg=0.65066 acc_top5_avg=0.89811 lr=0.00010 gn=36.72576 time=59.49it/s +====================Eval==================== +epoch=101 global_step=39882 loss=1.09526 test_loss_avg=0.91916 acc=0.71094 test_acc_avg=0.73921 test_acc_top5_avg=0.97507 time=231.82it/s +epoch=101 global_step=39882 loss=0.14973 test_loss_avg=1.49240 acc=0.94531 test_acc_avg=0.62500 test_acc_top5_avg=0.93431 time=238.71it/s +epoch=101 global_step=39882 loss=0.10769 test_loss_avg=1.35725 acc=0.93750 test_acc_avg=0.65734 test_acc_top5_avg=0.94066 time=511.13it/s +curr_acc 0.6573 +BEST_ACC 0.7056 +curr_acc_top5 0.9407 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=3.78929 loss_avg=3.73592 acc=0.64844 acc_top1_avg=0.64410 acc_top5_avg=0.90495 lr=0.00010 gn=33.65001 time=54.51it/s +epoch=102 global_step=39950 loss=3.79598 loss_avg=3.69798 acc=0.63281 acc_top1_avg=0.64545 acc_top5_avg=0.90211 lr=0.00010 gn=38.13201 time=44.34it/s +epoch=102 global_step=40000 loss=3.93751 loss_avg=3.68070 acc=0.61719 acc_top1_avg=0.64804 acc_top5_avg=0.90003 lr=0.00010 gn=30.26670 time=55.44it/s +epoch=102 global_step=40050 loss=3.60327 loss_avg=3.65015 acc=0.64844 acc_top1_avg=0.65090 acc_top5_avg=0.90007 lr=0.00010 gn=29.56573 time=58.08it/s +epoch=102 global_step=40100 loss=4.08434 loss_avg=3.62637 acc=0.60156 acc_top1_avg=0.65328 acc_top5_avg=0.90091 lr=0.00010 gn=35.61707 time=54.64it/s +epoch=102 global_step=40150 loss=4.20892 loss_avg=3.62868 acc=0.58594 acc_top1_avg=0.65290 acc_top5_avg=0.90068 lr=0.00010 gn=29.98579 time=54.72it/s +epoch=102 global_step=40200 loss=3.80392 loss_avg=3.63558 acc=0.63281 acc_top1_avg=0.65188 acc_top5_avg=0.90080 lr=0.00010 gn=39.26208 time=55.82it/s +epoch=102 global_step=40250 loss=3.62833 loss_avg=3.62869 acc=0.64844 acc_top1_avg=0.65275 acc_top5_avg=0.90018 lr=0.00010 gn=35.60989 time=48.73it/s +====================Eval==================== +epoch=102 global_step=40273 loss=1.11895 test_loss_avg=1.47624 acc=0.66406 test_acc_avg=0.62630 test_acc_top5_avg=0.90588 time=236.26it/s +epoch=102 global_step=40273 loss=0.14281 test_loss_avg=1.34890 acc=0.93750 test_acc_avg=0.66090 test_acc_top5_avg=0.94047 time=507.29it/s +curr_acc 0.6609 +BEST_ACC 0.7056 +curr_acc_top5 0.9405 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=4.05458 loss_avg=3.61258 acc=0.60938 acc_top1_avg=0.65567 acc_top5_avg=0.89149 lr=0.00010 gn=37.38956 time=53.28it/s +epoch=103 global_step=40350 loss=4.24420 loss_avg=3.67317 acc=0.58594 acc_top1_avg=0.64752 acc_top5_avg=0.89580 lr=0.00010 gn=31.29792 time=54.17it/s +epoch=103 global_step=40400 loss=3.95233 loss_avg=3.66666 acc=0.61719 acc_top1_avg=0.64770 acc_top5_avg=0.89647 lr=0.00010 gn=27.20944 time=62.04it/s +epoch=103 global_step=40450 loss=4.92698 loss_avg=3.63905 acc=0.50000 acc_top1_avg=0.65113 acc_top5_avg=0.89853 lr=0.00010 gn=32.16599 time=61.24it/s +epoch=103 global_step=40500 loss=3.56484 loss_avg=3.64741 acc=0.66406 acc_top1_avg=0.65061 acc_top5_avg=0.89858 lr=0.00010 gn=31.36402 time=57.37it/s +epoch=103 global_step=40550 loss=2.71290 loss_avg=3.63425 acc=0.75781 acc_top1_avg=0.65191 acc_top5_avg=0.89971 lr=0.00010 gn=39.05321 time=54.91it/s +epoch=103 global_step=40600 loss=3.23673 loss_avg=3.64589 acc=0.67969 acc_top1_avg=0.65078 acc_top5_avg=0.89877 lr=0.00010 gn=33.89887 time=62.76it/s +epoch=103 global_step=40650 loss=3.50964 loss_avg=3.64390 acc=0.66406 acc_top1_avg=0.65117 acc_top5_avg=0.89916 lr=0.00010 gn=28.31731 time=57.26it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.23415 test_loss_avg=0.95656 acc=0.92969 test_acc_avg=0.72476 test_acc_top5_avg=0.97897 time=231.61it/s +epoch=103 global_step=40664 loss=0.22587 test_loss_avg=1.65163 acc=0.92969 test_acc_avg=0.57974 test_acc_top5_avg=0.92572 time=235.37it/s +epoch=103 global_step=40664 loss=0.09125 test_loss_avg=1.34917 acc=0.93750 test_acc_avg=0.65467 test_acc_top5_avg=0.94007 time=801.05it/s +curr_acc 0.6547 +BEST_ACC 0.7056 +curr_acc_top5 0.9401 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=3.98302 loss_avg=3.61222 acc=0.60938 acc_top1_avg=0.65625 acc_top5_avg=0.89648 lr=0.00010 gn=29.05880 time=53.40it/s +epoch=104 global_step=40750 loss=3.33146 loss_avg=3.62422 acc=0.68750 acc_top1_avg=0.65389 acc_top5_avg=0.89771 lr=0.00010 gn=36.37956 time=62.04it/s +epoch=104 global_step=40800 loss=4.19541 loss_avg=3.61395 acc=0.59375 acc_top1_avg=0.65545 acc_top5_avg=0.89844 lr=0.00010 gn=38.60996 time=57.48it/s +epoch=104 global_step=40850 loss=3.17726 loss_avg=3.62027 acc=0.72656 acc_top1_avg=0.65411 acc_top5_avg=0.89848 lr=0.00010 gn=36.20146 time=61.45it/s +epoch=104 global_step=40900 loss=3.51804 loss_avg=3.61688 acc=0.65625 acc_top1_avg=0.65489 acc_top5_avg=0.89847 lr=0.00010 gn=29.59400 time=55.20it/s +epoch=104 global_step=40950 loss=3.22307 loss_avg=3.62844 acc=0.71094 acc_top1_avg=0.65325 acc_top5_avg=0.89887 lr=0.00010 gn=32.42805 time=43.09it/s +epoch=104 global_step=41000 loss=3.71006 loss_avg=3.62821 acc=0.66406 acc_top1_avg=0.65290 acc_top5_avg=0.89916 lr=0.00010 gn=42.48229 time=62.75it/s +epoch=104 global_step=41050 loss=3.29892 loss_avg=3.63420 acc=0.69531 acc_top1_avg=0.65242 acc_top5_avg=0.89842 lr=0.00010 gn=30.87026 time=54.49it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.27579 test_loss_avg=1.65884 acc=0.93750 test_acc_avg=0.58249 test_acc_top5_avg=0.89131 time=237.73it/s +epoch=104 global_step=41055 loss=0.08610 test_loss_avg=1.32941 acc=0.93750 test_acc_avg=0.66357 test_acc_top5_avg=0.94225 time=832.04it/s +curr_acc 0.6636 +BEST_ACC 0.7056 +curr_acc_top5 0.9422 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=3.53659 loss_avg=3.66982 acc=0.67188 acc_top1_avg=0.64809 acc_top5_avg=0.90174 lr=0.00010 gn=33.66591 time=53.33it/s +epoch=105 global_step=41150 loss=3.42386 loss_avg=3.61859 acc=0.66406 acc_top1_avg=0.65403 acc_top5_avg=0.90000 lr=0.00010 gn=30.04551 time=61.31it/s +epoch=105 global_step=41200 loss=3.44547 loss_avg=3.61544 acc=0.65625 acc_top1_avg=0.65415 acc_top5_avg=0.90070 lr=0.00010 gn=36.50952 time=60.18it/s +epoch=105 global_step=41250 loss=2.94723 loss_avg=3.61399 acc=0.72656 acc_top1_avg=0.65385 acc_top5_avg=0.90008 lr=0.00010 gn=33.63798 time=59.01it/s +epoch=105 global_step=41300 loss=2.94888 loss_avg=3.60459 acc=0.71875 acc_top1_avg=0.65482 acc_top5_avg=0.90064 lr=0.00010 gn=28.40840 time=52.51it/s +epoch=105 global_step=41350 loss=3.17506 loss_avg=3.60607 acc=0.69531 acc_top1_avg=0.65530 acc_top5_avg=0.90058 lr=0.00010 gn=26.97586 time=59.19it/s +epoch=105 global_step=41400 loss=3.66230 loss_avg=3.61999 acc=0.64844 acc_top1_avg=0.65401 acc_top5_avg=0.90070 lr=0.00010 gn=29.35464 time=58.25it/s +====================Eval==================== +epoch=105 global_step=41446 loss=1.07941 test_loss_avg=1.27167 acc=0.60938 test_acc_avg=0.61406 test_acc_top5_avg=0.97188 time=169.87it/s +epoch=105 global_step=41446 loss=2.96961 test_loss_avg=1.86723 acc=0.26562 test_acc_avg=0.53750 test_acc_top5_avg=0.91463 time=218.34it/s +epoch=105 global_step=41446 loss=0.13171 test_loss_avg=1.37469 acc=0.93750 test_acc_avg=0.65595 test_acc_top5_avg=0.93869 time=836.85it/s +curr_acc 0.6560 +BEST_ACC 0.7056 +curr_acc_top5 0.9387 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=3.32812 loss_avg=3.37427 acc=0.69531 acc_top1_avg=0.68359 acc_top5_avg=0.90625 lr=0.00010 gn=41.90260 time=49.16it/s +epoch=106 global_step=41500 loss=3.65368 loss_avg=3.59825 acc=0.64844 acc_top1_avg=0.65524 acc_top5_avg=0.89786 lr=0.00010 gn=32.68133 time=51.20it/s +epoch=106 global_step=41550 loss=3.62233 loss_avg=3.59511 acc=0.65625 acc_top1_avg=0.65663 acc_top5_avg=0.89956 lr=0.00010 gn=33.09479 time=45.42it/s +epoch=106 global_step=41600 loss=3.19810 loss_avg=3.60222 acc=0.68750 acc_top1_avg=0.65569 acc_top5_avg=0.89996 lr=0.00010 gn=32.17045 time=52.34it/s +epoch=106 global_step=41650 loss=2.83177 loss_avg=3.61052 acc=0.73438 acc_top1_avg=0.65518 acc_top5_avg=0.90062 lr=0.00010 gn=26.08219 time=55.58it/s +epoch=106 global_step=41700 loss=3.78019 loss_avg=3.62317 acc=0.60938 acc_top1_avg=0.65327 acc_top5_avg=0.90145 lr=0.00010 gn=30.11520 time=56.73it/s +epoch=106 global_step=41750 loss=4.14863 loss_avg=3.62044 acc=0.58594 acc_top1_avg=0.65376 acc_top5_avg=0.90070 lr=0.00010 gn=35.22212 time=52.43it/s +epoch=106 global_step=41800 loss=3.59234 loss_avg=3.62256 acc=0.67188 acc_top1_avg=0.65336 acc_top5_avg=0.90020 lr=0.00010 gn=40.93503 time=54.11it/s +====================Eval==================== +epoch=106 global_step=41837 loss=4.61621 test_loss_avg=1.32384 acc=0.00000 test_acc_avg=0.65294 test_acc_top5_avg=0.92969 time=238.88it/s +epoch=106 global_step=41837 loss=0.11591 test_loss_avg=1.43242 acc=0.95312 test_acc_avg=0.64237 test_acc_top5_avg=0.93668 time=243.36it/s +epoch=106 global_step=41837 loss=0.08746 test_loss_avg=1.38302 acc=0.93750 test_acc_avg=0.65378 test_acc_top5_avg=0.93898 time=510.01it/s +curr_acc 0.6538 +BEST_ACC 0.7056 +curr_acc_top5 0.9390 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=3.32796 loss_avg=3.51940 acc=0.68750 acc_top1_avg=0.66587 acc_top5_avg=0.89483 lr=0.00010 gn=35.06775 time=53.95it/s +epoch=107 global_step=41900 loss=3.85802 loss_avg=3.59309 acc=0.63281 acc_top1_avg=0.65489 acc_top5_avg=0.90228 lr=0.00010 gn=31.08708 time=52.81it/s +epoch=107 global_step=41950 loss=3.59865 loss_avg=3.58597 acc=0.67188 acc_top1_avg=0.65632 acc_top5_avg=0.90017 lr=0.00010 gn=38.34587 time=44.89it/s +epoch=107 global_step=42000 loss=3.97548 loss_avg=3.56114 acc=0.60938 acc_top1_avg=0.65869 acc_top5_avg=0.90059 lr=0.00010 gn=37.31058 time=53.57it/s +epoch=107 global_step=42050 loss=4.00315 loss_avg=3.56857 acc=0.60938 acc_top1_avg=0.65845 acc_top5_avg=0.90056 lr=0.00010 gn=33.98805 time=56.47it/s +epoch=107 global_step=42100 loss=3.52332 loss_avg=3.57404 acc=0.64844 acc_top1_avg=0.65782 acc_top5_avg=0.90043 lr=0.00010 gn=25.82202 time=53.63it/s +epoch=107 global_step=42150 loss=3.64121 loss_avg=3.59647 acc=0.63281 acc_top1_avg=0.65570 acc_top5_avg=0.89916 lr=0.00010 gn=32.56655 time=53.10it/s +epoch=107 global_step=42200 loss=3.31695 loss_avg=3.60790 acc=0.67969 acc_top1_avg=0.65444 acc_top5_avg=0.89938 lr=0.00010 gn=29.03403 time=55.42it/s +====================Eval==================== +epoch=107 global_step=42228 loss=1.47700 test_loss_avg=1.46256 acc=0.57031 test_acc_avg=0.62417 test_acc_top5_avg=0.90741 time=239.48it/s +epoch=107 global_step=42228 loss=0.15118 test_loss_avg=1.36759 acc=0.93750 test_acc_avg=0.65546 test_acc_top5_avg=0.93879 time=502.85it/s +curr_acc 0.6555 +BEST_ACC 0.7056 +curr_acc_top5 0.9388 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=3.78595 loss_avg=3.71460 acc=0.61719 acc_top1_avg=0.63885 acc_top5_avg=0.89347 lr=0.00010 gn=26.73773 time=54.57it/s +epoch=108 global_step=42300 loss=2.65277 loss_avg=3.62013 acc=0.75781 acc_top1_avg=0.65093 acc_top5_avg=0.89811 lr=0.00010 gn=37.51484 time=53.66it/s +epoch=108 global_step=42350 loss=3.71538 loss_avg=3.62055 acc=0.64062 acc_top1_avg=0.65196 acc_top5_avg=0.89761 lr=0.00010 gn=34.45390 time=59.09it/s +epoch=108 global_step=42400 loss=3.51365 loss_avg=3.61398 acc=0.65625 acc_top1_avg=0.65339 acc_top5_avg=0.89830 lr=0.00010 gn=34.80206 time=54.05it/s +epoch=108 global_step=42450 loss=4.33281 loss_avg=3.61514 acc=0.57031 acc_top1_avg=0.65386 acc_top5_avg=0.89791 lr=0.00010 gn=26.74415 time=54.38it/s +epoch=108 global_step=42500 loss=4.33129 loss_avg=3.63382 acc=0.57031 acc_top1_avg=0.65174 acc_top5_avg=0.89783 lr=0.00010 gn=37.39052 time=56.11it/s +epoch=108 global_step=42550 loss=3.77168 loss_avg=3.63109 acc=0.63281 acc_top1_avg=0.65213 acc_top5_avg=0.89832 lr=0.00010 gn=28.70163 time=45.19it/s +epoch=108 global_step=42600 loss=3.86509 loss_avg=3.61974 acc=0.62500 acc_top1_avg=0.65348 acc_top5_avg=0.89913 lr=0.00010 gn=35.91309 time=43.76it/s +====================Eval==================== +epoch=108 global_step=42619 loss=1.17401 test_loss_avg=0.90785 acc=0.66406 test_acc_avg=0.73785 test_acc_top5_avg=0.97656 time=229.98it/s +epoch=108 global_step=42619 loss=0.11998 test_loss_avg=1.54539 acc=0.96094 test_acc_avg=0.60685 test_acc_top5_avg=0.92831 time=247.69it/s +epoch=108 global_step=42619 loss=0.09833 test_loss_avg=1.35237 acc=0.93750 test_acc_avg=0.65427 test_acc_top5_avg=0.93790 time=697.89it/s +curr_acc 0.6543 +BEST_ACC 0.7056 +curr_acc_top5 0.9379 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=2.56899 loss_avg=3.56165 acc=0.77344 acc_top1_avg=0.66028 acc_top5_avg=0.90474 lr=0.00010 gn=29.60811 time=54.83it/s +epoch=109 global_step=42700 loss=3.62950 loss_avg=3.55794 acc=0.65625 acc_top1_avg=0.66107 acc_top5_avg=0.90538 lr=0.00010 gn=34.55250 time=52.37it/s +epoch=109 global_step=42750 loss=3.45735 loss_avg=3.56998 acc=0.66406 acc_top1_avg=0.65923 acc_top5_avg=0.90279 lr=0.00010 gn=33.47096 time=54.40it/s +epoch=109 global_step=42800 loss=3.22534 loss_avg=3.58791 acc=0.69531 acc_top1_avg=0.65720 acc_top5_avg=0.90167 lr=0.00010 gn=31.69352 time=57.19it/s +epoch=109 global_step=42850 loss=3.58750 loss_avg=3.60979 acc=0.67188 acc_top1_avg=0.65551 acc_top5_avg=0.90077 lr=0.00010 gn=31.86080 time=51.57it/s +epoch=109 global_step=42900 loss=3.73852 loss_avg=3.60559 acc=0.64062 acc_top1_avg=0.65606 acc_top5_avg=0.90080 lr=0.00010 gn=34.62425 time=49.29it/s +epoch=109 global_step=42950 loss=3.76718 loss_avg=3.60965 acc=0.62500 acc_top1_avg=0.65552 acc_top5_avg=0.90018 lr=0.00010 gn=32.05306 time=53.06it/s +epoch=109 global_step=43000 loss=3.36828 loss_avg=3.60530 acc=0.67969 acc_top1_avg=0.65580 acc_top5_avg=0.89981 lr=0.00010 gn=32.90836 time=59.94it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.34806 test_loss_avg=1.51884 acc=0.90625 test_acc_avg=0.61358 test_acc_top5_avg=0.90264 time=243.66it/s +epoch=109 global_step=43010 loss=0.10786 test_loss_avg=1.36226 acc=0.93750 test_acc_avg=0.65477 test_acc_top5_avg=0.94057 time=813.01it/s +curr_acc 0.6548 +BEST_ACC 0.7056 +curr_acc_top5 0.9406 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=3.88358 loss_avg=3.64272 acc=0.61719 acc_top1_avg=0.65020 acc_top5_avg=0.89414 lr=0.00010 gn=37.03481 time=55.06it/s +epoch=110 global_step=43100 loss=3.60056 loss_avg=3.57849 acc=0.65625 acc_top1_avg=0.65799 acc_top5_avg=0.89809 lr=0.00010 gn=30.32543 time=62.09it/s +epoch=110 global_step=43150 loss=3.48831 loss_avg=3.58528 acc=0.66406 acc_top1_avg=0.65714 acc_top5_avg=0.89939 lr=0.00010 gn=33.32835 time=53.96it/s +epoch=110 global_step=43200 loss=2.75038 loss_avg=3.57696 acc=0.74219 acc_top1_avg=0.65835 acc_top5_avg=0.89836 lr=0.00010 gn=27.52571 time=54.43it/s +epoch=110 global_step=43250 loss=3.78379 loss_avg=3.59452 acc=0.62500 acc_top1_avg=0.65648 acc_top5_avg=0.89915 lr=0.00010 gn=29.05318 time=57.10it/s +epoch=110 global_step=43300 loss=3.45780 loss_avg=3.59755 acc=0.65625 acc_top1_avg=0.65620 acc_top5_avg=0.89968 lr=0.00010 gn=26.44274 time=55.10it/s +epoch=110 global_step=43350 loss=3.47110 loss_avg=3.59370 acc=0.66406 acc_top1_avg=0.65676 acc_top5_avg=0.89931 lr=0.00010 gn=32.89110 time=53.05it/s +epoch=110 global_step=43400 loss=3.74362 loss_avg=3.59867 acc=0.63281 acc_top1_avg=0.65607 acc_top5_avg=0.89956 lr=0.00010 gn=29.08701 time=52.68it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.31698 test_loss_avg=1.04487 acc=0.91406 test_acc_avg=0.69453 test_acc_top5_avg=0.97578 time=246.20it/s +epoch=110 global_step=43401 loss=0.36827 test_loss_avg=1.70919 acc=0.89844 test_acc_avg=0.57044 test_acc_top5_avg=0.92318 time=228.98it/s +epoch=110 global_step=43401 loss=0.08642 test_loss_avg=1.34773 acc=0.93750 test_acc_avg=0.65922 test_acc_top5_avg=0.94047 time=839.70it/s +curr_acc 0.6592 +BEST_ACC 0.7056 +curr_acc_top5 0.9405 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=3.18364 loss_avg=3.62576 acc=0.71875 acc_top1_avg=0.65354 acc_top5_avg=0.89955 lr=0.00010 gn=36.73607 time=49.88it/s +epoch=111 global_step=43500 loss=3.63403 loss_avg=3.60598 acc=0.65625 acc_top1_avg=0.65609 acc_top5_avg=0.89796 lr=0.00010 gn=41.67567 time=51.76it/s +epoch=111 global_step=43550 loss=3.36288 loss_avg=3.58321 acc=0.67969 acc_top1_avg=0.65809 acc_top5_avg=0.89797 lr=0.00010 gn=36.77692 time=55.35it/s +epoch=111 global_step=43600 loss=3.68574 loss_avg=3.58575 acc=0.65625 acc_top1_avg=0.65825 acc_top5_avg=0.89777 lr=0.00010 gn=35.13436 time=53.88it/s +epoch=111 global_step=43650 loss=3.44375 loss_avg=3.59069 acc=0.66406 acc_top1_avg=0.65838 acc_top5_avg=0.89891 lr=0.00010 gn=31.44477 time=55.42it/s +epoch=111 global_step=43700 loss=3.73840 loss_avg=3.58480 acc=0.64062 acc_top1_avg=0.65860 acc_top5_avg=0.89933 lr=0.00010 gn=29.32240 time=54.82it/s +epoch=111 global_step=43750 loss=3.48560 loss_avg=3.59488 acc=0.67969 acc_top1_avg=0.65723 acc_top5_avg=0.89850 lr=0.00010 gn=28.45222 time=59.18it/s +====================Eval==================== +epoch=111 global_step=43792 loss=4.41157 test_loss_avg=1.77240 acc=0.00000 test_acc_avg=0.55192 test_acc_top5_avg=0.87424 time=237.13it/s +epoch=111 global_step=43792 loss=0.08531 test_loss_avg=1.33580 acc=0.93750 test_acc_avg=0.65932 test_acc_top5_avg=0.93691 time=849.91it/s +curr_acc 0.6593 +BEST_ACC 0.7056 +curr_acc_top5 0.9369 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=3.04307 loss_avg=3.43838 acc=0.71875 acc_top1_avg=0.67285 acc_top5_avg=0.90820 lr=0.00010 gn=35.93944 time=54.25it/s +epoch=112 global_step=43850 loss=3.61292 loss_avg=3.51862 acc=0.64844 acc_top1_avg=0.66298 acc_top5_avg=0.90423 lr=0.00010 gn=29.04298 time=51.03it/s +epoch=112 global_step=43900 loss=4.01904 loss_avg=3.53533 acc=0.60156 acc_top1_avg=0.66175 acc_top5_avg=0.90285 lr=0.00010 gn=38.58074 time=60.61it/s +epoch=112 global_step=43950 loss=4.23189 loss_avg=3.55902 acc=0.58594 acc_top1_avg=0.66055 acc_top5_avg=0.90249 lr=0.00010 gn=35.38814 time=59.44it/s +epoch=112 global_step=44000 loss=3.40826 loss_avg=3.57654 acc=0.66406 acc_top1_avg=0.65895 acc_top5_avg=0.90148 lr=0.00010 gn=30.99237 time=58.48it/s +epoch=112 global_step=44050 loss=3.50897 loss_avg=3.59365 acc=0.65625 acc_top1_avg=0.65692 acc_top5_avg=0.90016 lr=0.00010 gn=38.29341 time=55.70it/s +epoch=112 global_step=44100 loss=3.20717 loss_avg=3.58985 acc=0.70312 acc_top1_avg=0.65701 acc_top5_avg=0.89988 lr=0.00010 gn=33.80811 time=54.69it/s +epoch=112 global_step=44150 loss=3.28342 loss_avg=3.59360 acc=0.67969 acc_top1_avg=0.65656 acc_top5_avg=0.90066 lr=0.00010 gn=34.75863 time=63.41it/s +====================Eval==================== +epoch=112 global_step=44183 loss=1.60839 test_loss_avg=1.53740 acc=0.56250 test_acc_avg=0.57422 test_acc_top5_avg=0.96094 time=233.24it/s +epoch=112 global_step=44183 loss=4.29840 test_loss_avg=1.76567 acc=0.00000 test_acc_avg=0.55183 test_acc_top5_avg=0.91226 time=230.37it/s +epoch=112 global_step=44183 loss=0.12378 test_loss_avg=1.37835 acc=0.93750 test_acc_avg=0.64814 test_acc_top5_avg=0.93879 time=546.28it/s +curr_acc 0.6481 +BEST_ACC 0.7056 +curr_acc_top5 0.9388 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=3.84471 loss_avg=3.57593 acc=0.62500 acc_top1_avg=0.65993 acc_top5_avg=0.90441 lr=0.00010 gn=32.62175 time=47.48it/s +epoch=113 global_step=44250 loss=3.57209 loss_avg=3.59905 acc=0.65625 acc_top1_avg=0.65613 acc_top5_avg=0.90205 lr=0.00010 gn=38.59258 time=58.09it/s +epoch=113 global_step=44300 loss=4.14675 loss_avg=3.57786 acc=0.60938 acc_top1_avg=0.65799 acc_top5_avg=0.90338 lr=0.00010 gn=35.62329 time=53.26it/s +epoch=113 global_step=44350 loss=3.53450 loss_avg=3.60265 acc=0.67188 acc_top1_avg=0.65527 acc_top5_avg=0.90167 lr=0.00010 gn=32.95916 time=57.86it/s +epoch=113 global_step=44400 loss=3.68674 loss_avg=3.58942 acc=0.63281 acc_top1_avg=0.65715 acc_top5_avg=0.90035 lr=0.00010 gn=29.55508 time=52.64it/s +epoch=113 global_step=44450 loss=3.50590 loss_avg=3.57841 acc=0.67969 acc_top1_avg=0.65830 acc_top5_avg=0.90028 lr=0.00010 gn=38.92548 time=53.78it/s +epoch=113 global_step=44500 loss=3.73956 loss_avg=3.59178 acc=0.63281 acc_top1_avg=0.65719 acc_top5_avg=0.89878 lr=0.00010 gn=36.57784 time=53.54it/s +epoch=113 global_step=44550 loss=3.22273 loss_avg=3.59386 acc=0.71094 acc_top1_avg=0.65693 acc_top5_avg=0.89867 lr=0.00010 gn=37.89058 time=62.81it/s +====================Eval==================== +epoch=113 global_step=44574 loss=0.99768 test_loss_avg=0.99072 acc=0.69531 test_acc_avg=0.71807 test_acc_top5_avg=0.97011 time=234.65it/s +epoch=113 global_step=44574 loss=0.23175 test_loss_avg=1.49036 acc=0.92969 test_acc_avg=0.62158 test_acc_top5_avg=0.93022 time=242.68it/s +epoch=113 global_step=44574 loss=0.10212 test_loss_avg=1.38766 acc=0.93750 test_acc_avg=0.64666 test_acc_top5_avg=0.93523 time=497.49it/s +curr_acc 0.6467 +BEST_ACC 0.7056 +curr_acc_top5 0.9352 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=3.37413 loss_avg=3.61200 acc=0.70312 acc_top1_avg=0.65325 acc_top5_avg=0.89754 lr=0.00010 gn=34.74160 time=61.65it/s +epoch=114 global_step=44650 loss=3.64136 loss_avg=3.60072 acc=0.64844 acc_top1_avg=0.65604 acc_top5_avg=0.89638 lr=0.00010 gn=26.57047 time=58.90it/s +epoch=114 global_step=44700 loss=3.57845 loss_avg=3.56611 acc=0.65625 acc_top1_avg=0.66003 acc_top5_avg=0.89937 lr=0.00010 gn=31.09000 time=54.89it/s +epoch=114 global_step=44750 loss=3.08733 loss_avg=3.58785 acc=0.70312 acc_top1_avg=0.65723 acc_top5_avg=0.89893 lr=0.00010 gn=35.52617 time=56.44it/s +epoch=114 global_step=44800 loss=3.81825 loss_avg=3.58431 acc=0.63281 acc_top1_avg=0.65732 acc_top5_avg=0.89861 lr=0.00010 gn=36.24862 time=52.84it/s +epoch=114 global_step=44850 loss=4.05163 loss_avg=3.58967 acc=0.58594 acc_top1_avg=0.65701 acc_top5_avg=0.89815 lr=0.00010 gn=39.16057 time=51.97it/s +epoch=114 global_step=44900 loss=3.81843 loss_avg=3.58301 acc=0.63281 acc_top1_avg=0.65762 acc_top5_avg=0.89870 lr=0.00010 gn=32.17930 time=53.82it/s +epoch=114 global_step=44950 loss=3.99686 loss_avg=3.58266 acc=0.60938 acc_top1_avg=0.65789 acc_top5_avg=0.89840 lr=0.00010 gn=38.59248 time=57.61it/s +====================Eval==================== +epoch=114 global_step=44965 loss=1.07521 test_loss_avg=1.47359 acc=0.69531 test_acc_avg=0.61985 test_acc_top5_avg=0.90110 time=210.36it/s +epoch=114 global_step=44965 loss=0.10862 test_loss_avg=1.35959 acc=0.93750 test_acc_avg=0.65496 test_acc_top5_avg=0.93641 time=755.19it/s +curr_acc 0.6550 +BEST_ACC 0.7056 +curr_acc_top5 0.9364 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=3.07515 loss_avg=3.78752 acc=0.72656 acc_top1_avg=0.63661 acc_top5_avg=0.89531 lr=0.00010 gn=37.27404 time=55.17it/s +epoch=115 global_step=45050 loss=4.25968 loss_avg=3.67730 acc=0.60156 acc_top1_avg=0.64835 acc_top5_avg=0.90037 lr=0.00010 gn=39.47516 time=60.19it/s +epoch=115 global_step=45100 loss=3.83880 loss_avg=3.63570 acc=0.63281 acc_top1_avg=0.65249 acc_top5_avg=0.90069 lr=0.00010 gn=37.73216 time=57.90it/s +epoch=115 global_step=45150 loss=3.43367 loss_avg=3.62121 acc=0.67188 acc_top1_avg=0.65397 acc_top5_avg=0.90046 lr=0.00010 gn=34.20441 time=51.68it/s +epoch=115 global_step=45200 loss=4.16620 loss_avg=3.59711 acc=0.60156 acc_top1_avg=0.65662 acc_top5_avg=0.89927 lr=0.00010 gn=29.88575 time=54.94it/s +epoch=115 global_step=45250 loss=3.63384 loss_avg=3.61497 acc=0.63281 acc_top1_avg=0.65474 acc_top5_avg=0.89879 lr=0.00010 gn=30.62042 time=49.19it/s +epoch=115 global_step=45300 loss=3.76352 loss_avg=3.59085 acc=0.63281 acc_top1_avg=0.65746 acc_top5_avg=0.89974 lr=0.00010 gn=37.79619 time=58.49it/s +epoch=115 global_step=45350 loss=3.44064 loss_avg=3.58697 acc=0.67188 acc_top1_avg=0.65791 acc_top5_avg=0.89996 lr=0.00010 gn=29.90145 time=63.37it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.28576 test_loss_avg=0.85270 acc=0.92188 test_acc_avg=0.75417 test_acc_top5_avg=0.98177 time=234.03it/s +epoch=115 global_step=45356 loss=0.14869 test_loss_avg=1.62590 acc=0.94531 test_acc_avg=0.58918 test_acc_top5_avg=0.92620 time=236.57it/s +epoch=115 global_step=45356 loss=0.07766 test_loss_avg=1.36663 acc=0.93750 test_acc_avg=0.65249 test_acc_top5_avg=0.93898 time=736.75it/s +curr_acc 0.6525 +BEST_ACC 0.7056 +curr_acc_top5 0.9390 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=4.04177 loss_avg=3.59504 acc=0.59375 acc_top1_avg=0.65536 acc_top5_avg=0.91033 lr=0.00010 gn=33.36663 time=36.96it/s +epoch=116 global_step=45450 loss=3.96362 loss_avg=3.60986 acc=0.61719 acc_top1_avg=0.65409 acc_top5_avg=0.90342 lr=0.00010 gn=32.25657 time=56.43it/s +epoch=116 global_step=45500 loss=3.11857 loss_avg=3.59184 acc=0.71094 acc_top1_avg=0.65625 acc_top5_avg=0.90088 lr=0.00010 gn=31.28241 time=57.78it/s +epoch=116 global_step=45550 loss=4.01700 loss_avg=3.57630 acc=0.60938 acc_top1_avg=0.65806 acc_top5_avg=0.90122 lr=0.00010 gn=35.70766 time=55.06it/s +epoch=116 global_step=45600 loss=4.05394 loss_avg=3.57208 acc=0.58594 acc_top1_avg=0.65884 acc_top5_avg=0.89956 lr=0.00010 gn=34.16154 time=54.36it/s +epoch=116 global_step=45650 loss=3.78601 loss_avg=3.56945 acc=0.64844 acc_top1_avg=0.65962 acc_top5_avg=0.90024 lr=0.00010 gn=45.86797 time=63.14it/s +epoch=116 global_step=45700 loss=3.73549 loss_avg=3.57458 acc=0.64844 acc_top1_avg=0.65900 acc_top5_avg=0.89996 lr=0.00010 gn=44.58909 time=60.66it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.34925 test_loss_avg=1.65339 acc=0.89062 test_acc_avg=0.58333 test_acc_top5_avg=0.89301 time=212.48it/s +epoch=116 global_step=45747 loss=0.09614 test_loss_avg=1.36726 acc=0.93750 test_acc_avg=0.65348 test_acc_top5_avg=0.93948 time=851.12it/s +curr_acc 0.6535 +BEST_ACC 0.7056 +curr_acc_top5 0.9395 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=3.54773 loss_avg=3.45630 acc=0.64844 acc_top1_avg=0.67188 acc_top5_avg=0.90104 lr=0.00010 gn=33.85197 time=36.95it/s +epoch=117 global_step=45800 loss=3.76642 loss_avg=3.54767 acc=0.63281 acc_top1_avg=0.66303 acc_top5_avg=0.89755 lr=0.00010 gn=28.84255 time=54.05it/s +epoch=117 global_step=45850 loss=3.97522 loss_avg=3.56252 acc=0.60938 acc_top1_avg=0.66141 acc_top5_avg=0.89942 lr=0.00010 gn=33.18753 time=53.61it/s +epoch=117 global_step=45900 loss=3.53245 loss_avg=3.55711 acc=0.67188 acc_top1_avg=0.66176 acc_top5_avg=0.90017 lr=0.00010 gn=38.23279 time=54.65it/s +epoch=117 global_step=45950 loss=3.83683 loss_avg=3.57102 acc=0.63281 acc_top1_avg=0.65979 acc_top5_avg=0.90052 lr=0.00010 gn=42.36369 time=60.48it/s +epoch=117 global_step=46000 loss=3.81433 loss_avg=3.56466 acc=0.64062 acc_top1_avg=0.66017 acc_top5_avg=0.90004 lr=0.00010 gn=37.65592 time=57.60it/s +epoch=117 global_step=46050 loss=4.19302 loss_avg=3.56566 acc=0.61719 acc_top1_avg=0.65996 acc_top5_avg=0.89916 lr=0.00010 gn=36.70820 time=59.05it/s +epoch=117 global_step=46100 loss=4.15481 loss_avg=3.57084 acc=0.59375 acc_top1_avg=0.65946 acc_top5_avg=0.89921 lr=0.00010 gn=26.75277 time=55.49it/s +====================Eval==================== +epoch=117 global_step=46138 loss=1.51909 test_loss_avg=1.51862 acc=0.57031 test_acc_avg=0.57031 test_acc_top5_avg=0.96987 time=219.02it/s +epoch=117 global_step=46138 loss=0.32772 test_loss_avg=1.82680 acc=0.89062 test_acc_avg=0.54263 test_acc_top5_avg=0.91680 time=248.26it/s +epoch=117 global_step=46138 loss=0.10883 test_loss_avg=1.38031 acc=0.93750 test_acc_avg=0.65071 test_acc_top5_avg=0.93859 time=570.03it/s +curr_acc 0.6507 +BEST_ACC 0.7056 +curr_acc_top5 0.9386 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=3.95598 loss_avg=3.73636 acc=0.60156 acc_top1_avg=0.63997 acc_top5_avg=0.89648 lr=0.00010 gn=32.46780 time=59.62it/s +epoch=118 global_step=46200 loss=3.42249 loss_avg=3.65508 acc=0.67969 acc_top1_avg=0.65108 acc_top5_avg=0.89680 lr=0.00010 gn=35.22091 time=61.33it/s +epoch=118 global_step=46250 loss=3.53129 loss_avg=3.61907 acc=0.65625 acc_top1_avg=0.65527 acc_top5_avg=0.89823 lr=0.00010 gn=36.51244 time=58.51it/s +epoch=118 global_step=46300 loss=3.21152 loss_avg=3.59218 acc=0.69531 acc_top1_avg=0.65803 acc_top5_avg=0.89829 lr=0.00010 gn=28.87332 time=54.18it/s +epoch=118 global_step=46350 loss=3.25157 loss_avg=3.56598 acc=0.68750 acc_top1_avg=0.65997 acc_top5_avg=0.90043 lr=0.00010 gn=32.82455 time=50.42it/s +epoch=118 global_step=46400 loss=2.90445 loss_avg=3.56152 acc=0.74219 acc_top1_avg=0.66069 acc_top5_avg=0.89984 lr=0.00010 gn=40.12906 time=60.59it/s +epoch=118 global_step=46450 loss=3.96337 loss_avg=3.55852 acc=0.62500 acc_top1_avg=0.66101 acc_top5_avg=0.89984 lr=0.00010 gn=34.27802 time=56.80it/s +epoch=118 global_step=46500 loss=3.63584 loss_avg=3.56429 acc=0.64844 acc_top1_avg=0.66048 acc_top5_avg=0.89941 lr=0.00010 gn=35.95408 time=58.20it/s +====================Eval==================== +epoch=118 global_step=46529 loss=4.22506 test_loss_avg=1.57259 acc=0.00000 test_acc_avg=0.59152 test_acc_top5_avg=0.90848 time=240.98it/s +epoch=118 global_step=46529 loss=0.11320 test_loss_avg=1.40081 acc=0.96094 test_acc_avg=0.64403 test_acc_top5_avg=0.93820 time=248.23it/s +epoch=118 global_step=46529 loss=0.07978 test_loss_avg=1.38409 acc=0.93750 test_acc_avg=0.64775 test_acc_top5_avg=0.93898 time=557.01it/s +curr_acc 0.6477 +BEST_ACC 0.7056 +curr_acc_top5 0.9390 +BEST_ACC_top5 0.9609 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=3.45005 loss_avg=3.61287 acc=0.68750 acc_top1_avg=0.65551 acc_top5_avg=0.90327 lr=0.00010 gn=34.55029 time=64.33it/s +epoch=119 global_step=46600 loss=3.24670 loss_avg=3.53090 acc=0.71094 acc_top1_avg=0.66549 acc_top5_avg=0.90031 lr=0.00010 gn=41.92497 time=63.45it/s +epoch=119 global_step=46650 loss=3.07420 loss_avg=3.52377 acc=0.71094 acc_top1_avg=0.66561 acc_top5_avg=0.90309 lr=0.00010 gn=36.38361 time=62.42it/s +epoch=119 global_step=46700 loss=3.34729 loss_avg=3.54046 acc=0.68750 acc_top1_avg=0.66415 acc_top5_avg=0.90196 lr=0.00010 gn=35.98449 time=56.10it/s +epoch=119 global_step=46750 loss=3.20928 loss_avg=3.54890 acc=0.69531 acc_top1_avg=0.66336 acc_top5_avg=0.90119 lr=0.00010 gn=30.22024 time=54.72it/s +epoch=119 global_step=46800 loss=3.75447 loss_avg=3.56430 acc=0.64062 acc_top1_avg=0.66092 acc_top5_avg=0.90118 lr=0.00010 gn=35.99051 time=56.70it/s +epoch=119 global_step=46850 loss=4.12547 loss_avg=3.55006 acc=0.59375 acc_top1_avg=0.66243 acc_top5_avg=0.90058 lr=0.00010 gn=34.78335 time=57.41it/s +epoch=119 global_step=46900 loss=3.45657 loss_avg=3.55515 acc=0.66406 acc_top1_avg=0.66170 acc_top5_avg=0.90065 lr=0.00010 gn=26.05131 time=63.12it/s +====================Eval==================== +epoch=119 global_step=46920 loss=4.44573 test_loss_avg=1.60065 acc=0.00000 test_acc_avg=0.58658 test_acc_top5_avg=0.90753 time=251.59it/s +epoch=119 global_step=46920 loss=0.16403 test_loss_avg=1.36443 acc=0.93750 test_acc_avg=0.64883 test_acc_top5_avg=0.93750 time=867.67it/s +curr_acc 0.6488 +BEST_ACC 0.7056 +curr_acc_top5 0.9375 +BEST_ACC_top5 0.9609 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_4_6.log b/other_methods/sceloss/sceloss_results/out_4_6.log new file mode 100644 index 0000000..ae398e5 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_4_6.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.4__noise_amount__0.6.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.53269 loss_avg=7.73217 acc=0.24219 acc_top1_avg=0.21672 acc_top5_avg=0.66953 lr=0.01000 gn=5.82474 time=52.40it/s +epoch=0 global_step=100 loss=7.42368 loss_avg=7.57091 acc=0.25000 acc_top1_avg=0.23172 acc_top5_avg=0.69445 lr=0.01000 gn=5.22619 time=55.80it/s +epoch=0 global_step=150 loss=7.34277 loss_avg=7.49761 acc=0.27344 acc_top1_avg=0.23969 acc_top5_avg=0.70219 lr=0.01000 gn=5.33192 time=55.96it/s +epoch=0 global_step=200 loss=7.26329 loss_avg=7.43905 acc=0.25000 acc_top1_avg=0.24477 acc_top5_avg=0.70770 lr=0.01000 gn=3.92980 time=59.25it/s +epoch=0 global_step=250 loss=7.12443 loss_avg=7.40513 acc=0.28906 acc_top1_avg=0.24838 acc_top5_avg=0.71341 lr=0.01000 gn=3.21938 time=59.40it/s +epoch=0 global_step=300 loss=6.59655 loss_avg=7.35619 acc=0.32812 acc_top1_avg=0.25286 acc_top5_avg=0.72031 lr=0.01000 gn=4.12071 time=62.66it/s +epoch=0 global_step=350 loss=6.80107 loss_avg=7.31772 acc=0.32031 acc_top1_avg=0.25636 acc_top5_avg=0.72527 lr=0.01000 gn=3.88071 time=63.64it/s +====================Eval==================== +epoch=0 global_step=391 loss=5.06739 test_loss_avg=4.60728 acc=0.00000 test_acc_avg=0.08406 test_acc_top5_avg=0.59547 time=242.21it/s +epoch=0 global_step=391 loss=6.77021 test_loss_avg=4.27085 acc=0.00000 test_acc_avg=0.18374 test_acc_top5_avg=0.63894 time=35.95it/s +curr_acc 0.1837 +BEST_ACC 0.0000 +curr_acc_top5 0.6389 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=7.07929 loss_avg=6.98320 acc=0.26562 acc_top1_avg=0.28993 acc_top5_avg=0.74566 lr=0.01000 gn=3.21237 time=55.39it/s +epoch=1 global_step=450 loss=7.05588 loss_avg=7.00083 acc=0.28906 acc_top1_avg=0.28694 acc_top5_avg=0.75847 lr=0.01000 gn=3.60064 time=62.90it/s +epoch=1 global_step=500 loss=7.70979 loss_avg=6.99785 acc=0.21875 acc_top1_avg=0.28892 acc_top5_avg=0.76018 lr=0.01000 gn=3.08229 time=56.40it/s +epoch=1 global_step=550 loss=6.65804 loss_avg=6.98529 acc=0.31250 acc_top1_avg=0.29118 acc_top5_avg=0.76307 lr=0.01000 gn=3.71291 time=53.71it/s +epoch=1 global_step=600 loss=7.14993 loss_avg=6.97430 acc=0.27344 acc_top1_avg=0.29261 acc_top5_avg=0.76252 lr=0.01000 gn=3.85641 time=59.65it/s +epoch=1 global_step=650 loss=6.69701 loss_avg=6.95387 acc=0.31250 acc_top1_avg=0.29528 acc_top5_avg=0.76403 lr=0.01000 gn=3.60702 time=57.76it/s +epoch=1 global_step=700 loss=7.12722 loss_avg=6.94313 acc=0.28906 acc_top1_avg=0.29662 acc_top5_avg=0.76413 lr=0.01000 gn=2.94816 time=56.92it/s +epoch=1 global_step=750 loss=6.78199 loss_avg=6.93304 acc=0.32031 acc_top1_avg=0.29768 acc_top5_avg=0.76465 lr=0.01000 gn=3.34389 time=63.48it/s +====================Eval==================== +epoch=1 global_step=782 loss=5.04935 test_loss_avg=4.38974 acc=0.02344 test_acc_avg=0.07440 test_acc_top5_avg=0.66704 time=221.86it/s +epoch=1 global_step=782 loss=5.06605 test_loss_avg=3.23986 acc=0.25781 test_acc_avg=0.24373 test_acc_top5_avg=0.75825 time=247.85it/s +epoch=1 global_step=782 loss=6.61220 test_loss_avg=3.57434 acc=0.00000 test_acc_avg=0.21905 test_acc_top5_avg=0.71796 time=827.61it/s +curr_acc 0.2190 +BEST_ACC 0.1837 +curr_acc_top5 0.7180 +BEST_ACC_top5 0.6389 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=6.73487 loss_avg=6.72742 acc=0.34375 acc_top1_avg=0.32422 acc_top5_avg=0.78602 lr=0.01000 gn=4.93975 time=64.06it/s +epoch=2 global_step=850 loss=7.09361 loss_avg=6.83492 acc=0.28125 acc_top1_avg=0.30974 acc_top5_avg=0.77953 lr=0.01000 gn=3.46868 time=58.43it/s +epoch=2 global_step=900 loss=6.32044 loss_avg=6.84267 acc=0.38281 acc_top1_avg=0.30667 acc_top5_avg=0.78383 lr=0.01000 gn=3.62734 time=60.92it/s +epoch=2 global_step=950 loss=6.50375 loss_avg=6.80837 acc=0.36719 acc_top1_avg=0.30994 acc_top5_avg=0.78283 lr=0.01000 gn=3.41765 time=55.34it/s +epoch=2 global_step=1000 loss=6.41717 loss_avg=6.78678 acc=0.33594 acc_top1_avg=0.31228 acc_top5_avg=0.78315 lr=0.01000 gn=3.05396 time=63.01it/s +epoch=2 global_step=1050 loss=6.47566 loss_avg=6.77507 acc=0.33594 acc_top1_avg=0.31375 acc_top5_avg=0.78277 lr=0.01000 gn=3.61244 time=61.75it/s +epoch=2 global_step=1100 loss=6.77926 loss_avg=6.77007 acc=0.31250 acc_top1_avg=0.31390 acc_top5_avg=0.78270 lr=0.01000 gn=4.13896 time=51.18it/s +epoch=2 global_step=1150 loss=6.56506 loss_avg=6.77041 acc=0.32031 acc_top1_avg=0.31329 acc_top5_avg=0.78271 lr=0.01000 gn=3.86525 time=65.12it/s +====================Eval==================== +epoch=2 global_step=1173 loss=1.79924 test_loss_avg=5.14813 acc=0.43750 test_acc_avg=0.04985 test_acc_top5_avg=0.58501 time=221.45it/s +epoch=2 global_step=1173 loss=7.43566 test_loss_avg=4.31504 acc=0.00000 test_acc_avg=0.21499 test_acc_top5_avg=0.71618 time=633.10it/s +curr_acc 0.2150 +BEST_ACC 0.2190 +curr_acc_top5 0.7162 +BEST_ACC_top5 0.7180 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=6.72797 loss_avg=6.72312 acc=0.33594 acc_top1_avg=0.31655 acc_top5_avg=0.79369 lr=0.01000 gn=4.10015 time=63.96it/s +epoch=3 global_step=1250 loss=6.71639 loss_avg=6.67622 acc=0.31250 acc_top1_avg=0.32244 acc_top5_avg=0.78987 lr=0.01000 gn=3.66661 time=54.95it/s +epoch=3 global_step=1300 loss=6.64065 loss_avg=6.64556 acc=0.29688 acc_top1_avg=0.32542 acc_top5_avg=0.79281 lr=0.01000 gn=3.13461 time=61.65it/s +epoch=3 global_step=1350 loss=6.74869 loss_avg=6.66490 acc=0.30469 acc_top1_avg=0.32265 acc_top5_avg=0.79105 lr=0.01000 gn=2.74244 time=48.46it/s +epoch=3 global_step=1400 loss=7.09698 loss_avg=6.65763 acc=0.28125 acc_top1_avg=0.32341 acc_top5_avg=0.79226 lr=0.01000 gn=3.87797 time=60.63it/s +epoch=3 global_step=1450 loss=6.70344 loss_avg=6.66522 acc=0.32031 acc_top1_avg=0.32234 acc_top5_avg=0.79253 lr=0.01000 gn=3.96573 time=58.79it/s +epoch=3 global_step=1500 loss=6.94477 loss_avg=6.66317 acc=0.28125 acc_top1_avg=0.32253 acc_top5_avg=0.79324 lr=0.01000 gn=3.17223 time=56.60it/s +epoch=3 global_step=1550 loss=6.82835 loss_avg=6.65749 acc=0.29688 acc_top1_avg=0.32336 acc_top5_avg=0.79317 lr=0.01000 gn=3.18866 time=62.05it/s +====================Eval==================== +epoch=3 global_step=1564 loss=2.07061 test_loss_avg=4.13076 acc=0.32812 test_acc_avg=0.14483 test_acc_top5_avg=0.72476 time=253.46it/s +epoch=3 global_step=1564 loss=0.62650 test_loss_avg=3.43505 acc=0.85156 test_acc_avg=0.23239 test_acc_top5_avg=0.72631 time=246.74it/s +epoch=3 global_step=1564 loss=5.81262 test_loss_avg=3.51829 acc=0.00000 test_acc_avg=0.26147 test_acc_top5_avg=0.73299 time=695.57it/s +curr_acc 0.2615 +BEST_ACC 0.2190 +curr_acc_top5 0.7330 +BEST_ACC_top5 0.7180 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=6.38527 loss_avg=6.64020 acc=0.37500 acc_top1_avg=0.32769 acc_top5_avg=0.79579 lr=0.01000 gn=3.90090 time=52.74it/s +epoch=4 global_step=1650 loss=5.54094 loss_avg=6.60440 acc=0.48438 acc_top1_avg=0.33021 acc_top5_avg=0.79987 lr=0.01000 gn=4.88003 time=56.71it/s +epoch=4 global_step=1700 loss=6.72771 loss_avg=6.60448 acc=0.32812 acc_top1_avg=0.33048 acc_top5_avg=0.79860 lr=0.01000 gn=4.62358 time=57.55it/s +epoch=4 global_step=1750 loss=6.15449 loss_avg=6.59653 acc=0.36719 acc_top1_avg=0.33119 acc_top5_avg=0.79822 lr=0.01000 gn=3.71928 time=51.42it/s +epoch=4 global_step=1800 loss=6.19261 loss_avg=6.58876 acc=0.36719 acc_top1_avg=0.33213 acc_top5_avg=0.79740 lr=0.01000 gn=3.82553 time=64.56it/s +epoch=4 global_step=1850 loss=6.37888 loss_avg=6.58403 acc=0.36719 acc_top1_avg=0.33211 acc_top5_avg=0.79854 lr=0.01000 gn=4.11152 time=57.13it/s +epoch=4 global_step=1900 loss=6.18924 loss_avg=6.57511 acc=0.38281 acc_top1_avg=0.33340 acc_top5_avg=0.79771 lr=0.01000 gn=3.53743 time=54.29it/s +epoch=4 global_step=1950 loss=6.95196 loss_avg=6.56925 acc=0.29688 acc_top1_avg=0.33430 acc_top5_avg=0.79777 lr=0.01000 gn=4.02738 time=57.90it/s +====================Eval==================== +epoch=4 global_step=1955 loss=5.28240 test_loss_avg=4.12056 acc=0.00000 test_acc_avg=0.18107 test_acc_top5_avg=0.62638 time=240.96it/s +epoch=4 global_step=1955 loss=7.85855 test_loss_avg=3.81643 acc=0.00000 test_acc_avg=0.27284 test_acc_top5_avg=0.73625 time=581.81it/s +curr_acc 0.2728 +BEST_ACC 0.2615 +curr_acc_top5 0.7363 +BEST_ACC_top5 0.7330 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=6.37114 loss_avg=6.50433 acc=0.32812 acc_top1_avg=0.34323 acc_top5_avg=0.80955 lr=0.01000 gn=4.31519 time=53.14it/s +epoch=5 global_step=2050 loss=6.85819 loss_avg=6.51231 acc=0.28906 acc_top1_avg=0.33972 acc_top5_avg=0.80699 lr=0.01000 gn=4.23951 time=37.48it/s +epoch=5 global_step=2100 loss=6.93772 loss_avg=6.51471 acc=0.28906 acc_top1_avg=0.33998 acc_top5_avg=0.80474 lr=0.01000 gn=3.56409 time=62.64it/s +epoch=5 global_step=2150 loss=6.91929 loss_avg=6.50992 acc=0.29688 acc_top1_avg=0.34071 acc_top5_avg=0.80361 lr=0.01000 gn=3.49710 time=52.86it/s +epoch=5 global_step=2200 loss=6.29204 loss_avg=6.50704 acc=0.34375 acc_top1_avg=0.34082 acc_top5_avg=0.80281 lr=0.01000 gn=4.13479 time=62.92it/s +epoch=5 global_step=2250 loss=5.94343 loss_avg=6.50167 acc=0.40625 acc_top1_avg=0.34166 acc_top5_avg=0.80220 lr=0.01000 gn=4.00424 time=58.84it/s +epoch=5 global_step=2300 loss=6.67969 loss_avg=6.50904 acc=0.34375 acc_top1_avg=0.34072 acc_top5_avg=0.80322 lr=0.01000 gn=2.68627 time=45.21it/s +====================Eval==================== +epoch=5 global_step=2346 loss=5.65652 test_loss_avg=5.90550 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.45469 time=221.28it/s +epoch=5 global_step=2346 loss=4.68282 test_loss_avg=4.72144 acc=0.21094 test_acc_avg=0.13693 test_acc_top5_avg=0.67401 time=239.29it/s +epoch=5 global_step=2346 loss=5.88623 test_loss_avg=4.17027 acc=0.00000 test_acc_avg=0.24506 test_acc_top5_avg=0.70936 time=847.51it/s +curr_acc 0.2451 +BEST_ACC 0.2728 +curr_acc_top5 0.7094 +BEST_ACC_top5 0.7363 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=6.55633 loss_avg=6.36001 acc=0.33594 acc_top1_avg=0.35938 acc_top5_avg=0.81836 lr=0.01000 gn=3.66827 time=53.12it/s +epoch=6 global_step=2400 loss=6.38255 loss_avg=6.40056 acc=0.35938 acc_top1_avg=0.35475 acc_top5_avg=0.80830 lr=0.01000 gn=3.45454 time=58.29it/s +epoch=6 global_step=2450 loss=6.56032 loss_avg=6.43683 acc=0.34375 acc_top1_avg=0.35066 acc_top5_avg=0.81040 lr=0.01000 gn=3.86796 time=54.41it/s +epoch=6 global_step=2500 loss=6.93998 loss_avg=6.44802 acc=0.27344 acc_top1_avg=0.34842 acc_top5_avg=0.80930 lr=0.01000 gn=3.06380 time=31.70it/s +epoch=6 global_step=2550 loss=6.36180 loss_avg=6.44812 acc=0.33594 acc_top1_avg=0.34819 acc_top5_avg=0.80691 lr=0.01000 gn=4.71279 time=54.22it/s +epoch=6 global_step=2600 loss=7.02820 loss_avg=6.44848 acc=0.28125 acc_top1_avg=0.34818 acc_top5_avg=0.80721 lr=0.01000 gn=4.17849 time=64.72it/s +epoch=6 global_step=2650 loss=6.50331 loss_avg=6.45651 acc=0.33594 acc_top1_avg=0.34696 acc_top5_avg=0.80667 lr=0.01000 gn=4.54422 time=57.58it/s +epoch=6 global_step=2700 loss=6.50550 loss_avg=6.45820 acc=0.36719 acc_top1_avg=0.34684 acc_top5_avg=0.80687 lr=0.01000 gn=3.84224 time=58.19it/s +====================Eval==================== +epoch=6 global_step=2737 loss=3.04155 test_loss_avg=4.63971 acc=0.11719 test_acc_avg=0.06941 test_acc_top5_avg=0.57933 time=238.90it/s +epoch=6 global_step=2737 loss=9.15178 test_loss_avg=3.90355 acc=0.00000 test_acc_avg=0.22625 test_acc_top5_avg=0.70826 time=245.12it/s +epoch=6 global_step=2737 loss=8.44983 test_loss_avg=4.09617 acc=0.00000 test_acc_avg=0.21766 test_acc_top5_avg=0.71321 time=515.84it/s +curr_acc 0.2177 +BEST_ACC 0.2728 +curr_acc_top5 0.7132 +BEST_ACC_top5 0.7363 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=6.52886 loss_avg=6.46630 acc=0.30469 acc_top1_avg=0.34495 acc_top5_avg=0.80769 lr=0.01000 gn=3.68393 time=56.03it/s +epoch=7 global_step=2800 loss=6.61506 loss_avg=6.49343 acc=0.32031 acc_top1_avg=0.34375 acc_top5_avg=0.80407 lr=0.01000 gn=3.94486 time=58.62it/s +epoch=7 global_step=2850 loss=7.03851 loss_avg=6.48824 acc=0.27344 acc_top1_avg=0.34451 acc_top5_avg=0.80455 lr=0.01000 gn=3.15906 time=51.61it/s +epoch=7 global_step=2900 loss=5.93693 loss_avg=6.48825 acc=0.42969 acc_top1_avg=0.34341 acc_top5_avg=0.80521 lr=0.01000 gn=4.68142 time=54.15it/s +epoch=7 global_step=2950 loss=6.30718 loss_avg=6.49442 acc=0.35156 acc_top1_avg=0.34243 acc_top5_avg=0.80542 lr=0.01000 gn=3.56076 time=61.18it/s +epoch=7 global_step=3000 loss=5.70381 loss_avg=6.47223 acc=0.43750 acc_top1_avg=0.34518 acc_top5_avg=0.80579 lr=0.01000 gn=3.54794 time=56.91it/s +epoch=7 global_step=3050 loss=6.70048 loss_avg=6.45804 acc=0.32031 acc_top1_avg=0.34647 acc_top5_avg=0.80706 lr=0.01000 gn=3.95397 time=55.18it/s +epoch=7 global_step=3100 loss=6.51189 loss_avg=6.46279 acc=0.34375 acc_top1_avg=0.34601 acc_top5_avg=0.80746 lr=0.01000 gn=3.21500 time=62.04it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.34525 test_loss_avg=3.66087 acc=0.59375 test_acc_avg=0.24651 test_acc_top5_avg=0.66606 time=241.89it/s +epoch=7 global_step=3128 loss=7.79873 test_loss_avg=3.63105 acc=0.00000 test_acc_avg=0.29688 test_acc_top5_avg=0.76345 time=850.94it/s +curr_acc 0.2969 +BEST_ACC 0.2728 +curr_acc_top5 0.7634 +BEST_ACC_top5 0.7363 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=6.23643 loss_avg=6.35967 acc=0.38281 acc_top1_avg=0.35724 acc_top5_avg=0.80824 lr=0.01000 gn=3.25703 time=55.51it/s +epoch=8 global_step=3200 loss=6.10917 loss_avg=6.44550 acc=0.39844 acc_top1_avg=0.34711 acc_top5_avg=0.81087 lr=0.01000 gn=3.33547 time=47.22it/s +epoch=8 global_step=3250 loss=6.35057 loss_avg=6.43795 acc=0.35156 acc_top1_avg=0.34913 acc_top5_avg=0.81096 lr=0.01000 gn=3.32160 time=55.68it/s +epoch=8 global_step=3300 loss=6.37872 loss_avg=6.42782 acc=0.37500 acc_top1_avg=0.35047 acc_top5_avg=0.81064 lr=0.01000 gn=3.63272 time=54.54it/s +epoch=8 global_step=3350 loss=6.30361 loss_avg=6.41680 acc=0.37500 acc_top1_avg=0.35096 acc_top5_avg=0.81074 lr=0.01000 gn=4.42016 time=64.58it/s +epoch=8 global_step=3400 loss=6.69253 loss_avg=6.41974 acc=0.29688 acc_top1_avg=0.34995 acc_top5_avg=0.80911 lr=0.01000 gn=4.66138 time=65.21it/s +epoch=8 global_step=3450 loss=6.16198 loss_avg=6.42255 acc=0.39062 acc_top1_avg=0.34918 acc_top5_avg=0.81056 lr=0.01000 gn=4.47468 time=51.99it/s +epoch=8 global_step=3500 loss=6.79335 loss_avg=6.42154 acc=0.32812 acc_top1_avg=0.34938 acc_top5_avg=0.81185 lr=0.01000 gn=4.70900 time=57.91it/s +====================Eval==================== +epoch=8 global_step=3519 loss=5.02272 test_loss_avg=3.27335 acc=0.01562 test_acc_avg=0.32639 test_acc_top5_avg=0.77431 time=246.51it/s +epoch=8 global_step=3519 loss=0.67639 test_loss_avg=3.19014 acc=0.82031 test_acc_avg=0.31882 test_acc_top5_avg=0.74586 time=251.62it/s +epoch=8 global_step=3519 loss=6.45508 test_loss_avg=3.51195 acc=0.00000 test_acc_avg=0.30073 test_acc_top5_avg=0.74970 time=558.87it/s +curr_acc 0.3007 +BEST_ACC 0.2969 +curr_acc_top5 0.7497 +BEST_ACC_top5 0.7634 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=6.43491 loss_avg=6.36738 acc=0.35156 acc_top1_avg=0.35635 acc_top5_avg=0.81578 lr=0.01000 gn=4.47424 time=49.61it/s +epoch=9 global_step=3600 loss=6.46514 loss_avg=6.44124 acc=0.34375 acc_top1_avg=0.34587 acc_top5_avg=0.80990 lr=0.01000 gn=4.24128 time=65.79it/s +epoch=9 global_step=3650 loss=5.97878 loss_avg=6.43824 acc=0.39844 acc_top1_avg=0.34679 acc_top5_avg=0.80862 lr=0.01000 gn=4.26782 time=41.26it/s +epoch=9 global_step=3700 loss=6.47910 loss_avg=6.43328 acc=0.33594 acc_top1_avg=0.34612 acc_top5_avg=0.81159 lr=0.01000 gn=3.81262 time=57.45it/s +epoch=9 global_step=3750 loss=6.61759 loss_avg=6.43584 acc=0.35156 acc_top1_avg=0.34669 acc_top5_avg=0.81267 lr=0.01000 gn=4.67065 time=57.64it/s +epoch=9 global_step=3800 loss=6.69326 loss_avg=6.43333 acc=0.30469 acc_top1_avg=0.34703 acc_top5_avg=0.81192 lr=0.01000 gn=4.41210 time=55.48it/s +epoch=9 global_step=3850 loss=6.47813 loss_avg=6.43101 acc=0.33594 acc_top1_avg=0.34781 acc_top5_avg=0.81316 lr=0.01000 gn=4.66767 time=41.86it/s +epoch=9 global_step=3900 loss=6.49650 loss_avg=6.42410 acc=0.32812 acc_top1_avg=0.34840 acc_top5_avg=0.81277 lr=0.01000 gn=4.79586 time=57.29it/s +====================Eval==================== +epoch=9 global_step=3910 loss=4.54730 test_loss_avg=4.89102 acc=0.02344 test_acc_avg=0.12921 test_acc_top5_avg=0.64423 time=244.67it/s +epoch=9 global_step=3910 loss=5.91050 test_loss_avg=4.08873 acc=0.00000 test_acc_avg=0.24041 test_acc_top5_avg=0.72696 time=834.52it/s +curr_acc 0.2404 +BEST_ACC 0.3007 +curr_acc_top5 0.7270 +BEST_ACC_top5 0.7634 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=6.58047 loss_avg=6.36609 acc=0.32812 acc_top1_avg=0.35762 acc_top5_avg=0.81738 lr=0.01000 gn=4.26017 time=52.26it/s +epoch=10 global_step=4000 loss=6.04858 loss_avg=6.37758 acc=0.38281 acc_top1_avg=0.35582 acc_top5_avg=0.81745 lr=0.01000 gn=4.55211 time=50.84it/s +epoch=10 global_step=4050 loss=5.96463 loss_avg=6.39962 acc=0.40625 acc_top1_avg=0.35324 acc_top5_avg=0.81546 lr=0.01000 gn=4.35699 time=50.17it/s +epoch=10 global_step=4100 loss=6.30606 loss_avg=6.40435 acc=0.36719 acc_top1_avg=0.35358 acc_top5_avg=0.81287 lr=0.01000 gn=5.87580 time=53.63it/s +epoch=10 global_step=4150 loss=6.56146 loss_avg=6.41704 acc=0.32812 acc_top1_avg=0.35156 acc_top5_avg=0.81296 lr=0.01000 gn=3.92906 time=64.40it/s +epoch=10 global_step=4200 loss=7.20778 loss_avg=6.41583 acc=0.25781 acc_top1_avg=0.35119 acc_top5_avg=0.81272 lr=0.01000 gn=3.97850 time=57.81it/s +epoch=10 global_step=4250 loss=6.83118 loss_avg=6.41609 acc=0.31250 acc_top1_avg=0.35099 acc_top5_avg=0.81183 lr=0.01000 gn=3.83750 time=63.36it/s +epoch=10 global_step=4300 loss=6.09891 loss_avg=6.41152 acc=0.39844 acc_top1_avg=0.35174 acc_top5_avg=0.81272 lr=0.01000 gn=3.38934 time=47.50it/s +====================Eval==================== +epoch=10 global_step=4301 loss=1.91722 test_loss_avg=4.43922 acc=0.58594 test_acc_avg=0.12656 test_acc_top5_avg=0.49766 time=236.61it/s +epoch=10 global_step=4301 loss=0.46187 test_loss_avg=3.55360 acc=0.83594 test_acc_avg=0.22500 test_acc_top5_avg=0.67292 time=185.44it/s +epoch=10 global_step=4301 loss=8.76390 test_loss_avg=3.70024 acc=0.00000 test_acc_avg=0.28501 test_acc_top5_avg=0.73042 time=804.43it/s +curr_acc 0.2850 +BEST_ACC 0.3007 +curr_acc_top5 0.7304 +BEST_ACC_top5 0.7634 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=7.04689 loss_avg=6.40302 acc=0.27344 acc_top1_avg=0.34965 acc_top5_avg=0.81441 lr=0.01000 gn=5.28884 time=49.75it/s +epoch=11 global_step=4400 loss=6.08256 loss_avg=6.38736 acc=0.35938 acc_top1_avg=0.35125 acc_top5_avg=0.81226 lr=0.01000 gn=4.47504 time=58.36it/s +epoch=11 global_step=4450 loss=6.38355 loss_avg=6.37733 acc=0.35938 acc_top1_avg=0.35314 acc_top5_avg=0.81292 lr=0.01000 gn=5.09429 time=54.45it/s +epoch=11 global_step=4500 loss=6.39428 loss_avg=6.41326 acc=0.34375 acc_top1_avg=0.34905 acc_top5_avg=0.81254 lr=0.01000 gn=3.22105 time=56.68it/s +epoch=11 global_step=4550 loss=6.94640 loss_avg=6.40866 acc=0.32031 acc_top1_avg=0.35003 acc_top5_avg=0.81241 lr=0.01000 gn=4.46393 time=59.79it/s +epoch=11 global_step=4600 loss=6.24530 loss_avg=6.40506 acc=0.36719 acc_top1_avg=0.35060 acc_top5_avg=0.81326 lr=0.01000 gn=4.40478 time=56.92it/s +epoch=11 global_step=4650 loss=6.40328 loss_avg=6.39469 acc=0.34375 acc_top1_avg=0.35129 acc_top5_avg=0.81313 lr=0.01000 gn=4.54430 time=59.20it/s +====================Eval==================== +epoch=11 global_step=4692 loss=3.82181 test_loss_avg=3.91550 acc=0.09375 test_acc_avg=0.12878 test_acc_top5_avg=0.65549 time=227.73it/s +epoch=11 global_step=4692 loss=6.23274 test_loss_avg=3.52432 acc=0.00000 test_acc_avg=0.27027 test_acc_top5_avg=0.72112 time=495.20it/s +curr_acc 0.2703 +BEST_ACC 0.3007 +curr_acc_top5 0.7211 +BEST_ACC_top5 0.7634 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=6.27524 loss_avg=6.34732 acc=0.37500 acc_top1_avg=0.36230 acc_top5_avg=0.81348 lr=0.01000 gn=4.40148 time=62.82it/s +epoch=12 global_step=4750 loss=6.09557 loss_avg=6.32674 acc=0.39844 acc_top1_avg=0.35978 acc_top5_avg=0.81950 lr=0.01000 gn=4.04372 time=58.58it/s +epoch=12 global_step=4800 loss=6.00633 loss_avg=6.33640 acc=0.39062 acc_top1_avg=0.35670 acc_top5_avg=0.82038 lr=0.01000 gn=4.73429 time=54.40it/s +epoch=12 global_step=4850 loss=6.35447 loss_avg=6.34222 acc=0.32812 acc_top1_avg=0.35616 acc_top5_avg=0.81972 lr=0.01000 gn=4.71140 time=61.63it/s +epoch=12 global_step=4900 loss=6.52629 loss_avg=6.34114 acc=0.32812 acc_top1_avg=0.35799 acc_top5_avg=0.82024 lr=0.01000 gn=4.99103 time=59.08it/s +epoch=12 global_step=4950 loss=6.33433 loss_avg=6.34835 acc=0.35938 acc_top1_avg=0.35771 acc_top5_avg=0.82031 lr=0.01000 gn=5.40065 time=56.67it/s +epoch=12 global_step=5000 loss=6.70580 loss_avg=6.36336 acc=0.32031 acc_top1_avg=0.35648 acc_top5_avg=0.81945 lr=0.01000 gn=5.12021 time=53.83it/s +epoch=12 global_step=5050 loss=6.74806 loss_avg=6.37452 acc=0.30469 acc_top1_avg=0.35495 acc_top5_avg=0.81909 lr=0.01000 gn=4.23205 time=62.38it/s +====================Eval==================== +epoch=12 global_step=5083 loss=5.25911 test_loss_avg=5.16321 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.66016 time=239.99it/s +epoch=12 global_step=5083 loss=4.95920 test_loss_avg=3.69035 acc=0.00000 test_acc_avg=0.18585 test_acc_top5_avg=0.73708 time=233.93it/s +epoch=12 global_step=5083 loss=8.35692 test_loss_avg=3.67154 acc=0.00000 test_acc_avg=0.28521 test_acc_top5_avg=0.79648 time=830.72it/s +curr_acc 0.2852 +BEST_ACC 0.3007 +curr_acc_top5 0.7965 +BEST_ACC_top5 0.7634 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=6.63325 loss_avg=6.39922 acc=0.33594 acc_top1_avg=0.35524 acc_top5_avg=0.80101 lr=0.01000 gn=4.53686 time=56.74it/s +epoch=13 global_step=5150 loss=6.12389 loss_avg=6.35993 acc=0.38281 acc_top1_avg=0.35926 acc_top5_avg=0.80574 lr=0.01000 gn=4.33748 time=58.24it/s +epoch=13 global_step=5200 loss=6.31331 loss_avg=6.35869 acc=0.38281 acc_top1_avg=0.35917 acc_top5_avg=0.81157 lr=0.01000 gn=5.42491 time=52.13it/s +epoch=13 global_step=5250 loss=6.76203 loss_avg=6.36360 acc=0.31250 acc_top1_avg=0.35811 acc_top5_avg=0.81259 lr=0.01000 gn=4.54839 time=60.32it/s +epoch=13 global_step=5300 loss=6.13287 loss_avg=6.36344 acc=0.39062 acc_top1_avg=0.35765 acc_top5_avg=0.81462 lr=0.01000 gn=5.42736 time=63.63it/s +epoch=13 global_step=5350 loss=6.66702 loss_avg=6.37022 acc=0.32031 acc_top1_avg=0.35686 acc_top5_avg=0.81519 lr=0.01000 gn=4.21427 time=61.69it/s +epoch=13 global_step=5400 loss=6.36732 loss_avg=6.38319 acc=0.35938 acc_top1_avg=0.35536 acc_top5_avg=0.81467 lr=0.01000 gn=5.00061 time=53.61it/s +epoch=13 global_step=5450 loss=5.96721 loss_avg=6.38212 acc=0.41406 acc_top1_avg=0.35539 acc_top5_avg=0.81444 lr=0.01000 gn=4.85981 time=59.78it/s +====================Eval==================== +epoch=13 global_step=5474 loss=4.16643 test_loss_avg=3.50548 acc=0.08594 test_acc_avg=0.26087 test_acc_top5_avg=0.67052 time=245.19it/s +epoch=13 global_step=5474 loss=9.38137 test_loss_avg=3.30844 acc=0.00000 test_acc_avg=0.30554 test_acc_top5_avg=0.75995 time=235.86it/s +epoch=13 global_step=5474 loss=8.78140 test_loss_avg=3.76062 acc=0.00000 test_acc_avg=0.28234 test_acc_top5_avg=0.76800 time=482.77it/s +curr_acc 0.2823 +BEST_ACC 0.3007 +curr_acc_top5 0.7680 +BEST_ACC_top5 0.7965 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=6.56889 loss_avg=6.31809 acc=0.35156 acc_top1_avg=0.36508 acc_top5_avg=0.82602 lr=0.01000 gn=4.98147 time=57.97it/s +epoch=14 global_step=5550 loss=6.14666 loss_avg=6.40068 acc=0.38281 acc_top1_avg=0.35495 acc_top5_avg=0.81713 lr=0.01000 gn=4.26428 time=61.55it/s +epoch=14 global_step=5600 loss=6.25194 loss_avg=6.35268 acc=0.35938 acc_top1_avg=0.35969 acc_top5_avg=0.81808 lr=0.01000 gn=4.97290 time=54.43it/s +epoch=14 global_step=5650 loss=6.72295 loss_avg=6.38954 acc=0.30469 acc_top1_avg=0.35525 acc_top5_avg=0.81836 lr=0.01000 gn=4.35779 time=58.48it/s +epoch=14 global_step=5700 loss=5.72993 loss_avg=6.37454 acc=0.43750 acc_top1_avg=0.35626 acc_top5_avg=0.81890 lr=0.01000 gn=4.76978 time=57.38it/s +epoch=14 global_step=5750 loss=6.54503 loss_avg=6.38168 acc=0.34375 acc_top1_avg=0.35519 acc_top5_avg=0.81808 lr=0.01000 gn=4.21115 time=61.26it/s +epoch=14 global_step=5800 loss=6.34288 loss_avg=6.37891 acc=0.32812 acc_top1_avg=0.35528 acc_top5_avg=0.81787 lr=0.01000 gn=4.15587 time=54.65it/s +epoch=14 global_step=5850 loss=6.57128 loss_avg=6.37959 acc=0.34375 acc_top1_avg=0.35532 acc_top5_avg=0.81757 lr=0.01000 gn=3.53845 time=57.39it/s +====================Eval==================== +epoch=14 global_step=5865 loss=1.18184 test_loss_avg=4.15630 acc=0.62500 test_acc_avg=0.22053 test_acc_top5_avg=0.63778 time=228.78it/s +epoch=14 global_step=5865 loss=6.99825 test_loss_avg=3.94626 acc=0.00000 test_acc_avg=0.29668 test_acc_top5_avg=0.75564 time=836.69it/s +curr_acc 0.2967 +BEST_ACC 0.3007 +curr_acc_top5 0.7556 +BEST_ACC_top5 0.7965 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=6.69506 loss_avg=6.21265 acc=0.32031 acc_top1_avg=0.37612 acc_top5_avg=0.83170 lr=0.01000 gn=4.18341 time=54.19it/s +epoch=15 global_step=5950 loss=6.64247 loss_avg=6.32222 acc=0.34375 acc_top1_avg=0.36029 acc_top5_avg=0.82243 lr=0.01000 gn=5.14891 time=56.99it/s +epoch=15 global_step=6000 loss=6.47930 loss_avg=6.34185 acc=0.35938 acc_top1_avg=0.35897 acc_top5_avg=0.82043 lr=0.01000 gn=5.14922 time=58.62it/s +epoch=15 global_step=6050 loss=6.63626 loss_avg=6.36362 acc=0.31250 acc_top1_avg=0.35621 acc_top5_avg=0.81862 lr=0.01000 gn=4.71429 time=57.51it/s +epoch=15 global_step=6100 loss=6.16990 loss_avg=6.35069 acc=0.37500 acc_top1_avg=0.35781 acc_top5_avg=0.81805 lr=0.01000 gn=4.57442 time=63.03it/s +epoch=15 global_step=6150 loss=5.94040 loss_avg=6.35766 acc=0.42188 acc_top1_avg=0.35666 acc_top5_avg=0.81705 lr=0.01000 gn=4.83472 time=62.99it/s +epoch=15 global_step=6200 loss=6.12355 loss_avg=6.35610 acc=0.39062 acc_top1_avg=0.35686 acc_top5_avg=0.81691 lr=0.01000 gn=4.04830 time=64.79it/s +epoch=15 global_step=6250 loss=5.93088 loss_avg=6.36608 acc=0.40625 acc_top1_avg=0.35605 acc_top5_avg=0.81727 lr=0.01000 gn=4.59742 time=45.66it/s +====================Eval==================== +epoch=15 global_step=6256 loss=2.57263 test_loss_avg=4.41584 acc=0.39062 test_acc_avg=0.21667 test_acc_top5_avg=0.80781 time=120.62it/s +epoch=15 global_step=6256 loss=0.08498 test_loss_avg=3.50938 acc=0.98438 test_acc_avg=0.26731 test_acc_top5_avg=0.79435 time=239.85it/s +epoch=15 global_step=6256 loss=8.23640 test_loss_avg=3.78587 acc=0.00000 test_acc_avg=0.28570 test_acc_top5_avg=0.79984 time=600.39it/s +curr_acc 0.2857 +BEST_ACC 0.3007 +curr_acc_top5 0.7998 +BEST_ACC_top5 0.7965 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=6.10646 loss_avg=6.32770 acc=0.38281 acc_top1_avg=0.36275 acc_top5_avg=0.81836 lr=0.01000 gn=4.27323 time=57.58it/s +epoch=16 global_step=6350 loss=6.41944 loss_avg=6.31101 acc=0.35156 acc_top1_avg=0.36386 acc_top5_avg=0.81765 lr=0.01000 gn=4.54438 time=57.47it/s +epoch=16 global_step=6400 loss=6.41417 loss_avg=6.32635 acc=0.36719 acc_top1_avg=0.36192 acc_top5_avg=0.81966 lr=0.01000 gn=4.42695 time=51.84it/s +epoch=16 global_step=6450 loss=6.22080 loss_avg=6.33666 acc=0.36719 acc_top1_avg=0.36143 acc_top5_avg=0.81802 lr=0.01000 gn=4.30126 time=55.68it/s +epoch=16 global_step=6500 loss=6.01021 loss_avg=6.34334 acc=0.41406 acc_top1_avg=0.36043 acc_top5_avg=0.81749 lr=0.01000 gn=5.07023 time=52.26it/s +epoch=16 global_step=6550 loss=6.07132 loss_avg=6.35181 acc=0.38281 acc_top1_avg=0.35935 acc_top5_avg=0.81670 lr=0.01000 gn=3.30762 time=57.69it/s +epoch=16 global_step=6600 loss=6.15710 loss_avg=6.35632 acc=0.36719 acc_top1_avg=0.35838 acc_top5_avg=0.81682 lr=0.01000 gn=3.91484 time=61.00it/s +====================Eval==================== +epoch=16 global_step=6647 loss=5.65135 test_loss_avg=4.41477 acc=0.00000 test_acc_avg=0.15408 test_acc_top5_avg=0.66992 time=244.52it/s +epoch=16 global_step=6647 loss=7.05403 test_loss_avg=3.90334 acc=0.00000 test_acc_avg=0.29173 test_acc_top5_avg=0.78857 time=499.02it/s +curr_acc 0.2917 +BEST_ACC 0.3007 +curr_acc_top5 0.7886 +BEST_ACC_top5 0.7998 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=6.54136 loss_avg=6.50392 acc=0.35938 acc_top1_avg=0.34635 acc_top5_avg=0.82552 lr=0.01000 gn=3.61585 time=54.40it/s +epoch=17 global_step=6700 loss=6.31105 loss_avg=6.32831 acc=0.35938 acc_top1_avg=0.36026 acc_top5_avg=0.82149 lr=0.01000 gn=4.85173 time=57.28it/s +epoch=17 global_step=6750 loss=6.65928 loss_avg=6.37032 acc=0.33594 acc_top1_avg=0.35558 acc_top5_avg=0.81933 lr=0.01000 gn=4.05725 time=57.26it/s +epoch=17 global_step=6800 loss=6.10899 loss_avg=6.35567 acc=0.39844 acc_top1_avg=0.35662 acc_top5_avg=0.81873 lr=0.01000 gn=5.35002 time=32.28it/s +epoch=17 global_step=6850 loss=6.89701 loss_avg=6.35839 acc=0.29688 acc_top1_avg=0.35618 acc_top5_avg=0.82043 lr=0.01000 gn=4.67541 time=62.79it/s +epoch=17 global_step=6900 loss=6.26630 loss_avg=6.35774 acc=0.35156 acc_top1_avg=0.35626 acc_top5_avg=0.82081 lr=0.01000 gn=4.03157 time=57.09it/s +epoch=17 global_step=6950 loss=6.37462 loss_avg=6.36198 acc=0.35938 acc_top1_avg=0.35543 acc_top5_avg=0.81982 lr=0.01000 gn=4.48295 time=57.07it/s +epoch=17 global_step=7000 loss=6.74819 loss_avg=6.35773 acc=0.28906 acc_top1_avg=0.35566 acc_top5_avg=0.81943 lr=0.01000 gn=3.14616 time=55.52it/s +====================Eval==================== +epoch=17 global_step=7038 loss=5.22783 test_loss_avg=5.19130 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.50335 time=238.15it/s +epoch=17 global_step=7038 loss=1.34820 test_loss_avg=4.30474 acc=0.68750 test_acc_avg=0.20861 test_acc_top5_avg=0.69271 time=242.78it/s +epoch=17 global_step=7038 loss=5.22984 test_loss_avg=3.85608 acc=0.00000 test_acc_avg=0.28204 test_acc_top5_avg=0.67751 time=504.79it/s +curr_acc 0.2820 +BEST_ACC 0.3007 +curr_acc_top5 0.6775 +BEST_ACC_top5 0.7998 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=6.67467 loss_avg=6.52175 acc=0.32031 acc_top1_avg=0.33984 acc_top5_avg=0.79883 lr=0.01000 gn=4.00827 time=64.93it/s +epoch=18 global_step=7100 loss=6.99847 loss_avg=6.36779 acc=0.28906 acc_top1_avg=0.35459 acc_top5_avg=0.81603 lr=0.01000 gn=3.82893 time=59.58it/s +epoch=18 global_step=7150 loss=5.76885 loss_avg=6.34603 acc=0.42969 acc_top1_avg=0.35861 acc_top5_avg=0.81906 lr=0.01000 gn=3.49584 time=57.63it/s +epoch=18 global_step=7200 loss=6.10727 loss_avg=6.34872 acc=0.38281 acc_top1_avg=0.35812 acc_top5_avg=0.81766 lr=0.01000 gn=4.00472 time=45.88it/s +epoch=18 global_step=7250 loss=5.89188 loss_avg=6.31150 acc=0.41406 acc_top1_avg=0.36236 acc_top5_avg=0.81950 lr=0.01000 gn=5.27149 time=34.95it/s +epoch=18 global_step=7300 loss=5.28785 loss_avg=6.31492 acc=0.49219 acc_top1_avg=0.36188 acc_top5_avg=0.82124 lr=0.01000 gn=5.58762 time=56.33it/s +epoch=18 global_step=7350 loss=6.00740 loss_avg=6.32182 acc=0.39062 acc_top1_avg=0.36103 acc_top5_avg=0.82089 lr=0.01000 gn=6.32853 time=55.67it/s +epoch=18 global_step=7400 loss=6.45572 loss_avg=6.34052 acc=0.35156 acc_top1_avg=0.35879 acc_top5_avg=0.81971 lr=0.01000 gn=4.28864 time=56.42it/s +====================Eval==================== +epoch=18 global_step=7429 loss=3.68240 test_loss_avg=3.86747 acc=0.05469 test_acc_avg=0.23884 test_acc_top5_avg=0.68834 time=242.12it/s +epoch=18 global_step=7429 loss=5.93190 test_loss_avg=3.27730 acc=0.00000 test_acc_avg=0.32903 test_acc_top5_avg=0.74609 time=245.47it/s +epoch=18 global_step=7429 loss=5.65481 test_loss_avg=3.30739 acc=0.00000 test_acc_avg=0.32486 test_acc_top5_avg=0.74140 time=489.65it/s +curr_acc 0.3249 +BEST_ACC 0.3007 +curr_acc_top5 0.7414 +BEST_ACC_top5 0.7998 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=6.85278 loss_avg=6.30979 acc=0.29688 acc_top1_avg=0.36384 acc_top5_avg=0.81994 lr=0.01000 gn=3.16720 time=61.65it/s +epoch=19 global_step=7500 loss=5.42373 loss_avg=6.32610 acc=0.47656 acc_top1_avg=0.36147 acc_top5_avg=0.81745 lr=0.01000 gn=4.59679 time=61.83it/s +epoch=19 global_step=7550 loss=6.66857 loss_avg=6.32117 acc=0.32031 acc_top1_avg=0.36241 acc_top5_avg=0.81721 lr=0.01000 gn=4.07688 time=55.96it/s +epoch=19 global_step=7600 loss=6.43598 loss_avg=6.35289 acc=0.35156 acc_top1_avg=0.35910 acc_top5_avg=0.81725 lr=0.01000 gn=4.77096 time=57.54it/s +epoch=19 global_step=7650 loss=6.03808 loss_avg=6.33032 acc=0.41406 acc_top1_avg=0.36157 acc_top5_avg=0.81847 lr=0.01000 gn=4.59726 time=63.44it/s +epoch=19 global_step=7700 loss=6.15173 loss_avg=6.34528 acc=0.36719 acc_top1_avg=0.35926 acc_top5_avg=0.81821 lr=0.01000 gn=3.64122 time=56.31it/s +epoch=19 global_step=7750 loss=6.26361 loss_avg=6.33946 acc=0.34375 acc_top1_avg=0.35959 acc_top5_avg=0.81931 lr=0.01000 gn=4.16912 time=56.18it/s +epoch=19 global_step=7800 loss=5.78282 loss_avg=6.34918 acc=0.42188 acc_top1_avg=0.35807 acc_top5_avg=0.81854 lr=0.01000 gn=4.59694 time=57.67it/s +====================Eval==================== +epoch=19 global_step=7820 loss=4.07829 test_loss_avg=3.62619 acc=0.00000 test_acc_avg=0.23246 test_acc_top5_avg=0.82254 time=241.54it/s +epoch=19 global_step=7820 loss=5.33859 test_loss_avg=3.34573 acc=0.00000 test_acc_avg=0.28985 test_acc_top5_avg=0.80429 time=434.42it/s +curr_acc 0.2899 +BEST_ACC 0.3249 +curr_acc_top5 0.8043 +BEST_ACC_top5 0.7998 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=6.26859 loss_avg=6.24719 acc=0.35938 acc_top1_avg=0.36875 acc_top5_avg=0.81745 lr=0.01000 gn=5.08291 time=62.48it/s +epoch=20 global_step=7900 loss=5.82038 loss_avg=6.31007 acc=0.41406 acc_top1_avg=0.36250 acc_top5_avg=0.81855 lr=0.01000 gn=3.83774 time=63.29it/s +epoch=20 global_step=7950 loss=6.76177 loss_avg=6.30146 acc=0.30469 acc_top1_avg=0.36388 acc_top5_avg=0.81947 lr=0.01000 gn=5.57403 time=61.62it/s +epoch=20 global_step=8000 loss=6.13453 loss_avg=6.29761 acc=0.37500 acc_top1_avg=0.36389 acc_top5_avg=0.82122 lr=0.01000 gn=3.82425 time=56.85it/s +epoch=20 global_step=8050 loss=6.32273 loss_avg=6.30974 acc=0.37500 acc_top1_avg=0.36304 acc_top5_avg=0.82086 lr=0.01000 gn=4.16409 time=53.65it/s +epoch=20 global_step=8100 loss=6.32445 loss_avg=6.32430 acc=0.32031 acc_top1_avg=0.36124 acc_top5_avg=0.82070 lr=0.01000 gn=4.81579 time=60.22it/s +epoch=20 global_step=8150 loss=5.93151 loss_avg=6.31648 acc=0.40625 acc_top1_avg=0.36158 acc_top5_avg=0.82069 lr=0.01000 gn=4.66740 time=56.60it/s +epoch=20 global_step=8200 loss=5.88236 loss_avg=6.33256 acc=0.39844 acc_top1_avg=0.35991 acc_top5_avg=0.82046 lr=0.01000 gn=4.06572 time=57.65it/s +====================Eval==================== +epoch=20 global_step=8211 loss=4.16779 test_loss_avg=3.06974 acc=0.09375 test_acc_avg=0.36953 test_acc_top5_avg=0.79258 time=235.38it/s +epoch=20 global_step=8211 loss=0.46203 test_loss_avg=2.96817 acc=0.89062 test_acc_avg=0.33371 test_acc_top5_avg=0.80458 time=241.44it/s +epoch=20 global_step=8211 loss=6.95945 test_loss_avg=3.47558 acc=0.00000 test_acc_avg=0.29885 test_acc_top5_avg=0.80340 time=481.33it/s +curr_acc 0.2989 +BEST_ACC 0.3249 +curr_acc_top5 0.8034 +BEST_ACC_top5 0.8043 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=6.57668 loss_avg=6.31766 acc=0.32812 acc_top1_avg=0.35897 acc_top5_avg=0.82392 lr=0.01000 gn=5.56578 time=49.79it/s +epoch=21 global_step=8300 loss=6.85226 loss_avg=6.35090 acc=0.29688 acc_top1_avg=0.35779 acc_top5_avg=0.81663 lr=0.01000 gn=4.78561 time=57.86it/s +epoch=21 global_step=8350 loss=6.71638 loss_avg=6.33488 acc=0.30469 acc_top1_avg=0.36078 acc_top5_avg=0.81981 lr=0.01000 gn=4.50033 time=55.96it/s +epoch=21 global_step=8400 loss=6.75796 loss_avg=6.31683 acc=0.30469 acc_top1_avg=0.36256 acc_top5_avg=0.81953 lr=0.01000 gn=3.76504 time=64.56it/s +epoch=21 global_step=8450 loss=6.17386 loss_avg=6.33137 acc=0.38281 acc_top1_avg=0.36058 acc_top5_avg=0.81858 lr=0.01000 gn=5.30160 time=50.61it/s +epoch=21 global_step=8500 loss=6.62350 loss_avg=6.34123 acc=0.30469 acc_top1_avg=0.35924 acc_top5_avg=0.81858 lr=0.01000 gn=3.23433 time=56.62it/s +epoch=21 global_step=8550 loss=6.37930 loss_avg=6.34044 acc=0.34375 acc_top1_avg=0.35889 acc_top5_avg=0.81877 lr=0.01000 gn=4.46910 time=64.96it/s +epoch=21 global_step=8600 loss=6.47801 loss_avg=6.33340 acc=0.32031 acc_top1_avg=0.36012 acc_top5_avg=0.81935 lr=0.01000 gn=3.99155 time=55.61it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.88482 test_loss_avg=4.26973 acc=0.44531 test_acc_avg=0.13834 test_acc_top5_avg=0.68579 time=237.17it/s +epoch=21 global_step=8602 loss=8.65591 test_loss_avg=4.00464 acc=0.00000 test_acc_avg=0.25485 test_acc_top5_avg=0.80657 time=460.00it/s +curr_acc 0.2548 +BEST_ACC 0.3249 +curr_acc_top5 0.8066 +BEST_ACC_top5 0.8043 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=6.59490 loss_avg=6.28482 acc=0.34375 acc_top1_avg=0.36312 acc_top5_avg=0.82438 lr=0.01000 gn=4.37066 time=60.40it/s +epoch=22 global_step=8700 loss=6.18985 loss_avg=6.27579 acc=0.39844 acc_top1_avg=0.36519 acc_top5_avg=0.81864 lr=0.01000 gn=4.43411 time=60.01it/s +epoch=22 global_step=8750 loss=6.19626 loss_avg=6.29168 acc=0.39062 acc_top1_avg=0.36439 acc_top5_avg=0.81773 lr=0.01000 gn=4.52472 time=45.39it/s +epoch=22 global_step=8800 loss=6.39972 loss_avg=6.28453 acc=0.35156 acc_top1_avg=0.36506 acc_top5_avg=0.81940 lr=0.01000 gn=4.49139 time=56.62it/s +epoch=22 global_step=8850 loss=6.79487 loss_avg=6.29254 acc=0.30469 acc_top1_avg=0.36442 acc_top5_avg=0.81867 lr=0.01000 gn=4.54766 time=56.80it/s +epoch=22 global_step=8900 loss=5.69586 loss_avg=6.30747 acc=0.45312 acc_top1_avg=0.36249 acc_top5_avg=0.81916 lr=0.01000 gn=5.86336 time=55.97it/s +epoch=22 global_step=8950 loss=6.62797 loss_avg=6.31460 acc=0.32031 acc_top1_avg=0.36169 acc_top5_avg=0.81892 lr=0.01000 gn=3.95032 time=55.52it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.63102 test_loss_avg=3.92089 acc=0.82031 test_acc_avg=0.28776 test_acc_top5_avg=0.82878 time=233.87it/s +epoch=22 global_step=8993 loss=1.29780 test_loss_avg=3.32689 acc=0.64844 test_acc_avg=0.27961 test_acc_top5_avg=0.80847 time=236.18it/s +epoch=22 global_step=8993 loss=5.37540 test_loss_avg=3.32915 acc=0.00000 test_acc_avg=0.30924 test_acc_top5_avg=0.79460 time=495.60it/s +curr_acc 0.3092 +BEST_ACC 0.3249 +curr_acc_top5 0.7946 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=6.56464 loss_avg=6.23467 acc=0.35938 acc_top1_avg=0.38058 acc_top5_avg=0.82701 lr=0.01000 gn=4.93120 time=56.46it/s +epoch=23 global_step=9050 loss=6.57468 loss_avg=6.36440 acc=0.35156 acc_top1_avg=0.35855 acc_top5_avg=0.82319 lr=0.01000 gn=4.43860 time=57.33it/s +epoch=23 global_step=9100 loss=7.04236 loss_avg=6.36902 acc=0.29688 acc_top1_avg=0.35682 acc_top5_avg=0.82477 lr=0.01000 gn=4.81249 time=65.45it/s +epoch=23 global_step=9150 loss=6.75895 loss_avg=6.32233 acc=0.30469 acc_top1_avg=0.36117 acc_top5_avg=0.82335 lr=0.01000 gn=4.96221 time=51.70it/s +epoch=23 global_step=9200 loss=6.23269 loss_avg=6.30087 acc=0.35938 acc_top1_avg=0.36394 acc_top5_avg=0.82465 lr=0.01000 gn=4.42693 time=55.99it/s +epoch=23 global_step=9250 loss=6.21346 loss_avg=6.30071 acc=0.37500 acc_top1_avg=0.36345 acc_top5_avg=0.82521 lr=0.01000 gn=4.82255 time=56.57it/s +epoch=23 global_step=9300 loss=6.32826 loss_avg=6.29922 acc=0.35156 acc_top1_avg=0.36383 acc_top5_avg=0.82446 lr=0.01000 gn=4.62088 time=56.48it/s +epoch=23 global_step=9350 loss=6.59833 loss_avg=6.32437 acc=0.34375 acc_top1_avg=0.36104 acc_top5_avg=0.82364 lr=0.01000 gn=5.87775 time=54.45it/s +====================Eval==================== +epoch=23 global_step=9384 loss=5.25371 test_loss_avg=3.65499 acc=0.00781 test_acc_avg=0.18419 test_acc_top5_avg=0.65246 time=240.46it/s +epoch=23 global_step=9384 loss=5.90188 test_loss_avg=3.22669 acc=0.00000 test_acc_avg=0.31893 test_acc_top5_avg=0.74664 time=860.37it/s +curr_acc 0.3189 +BEST_ACC 0.3249 +curr_acc_top5 0.7466 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=5.80789 loss_avg=6.23787 acc=0.41406 acc_top1_avg=0.36768 acc_top5_avg=0.80957 lr=0.01000 gn=4.36019 time=54.12it/s +epoch=24 global_step=9450 loss=5.62647 loss_avg=6.25091 acc=0.43750 acc_top1_avg=0.36802 acc_top5_avg=0.82256 lr=0.01000 gn=5.14375 time=48.02it/s +epoch=24 global_step=9500 loss=6.21378 loss_avg=6.27765 acc=0.36719 acc_top1_avg=0.36510 acc_top5_avg=0.82294 lr=0.01000 gn=3.62848 time=64.94it/s +epoch=24 global_step=9550 loss=6.70855 loss_avg=6.29304 acc=0.31250 acc_top1_avg=0.36309 acc_top5_avg=0.82215 lr=0.01000 gn=3.88552 time=61.47it/s +epoch=24 global_step=9600 loss=6.40451 loss_avg=6.31979 acc=0.36719 acc_top1_avg=0.36089 acc_top5_avg=0.82114 lr=0.01000 gn=5.96064 time=57.94it/s +epoch=24 global_step=9650 loss=6.36803 loss_avg=6.32849 acc=0.35938 acc_top1_avg=0.36034 acc_top5_avg=0.82081 lr=0.01000 gn=5.29206 time=49.93it/s +epoch=24 global_step=9700 loss=5.93609 loss_avg=6.33659 acc=0.41406 acc_top1_avg=0.35945 acc_top5_avg=0.82105 lr=0.01000 gn=4.57548 time=60.77it/s +epoch=24 global_step=9750 loss=6.36180 loss_avg=6.33390 acc=0.34375 acc_top1_avg=0.35987 acc_top5_avg=0.81997 lr=0.01000 gn=3.52043 time=60.82it/s +====================Eval==================== +epoch=24 global_step=9775 loss=4.85600 test_loss_avg=5.04693 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.60742 time=255.30it/s +epoch=24 global_step=9775 loss=4.84461 test_loss_avg=3.98268 acc=0.00000 test_acc_avg=0.22627 test_acc_top5_avg=0.66435 time=243.04it/s +epoch=24 global_step=9775 loss=5.34330 test_loss_avg=3.46126 acc=0.00000 test_acc_avg=0.32911 test_acc_top5_avg=0.70491 time=553.27it/s +curr_acc 0.3291 +BEST_ACC 0.3249 +curr_acc_top5 0.7049 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=6.16777 loss_avg=6.27280 acc=0.38281 acc_top1_avg=0.36406 acc_top5_avg=0.83219 lr=0.01000 gn=5.11882 time=53.76it/s +epoch=25 global_step=9850 loss=6.43775 loss_avg=6.29129 acc=0.36719 acc_top1_avg=0.36396 acc_top5_avg=0.82396 lr=0.01000 gn=4.44531 time=48.14it/s +epoch=25 global_step=9900 loss=6.44386 loss_avg=6.30367 acc=0.34375 acc_top1_avg=0.36331 acc_top5_avg=0.82725 lr=0.01000 gn=4.77929 time=56.20it/s +epoch=25 global_step=9950 loss=6.83957 loss_avg=6.32553 acc=0.30469 acc_top1_avg=0.36022 acc_top5_avg=0.82567 lr=0.01000 gn=4.56814 time=30.62it/s +epoch=25 global_step=10000 loss=5.74171 loss_avg=6.32428 acc=0.42188 acc_top1_avg=0.36014 acc_top5_avg=0.82424 lr=0.01000 gn=3.92736 time=56.00it/s +epoch=25 global_step=10050 loss=5.61702 loss_avg=6.29912 acc=0.43750 acc_top1_avg=0.36338 acc_top5_avg=0.82506 lr=0.01000 gn=4.62835 time=56.85it/s +epoch=25 global_step=10100 loss=6.53874 loss_avg=6.30245 acc=0.35938 acc_top1_avg=0.36329 acc_top5_avg=0.82483 lr=0.01000 gn=5.93438 time=55.79it/s +epoch=25 global_step=10150 loss=6.04814 loss_avg=6.32002 acc=0.38281 acc_top1_avg=0.36117 acc_top5_avg=0.82398 lr=0.01000 gn=3.89890 time=56.59it/s +====================Eval==================== +epoch=25 global_step=10166 loss=5.52590 test_loss_avg=3.62971 acc=0.00781 test_acc_avg=0.27969 test_acc_top5_avg=0.69312 time=221.08it/s +epoch=25 global_step=10166 loss=6.49474 test_loss_avg=3.76791 acc=0.00000 test_acc_avg=0.28813 test_acc_top5_avg=0.75927 time=258.72it/s +epoch=25 global_step=10166 loss=5.72367 test_loss_avg=3.87759 acc=0.00000 test_acc_avg=0.27354 test_acc_top5_avg=0.74990 time=489.76it/s +curr_acc 0.2735 +BEST_ACC 0.3291 +curr_acc_top5 0.7499 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=6.60315 loss_avg=6.32019 acc=0.33594 acc_top1_avg=0.36029 acc_top5_avg=0.82261 lr=0.01000 gn=4.65276 time=54.23it/s +epoch=26 global_step=10250 loss=6.56139 loss_avg=6.26076 acc=0.35156 acc_top1_avg=0.36812 acc_top5_avg=0.82366 lr=0.01000 gn=4.43817 time=57.21it/s +epoch=26 global_step=10300 loss=6.02081 loss_avg=6.26808 acc=0.39844 acc_top1_avg=0.36625 acc_top5_avg=0.82311 lr=0.01000 gn=4.61237 time=59.40it/s +epoch=26 global_step=10350 loss=6.18980 loss_avg=6.26970 acc=0.36719 acc_top1_avg=0.36659 acc_top5_avg=0.82409 lr=0.01000 gn=4.01910 time=55.73it/s +epoch=26 global_step=10400 loss=6.19227 loss_avg=6.29274 acc=0.37500 acc_top1_avg=0.36412 acc_top5_avg=0.82419 lr=0.01000 gn=4.57130 time=55.63it/s +epoch=26 global_step=10450 loss=6.39227 loss_avg=6.31018 acc=0.35156 acc_top1_avg=0.36191 acc_top5_avg=0.82408 lr=0.01000 gn=5.10294 time=53.15it/s +epoch=26 global_step=10500 loss=6.28806 loss_avg=6.31342 acc=0.35156 acc_top1_avg=0.36155 acc_top5_avg=0.82352 lr=0.01000 gn=4.40462 time=56.70it/s +epoch=26 global_step=10550 loss=6.26864 loss_avg=6.31447 acc=0.36719 acc_top1_avg=0.36127 acc_top5_avg=0.82324 lr=0.01000 gn=5.11070 time=60.15it/s +====================Eval==================== +epoch=26 global_step=10557 loss=0.61544 test_loss_avg=3.90075 acc=0.82031 test_acc_avg=0.23913 test_acc_top5_avg=0.69769 time=238.62it/s +epoch=26 global_step=10557 loss=7.80360 test_loss_avg=3.94913 acc=0.00000 test_acc_avg=0.28412 test_acc_top5_avg=0.79470 time=749.79it/s +curr_acc 0.2841 +BEST_ACC 0.3291 +curr_acc_top5 0.7947 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=6.45986 loss_avg=6.16833 acc=0.35938 acc_top1_avg=0.37700 acc_top5_avg=0.82722 lr=0.01000 gn=4.86159 time=61.73it/s +epoch=27 global_step=10650 loss=5.93480 loss_avg=6.21347 acc=0.42969 acc_top1_avg=0.37248 acc_top5_avg=0.82636 lr=0.01000 gn=5.12746 time=62.79it/s +epoch=27 global_step=10700 loss=6.33860 loss_avg=6.24818 acc=0.34375 acc_top1_avg=0.36948 acc_top5_avg=0.82627 lr=0.01000 gn=4.39744 time=61.10it/s +epoch=27 global_step=10750 loss=6.85069 loss_avg=6.28493 acc=0.31250 acc_top1_avg=0.36468 acc_top5_avg=0.82375 lr=0.01000 gn=5.23028 time=55.05it/s +epoch=27 global_step=10800 loss=6.40674 loss_avg=6.29005 acc=0.33594 acc_top1_avg=0.36410 acc_top5_avg=0.82456 lr=0.01000 gn=6.01942 time=54.58it/s +epoch=27 global_step=10850 loss=6.52483 loss_avg=6.30902 acc=0.36719 acc_top1_avg=0.36225 acc_top5_avg=0.82311 lr=0.01000 gn=3.95992 time=56.41it/s +epoch=27 global_step=10900 loss=5.99802 loss_avg=6.30328 acc=0.42188 acc_top1_avg=0.36306 acc_top5_avg=0.82412 lr=0.01000 gn=4.70539 time=56.24it/s +====================Eval==================== +epoch=27 global_step=10948 loss=4.60595 test_loss_avg=4.02497 acc=0.03906 test_acc_avg=0.18980 test_acc_top5_avg=0.77987 time=244.88it/s +epoch=27 global_step=10948 loss=0.15435 test_loss_avg=3.09549 acc=0.96094 test_acc_avg=0.30014 test_acc_top5_avg=0.82078 time=245.47it/s +epoch=27 global_step=10948 loss=7.10926 test_loss_avg=3.42661 acc=0.00000 test_acc_avg=0.29391 test_acc_top5_avg=0.82516 time=592.92it/s +curr_acc 0.2939 +BEST_ACC 0.3291 +curr_acc_top5 0.8252 +BEST_ACC_top5 0.8066 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=6.04816 loss_avg=5.68670 acc=0.38281 acc_top1_avg=0.42578 acc_top5_avg=0.85156 lr=0.01000 gn=5.27176 time=40.95it/s +epoch=28 global_step=11000 loss=6.33395 loss_avg=6.23652 acc=0.37500 acc_top1_avg=0.37019 acc_top5_avg=0.82843 lr=0.01000 gn=4.70374 time=61.82it/s +epoch=28 global_step=11050 loss=6.01388 loss_avg=6.23084 acc=0.38281 acc_top1_avg=0.37217 acc_top5_avg=0.82958 lr=0.01000 gn=5.72657 time=55.37it/s +epoch=28 global_step=11100 loss=5.73149 loss_avg=6.26270 acc=0.43750 acc_top1_avg=0.36842 acc_top5_avg=0.82602 lr=0.01000 gn=4.89526 time=57.46it/s +epoch=28 global_step=11150 loss=5.96239 loss_avg=6.26346 acc=0.39844 acc_top1_avg=0.36881 acc_top5_avg=0.82596 lr=0.01000 gn=4.43894 time=57.12it/s +epoch=28 global_step=11200 loss=5.99781 loss_avg=6.27657 acc=0.41406 acc_top1_avg=0.36651 acc_top5_avg=0.82555 lr=0.01000 gn=4.14653 time=60.34it/s +epoch=28 global_step=11250 loss=7.00042 loss_avg=6.28734 acc=0.25781 acc_top1_avg=0.36470 acc_top5_avg=0.82471 lr=0.01000 gn=6.13427 time=50.83it/s +epoch=28 global_step=11300 loss=5.68384 loss_avg=6.29891 acc=0.43750 acc_top1_avg=0.36310 acc_top5_avg=0.82382 lr=0.01000 gn=4.43388 time=60.25it/s +====================Eval==================== +epoch=28 global_step=11339 loss=5.11526 test_loss_avg=3.94117 acc=0.00000 test_acc_avg=0.15029 test_acc_top5_avg=0.68832 time=233.21it/s +epoch=28 global_step=11339 loss=10.33265 test_loss_avg=3.94697 acc=0.00000 test_acc_avg=0.27769 test_acc_top5_avg=0.81290 time=744.99it/s +curr_acc 0.2777 +BEST_ACC 0.3291 +curr_acc_top5 0.8129 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=6.40338 loss_avg=6.30245 acc=0.35938 acc_top1_avg=0.36222 acc_top5_avg=0.81108 lr=0.01000 gn=4.86074 time=53.86it/s +epoch=29 global_step=11400 loss=6.43772 loss_avg=6.29275 acc=0.33594 acc_top1_avg=0.36463 acc_top5_avg=0.82006 lr=0.01000 gn=3.53619 time=56.03it/s +epoch=29 global_step=11450 loss=6.28692 loss_avg=6.30210 acc=0.35938 acc_top1_avg=0.36430 acc_top5_avg=0.82221 lr=0.01000 gn=4.35476 time=54.98it/s +epoch=29 global_step=11500 loss=6.78883 loss_avg=6.29133 acc=0.31250 acc_top1_avg=0.36568 acc_top5_avg=0.82516 lr=0.01000 gn=4.51350 time=60.79it/s +epoch=29 global_step=11550 loss=6.43914 loss_avg=6.29286 acc=0.34375 acc_top1_avg=0.36545 acc_top5_avg=0.82564 lr=0.01000 gn=5.31094 time=62.04it/s +epoch=29 global_step=11600 loss=6.26763 loss_avg=6.30587 acc=0.35938 acc_top1_avg=0.36384 acc_top5_avg=0.82483 lr=0.01000 gn=4.65796 time=64.91it/s +epoch=29 global_step=11650 loss=5.79827 loss_avg=6.29422 acc=0.42188 acc_top1_avg=0.36530 acc_top5_avg=0.82554 lr=0.01000 gn=5.79709 time=61.15it/s +epoch=29 global_step=11700 loss=5.91339 loss_avg=6.28672 acc=0.40625 acc_top1_avg=0.36611 acc_top5_avg=0.82548 lr=0.01000 gn=4.92049 time=61.14it/s +====================Eval==================== +epoch=29 global_step=11730 loss=3.29773 test_loss_avg=4.91968 acc=0.34375 test_acc_avg=0.04427 test_acc_top5_avg=0.66146 time=241.34it/s +epoch=29 global_step=11730 loss=1.18502 test_loss_avg=3.57204 acc=0.74219 test_acc_avg=0.21014 test_acc_top5_avg=0.75252 time=230.34it/s +epoch=29 global_step=11730 loss=10.41683 test_loss_avg=3.87273 acc=0.00000 test_acc_avg=0.27967 test_acc_top5_avg=0.80459 time=826.63it/s +curr_acc 0.2797 +BEST_ACC 0.3291 +curr_acc_top5 0.8046 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=6.90164 loss_avg=6.29978 acc=0.28125 acc_top1_avg=0.36367 acc_top5_avg=0.82734 lr=0.01000 gn=4.26266 time=56.81it/s +epoch=30 global_step=11800 loss=6.12789 loss_avg=6.35558 acc=0.39844 acc_top1_avg=0.35725 acc_top5_avg=0.82009 lr=0.01000 gn=5.68146 time=61.97it/s +epoch=30 global_step=11850 loss=6.14408 loss_avg=6.32321 acc=0.37500 acc_top1_avg=0.36081 acc_top5_avg=0.82096 lr=0.01000 gn=4.86631 time=51.79it/s +epoch=30 global_step=11900 loss=6.91521 loss_avg=6.32256 acc=0.28125 acc_top1_avg=0.36121 acc_top5_avg=0.82279 lr=0.01000 gn=3.97930 time=62.24it/s +epoch=30 global_step=11950 loss=6.33655 loss_avg=6.32762 acc=0.39062 acc_top1_avg=0.36040 acc_top5_avg=0.82305 lr=0.01000 gn=5.53775 time=58.32it/s +epoch=30 global_step=12000 loss=6.30085 loss_avg=6.32299 acc=0.36719 acc_top1_avg=0.36073 acc_top5_avg=0.82274 lr=0.01000 gn=4.15072 time=60.07it/s +epoch=30 global_step=12050 loss=6.41894 loss_avg=6.32120 acc=0.36719 acc_top1_avg=0.36091 acc_top5_avg=0.82207 lr=0.01000 gn=6.48874 time=61.96it/s +epoch=30 global_step=12100 loss=6.28902 loss_avg=6.32179 acc=0.35938 acc_top1_avg=0.36100 acc_top5_avg=0.82181 lr=0.01000 gn=6.22980 time=56.83it/s +====================Eval==================== +epoch=30 global_step=12121 loss=4.26634 test_loss_avg=3.80941 acc=0.03906 test_acc_avg=0.15260 test_acc_top5_avg=0.72760 time=241.54it/s +epoch=30 global_step=12121 loss=9.20693 test_loss_avg=3.73165 acc=0.00000 test_acc_avg=0.30756 test_acc_top5_avg=0.78283 time=467.70it/s +curr_acc 0.3076 +BEST_ACC 0.3291 +curr_acc_top5 0.7828 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=6.12876 loss_avg=6.27864 acc=0.37500 acc_top1_avg=0.36234 acc_top5_avg=0.83001 lr=0.01000 gn=4.73306 time=56.08it/s +epoch=31 global_step=12200 loss=5.71450 loss_avg=6.21234 acc=0.44531 acc_top1_avg=0.37184 acc_top5_avg=0.82763 lr=0.01000 gn=5.12725 time=51.21it/s +epoch=31 global_step=12250 loss=5.60135 loss_avg=6.23554 acc=0.43750 acc_top1_avg=0.36967 acc_top5_avg=0.82540 lr=0.01000 gn=4.24204 time=57.22it/s +epoch=31 global_step=12300 loss=6.16149 loss_avg=6.26208 acc=0.38281 acc_top1_avg=0.36714 acc_top5_avg=0.82485 lr=0.01000 gn=4.40388 time=64.18it/s +epoch=31 global_step=12350 loss=6.49164 loss_avg=6.28355 acc=0.33594 acc_top1_avg=0.36490 acc_top5_avg=0.82379 lr=0.01000 gn=5.07976 time=61.00it/s +epoch=31 global_step=12400 loss=5.84044 loss_avg=6.29156 acc=0.41406 acc_top1_avg=0.36402 acc_top5_avg=0.82224 lr=0.01000 gn=4.82376 time=61.64it/s +epoch=31 global_step=12450 loss=6.11117 loss_avg=6.29393 acc=0.38281 acc_top1_avg=0.36379 acc_top5_avg=0.82278 lr=0.01000 gn=4.41246 time=56.53it/s +epoch=31 global_step=12500 loss=6.26396 loss_avg=6.30751 acc=0.38281 acc_top1_avg=0.36243 acc_top5_avg=0.82330 lr=0.01000 gn=5.50219 time=61.55it/s +====================Eval==================== +epoch=31 global_step=12512 loss=6.19223 test_loss_avg=6.19223 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.39062 time=212.63it/s +epoch=31 global_step=12512 loss=3.97923 test_loss_avg=4.08461 acc=0.00000 test_acc_avg=0.13741 test_acc_top5_avg=0.68168 time=216.70it/s +epoch=31 global_step=12512 loss=9.03501 test_loss_avg=4.08735 acc=0.00000 test_acc_avg=0.22943 test_acc_top5_avg=0.75761 time=842.57it/s +curr_acc 0.2294 +BEST_ACC 0.3291 +curr_acc_top5 0.7576 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=6.27614 loss_avg=6.26217 acc=0.37500 acc_top1_avg=0.36657 acc_top5_avg=0.82360 lr=0.01000 gn=4.82448 time=56.58it/s +epoch=32 global_step=12600 loss=5.96318 loss_avg=6.27006 acc=0.40625 acc_top1_avg=0.36541 acc_top5_avg=0.82102 lr=0.01000 gn=4.82749 time=57.88it/s +epoch=32 global_step=12650 loss=6.50611 loss_avg=6.23052 acc=0.32031 acc_top1_avg=0.37007 acc_top5_avg=0.82394 lr=0.01000 gn=4.28995 time=56.84it/s +epoch=32 global_step=12700 loss=6.48821 loss_avg=6.27707 acc=0.35156 acc_top1_avg=0.36565 acc_top5_avg=0.82243 lr=0.01000 gn=4.39929 time=62.10it/s +epoch=32 global_step=12750 loss=6.43065 loss_avg=6.28365 acc=0.34375 acc_top1_avg=0.36463 acc_top5_avg=0.82189 lr=0.01000 gn=4.51798 time=53.81it/s +epoch=32 global_step=12800 loss=6.16737 loss_avg=6.27740 acc=0.37500 acc_top1_avg=0.36542 acc_top5_avg=0.82286 lr=0.01000 gn=3.55779 time=52.03it/s +epoch=32 global_step=12850 loss=6.41619 loss_avg=6.29351 acc=0.36719 acc_top1_avg=0.36379 acc_top5_avg=0.82290 lr=0.01000 gn=3.99559 time=54.82it/s +epoch=32 global_step=12900 loss=6.21914 loss_avg=6.29944 acc=0.36719 acc_top1_avg=0.36322 acc_top5_avg=0.82219 lr=0.01000 gn=4.31298 time=64.83it/s +====================Eval==================== +epoch=32 global_step=12903 loss=6.09570 test_loss_avg=4.23127 acc=0.00781 test_acc_avg=0.25817 test_acc_top5_avg=0.68004 time=235.48it/s +epoch=32 global_step=12903 loss=6.48505 test_loss_avg=3.47522 acc=0.00000 test_acc_avg=0.32227 test_acc_top5_avg=0.79253 time=237.58it/s +epoch=32 global_step=12903 loss=6.70366 test_loss_avg=3.76161 acc=0.00000 test_acc_avg=0.29371 test_acc_top5_avg=0.78916 time=436.18it/s +curr_acc 0.2937 +BEST_ACC 0.3291 +curr_acc_top5 0.7892 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=6.28125 loss_avg=6.26953 acc=0.40625 acc_top1_avg=0.36818 acc_top5_avg=0.82812 lr=0.01000 gn=5.10198 time=55.52it/s +epoch=33 global_step=13000 loss=6.54405 loss_avg=6.27281 acc=0.35156 acc_top1_avg=0.36630 acc_top5_avg=0.82555 lr=0.01000 gn=4.18784 time=54.84it/s +epoch=33 global_step=13050 loss=6.31699 loss_avg=6.29653 acc=0.34375 acc_top1_avg=0.36352 acc_top5_avg=0.82451 lr=0.01000 gn=4.95876 time=56.90it/s +epoch=33 global_step=13100 loss=7.01219 loss_avg=6.31048 acc=0.28125 acc_top1_avg=0.36152 acc_top5_avg=0.82321 lr=0.01000 gn=2.70861 time=61.52it/s +epoch=33 global_step=13150 loss=5.62766 loss_avg=6.30300 acc=0.45312 acc_top1_avg=0.36175 acc_top5_avg=0.82335 lr=0.01000 gn=5.05128 time=55.63it/s +epoch=33 global_step=13200 loss=6.50435 loss_avg=6.30764 acc=0.35938 acc_top1_avg=0.36219 acc_top5_avg=0.82347 lr=0.01000 gn=4.40444 time=64.43it/s +epoch=33 global_step=13250 loss=6.35104 loss_avg=6.30910 acc=0.34375 acc_top1_avg=0.36217 acc_top5_avg=0.82331 lr=0.01000 gn=4.18257 time=57.35it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.99820 test_loss_avg=3.69398 acc=0.64844 test_acc_avg=0.24146 test_acc_top5_avg=0.74800 time=229.75it/s +epoch=33 global_step=13294 loss=6.45754 test_loss_avg=3.43913 acc=0.00000 test_acc_avg=0.32644 test_acc_top5_avg=0.81171 time=833.69it/s +curr_acc 0.3264 +BEST_ACC 0.3291 +curr_acc_top5 0.8117 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=6.19913 loss_avg=6.11063 acc=0.36719 acc_top1_avg=0.38542 acc_top5_avg=0.82161 lr=0.01000 gn=4.42693 time=56.23it/s +epoch=34 global_step=13350 loss=6.71588 loss_avg=6.20223 acc=0.30469 acc_top1_avg=0.37347 acc_top5_avg=0.82966 lr=0.01000 gn=5.94957 time=56.49it/s +epoch=34 global_step=13400 loss=6.65109 loss_avg=6.26918 acc=0.34375 acc_top1_avg=0.36601 acc_top5_avg=0.82849 lr=0.01000 gn=5.52593 time=58.31it/s +epoch=34 global_step=13450 loss=6.12193 loss_avg=6.30207 acc=0.38281 acc_top1_avg=0.36258 acc_top5_avg=0.82577 lr=0.01000 gn=4.56121 time=52.46it/s +epoch=34 global_step=13500 loss=6.33405 loss_avg=6.28333 acc=0.35156 acc_top1_avg=0.36480 acc_top5_avg=0.82611 lr=0.01000 gn=4.75499 time=56.37it/s +epoch=34 global_step=13550 loss=6.48703 loss_avg=6.29253 acc=0.32812 acc_top1_avg=0.36395 acc_top5_avg=0.82462 lr=0.01000 gn=4.21556 time=61.00it/s +epoch=34 global_step=13600 loss=5.80428 loss_avg=6.28692 acc=0.40625 acc_top1_avg=0.36448 acc_top5_avg=0.82547 lr=0.01000 gn=4.60588 time=56.83it/s +epoch=34 global_step=13650 loss=6.77435 loss_avg=6.29747 acc=0.31250 acc_top1_avg=0.36339 acc_top5_avg=0.82422 lr=0.01000 gn=4.97141 time=61.03it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.98420 test_loss_avg=3.40574 acc=0.64062 test_acc_avg=0.25837 test_acc_top5_avg=0.78460 time=233.71it/s +epoch=34 global_step=13685 loss=0.52262 test_loss_avg=3.84165 acc=0.85156 test_acc_avg=0.25623 test_acc_top5_avg=0.70911 time=241.00it/s +epoch=34 global_step=13685 loss=6.30191 test_loss_avg=3.86464 acc=0.00000 test_acc_avg=0.27571 test_acc_top5_avg=0.72369 time=811.12it/s +curr_acc 0.2757 +BEST_ACC 0.3291 +curr_acc_top5 0.7237 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=6.20770 loss_avg=6.34784 acc=0.38281 acc_top1_avg=0.35990 acc_top5_avg=0.81458 lr=0.01000 gn=5.22976 time=58.89it/s +epoch=35 global_step=13750 loss=6.98534 loss_avg=6.32153 acc=0.29688 acc_top1_avg=0.35986 acc_top5_avg=0.82668 lr=0.01000 gn=4.31727 time=56.42it/s +epoch=35 global_step=13800 loss=6.36738 loss_avg=6.29605 acc=0.35938 acc_top1_avg=0.36257 acc_top5_avg=0.82880 lr=0.01000 gn=5.03949 time=58.15it/s +epoch=35 global_step=13850 loss=5.77223 loss_avg=6.28347 acc=0.40625 acc_top1_avg=0.36482 acc_top5_avg=0.82779 lr=0.01000 gn=4.62753 time=56.52it/s +epoch=35 global_step=13900 loss=6.77045 loss_avg=6.30350 acc=0.32031 acc_top1_avg=0.36374 acc_top5_avg=0.82594 lr=0.01000 gn=3.98120 time=61.38it/s +epoch=35 global_step=13950 loss=5.52591 loss_avg=6.30246 acc=0.47656 acc_top1_avg=0.36359 acc_top5_avg=0.82382 lr=0.01000 gn=5.37497 time=53.86it/s +epoch=35 global_step=14000 loss=5.64931 loss_avg=6.28529 acc=0.42969 acc_top1_avg=0.36550 acc_top5_avg=0.82589 lr=0.01000 gn=4.37396 time=61.30it/s +epoch=35 global_step=14050 loss=6.44031 loss_avg=6.29591 acc=0.34375 acc_top1_avg=0.36361 acc_top5_avg=0.82562 lr=0.01000 gn=5.79616 time=57.20it/s +====================Eval==================== +epoch=35 global_step=14076 loss=4.33364 test_loss_avg=4.11372 acc=0.01562 test_acc_avg=0.15491 test_acc_top5_avg=0.72254 time=239.73it/s +epoch=35 global_step=14076 loss=6.65193 test_loss_avg=3.60624 acc=0.00000 test_acc_avg=0.29272 test_acc_top5_avg=0.81606 time=829.73it/s +curr_acc 0.2927 +BEST_ACC 0.3291 +curr_acc_top5 0.8161 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=6.94991 loss_avg=6.27303 acc=0.29688 acc_top1_avg=0.36426 acc_top5_avg=0.82422 lr=0.01000 gn=4.47235 time=57.05it/s +epoch=36 global_step=14150 loss=6.70686 loss_avg=6.27110 acc=0.32031 acc_top1_avg=0.36666 acc_top5_avg=0.83161 lr=0.01000 gn=5.57119 time=63.81it/s +epoch=36 global_step=14200 loss=6.81250 loss_avg=6.29970 acc=0.32031 acc_top1_avg=0.36353 acc_top5_avg=0.82466 lr=0.01000 gn=5.62351 time=50.29it/s +epoch=36 global_step=14250 loss=6.75803 loss_avg=6.29703 acc=0.28906 acc_top1_avg=0.36413 acc_top5_avg=0.82413 lr=0.01000 gn=5.47686 time=57.15it/s +epoch=36 global_step=14300 loss=6.77366 loss_avg=6.26508 acc=0.29688 acc_top1_avg=0.36743 acc_top5_avg=0.82586 lr=0.01000 gn=5.62275 time=61.89it/s +epoch=36 global_step=14350 loss=6.42530 loss_avg=6.28050 acc=0.34375 acc_top1_avg=0.36536 acc_top5_avg=0.82582 lr=0.01000 gn=6.39457 time=55.53it/s +epoch=36 global_step=14400 loss=5.69761 loss_avg=6.29667 acc=0.40625 acc_top1_avg=0.36352 acc_top5_avg=0.82489 lr=0.01000 gn=4.31911 time=61.48it/s +epoch=36 global_step=14450 loss=6.06925 loss_avg=6.30226 acc=0.38281 acc_top1_avg=0.36280 acc_top5_avg=0.82430 lr=0.01000 gn=4.56032 time=59.99it/s +====================Eval==================== +epoch=36 global_step=14467 loss=4.85636 test_loss_avg=4.93089 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.61589 time=239.07it/s +epoch=36 global_step=14467 loss=0.48923 test_loss_avg=3.77070 acc=0.87500 test_acc_avg=0.16476 test_acc_top5_avg=0.69657 time=235.28it/s +epoch=36 global_step=14467 loss=7.79220 test_loss_avg=3.66407 acc=0.00000 test_acc_avg=0.27403 test_acc_top5_avg=0.76543 time=836.02it/s +curr_acc 0.2740 +BEST_ACC 0.3291 +curr_acc_top5 0.7654 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=5.86388 loss_avg=6.21440 acc=0.41406 acc_top1_avg=0.37689 acc_top5_avg=0.83381 lr=0.01000 gn=4.78083 time=61.43it/s +epoch=37 global_step=14550 loss=6.30216 loss_avg=6.26426 acc=0.35156 acc_top1_avg=0.36803 acc_top5_avg=0.82107 lr=0.01000 gn=4.93128 time=58.90it/s +epoch=37 global_step=14600 loss=6.70192 loss_avg=6.26717 acc=0.33594 acc_top1_avg=0.36666 acc_top5_avg=0.82331 lr=0.01000 gn=4.36859 time=64.66it/s +epoch=37 global_step=14650 loss=6.69118 loss_avg=6.27368 acc=0.32031 acc_top1_avg=0.36608 acc_top5_avg=0.82270 lr=0.01000 gn=5.03848 time=62.53it/s +epoch=37 global_step=14700 loss=5.72163 loss_avg=6.27481 acc=0.46875 acc_top1_avg=0.36638 acc_top5_avg=0.82289 lr=0.01000 gn=6.06535 time=57.07it/s +epoch=37 global_step=14750 loss=6.34712 loss_avg=6.29210 acc=0.35156 acc_top1_avg=0.36470 acc_top5_avg=0.82346 lr=0.01000 gn=4.14765 time=60.19it/s +epoch=37 global_step=14800 loss=6.99036 loss_avg=6.29651 acc=0.30469 acc_top1_avg=0.36407 acc_top5_avg=0.82212 lr=0.01000 gn=4.93352 time=58.59it/s +epoch=37 global_step=14850 loss=6.30572 loss_avg=6.29082 acc=0.37500 acc_top1_avg=0.36519 acc_top5_avg=0.82294 lr=0.01000 gn=4.86204 time=60.49it/s +====================Eval==================== +epoch=37 global_step=14858 loss=4.68401 test_loss_avg=4.25885 acc=0.03906 test_acc_avg=0.22396 test_acc_top5_avg=0.63108 time=240.03it/s +epoch=37 global_step=14858 loss=8.19715 test_loss_avg=4.18457 acc=0.00000 test_acc_avg=0.28024 test_acc_top5_avg=0.75061 time=241.64it/s +epoch=37 global_step=14858 loss=8.33284 test_loss_avg=4.29028 acc=0.00000 test_acc_avg=0.27314 test_acc_top5_avg=0.75485 time=490.56it/s +curr_acc 0.2731 +BEST_ACC 0.3291 +curr_acc_top5 0.7548 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=5.39208 loss_avg=6.24192 acc=0.45312 acc_top1_avg=0.36626 acc_top5_avg=0.83110 lr=0.01000 gn=3.95751 time=55.85it/s +epoch=38 global_step=14950 loss=6.17343 loss_avg=6.25636 acc=0.39062 acc_top1_avg=0.36812 acc_top5_avg=0.82235 lr=0.01000 gn=4.92538 time=56.72it/s +epoch=38 global_step=15000 loss=6.31316 loss_avg=6.25227 acc=0.37500 acc_top1_avg=0.36807 acc_top5_avg=0.82119 lr=0.01000 gn=5.41527 time=54.61it/s +epoch=38 global_step=15050 loss=5.98982 loss_avg=6.27231 acc=0.37500 acc_top1_avg=0.36499 acc_top5_avg=0.82345 lr=0.01000 gn=4.35528 time=61.61it/s +epoch=38 global_step=15100 loss=6.31872 loss_avg=6.26498 acc=0.35938 acc_top1_avg=0.36573 acc_top5_avg=0.82464 lr=0.01000 gn=5.79792 time=60.37it/s +epoch=38 global_step=15150 loss=6.83628 loss_avg=6.25877 acc=0.28906 acc_top1_avg=0.36679 acc_top5_avg=0.82521 lr=0.01000 gn=4.03456 time=59.14it/s +epoch=38 global_step=15200 loss=6.56633 loss_avg=6.27724 acc=0.33594 acc_top1_avg=0.36509 acc_top5_avg=0.82465 lr=0.01000 gn=5.07083 time=56.14it/s +====================Eval==================== +epoch=38 global_step=15249 loss=6.96017 test_loss_avg=4.45666 acc=0.00000 test_acc_avg=0.25537 test_acc_top5_avg=0.59245 time=237.87it/s +epoch=38 global_step=15249 loss=6.12106 test_loss_avg=4.37198 acc=0.00000 test_acc_avg=0.25781 test_acc_top5_avg=0.63736 time=810.65it/s +curr_acc 0.2578 +BEST_ACC 0.3291 +curr_acc_top5 0.6374 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=6.01160 loss_avg=6.01160 acc=0.42188 acc_top1_avg=0.42188 acc_top5_avg=0.84375 lr=0.01000 gn=5.09041 time=48.57it/s +epoch=39 global_step=15300 loss=6.34519 loss_avg=6.31951 acc=0.35938 acc_top1_avg=0.36351 acc_top5_avg=0.81618 lr=0.01000 gn=3.70269 time=60.34it/s +epoch=39 global_step=15350 loss=6.80523 loss_avg=6.23110 acc=0.30469 acc_top1_avg=0.37144 acc_top5_avg=0.82681 lr=0.01000 gn=5.24425 time=59.43it/s +epoch=39 global_step=15400 loss=6.19339 loss_avg=6.25308 acc=0.39062 acc_top1_avg=0.36822 acc_top5_avg=0.82580 lr=0.01000 gn=5.93888 time=54.64it/s +epoch=39 global_step=15450 loss=6.27591 loss_avg=6.27718 acc=0.36719 acc_top1_avg=0.36466 acc_top5_avg=0.82463 lr=0.01000 gn=5.06194 time=58.14it/s +epoch=39 global_step=15500 loss=5.73475 loss_avg=6.26567 acc=0.43750 acc_top1_avg=0.36579 acc_top5_avg=0.82582 lr=0.01000 gn=4.89838 time=62.67it/s +epoch=39 global_step=15550 loss=6.22972 loss_avg=6.26877 acc=0.35938 acc_top1_avg=0.36589 acc_top5_avg=0.82729 lr=0.01000 gn=4.56250 time=60.89it/s +epoch=39 global_step=15600 loss=6.24924 loss_avg=6.27048 acc=0.37500 acc_top1_avg=0.36556 acc_top5_avg=0.82679 lr=0.01000 gn=4.69154 time=54.55it/s +====================Eval==================== +epoch=39 global_step=15640 loss=4.21333 test_loss_avg=3.06534 acc=0.07812 test_acc_avg=0.40543 test_acc_top5_avg=0.76151 time=224.74it/s +epoch=39 global_step=15640 loss=0.48213 test_loss_avg=3.06800 acc=0.86719 test_acc_avg=0.32733 test_acc_top5_avg=0.78657 time=233.55it/s +epoch=39 global_step=15640 loss=6.87145 test_loss_avg=3.48514 acc=0.00000 test_acc_avg=0.30024 test_acc_top5_avg=0.76533 time=482.33it/s +curr_acc 0.3002 +BEST_ACC 0.3291 +curr_acc_top5 0.7653 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=6.29853 loss_avg=6.29286 acc=0.34375 acc_top1_avg=0.36016 acc_top5_avg=0.82969 lr=0.00100 gn=5.02024 time=55.24it/s +epoch=40 global_step=15700 loss=6.15961 loss_avg=6.17281 acc=0.36719 acc_top1_avg=0.37604 acc_top5_avg=0.83893 lr=0.00100 gn=4.49928 time=55.78it/s +epoch=40 global_step=15750 loss=5.95216 loss_avg=6.19272 acc=0.39062 acc_top1_avg=0.37401 acc_top5_avg=0.83594 lr=0.00100 gn=3.87184 time=54.14it/s +epoch=40 global_step=15800 loss=6.07682 loss_avg=6.12225 acc=0.39844 acc_top1_avg=0.38247 acc_top5_avg=0.83755 lr=0.00100 gn=4.29305 time=63.13it/s +epoch=40 global_step=15850 loss=5.84234 loss_avg=6.09554 acc=0.39844 acc_top1_avg=0.38557 acc_top5_avg=0.83772 lr=0.00100 gn=3.39908 time=64.54it/s +epoch=40 global_step=15900 loss=6.36425 loss_avg=6.07422 acc=0.36719 acc_top1_avg=0.38783 acc_top5_avg=0.83846 lr=0.00100 gn=4.80981 time=53.81it/s +epoch=40 global_step=15950 loss=6.28354 loss_avg=6.06907 acc=0.34375 acc_top1_avg=0.38838 acc_top5_avg=0.83793 lr=0.00100 gn=5.16616 time=50.92it/s +epoch=40 global_step=16000 loss=6.35011 loss_avg=6.08105 acc=0.35156 acc_top1_avg=0.38670 acc_top5_avg=0.83874 lr=0.00100 gn=4.54432 time=59.58it/s +====================Eval==================== +epoch=40 global_step=16031 loss=1.28245 test_loss_avg=3.88557 acc=0.60156 test_acc_avg=0.19453 test_acc_top5_avg=0.70703 time=143.45it/s +epoch=40 global_step=16031 loss=7.91175 test_loss_avg=3.53743 acc=0.00000 test_acc_avg=0.33831 test_acc_top5_avg=0.83139 time=818.88it/s +curr_acc 0.3383 +BEST_ACC 0.3291 +curr_acc_top5 0.8314 +BEST_ACC_top5 0.8252 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=5.92378 loss_avg=6.15077 acc=0.41406 acc_top1_avg=0.38076 acc_top5_avg=0.83799 lr=0.00100 gn=4.88099 time=63.66it/s +epoch=41 global_step=16100 loss=6.06069 loss_avg=6.00735 acc=0.38281 acc_top1_avg=0.39640 acc_top5_avg=0.83854 lr=0.00100 gn=4.90106 time=57.91it/s +epoch=41 global_step=16150 loss=5.81949 loss_avg=6.03860 acc=0.41406 acc_top1_avg=0.39174 acc_top5_avg=0.84309 lr=0.00100 gn=5.07196 time=49.86it/s +epoch=41 global_step=16200 loss=6.52874 loss_avg=6.03050 acc=0.33594 acc_top1_avg=0.39284 acc_top5_avg=0.84343 lr=0.00100 gn=4.33980 time=28.01it/s +epoch=41 global_step=16250 loss=5.33127 loss_avg=6.01761 acc=0.46094 acc_top1_avg=0.39412 acc_top5_avg=0.84553 lr=0.00100 gn=4.38110 time=59.11it/s +epoch=41 global_step=16300 loss=5.89878 loss_avg=6.00717 acc=0.39062 acc_top1_avg=0.39553 acc_top5_avg=0.84657 lr=0.00100 gn=4.94631 time=61.73it/s +epoch=41 global_step=16350 loss=6.11213 loss_avg=5.99544 acc=0.36719 acc_top1_avg=0.39663 acc_top5_avg=0.84745 lr=0.00100 gn=5.63997 time=52.41it/s +epoch=41 global_step=16400 loss=5.38078 loss_avg=5.99896 acc=0.47656 acc_top1_avg=0.39619 acc_top5_avg=0.84597 lr=0.00100 gn=5.72973 time=53.14it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.55371 test_loss_avg=3.94404 acc=0.82031 test_acc_avg=0.23864 test_acc_top5_avg=0.78693 time=231.52it/s +epoch=41 global_step=16422 loss=0.38306 test_loss_avg=3.41679 acc=0.90625 test_acc_avg=0.29880 test_acc_top5_avg=0.82390 time=215.38it/s +epoch=41 global_step=16422 loss=7.21037 test_loss_avg=3.48909 acc=0.00000 test_acc_avg=0.34108 test_acc_top5_avg=0.84464 time=808.93it/s +curr_acc 0.3411 +BEST_ACC 0.3383 +curr_acc_top5 0.8446 +BEST_ACC_top5 0.8314 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=5.64576 loss_avg=6.03285 acc=0.42969 acc_top1_avg=0.38867 acc_top5_avg=0.84263 lr=0.00100 gn=4.04346 time=55.49it/s +epoch=42 global_step=16500 loss=6.07751 loss_avg=5.94627 acc=0.35156 acc_top1_avg=0.39944 acc_top5_avg=0.84786 lr=0.00100 gn=4.99760 time=62.68it/s +epoch=42 global_step=16550 loss=6.58160 loss_avg=5.94850 acc=0.32031 acc_top1_avg=0.39996 acc_top5_avg=0.84912 lr=0.00100 gn=5.14072 time=51.69it/s +epoch=42 global_step=16600 loss=6.41361 loss_avg=5.94182 acc=0.33594 acc_top1_avg=0.40050 acc_top5_avg=0.84985 lr=0.00100 gn=5.24740 time=60.37it/s +epoch=42 global_step=16650 loss=5.66131 loss_avg=5.92565 acc=0.42188 acc_top1_avg=0.40296 acc_top5_avg=0.84954 lr=0.00100 gn=5.16898 time=63.26it/s +epoch=42 global_step=16700 loss=6.11065 loss_avg=5.94955 acc=0.38281 acc_top1_avg=0.40010 acc_top5_avg=0.84853 lr=0.00100 gn=3.98971 time=54.52it/s +epoch=42 global_step=16750 loss=5.55514 loss_avg=5.96422 acc=0.44531 acc_top1_avg=0.39875 acc_top5_avg=0.84816 lr=0.00100 gn=5.56853 time=52.76it/s +epoch=42 global_step=16800 loss=5.54559 loss_avg=5.95181 acc=0.44531 acc_top1_avg=0.39990 acc_top5_avg=0.84896 lr=0.00100 gn=5.54632 time=56.18it/s +====================Eval==================== +epoch=42 global_step=16813 loss=5.17753 test_loss_avg=3.69892 acc=0.00781 test_acc_avg=0.20068 test_acc_top5_avg=0.81616 time=237.21it/s +epoch=42 global_step=16813 loss=8.36783 test_loss_avg=3.61469 acc=0.00000 test_acc_avg=0.32902 test_acc_top5_avg=0.84246 time=500.04it/s +curr_acc 0.3290 +BEST_ACC 0.3411 +curr_acc_top5 0.8425 +BEST_ACC_top5 0.8446 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=5.29540 loss_avg=5.92126 acc=0.46094 acc_top1_avg=0.40625 acc_top5_avg=0.85008 lr=0.00100 gn=5.53878 time=55.20it/s +epoch=43 global_step=16900 loss=5.67908 loss_avg=5.90984 acc=0.42969 acc_top1_avg=0.40751 acc_top5_avg=0.85120 lr=0.00100 gn=5.66369 time=54.72it/s +epoch=43 global_step=16950 loss=6.05584 loss_avg=5.90827 acc=0.39062 acc_top1_avg=0.40682 acc_top5_avg=0.84962 lr=0.00100 gn=4.85418 time=57.73it/s +epoch=43 global_step=17000 loss=5.94871 loss_avg=5.90946 acc=0.39844 acc_top1_avg=0.40642 acc_top5_avg=0.84901 lr=0.00100 gn=5.41963 time=63.37it/s +epoch=43 global_step=17050 loss=5.58855 loss_avg=5.90517 acc=0.45312 acc_top1_avg=0.40671 acc_top5_avg=0.84879 lr=0.00100 gn=6.46955 time=57.15it/s +epoch=43 global_step=17100 loss=5.60436 loss_avg=5.90930 acc=0.44531 acc_top1_avg=0.40606 acc_top5_avg=0.84963 lr=0.00100 gn=5.72774 time=56.59it/s +epoch=43 global_step=17150 loss=6.07777 loss_avg=5.91356 acc=0.39062 acc_top1_avg=0.40500 acc_top5_avg=0.84985 lr=0.00100 gn=5.07999 time=54.94it/s +epoch=43 global_step=17200 loss=5.80158 loss_avg=5.92577 acc=0.42969 acc_top1_avg=0.40361 acc_top5_avg=0.84938 lr=0.00100 gn=6.84993 time=63.49it/s +====================Eval==================== +epoch=43 global_step=17204 loss=5.18084 test_loss_avg=5.25894 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.75000 time=113.46it/s +epoch=43 global_step=17204 loss=5.82295 test_loss_avg=3.83514 acc=0.00000 test_acc_avg=0.20843 test_acc_top5_avg=0.80557 time=236.25it/s +epoch=43 global_step=17204 loss=7.87403 test_loss_avg=3.61224 acc=0.00000 test_acc_avg=0.32288 test_acc_top5_avg=0.85384 time=510.07it/s +curr_acc 0.3229 +BEST_ACC 0.3411 +curr_acc_top5 0.8538 +BEST_ACC_top5 0.8446 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=5.78978 loss_avg=5.89379 acc=0.42969 acc_top1_avg=0.40863 acc_top5_avg=0.85564 lr=0.00100 gn=5.99865 time=59.90it/s +epoch=44 global_step=17300 loss=5.89537 loss_avg=5.87789 acc=0.41406 acc_top1_avg=0.40861 acc_top5_avg=0.85213 lr=0.00100 gn=6.45076 time=57.90it/s +epoch=44 global_step=17350 loss=5.91885 loss_avg=5.86900 acc=0.41406 acc_top1_avg=0.41048 acc_top5_avg=0.85418 lr=0.00100 gn=6.12589 time=54.16it/s +epoch=44 global_step=17400 loss=6.56493 loss_avg=5.89273 acc=0.32812 acc_top1_avg=0.40796 acc_top5_avg=0.85276 lr=0.00100 gn=6.24711 time=50.22it/s +epoch=44 global_step=17450 loss=6.30367 loss_avg=5.90549 acc=0.36719 acc_top1_avg=0.40631 acc_top5_avg=0.85163 lr=0.00100 gn=6.30962 time=61.33it/s +epoch=44 global_step=17500 loss=6.04934 loss_avg=5.90562 acc=0.38281 acc_top1_avg=0.40580 acc_top5_avg=0.85130 lr=0.00100 gn=5.80937 time=57.02it/s +epoch=44 global_step=17550 loss=5.58109 loss_avg=5.90424 acc=0.44531 acc_top1_avg=0.40630 acc_top5_avg=0.85129 lr=0.00100 gn=5.92176 time=56.11it/s +====================Eval==================== +epoch=44 global_step=17595 loss=3.93318 test_loss_avg=3.29888 acc=0.09375 test_acc_avg=0.26921 test_acc_top5_avg=0.82389 time=231.68it/s +epoch=44 global_step=17595 loss=7.86986 test_loss_avg=3.13133 acc=0.00000 test_acc_avg=0.36391 test_acc_top5_avg=0.84956 time=255.00it/s +epoch=44 global_step=17595 loss=7.44673 test_loss_avg=3.41393 acc=0.00000 test_acc_avg=0.34088 test_acc_top5_avg=0.85255 time=543.02it/s +curr_acc 0.3409 +BEST_ACC 0.3411 +curr_acc_top5 0.8526 +BEST_ACC_top5 0.8538 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=5.38716 loss_avg=5.80542 acc=0.44531 acc_top1_avg=0.41563 acc_top5_avg=0.86094 lr=0.00100 gn=6.24863 time=56.26it/s +epoch=45 global_step=17650 loss=5.72285 loss_avg=5.87601 acc=0.41406 acc_top1_avg=0.40753 acc_top5_avg=0.85128 lr=0.00100 gn=7.20959 time=62.57it/s +epoch=45 global_step=17700 loss=5.88868 loss_avg=5.84356 acc=0.40625 acc_top1_avg=0.41250 acc_top5_avg=0.85015 lr=0.00100 gn=5.37795 time=61.79it/s +epoch=45 global_step=17750 loss=6.00133 loss_avg=5.86460 acc=0.39062 acc_top1_avg=0.41013 acc_top5_avg=0.85066 lr=0.00100 gn=5.82379 time=52.27it/s +epoch=45 global_step=17800 loss=5.72474 loss_avg=5.87187 acc=0.43750 acc_top1_avg=0.40930 acc_top5_avg=0.85053 lr=0.00100 gn=6.15713 time=57.26it/s +epoch=45 global_step=17850 loss=6.00503 loss_avg=5.87600 acc=0.41406 acc_top1_avg=0.40907 acc_top5_avg=0.85101 lr=0.00100 gn=7.58618 time=48.52it/s +epoch=45 global_step=17900 loss=5.80888 loss_avg=5.87321 acc=0.42969 acc_top1_avg=0.40945 acc_top5_avg=0.85133 lr=0.00100 gn=8.02336 time=60.32it/s +epoch=45 global_step=17950 loss=5.84122 loss_avg=5.87636 acc=0.40625 acc_top1_avg=0.40858 acc_top5_avg=0.85150 lr=0.00100 gn=6.69155 time=55.88it/s +====================Eval==================== +epoch=45 global_step=17986 loss=1.26936 test_loss_avg=3.57073 acc=0.62500 test_acc_avg=0.23038 test_acc_top5_avg=0.79115 time=214.41it/s +epoch=45 global_step=17986 loss=7.51668 test_loss_avg=3.53378 acc=0.00000 test_acc_avg=0.32496 test_acc_top5_avg=0.86472 time=610.17it/s +curr_acc 0.3250 +BEST_ACC 0.3411 +curr_acc_top5 0.8647 +BEST_ACC_top5 0.8538 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=6.31335 loss_avg=5.69158 acc=0.34375 acc_top1_avg=0.43304 acc_top5_avg=0.86161 lr=0.00100 gn=6.02433 time=55.11it/s +epoch=46 global_step=18050 loss=6.02077 loss_avg=5.72915 acc=0.38281 acc_top1_avg=0.42688 acc_top5_avg=0.86182 lr=0.00100 gn=8.17224 time=54.65it/s +epoch=46 global_step=18100 loss=5.93620 loss_avg=5.78709 acc=0.42969 acc_top1_avg=0.42057 acc_top5_avg=0.85876 lr=0.00100 gn=6.73806 time=60.72it/s +epoch=46 global_step=18150 loss=6.14670 loss_avg=5.81747 acc=0.39062 acc_top1_avg=0.41659 acc_top5_avg=0.85499 lr=0.00100 gn=6.86838 time=55.58it/s +epoch=46 global_step=18200 loss=6.08786 loss_avg=5.83149 acc=0.39062 acc_top1_avg=0.41516 acc_top5_avg=0.85310 lr=0.00100 gn=7.04716 time=51.31it/s +epoch=46 global_step=18250 loss=6.42704 loss_avg=5.84487 acc=0.34375 acc_top1_avg=0.41282 acc_top5_avg=0.85257 lr=0.00100 gn=7.63418 time=39.08it/s +epoch=46 global_step=18300 loss=6.35991 loss_avg=5.83175 acc=0.34375 acc_top1_avg=0.41429 acc_top5_avg=0.85335 lr=0.00100 gn=7.11984 time=59.32it/s +epoch=46 global_step=18350 loss=5.56443 loss_avg=5.83179 acc=0.44531 acc_top1_avg=0.41402 acc_top5_avg=0.85356 lr=0.00100 gn=7.62438 time=56.44it/s +====================Eval==================== +epoch=46 global_step=18377 loss=2.31928 test_loss_avg=3.22207 acc=0.39844 test_acc_avg=0.33008 test_acc_top5_avg=0.82959 time=241.22it/s +epoch=46 global_step=18377 loss=0.20040 test_loss_avg=3.21308 acc=0.93750 test_acc_avg=0.32576 test_acc_top5_avg=0.84399 time=240.25it/s +epoch=46 global_step=18377 loss=8.24289 test_loss_avg=3.60633 acc=0.00000 test_acc_avg=0.32358 test_acc_top5_avg=0.86254 time=491.89it/s +curr_acc 0.3236 +BEST_ACC 0.3411 +curr_acc_top5 0.8625 +BEST_ACC_top5 0.8647 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=5.19515 loss_avg=5.67290 acc=0.46875 acc_top1_avg=0.42799 acc_top5_avg=0.86413 lr=0.00100 gn=8.15706 time=56.17it/s +epoch=47 global_step=18450 loss=5.40347 loss_avg=5.72800 acc=0.45312 acc_top1_avg=0.42455 acc_top5_avg=0.85713 lr=0.00100 gn=6.76848 time=51.39it/s +epoch=47 global_step=18500 loss=6.21794 loss_avg=5.76190 acc=0.37500 acc_top1_avg=0.42137 acc_top5_avg=0.85448 lr=0.00100 gn=8.13885 time=56.35it/s +epoch=47 global_step=18550 loss=5.13704 loss_avg=5.78236 acc=0.49219 acc_top1_avg=0.41971 acc_top5_avg=0.85721 lr=0.00100 gn=6.31327 time=54.22it/s +epoch=47 global_step=18600 loss=6.45157 loss_avg=5.78614 acc=0.35938 acc_top1_avg=0.41918 acc_top5_avg=0.85426 lr=0.00100 gn=7.61454 time=54.92it/s +epoch=47 global_step=18650 loss=6.16271 loss_avg=5.79293 acc=0.39062 acc_top1_avg=0.41901 acc_top5_avg=0.85417 lr=0.00100 gn=7.91353 time=56.20it/s +epoch=47 global_step=18700 loss=5.51947 loss_avg=5.80127 acc=0.43750 acc_top1_avg=0.41813 acc_top5_avg=0.85413 lr=0.00100 gn=7.67218 time=55.22it/s +epoch=47 global_step=18750 loss=6.12519 loss_avg=5.80398 acc=0.39844 acc_top1_avg=0.41752 acc_top5_avg=0.85502 lr=0.00100 gn=7.43876 time=59.68it/s +====================Eval==================== +epoch=47 global_step=18768 loss=5.25380 test_loss_avg=3.79462 acc=0.01562 test_acc_avg=0.15329 test_acc_top5_avg=0.78378 time=227.72it/s +epoch=47 global_step=18768 loss=8.08012 test_loss_avg=3.44671 acc=0.00000 test_acc_avg=0.31784 test_acc_top5_avg=0.86956 time=621.75it/s +curr_acc 0.3178 +BEST_ACC 0.3411 +curr_acc_top5 0.8696 +BEST_ACC_top5 0.8647 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=5.84313 loss_avg=5.72303 acc=0.42188 acc_top1_avg=0.42676 acc_top5_avg=0.85693 lr=0.00100 gn=8.72459 time=63.55it/s +epoch=48 global_step=18850 loss=5.69093 loss_avg=5.80703 acc=0.42188 acc_top1_avg=0.41825 acc_top5_avg=0.85356 lr=0.00100 gn=7.98246 time=62.47it/s +epoch=48 global_step=18900 loss=5.85836 loss_avg=5.78071 acc=0.40625 acc_top1_avg=0.42105 acc_top5_avg=0.85357 lr=0.00100 gn=6.77696 time=57.01it/s +epoch=48 global_step=18950 loss=5.27678 loss_avg=5.77930 acc=0.48438 acc_top1_avg=0.42042 acc_top5_avg=0.85332 lr=0.00100 gn=7.96905 time=57.74it/s +epoch=48 global_step=19000 loss=5.93099 loss_avg=5.76004 acc=0.41406 acc_top1_avg=0.42245 acc_top5_avg=0.85395 lr=0.00100 gn=9.95389 time=53.70it/s +epoch=48 global_step=19050 loss=4.85173 loss_avg=5.77159 acc=0.53125 acc_top1_avg=0.42146 acc_top5_avg=0.85464 lr=0.00100 gn=10.39748 time=58.44it/s +epoch=48 global_step=19100 loss=6.02613 loss_avg=5.77426 acc=0.41406 acc_top1_avg=0.42089 acc_top5_avg=0.85497 lr=0.00100 gn=9.72709 time=60.02it/s +epoch=48 global_step=19150 loss=5.76505 loss_avg=5.77300 acc=0.43750 acc_top1_avg=0.42122 acc_top5_avg=0.85584 lr=0.00100 gn=9.83017 time=55.73it/s +====================Eval==================== +epoch=48 global_step=19159 loss=4.59326 test_loss_avg=5.13628 acc=0.12500 test_acc_avg=0.01562 test_acc_top5_avg=0.75195 time=235.57it/s +epoch=48 global_step=19159 loss=0.30928 test_loss_avg=3.47654 acc=0.92969 test_acc_avg=0.24892 test_acc_top5_avg=0.83998 time=237.42it/s +epoch=48 global_step=19159 loss=7.80662 test_loss_avg=3.42062 acc=0.00000 test_acc_avg=0.32773 test_acc_top5_avg=0.87342 time=789.29it/s +curr_acc 0.3277 +BEST_ACC 0.3411 +curr_acc_top5 0.8734 +BEST_ACC_top5 0.8696 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=5.80397 loss_avg=5.79002 acc=0.42969 acc_top1_avg=0.42054 acc_top5_avg=0.84870 lr=0.00100 gn=8.01333 time=60.31it/s +epoch=49 global_step=19250 loss=6.13575 loss_avg=5.72733 acc=0.34375 acc_top1_avg=0.42780 acc_top5_avg=0.85560 lr=0.00100 gn=9.77763 time=63.10it/s +epoch=49 global_step=19300 loss=5.09593 loss_avg=5.73917 acc=0.51562 acc_top1_avg=0.42570 acc_top5_avg=0.85500 lr=0.00100 gn=9.04634 time=60.50it/s +epoch=49 global_step=19350 loss=6.04862 loss_avg=5.74215 acc=0.39062 acc_top1_avg=0.42486 acc_top5_avg=0.85426 lr=0.00100 gn=9.09962 time=62.97it/s +epoch=49 global_step=19400 loss=6.05934 loss_avg=5.74479 acc=0.39062 acc_top1_avg=0.42431 acc_top5_avg=0.85367 lr=0.00100 gn=7.78589 time=58.67it/s +epoch=49 global_step=19450 loss=5.68554 loss_avg=5.75570 acc=0.42969 acc_top1_avg=0.42351 acc_top5_avg=0.85358 lr=0.00100 gn=7.74222 time=54.57it/s +epoch=49 global_step=19500 loss=5.57666 loss_avg=5.76885 acc=0.44531 acc_top1_avg=0.42135 acc_top5_avg=0.85385 lr=0.00100 gn=10.46338 time=56.25it/s +epoch=49 global_step=19550 loss=6.14857 loss_avg=5.76252 acc=0.38750 acc_top1_avg=0.42249 acc_top5_avg=0.85459 lr=0.00100 gn=16.95053 time=80.46it/s +====================Eval==================== +epoch=49 global_step=19550 loss=4.28861 test_loss_avg=3.48920 acc=0.03125 test_acc_avg=0.21713 test_acc_top5_avg=0.84510 time=237.62it/s +epoch=49 global_step=19550 loss=7.80709 test_loss_avg=3.47953 acc=0.00000 test_acc_avg=0.32338 test_acc_top5_avg=0.86986 time=693.16it/s +epoch=49 global_step=19550 loss=7.80709 test_loss_avg=3.47953 acc=0.00000 test_acc_avg=0.32338 test_acc_top5_avg=0.86986 time=693.16it/s +curr_acc 0.3234 +BEST_ACC 0.3411 +curr_acc_top5 0.8699 +BEST_ACC_top5 0.8734 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=6.15525 loss_avg=5.66511 acc=0.38281 acc_top1_avg=0.43688 acc_top5_avg=0.85594 lr=0.00100 gn=7.68803 time=62.64it/s +epoch=50 global_step=19650 loss=6.35258 loss_avg=5.71365 acc=0.34375 acc_top1_avg=0.43031 acc_top5_avg=0.85664 lr=0.00100 gn=11.62814 time=55.59it/s +epoch=50 global_step=19700 loss=5.85513 loss_avg=5.71698 acc=0.40625 acc_top1_avg=0.42885 acc_top5_avg=0.85688 lr=0.00100 gn=10.15187 time=57.46it/s +epoch=50 global_step=19750 loss=5.26834 loss_avg=5.71047 acc=0.46875 acc_top1_avg=0.42969 acc_top5_avg=0.85777 lr=0.00100 gn=11.84725 time=60.62it/s +epoch=50 global_step=19800 loss=5.80692 loss_avg=5.72073 acc=0.40625 acc_top1_avg=0.42844 acc_top5_avg=0.85644 lr=0.00100 gn=7.55686 time=41.13it/s +epoch=50 global_step=19850 loss=5.69277 loss_avg=5.71784 acc=0.42188 acc_top1_avg=0.42878 acc_top5_avg=0.85560 lr=0.00100 gn=10.15477 time=58.22it/s +epoch=50 global_step=19900 loss=4.80354 loss_avg=5.72613 acc=0.52344 acc_top1_avg=0.42746 acc_top5_avg=0.85489 lr=0.00100 gn=9.62843 time=51.81it/s +====================Eval==================== +epoch=50 global_step=19941 loss=5.45228 test_loss_avg=3.68596 acc=0.00000 test_acc_avg=0.21906 test_acc_top5_avg=0.80937 time=234.00it/s +epoch=50 global_step=19941 loss=7.39710 test_loss_avg=3.51464 acc=0.00000 test_acc_avg=0.31962 test_acc_top5_avg=0.86669 time=730.08it/s +curr_acc 0.3196 +BEST_ACC 0.3411 +curr_acc_top5 0.8667 +BEST_ACC_top5 0.8734 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=5.74838 loss_avg=5.52944 acc=0.41406 acc_top1_avg=0.44531 acc_top5_avg=0.86024 lr=0.00100 gn=8.58333 time=58.88it/s +epoch=51 global_step=20000 loss=6.42034 loss_avg=5.67438 acc=0.35938 acc_top1_avg=0.43154 acc_top5_avg=0.85540 lr=0.00100 gn=9.82328 time=56.19it/s +epoch=51 global_step=20050 loss=5.73565 loss_avg=5.69965 acc=0.39844 acc_top1_avg=0.42782 acc_top5_avg=0.85536 lr=0.00100 gn=10.28168 time=57.04it/s +epoch=51 global_step=20100 loss=6.13436 loss_avg=5.68865 acc=0.38281 acc_top1_avg=0.42939 acc_top5_avg=0.85436 lr=0.00100 gn=12.38522 time=53.78it/s +epoch=51 global_step=20150 loss=6.36244 loss_avg=5.68435 acc=0.35156 acc_top1_avg=0.43062 acc_top5_avg=0.85500 lr=0.00100 gn=10.12715 time=60.32it/s +epoch=51 global_step=20200 loss=6.16643 loss_avg=5.69535 acc=0.37500 acc_top1_avg=0.42963 acc_top5_avg=0.85524 lr=0.00100 gn=11.12067 time=56.46it/s +epoch=51 global_step=20250 loss=5.52166 loss_avg=5.70038 acc=0.46094 acc_top1_avg=0.42956 acc_top5_avg=0.85513 lr=0.00100 gn=9.94670 time=61.71it/s +epoch=51 global_step=20300 loss=6.39102 loss_avg=5.70669 acc=0.37500 acc_top1_avg=0.42880 acc_top5_avg=0.85485 lr=0.00100 gn=10.24230 time=56.13it/s +====================Eval==================== +epoch=51 global_step=20332 loss=3.80049 test_loss_avg=3.49746 acc=0.10156 test_acc_avg=0.23103 test_acc_top5_avg=0.83557 time=231.00it/s +epoch=51 global_step=20332 loss=5.19243 test_loss_avg=3.06651 acc=0.29688 test_acc_avg=0.35299 test_acc_top5_avg=0.85409 time=240.07it/s +epoch=51 global_step=20332 loss=8.01017 test_loss_avg=3.53485 acc=0.00000 test_acc_avg=0.31725 test_acc_top5_avg=0.86353 time=839.36it/s +curr_acc 0.3172 +BEST_ACC 0.3411 +curr_acc_top5 0.8635 +BEST_ACC_top5 0.8734 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=5.34040 loss_avg=5.71709 acc=0.46094 acc_top1_avg=0.42882 acc_top5_avg=0.86111 lr=0.00100 gn=10.87314 time=53.97it/s +epoch=52 global_step=20400 loss=5.63212 loss_avg=5.63599 acc=0.45312 acc_top1_avg=0.43704 acc_top5_avg=0.85938 lr=0.00100 gn=10.08903 time=53.71it/s +epoch=52 global_step=20450 loss=5.76172 loss_avg=5.68916 acc=0.45312 acc_top1_avg=0.43061 acc_top5_avg=0.85765 lr=0.00100 gn=12.10685 time=62.82it/s +epoch=52 global_step=20500 loss=5.87170 loss_avg=5.66282 acc=0.42188 acc_top1_avg=0.43341 acc_top5_avg=0.85863 lr=0.00100 gn=11.81587 time=50.83it/s +epoch=52 global_step=20550 loss=6.21363 loss_avg=5.67444 acc=0.35156 acc_top1_avg=0.43252 acc_top5_avg=0.85529 lr=0.00100 gn=7.82174 time=55.12it/s +epoch=52 global_step=20600 loss=6.33764 loss_avg=5.67297 acc=0.35156 acc_top1_avg=0.43345 acc_top5_avg=0.85544 lr=0.00100 gn=10.63431 time=56.63it/s +epoch=52 global_step=20650 loss=5.33199 loss_avg=5.68854 acc=0.46094 acc_top1_avg=0.43187 acc_top5_avg=0.85471 lr=0.00100 gn=11.27327 time=57.03it/s +epoch=52 global_step=20700 loss=5.25602 loss_avg=5.68277 acc=0.49219 acc_top1_avg=0.43224 acc_top5_avg=0.85541 lr=0.00100 gn=12.33996 time=53.15it/s +====================Eval==================== +epoch=52 global_step=20723 loss=1.45944 test_loss_avg=3.78242 acc=0.57031 test_acc_avg=0.15588 test_acc_top5_avg=0.75558 time=235.36it/s +epoch=52 global_step=20723 loss=8.25755 test_loss_avg=3.55644 acc=0.00000 test_acc_avg=0.29490 test_acc_top5_avg=0.85661 time=503.94it/s +curr_acc 0.2949 +BEST_ACC 0.3411 +curr_acc_top5 0.8566 +BEST_ACC_top5 0.8734 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=5.89431 loss_avg=5.59288 acc=0.40625 acc_top1_avg=0.44473 acc_top5_avg=0.85735 lr=0.00100 gn=12.90602 time=55.05it/s +epoch=53 global_step=20800 loss=5.66653 loss_avg=5.62725 acc=0.45312 acc_top1_avg=0.43933 acc_top5_avg=0.85998 lr=0.00100 gn=11.71263 time=49.60it/s +epoch=53 global_step=20850 loss=5.52283 loss_avg=5.66197 acc=0.45312 acc_top1_avg=0.43504 acc_top5_avg=0.85759 lr=0.00100 gn=12.22356 time=57.78it/s +epoch=53 global_step=20900 loss=6.35235 loss_avg=5.64258 acc=0.35156 acc_top1_avg=0.43759 acc_top5_avg=0.85606 lr=0.00100 gn=10.50470 time=54.29it/s +epoch=53 global_step=20950 loss=5.01041 loss_avg=5.64119 acc=0.51562 acc_top1_avg=0.43815 acc_top5_avg=0.85635 lr=0.00100 gn=14.84995 time=63.72it/s +epoch=53 global_step=21000 loss=5.03129 loss_avg=5.64298 acc=0.49219 acc_top1_avg=0.43781 acc_top5_avg=0.85771 lr=0.00100 gn=11.81295 time=62.51it/s +epoch=53 global_step=21050 loss=5.31072 loss_avg=5.66401 acc=0.46094 acc_top1_avg=0.43523 acc_top5_avg=0.85682 lr=0.00100 gn=12.97945 time=55.61it/s +epoch=53 global_step=21100 loss=6.12481 loss_avg=5.66736 acc=0.37500 acc_top1_avg=0.43433 acc_top5_avg=0.85664 lr=0.00100 gn=12.21878 time=63.08it/s +====================Eval==================== +epoch=53 global_step=21114 loss=1.34193 test_loss_avg=4.05613 acc=0.53125 test_acc_avg=0.20312 test_acc_top5_avg=0.81490 time=237.56it/s +epoch=53 global_step=21114 loss=0.19730 test_loss_avg=3.27236 acc=0.94531 test_acc_avg=0.29365 test_acc_top5_avg=0.85379 time=231.42it/s +epoch=53 global_step=21114 loss=7.62919 test_loss_avg=3.42689 acc=0.00000 test_acc_avg=0.32180 test_acc_top5_avg=0.87441 time=490.96it/s +curr_acc 0.3218 +BEST_ACC 0.3411 +curr_acc_top5 0.8744 +BEST_ACC_top5 0.8734 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=5.21852 loss_avg=5.63839 acc=0.50781 acc_top1_avg=0.44010 acc_top5_avg=0.85829 lr=0.00100 gn=13.00783 time=59.00it/s +epoch=54 global_step=21200 loss=5.90820 loss_avg=5.66619 acc=0.40625 acc_top1_avg=0.43487 acc_top5_avg=0.85865 lr=0.00100 gn=10.93981 time=55.52it/s +epoch=54 global_step=21250 loss=5.10094 loss_avg=5.61388 acc=0.51562 acc_top1_avg=0.43974 acc_top5_avg=0.85777 lr=0.00100 gn=13.38538 time=48.17it/s +epoch=54 global_step=21300 loss=5.63044 loss_avg=5.62297 acc=0.42969 acc_top1_avg=0.43952 acc_top5_avg=0.85576 lr=0.00100 gn=13.17212 time=55.73it/s +epoch=54 global_step=21350 loss=5.60447 loss_avg=5.62806 acc=0.45312 acc_top1_avg=0.43929 acc_top5_avg=0.85510 lr=0.00100 gn=12.70577 time=61.93it/s +epoch=54 global_step=21400 loss=6.16909 loss_avg=5.62816 acc=0.36719 acc_top1_avg=0.43881 acc_top5_avg=0.85503 lr=0.00100 gn=12.10039 time=63.06it/s +epoch=54 global_step=21450 loss=5.15208 loss_avg=5.64119 acc=0.50000 acc_top1_avg=0.43776 acc_top5_avg=0.85514 lr=0.00100 gn=13.59951 time=52.82it/s +epoch=54 global_step=21500 loss=5.05976 loss_avg=5.63726 acc=0.48438 acc_top1_avg=0.43825 acc_top5_avg=0.85541 lr=0.00100 gn=12.23479 time=56.30it/s +====================Eval==================== +epoch=54 global_step=21505 loss=5.48791 test_loss_avg=3.77254 acc=0.00781 test_acc_avg=0.12339 test_acc_top5_avg=0.81411 time=235.01it/s +epoch=54 global_step=21505 loss=6.65557 test_loss_avg=3.42945 acc=0.00000 test_acc_avg=0.29124 test_acc_top5_avg=0.86640 time=498.08it/s +curr_acc 0.2912 +BEST_ACC 0.3411 +curr_acc_top5 0.8664 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=5.54492 loss_avg=5.60920 acc=0.45312 acc_top1_avg=0.44253 acc_top5_avg=0.85660 lr=0.00100 gn=13.86670 time=63.08it/s +epoch=55 global_step=21600 loss=5.94418 loss_avg=5.65816 acc=0.39844 acc_top1_avg=0.43569 acc_top5_avg=0.85452 lr=0.00100 gn=11.98513 time=50.34it/s +epoch=55 global_step=21650 loss=6.01538 loss_avg=5.60502 acc=0.38281 acc_top1_avg=0.44127 acc_top5_avg=0.85431 lr=0.00100 gn=9.47758 time=29.47it/s +epoch=55 global_step=21700 loss=5.58732 loss_avg=5.61969 acc=0.41406 acc_top1_avg=0.43894 acc_top5_avg=0.85489 lr=0.00100 gn=15.84731 time=52.20it/s +epoch=55 global_step=21750 loss=4.98778 loss_avg=5.62837 acc=0.50781 acc_top1_avg=0.43836 acc_top5_avg=0.85376 lr=0.00100 gn=13.89938 time=52.33it/s +epoch=55 global_step=21800 loss=5.68640 loss_avg=5.63409 acc=0.42969 acc_top1_avg=0.43784 acc_top5_avg=0.85416 lr=0.00100 gn=13.71821 time=60.15it/s +epoch=55 global_step=21850 loss=5.12723 loss_avg=5.62477 acc=0.50000 acc_top1_avg=0.43902 acc_top5_avg=0.85541 lr=0.00100 gn=14.62614 time=55.68it/s +====================Eval==================== +epoch=55 global_step=21896 loss=5.19515 test_loss_avg=5.27051 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.62813 time=235.61it/s +epoch=55 global_step=21896 loss=3.77504 test_loss_avg=3.66484 acc=0.25781 test_acc_avg=0.23011 test_acc_top5_avg=0.81108 time=235.15it/s +epoch=55 global_step=21896 loss=7.28638 test_loss_avg=3.39441 acc=0.00000 test_acc_avg=0.33979 test_acc_top5_avg=0.85641 time=497.60it/s +curr_acc 0.3398 +BEST_ACC 0.3411 +curr_acc_top5 0.8564 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=5.30981 loss_avg=5.50779 acc=0.47656 acc_top1_avg=0.45312 acc_top5_avg=0.87500 lr=0.00100 gn=13.33849 time=49.75it/s +epoch=56 global_step=21950 loss=4.96846 loss_avg=5.63558 acc=0.51562 acc_top1_avg=0.43649 acc_top5_avg=0.85894 lr=0.00100 gn=12.36398 time=57.24it/s +epoch=56 global_step=22000 loss=5.37264 loss_avg=5.58867 acc=0.47656 acc_top1_avg=0.44291 acc_top5_avg=0.85990 lr=0.00100 gn=16.08240 time=52.41it/s +epoch=56 global_step=22050 loss=5.40867 loss_avg=5.57129 acc=0.45312 acc_top1_avg=0.44486 acc_top5_avg=0.86034 lr=0.00100 gn=14.58959 time=60.69it/s +epoch=56 global_step=22100 loss=5.75234 loss_avg=5.57203 acc=0.44531 acc_top1_avg=0.44543 acc_top5_avg=0.85869 lr=0.00100 gn=15.97054 time=53.91it/s +epoch=56 global_step=22150 loss=5.26504 loss_avg=5.56531 acc=0.47656 acc_top1_avg=0.44617 acc_top5_avg=0.85888 lr=0.00100 gn=14.70803 time=55.03it/s +epoch=56 global_step=22200 loss=5.47787 loss_avg=5.56375 acc=0.46875 acc_top1_avg=0.44637 acc_top5_avg=0.85824 lr=0.00100 gn=12.94105 time=63.37it/s +epoch=56 global_step=22250 loss=5.82981 loss_avg=5.56920 acc=0.43750 acc_top1_avg=0.44544 acc_top5_avg=0.85807 lr=0.00100 gn=19.09876 time=58.13it/s +====================Eval==================== +epoch=56 global_step=22287 loss=4.31727 test_loss_avg=3.60674 acc=0.03906 test_acc_avg=0.17428 test_acc_top5_avg=0.83413 time=240.82it/s +epoch=56 global_step=22287 loss=7.07368 test_loss_avg=3.39630 acc=0.00000 test_acc_avg=0.30757 test_acc_top5_avg=0.85526 time=244.17it/s +epoch=56 global_step=22287 loss=7.36838 test_loss_avg=3.54003 acc=0.00000 test_acc_avg=0.29589 test_acc_top5_avg=0.85779 time=660.73it/s +curr_acc 0.2959 +BEST_ACC 0.3411 +curr_acc_top5 0.8578 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=6.04060 loss_avg=5.47194 acc=0.39062 acc_top1_avg=0.45793 acc_top5_avg=0.85637 lr=0.00100 gn=16.11247 time=54.94it/s +epoch=57 global_step=22350 loss=6.42736 loss_avg=5.48979 acc=0.35938 acc_top1_avg=0.45573 acc_top5_avg=0.85813 lr=0.00100 gn=15.56954 time=55.36it/s +epoch=57 global_step=22400 loss=5.97319 loss_avg=5.54186 acc=0.40625 acc_top1_avg=0.44981 acc_top5_avg=0.85765 lr=0.00100 gn=15.04871 time=59.36it/s +epoch=57 global_step=22450 loss=5.42945 loss_avg=5.56529 acc=0.46094 acc_top1_avg=0.44675 acc_top5_avg=0.85985 lr=0.00100 gn=16.20178 time=54.92it/s +epoch=57 global_step=22500 loss=5.85629 loss_avg=5.57273 acc=0.39062 acc_top1_avg=0.44553 acc_top5_avg=0.85893 lr=0.00100 gn=16.38404 time=55.35it/s +epoch=57 global_step=22550 loss=5.42625 loss_avg=5.56166 acc=0.49219 acc_top1_avg=0.44704 acc_top5_avg=0.85851 lr=0.00100 gn=18.07057 time=58.65it/s +epoch=57 global_step=22600 loss=6.14381 loss_avg=5.57392 acc=0.39844 acc_top1_avg=0.44594 acc_top5_avg=0.85750 lr=0.00100 gn=16.29135 time=63.23it/s +epoch=57 global_step=22650 loss=5.53955 loss_avg=5.56529 acc=0.45312 acc_top1_avg=0.44654 acc_top5_avg=0.85802 lr=0.00100 gn=19.46930 time=57.26it/s +====================Eval==================== +epoch=57 global_step=22678 loss=1.75447 test_loss_avg=3.42260 acc=0.58594 test_acc_avg=0.20562 test_acc_top5_avg=0.80203 time=196.39it/s +epoch=57 global_step=22678 loss=7.48242 test_loss_avg=3.39480 acc=0.00000 test_acc_avg=0.30597 test_acc_top5_avg=0.87025 time=496.72it/s +curr_acc 0.3060 +BEST_ACC 0.3411 +curr_acc_top5 0.8703 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=5.33495 loss_avg=5.46921 acc=0.46875 acc_top1_avg=0.46094 acc_top5_avg=0.85653 lr=0.00100 gn=13.52197 time=55.80it/s +epoch=58 global_step=22750 loss=5.21521 loss_avg=5.48976 acc=0.47656 acc_top1_avg=0.45443 acc_top5_avg=0.86111 lr=0.00100 gn=16.73178 time=49.88it/s +epoch=58 global_step=22800 loss=5.25845 loss_avg=5.47013 acc=0.48438 acc_top1_avg=0.45678 acc_top5_avg=0.86194 lr=0.00100 gn=16.28543 time=62.53it/s +epoch=58 global_step=22850 loss=5.46023 loss_avg=5.50625 acc=0.46094 acc_top1_avg=0.45435 acc_top5_avg=0.86201 lr=0.00100 gn=17.94379 time=55.25it/s +epoch=58 global_step=22900 loss=5.41938 loss_avg=5.49440 acc=0.46094 acc_top1_avg=0.45467 acc_top5_avg=0.86001 lr=0.00100 gn=16.38466 time=55.32it/s +epoch=58 global_step=22950 loss=5.34712 loss_avg=5.51885 acc=0.47656 acc_top1_avg=0.45203 acc_top5_avg=0.85785 lr=0.00100 gn=16.33298 time=55.03it/s +epoch=58 global_step=23000 loss=4.85931 loss_avg=5.52218 acc=0.53906 acc_top1_avg=0.45162 acc_top5_avg=0.85726 lr=0.00100 gn=17.25368 time=56.40it/s +epoch=58 global_step=23050 loss=6.05198 loss_avg=5.53927 acc=0.39844 acc_top1_avg=0.44979 acc_top5_avg=0.85673 lr=0.00100 gn=17.72456 time=61.10it/s +====================Eval==================== +epoch=58 global_step=23069 loss=3.45080 test_loss_avg=3.29894 acc=0.15625 test_acc_avg=0.31424 test_acc_top5_avg=0.84201 time=238.96it/s +epoch=58 global_step=23069 loss=0.37565 test_loss_avg=3.04326 acc=0.92188 test_acc_avg=0.35915 test_acc_top5_avg=0.85581 time=240.38it/s +epoch=58 global_step=23069 loss=7.34013 test_loss_avg=3.37883 acc=0.00000 test_acc_avg=0.33703 test_acc_top5_avg=0.86472 time=586.21it/s +curr_acc 0.3370 +BEST_ACC 0.3411 +curr_acc_top5 0.8647 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=4.81625 loss_avg=5.65335 acc=0.52344 acc_top1_avg=0.43876 acc_top5_avg=0.85786 lr=0.00100 gn=16.84642 time=60.16it/s +epoch=59 global_step=23150 loss=5.99503 loss_avg=5.57791 acc=0.39844 acc_top1_avg=0.44551 acc_top5_avg=0.85774 lr=0.00100 gn=17.30287 time=59.93it/s +epoch=59 global_step=23200 loss=5.91743 loss_avg=5.55626 acc=0.41406 acc_top1_avg=0.44704 acc_top5_avg=0.85717 lr=0.00100 gn=16.60887 time=60.75it/s +epoch=59 global_step=23250 loss=5.36409 loss_avg=5.53678 acc=0.46875 acc_top1_avg=0.44911 acc_top5_avg=0.85739 lr=0.00100 gn=20.25311 time=52.56it/s +epoch=59 global_step=23300 loss=5.63279 loss_avg=5.53055 acc=0.45312 acc_top1_avg=0.45001 acc_top5_avg=0.85792 lr=0.00100 gn=17.71392 time=59.72it/s +epoch=59 global_step=23350 loss=4.73616 loss_avg=5.50560 acc=0.53906 acc_top1_avg=0.45304 acc_top5_avg=0.85857 lr=0.00100 gn=20.18093 time=54.98it/s +epoch=59 global_step=23400 loss=5.37783 loss_avg=5.51104 acc=0.48438 acc_top1_avg=0.45246 acc_top5_avg=0.85763 lr=0.00100 gn=16.64446 time=57.96it/s +epoch=59 global_step=23450 loss=5.68348 loss_avg=5.53273 acc=0.42188 acc_top1_avg=0.45013 acc_top5_avg=0.85696 lr=0.00100 gn=21.44931 time=64.15it/s +====================Eval==================== +epoch=59 global_step=23460 loss=4.87377 test_loss_avg=3.93652 acc=0.00000 test_acc_avg=0.13482 test_acc_top5_avg=0.74880 time=240.83it/s +epoch=59 global_step=23460 loss=7.77891 test_loss_avg=3.56685 acc=0.00000 test_acc_avg=0.29628 test_acc_top5_avg=0.85858 time=811.59it/s +curr_acc 0.2963 +BEST_ACC 0.3411 +curr_acc_top5 0.8586 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=5.68777 loss_avg=5.44589 acc=0.42188 acc_top1_avg=0.45781 acc_top5_avg=0.85996 lr=0.00100 gn=16.49465 time=56.69it/s +epoch=60 global_step=23550 loss=5.68828 loss_avg=5.44756 acc=0.42188 acc_top1_avg=0.45920 acc_top5_avg=0.85938 lr=0.00100 gn=16.85484 time=61.84it/s +epoch=60 global_step=23600 loss=5.63761 loss_avg=5.48130 acc=0.42969 acc_top1_avg=0.45547 acc_top5_avg=0.86027 lr=0.00100 gn=16.77121 time=57.38it/s +epoch=60 global_step=23650 loss=5.56974 loss_avg=5.50037 acc=0.43750 acc_top1_avg=0.45362 acc_top5_avg=0.85909 lr=0.00100 gn=21.69189 time=63.98it/s +epoch=60 global_step=23700 loss=5.11680 loss_avg=5.48916 acc=0.48438 acc_top1_avg=0.45430 acc_top5_avg=0.85996 lr=0.00100 gn=16.84248 time=56.35it/s +epoch=60 global_step=23750 loss=5.86980 loss_avg=5.50379 acc=0.41406 acc_top1_avg=0.45261 acc_top5_avg=0.85884 lr=0.00100 gn=17.83610 time=54.58it/s +epoch=60 global_step=23800 loss=5.62890 loss_avg=5.49949 acc=0.43750 acc_top1_avg=0.45347 acc_top5_avg=0.85855 lr=0.00100 gn=19.94406 time=55.06it/s +epoch=60 global_step=23850 loss=5.95789 loss_avg=5.49642 acc=0.39844 acc_top1_avg=0.45383 acc_top5_avg=0.85841 lr=0.00100 gn=14.81742 time=62.44it/s +====================Eval==================== +epoch=60 global_step=23851 loss=1.31871 test_loss_avg=4.61383 acc=0.58594 test_acc_avg=0.12578 test_acc_top5_avg=0.70937 time=240.57it/s +epoch=60 global_step=23851 loss=0.39098 test_loss_avg=3.53290 acc=0.89844 test_acc_avg=0.25898 test_acc_top5_avg=0.82500 time=242.07it/s +epoch=60 global_step=23851 loss=7.52792 test_loss_avg=3.50792 acc=0.00000 test_acc_avg=0.31754 test_acc_top5_avg=0.85601 time=850.94it/s +curr_acc 0.3175 +BEST_ACC 0.3411 +curr_acc_top5 0.8560 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=5.31909 loss_avg=5.47404 acc=0.48438 acc_top1_avg=0.45647 acc_top5_avg=0.85619 lr=0.00100 gn=18.74374 time=56.23it/s +epoch=61 global_step=23950 loss=5.44530 loss_avg=5.44707 acc=0.45312 acc_top1_avg=0.46054 acc_top5_avg=0.85724 lr=0.00100 gn=20.11924 time=63.13it/s +epoch=61 global_step=24000 loss=5.57126 loss_avg=5.46149 acc=0.46094 acc_top1_avg=0.45832 acc_top5_avg=0.85450 lr=0.00100 gn=22.85368 time=61.14it/s +epoch=61 global_step=24050 loss=5.27424 loss_avg=5.45822 acc=0.46875 acc_top1_avg=0.45905 acc_top5_avg=0.85623 lr=0.00100 gn=19.15302 time=56.70it/s +epoch=61 global_step=24100 loss=5.44010 loss_avg=5.44688 acc=0.44531 acc_top1_avg=0.46059 acc_top5_avg=0.85809 lr=0.00100 gn=16.39926 time=63.77it/s +epoch=61 global_step=24150 loss=6.02784 loss_avg=5.45706 acc=0.40625 acc_top1_avg=0.45979 acc_top5_avg=0.85823 lr=0.00100 gn=14.86003 time=57.16it/s +epoch=61 global_step=24200 loss=5.03791 loss_avg=5.46668 acc=0.49219 acc_top1_avg=0.45843 acc_top5_avg=0.85783 lr=0.00100 gn=16.43600 time=58.04it/s +====================Eval==================== +epoch=61 global_step=24242 loss=2.89089 test_loss_avg=3.19648 acc=0.20312 test_acc_avg=0.22077 test_acc_top5_avg=0.86341 time=240.31it/s +epoch=61 global_step=24242 loss=7.36252 test_loss_avg=3.23356 acc=0.00000 test_acc_avg=0.33040 test_acc_top5_avg=0.87144 time=832.04it/s +curr_acc 0.3304 +BEST_ACC 0.3411 +curr_acc_top5 0.8714 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=5.95284 loss_avg=5.66018 acc=0.41406 acc_top1_avg=0.43945 acc_top5_avg=0.85645 lr=0.00100 gn=19.94303 time=50.47it/s +epoch=62 global_step=24300 loss=4.80401 loss_avg=5.44159 acc=0.52344 acc_top1_avg=0.45999 acc_top5_avg=0.86247 lr=0.00100 gn=17.21057 time=55.99it/s +epoch=62 global_step=24350 loss=5.03625 loss_avg=5.46039 acc=0.51562 acc_top1_avg=0.45841 acc_top5_avg=0.86010 lr=0.00100 gn=17.52223 time=52.45it/s +epoch=62 global_step=24400 loss=5.64073 loss_avg=5.44951 acc=0.46094 acc_top1_avg=0.46064 acc_top5_avg=0.85868 lr=0.00100 gn=21.43439 time=55.13it/s +epoch=62 global_step=24450 loss=5.87926 loss_avg=5.45516 acc=0.41406 acc_top1_avg=0.45936 acc_top5_avg=0.85738 lr=0.00100 gn=19.37874 time=58.50it/s +epoch=62 global_step=24500 loss=5.47800 loss_avg=5.44799 acc=0.44531 acc_top1_avg=0.45991 acc_top5_avg=0.85759 lr=0.00100 gn=19.27387 time=61.18it/s +epoch=62 global_step=24550 loss=5.44120 loss_avg=5.44195 acc=0.46094 acc_top1_avg=0.46020 acc_top5_avg=0.85768 lr=0.00100 gn=21.61423 time=59.31it/s +epoch=62 global_step=24600 loss=5.76678 loss_avg=5.45634 acc=0.45312 acc_top1_avg=0.45886 acc_top5_avg=0.85689 lr=0.00100 gn=24.21604 time=55.20it/s +====================Eval==================== +epoch=62 global_step=24633 loss=5.17862 test_loss_avg=5.28299 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.62891 time=232.71it/s +epoch=62 global_step=24633 loss=5.31730 test_loss_avg=3.64643 acc=0.00000 test_acc_avg=0.20944 test_acc_top5_avg=0.80544 time=144.55it/s +epoch=62 global_step=24633 loss=7.05614 test_loss_avg=3.38333 acc=0.00000 test_acc_avg=0.31675 test_acc_top5_avg=0.85710 time=514.95it/s +curr_acc 0.3168 +BEST_ACC 0.3411 +curr_acc_top5 0.8571 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=4.70768 loss_avg=5.47247 acc=0.53906 acc_top1_avg=0.45267 acc_top5_avg=0.85708 lr=0.00100 gn=16.85586 time=53.94it/s +epoch=63 global_step=24700 loss=5.93193 loss_avg=5.43039 acc=0.41406 acc_top1_avg=0.46117 acc_top5_avg=0.85494 lr=0.00100 gn=22.25129 time=55.41it/s +epoch=63 global_step=24750 loss=5.63460 loss_avg=5.42669 acc=0.42188 acc_top1_avg=0.46274 acc_top5_avg=0.85597 lr=0.00100 gn=17.76276 time=56.49it/s +epoch=63 global_step=24800 loss=5.75747 loss_avg=5.42688 acc=0.44531 acc_top1_avg=0.46323 acc_top5_avg=0.85615 lr=0.00100 gn=25.34562 time=58.01it/s +epoch=63 global_step=24850 loss=6.05949 loss_avg=5.44557 acc=0.41406 acc_top1_avg=0.46195 acc_top5_avg=0.85563 lr=0.00100 gn=22.90555 time=54.13it/s +epoch=63 global_step=24900 loss=5.15172 loss_avg=5.43003 acc=0.49219 acc_top1_avg=0.46281 acc_top5_avg=0.85654 lr=0.00100 gn=19.17101 time=62.96it/s +epoch=63 global_step=24950 loss=5.55970 loss_avg=5.43186 acc=0.46094 acc_top1_avg=0.46281 acc_top5_avg=0.85632 lr=0.00100 gn=19.63258 time=53.23it/s +epoch=63 global_step=25000 loss=5.24204 loss_avg=5.44005 acc=0.50000 acc_top1_avg=0.46194 acc_top5_avg=0.85688 lr=0.00100 gn=23.50390 time=55.21it/s +====================Eval==================== +epoch=63 global_step=25024 loss=3.01629 test_loss_avg=3.47707 acc=0.20312 test_acc_avg=0.17833 test_acc_top5_avg=0.83152 time=63.05it/s +epoch=63 global_step=25024 loss=6.73385 test_loss_avg=3.10359 acc=0.00000 test_acc_avg=0.31443 test_acc_top5_avg=0.86826 time=246.17it/s +epoch=63 global_step=25024 loss=7.09868 test_loss_avg=3.37852 acc=0.00000 test_acc_avg=0.29055 test_acc_top5_avg=0.87154 time=539.74it/s +curr_acc 0.2905 +BEST_ACC 0.3411 +curr_acc_top5 0.8715 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=5.52602 loss_avg=5.42489 acc=0.46094 acc_top1_avg=0.45763 acc_top5_avg=0.85847 lr=0.00100 gn=20.80666 time=62.27it/s +epoch=64 global_step=25100 loss=5.48675 loss_avg=5.43641 acc=0.45312 acc_top1_avg=0.45765 acc_top5_avg=0.86061 lr=0.00100 gn=23.33091 time=57.35it/s +epoch=64 global_step=25150 loss=6.10266 loss_avg=5.44562 acc=0.39062 acc_top1_avg=0.45802 acc_top5_avg=0.85962 lr=0.00100 gn=23.75396 time=51.46it/s +epoch=64 global_step=25200 loss=5.31208 loss_avg=5.46281 acc=0.47656 acc_top1_avg=0.45765 acc_top5_avg=0.85538 lr=0.00100 gn=21.95643 time=63.47it/s +epoch=64 global_step=25250 loss=5.07008 loss_avg=5.43562 acc=0.50000 acc_top1_avg=0.46070 acc_top5_avg=0.85723 lr=0.00100 gn=21.52784 time=55.30it/s +epoch=64 global_step=25300 loss=5.55570 loss_avg=5.41690 acc=0.43750 acc_top1_avg=0.46266 acc_top5_avg=0.85861 lr=0.00100 gn=23.10746 time=47.77it/s +epoch=64 global_step=25350 loss=5.75971 loss_avg=5.41611 acc=0.42969 acc_top1_avg=0.46302 acc_top5_avg=0.85796 lr=0.00100 gn=20.94793 time=61.00it/s +epoch=64 global_step=25400 loss=5.94484 loss_avg=5.41674 acc=0.39844 acc_top1_avg=0.46314 acc_top5_avg=0.85800 lr=0.00100 gn=21.15374 time=56.79it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.93442 test_loss_avg=3.51063 acc=0.69531 test_acc_avg=0.24112 test_acc_top5_avg=0.77362 time=229.36it/s +epoch=64 global_step=25415 loss=7.13301 test_loss_avg=3.34599 acc=0.00000 test_acc_avg=0.33515 test_acc_top5_avg=0.85769 time=815.06it/s +curr_acc 0.3351 +BEST_ACC 0.3411 +curr_acc_top5 0.8577 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=5.21371 loss_avg=5.40856 acc=0.48438 acc_top1_avg=0.46362 acc_top5_avg=0.86116 lr=0.00100 gn=18.24068 time=56.58it/s +epoch=65 global_step=25500 loss=5.82300 loss_avg=5.32947 acc=0.41406 acc_top1_avg=0.47298 acc_top5_avg=0.86204 lr=0.00100 gn=20.85339 time=54.26it/s +epoch=65 global_step=25550 loss=5.72903 loss_avg=5.35631 acc=0.43750 acc_top1_avg=0.47066 acc_top5_avg=0.86082 lr=0.00100 gn=24.55770 time=54.66it/s +epoch=65 global_step=25600 loss=5.13654 loss_avg=5.37264 acc=0.49219 acc_top1_avg=0.46905 acc_top5_avg=0.85954 lr=0.00100 gn=21.78999 time=54.55it/s +epoch=65 global_step=25650 loss=5.88629 loss_avg=5.39183 acc=0.41406 acc_top1_avg=0.46682 acc_top5_avg=0.85814 lr=0.00100 gn=25.40665 time=57.31it/s +epoch=65 global_step=25700 loss=5.27098 loss_avg=5.39200 acc=0.48438 acc_top1_avg=0.46702 acc_top5_avg=0.85641 lr=0.00100 gn=22.42018 time=55.23it/s +epoch=65 global_step=25750 loss=5.70436 loss_avg=5.39934 acc=0.43750 acc_top1_avg=0.46588 acc_top5_avg=0.85536 lr=0.00100 gn=20.06534 time=57.49it/s +epoch=65 global_step=25800 loss=5.35764 loss_avg=5.39774 acc=0.49219 acc_top1_avg=0.46603 acc_top5_avg=0.85655 lr=0.00100 gn=22.33621 time=55.54it/s +====================Eval==================== +epoch=65 global_step=25806 loss=1.66573 test_loss_avg=3.58615 acc=0.46094 test_acc_avg=0.22552 test_acc_top5_avg=0.82969 time=239.73it/s +epoch=65 global_step=25806 loss=0.17314 test_loss_avg=3.23279 acc=0.94531 test_acc_avg=0.29183 test_acc_top5_avg=0.84964 time=232.60it/s +epoch=65 global_step=25806 loss=7.18500 test_loss_avg=3.43735 acc=0.00000 test_acc_avg=0.30340 test_acc_top5_avg=0.86333 time=830.06it/s +curr_acc 0.3034 +BEST_ACC 0.3411 +curr_acc_top5 0.8633 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=5.29936 loss_avg=5.30374 acc=0.48438 acc_top1_avg=0.47674 acc_top5_avg=0.85973 lr=0.00100 gn=19.95846 time=55.74it/s +epoch=66 global_step=25900 loss=5.49508 loss_avg=5.31868 acc=0.43750 acc_top1_avg=0.47731 acc_top5_avg=0.85680 lr=0.00100 gn=20.08676 time=60.63it/s +epoch=66 global_step=25950 loss=5.32398 loss_avg=5.32084 acc=0.47656 acc_top1_avg=0.47678 acc_top5_avg=0.85856 lr=0.00100 gn=26.18741 time=51.30it/s +epoch=66 global_step=26000 loss=5.91693 loss_avg=5.33820 acc=0.44531 acc_top1_avg=0.47499 acc_top5_avg=0.85672 lr=0.00100 gn=24.41509 time=63.44it/s +epoch=66 global_step=26050 loss=4.88528 loss_avg=5.33014 acc=0.54688 acc_top1_avg=0.47583 acc_top5_avg=0.85572 lr=0.00100 gn=25.69545 time=54.94it/s +epoch=66 global_step=26100 loss=5.82369 loss_avg=5.33984 acc=0.42188 acc_top1_avg=0.47436 acc_top5_avg=0.85480 lr=0.00100 gn=27.39409 time=53.21it/s +epoch=66 global_step=26150 loss=5.49701 loss_avg=5.35453 acc=0.46094 acc_top1_avg=0.47259 acc_top5_avg=0.85526 lr=0.00100 gn=22.99374 time=60.46it/s +====================Eval==================== +epoch=66 global_step=26197 loss=5.08889 test_loss_avg=3.82265 acc=0.00781 test_acc_avg=0.16385 test_acc_top5_avg=0.76801 time=241.59it/s +epoch=66 global_step=26197 loss=7.02021 test_loss_avg=3.44331 acc=0.00000 test_acc_avg=0.31853 test_acc_top5_avg=0.85008 time=817.92it/s +curr_acc 0.3185 +BEST_ACC 0.3411 +curr_acc_top5 0.8501 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=5.04922 loss_avg=5.51075 acc=0.50781 acc_top1_avg=0.46354 acc_top5_avg=0.88021 lr=0.00100 gn=21.83801 time=60.87it/s +epoch=67 global_step=26250 loss=5.34799 loss_avg=5.37204 acc=0.47656 acc_top1_avg=0.47317 acc_top5_avg=0.85938 lr=0.00100 gn=22.04470 time=53.66it/s +epoch=67 global_step=26300 loss=5.48749 loss_avg=5.34972 acc=0.43750 acc_top1_avg=0.47436 acc_top5_avg=0.85824 lr=0.00100 gn=19.24410 time=60.36it/s +epoch=67 global_step=26350 loss=5.46083 loss_avg=5.33905 acc=0.46875 acc_top1_avg=0.47452 acc_top5_avg=0.85800 lr=0.00100 gn=24.81956 time=54.74it/s +epoch=67 global_step=26400 loss=5.08828 loss_avg=5.36027 acc=0.48438 acc_top1_avg=0.47179 acc_top5_avg=0.85610 lr=0.00100 gn=29.02821 time=59.06it/s +epoch=67 global_step=26450 loss=5.21016 loss_avg=5.36459 acc=0.48438 acc_top1_avg=0.47131 acc_top5_avg=0.85678 lr=0.00100 gn=23.10649 time=52.90it/s +epoch=67 global_step=26500 loss=5.12893 loss_avg=5.37097 acc=0.52344 acc_top1_avg=0.47076 acc_top5_avg=0.85739 lr=0.00100 gn=25.46521 time=57.26it/s +epoch=67 global_step=26550 loss=5.96194 loss_avg=5.36589 acc=0.40625 acc_top1_avg=0.47152 acc_top5_avg=0.85811 lr=0.00100 gn=28.04892 time=58.92it/s +====================Eval==================== +epoch=67 global_step=26588 loss=5.12915 test_loss_avg=5.14379 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.71205 time=226.76it/s +epoch=67 global_step=26588 loss=1.00348 test_loss_avg=3.51626 acc=0.75000 test_acc_avg=0.23671 test_acc_top5_avg=0.84690 time=231.64it/s +epoch=67 global_step=26588 loss=7.07134 test_loss_avg=3.34593 acc=0.00000 test_acc_avg=0.31744 test_acc_top5_avg=0.87203 time=495.72it/s +curr_acc 0.3174 +BEST_ACC 0.3411 +curr_acc_top5 0.8720 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=5.14277 loss_avg=5.00895 acc=0.48438 acc_top1_avg=0.51237 acc_top5_avg=0.87240 lr=0.00100 gn=23.02058 time=52.35it/s +epoch=68 global_step=26650 loss=5.41299 loss_avg=5.24074 acc=0.47656 acc_top1_avg=0.48412 acc_top5_avg=0.86341 lr=0.00100 gn=25.98629 time=54.69it/s +epoch=68 global_step=26700 loss=4.91146 loss_avg=5.29219 acc=0.54688 acc_top1_avg=0.47914 acc_top5_avg=0.86042 lr=0.00100 gn=26.59623 time=56.66it/s +epoch=68 global_step=26750 loss=5.56579 loss_avg=5.29739 acc=0.44531 acc_top1_avg=0.47868 acc_top5_avg=0.85976 lr=0.00100 gn=25.91330 time=51.85it/s +epoch=68 global_step=26800 loss=5.65189 loss_avg=5.30748 acc=0.42969 acc_top1_avg=0.47745 acc_top5_avg=0.85853 lr=0.00100 gn=22.77352 time=60.36it/s +epoch=68 global_step=26850 loss=5.48892 loss_avg=5.33017 acc=0.47656 acc_top1_avg=0.47492 acc_top5_avg=0.85961 lr=0.00100 gn=30.91803 time=54.15it/s +epoch=68 global_step=26900 loss=5.32151 loss_avg=5.32970 acc=0.47656 acc_top1_avg=0.47483 acc_top5_avg=0.85907 lr=0.00100 gn=24.06817 time=23.11it/s +epoch=68 global_step=26950 loss=5.30428 loss_avg=5.33025 acc=0.47656 acc_top1_avg=0.47458 acc_top5_avg=0.85979 lr=0.00100 gn=26.80751 time=54.80it/s +====================Eval==================== +epoch=68 global_step=26979 loss=3.97292 test_loss_avg=3.34192 acc=0.11719 test_acc_avg=0.22573 test_acc_top5_avg=0.82227 time=240.51it/s +epoch=68 global_step=26979 loss=6.78653 test_loss_avg=3.36956 acc=0.00000 test_acc_avg=0.32342 test_acc_top5_avg=0.85377 time=241.93it/s +epoch=68 global_step=26979 loss=7.34548 test_loss_avg=3.41988 acc=0.00000 test_acc_avg=0.31932 test_acc_top5_avg=0.85483 time=495.72it/s +curr_acc 0.3193 +BEST_ACC 0.3411 +curr_acc_top5 0.8548 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=5.16878 loss_avg=5.39684 acc=0.49219 acc_top1_avg=0.46801 acc_top5_avg=0.84598 lr=0.00100 gn=28.99750 time=60.51it/s +epoch=69 global_step=27050 loss=5.31341 loss_avg=5.31782 acc=0.46094 acc_top1_avg=0.47799 acc_top5_avg=0.85497 lr=0.00100 gn=19.65605 time=53.94it/s +epoch=69 global_step=27100 loss=5.73043 loss_avg=5.27958 acc=0.44531 acc_top1_avg=0.48270 acc_top5_avg=0.86015 lr=0.00100 gn=26.28098 time=55.49it/s +epoch=69 global_step=27150 loss=5.20082 loss_avg=5.27534 acc=0.50781 acc_top1_avg=0.48259 acc_top5_avg=0.85851 lr=0.00100 gn=27.87618 time=54.55it/s +epoch=69 global_step=27200 loss=5.67278 loss_avg=5.29591 acc=0.42969 acc_top1_avg=0.48052 acc_top5_avg=0.85761 lr=0.00100 gn=25.79258 time=55.15it/s +epoch=69 global_step=27250 loss=5.03811 loss_avg=5.30695 acc=0.50781 acc_top1_avg=0.47896 acc_top5_avg=0.85704 lr=0.00100 gn=24.61270 time=46.46it/s +epoch=69 global_step=27300 loss=5.18448 loss_avg=5.31299 acc=0.51562 acc_top1_avg=0.47788 acc_top5_avg=0.85638 lr=0.00100 gn=28.49435 time=54.12it/s +epoch=69 global_step=27350 loss=5.04733 loss_avg=5.31677 acc=0.49219 acc_top1_avg=0.47738 acc_top5_avg=0.85660 lr=0.00100 gn=28.30827 time=55.45it/s +====================Eval==================== +epoch=69 global_step=27370 loss=5.37693 test_loss_avg=3.40134 acc=0.00000 test_acc_avg=0.24171 test_acc_top5_avg=0.81362 time=61.45it/s +epoch=69 global_step=27370 loss=6.86013 test_loss_avg=3.33563 acc=0.00000 test_acc_avg=0.32447 test_acc_top5_avg=0.86689 time=506.56it/s +curr_acc 0.3245 +BEST_ACC 0.3411 +curr_acc_top5 0.8669 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=4.77523 loss_avg=5.26816 acc=0.53906 acc_top1_avg=0.47995 acc_top5_avg=0.85573 lr=0.00100 gn=24.22504 time=54.32it/s +epoch=70 global_step=27450 loss=4.91622 loss_avg=5.31628 acc=0.53125 acc_top1_avg=0.47773 acc_top5_avg=0.85410 lr=0.00100 gn=30.07370 time=54.75it/s +epoch=70 global_step=27500 loss=4.92950 loss_avg=5.30173 acc=0.50781 acc_top1_avg=0.47945 acc_top5_avg=0.85589 lr=0.00100 gn=34.79846 time=54.52it/s +epoch=70 global_step=27550 loss=5.01511 loss_avg=5.28836 acc=0.51562 acc_top1_avg=0.48051 acc_top5_avg=0.85577 lr=0.00100 gn=23.23511 time=59.17it/s +epoch=70 global_step=27600 loss=5.49072 loss_avg=5.28052 acc=0.45312 acc_top1_avg=0.48203 acc_top5_avg=0.85754 lr=0.00100 gn=30.71275 time=57.43it/s +epoch=70 global_step=27650 loss=5.86479 loss_avg=5.27881 acc=0.42188 acc_top1_avg=0.48195 acc_top5_avg=0.85898 lr=0.00100 gn=27.45986 time=57.71it/s +epoch=70 global_step=27700 loss=5.39715 loss_avg=5.28076 acc=0.46094 acc_top1_avg=0.48177 acc_top5_avg=0.85793 lr=0.00100 gn=23.56107 time=51.12it/s +epoch=70 global_step=27750 loss=6.43597 loss_avg=5.28981 acc=0.36719 acc_top1_avg=0.48072 acc_top5_avg=0.85816 lr=0.00100 gn=23.56036 time=61.70it/s +====================Eval==================== +epoch=70 global_step=27761 loss=3.30848 test_loss_avg=3.41386 acc=0.18750 test_acc_avg=0.20820 test_acc_top5_avg=0.81250 time=238.39it/s +epoch=70 global_step=27761 loss=0.11176 test_loss_avg=2.97973 acc=0.96875 test_acc_avg=0.33058 test_acc_top5_avg=0.86741 time=253.63it/s +epoch=70 global_step=27761 loss=6.57150 test_loss_avg=3.35726 acc=0.00000 test_acc_avg=0.29678 test_acc_top5_avg=0.86877 time=801.36it/s +curr_acc 0.2968 +BEST_ACC 0.3411 +curr_acc_top5 0.8688 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=5.01862 loss_avg=5.25232 acc=0.53125 acc_top1_avg=0.48718 acc_top5_avg=0.85677 lr=0.00100 gn=31.14159 time=55.45it/s +epoch=71 global_step=27850 loss=5.69437 loss_avg=5.20683 acc=0.40625 acc_top1_avg=0.49184 acc_top5_avg=0.86104 lr=0.00100 gn=22.66588 time=55.44it/s +epoch=71 global_step=27900 loss=5.62938 loss_avg=5.23950 acc=0.42969 acc_top1_avg=0.48730 acc_top5_avg=0.86016 lr=0.00100 gn=22.05555 time=57.51it/s +epoch=71 global_step=27950 loss=5.20276 loss_avg=5.25663 acc=0.50781 acc_top1_avg=0.48508 acc_top5_avg=0.85904 lr=0.00100 gn=28.47388 time=52.93it/s +epoch=71 global_step=28000 loss=4.89161 loss_avg=5.26875 acc=0.50781 acc_top1_avg=0.48408 acc_top5_avg=0.85794 lr=0.00100 gn=25.92437 time=62.37it/s +epoch=71 global_step=28050 loss=5.32775 loss_avg=5.26809 acc=0.48438 acc_top1_avg=0.48348 acc_top5_avg=0.85878 lr=0.00100 gn=28.29742 time=63.14it/s +epoch=71 global_step=28100 loss=4.80397 loss_avg=5.27389 acc=0.52344 acc_top1_avg=0.48295 acc_top5_avg=0.85850 lr=0.00100 gn=23.10797 time=58.52it/s +epoch=71 global_step=28150 loss=5.55324 loss_avg=5.27096 acc=0.42969 acc_top1_avg=0.48327 acc_top5_avg=0.85825 lr=0.00100 gn=26.54006 time=61.14it/s +====================Eval==================== +epoch=71 global_step=28152 loss=1.34929 test_loss_avg=3.64816 acc=0.61719 test_acc_avg=0.20179 test_acc_top5_avg=0.75743 time=181.51it/s +epoch=71 global_step=28152 loss=6.86739 test_loss_avg=3.29887 acc=0.00000 test_acc_avg=0.32417 test_acc_top5_avg=0.85591 time=514.89it/s +curr_acc 0.3242 +BEST_ACC 0.3411 +curr_acc_top5 0.8559 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=5.73228 loss_avg=5.24367 acc=0.42188 acc_top1_avg=0.48454 acc_top5_avg=0.86117 lr=0.00100 gn=31.66214 time=55.95it/s +epoch=72 global_step=28250 loss=5.39061 loss_avg=5.24411 acc=0.46094 acc_top1_avg=0.48533 acc_top5_avg=0.86400 lr=0.00100 gn=26.54659 time=56.26it/s +epoch=72 global_step=28300 loss=5.33721 loss_avg=5.24728 acc=0.47656 acc_top1_avg=0.48596 acc_top5_avg=0.86186 lr=0.00100 gn=32.01885 time=53.63it/s +epoch=72 global_step=28350 loss=5.77561 loss_avg=5.27253 acc=0.42188 acc_top1_avg=0.48292 acc_top5_avg=0.86103 lr=0.00100 gn=29.80984 time=58.17it/s +epoch=72 global_step=28400 loss=5.39399 loss_avg=5.24780 acc=0.47656 acc_top1_avg=0.48592 acc_top5_avg=0.86045 lr=0.00100 gn=29.03762 time=56.03it/s +epoch=72 global_step=28450 loss=5.29674 loss_avg=5.24359 acc=0.49219 acc_top1_avg=0.48666 acc_top5_avg=0.86019 lr=0.00100 gn=27.93592 time=63.16it/s +epoch=72 global_step=28500 loss=4.70192 loss_avg=5.24213 acc=0.57031 acc_top1_avg=0.48660 acc_top5_avg=0.85938 lr=0.00100 gn=27.71434 time=57.95it/s +====================Eval==================== +epoch=72 global_step=28543 loss=1.53172 test_loss_avg=3.94477 acc=0.50781 test_acc_avg=0.17383 test_acc_top5_avg=0.74870 time=237.42it/s +epoch=72 global_step=28543 loss=0.65277 test_loss_avg=3.38953 acc=0.79688 test_acc_avg=0.26651 test_acc_top5_avg=0.84992 time=229.71it/s +epoch=72 global_step=28543 loss=7.00075 test_loss_avg=3.42853 acc=0.00000 test_acc_avg=0.30845 test_acc_top5_avg=0.86907 time=493.62it/s +curr_acc 0.3084 +BEST_ACC 0.3411 +curr_acc_top5 0.8691 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=5.33684 loss_avg=5.12089 acc=0.47656 acc_top1_avg=0.49665 acc_top5_avg=0.82812 lr=0.00100 gn=24.06892 time=55.74it/s +epoch=73 global_step=28600 loss=5.75699 loss_avg=5.08164 acc=0.43750 acc_top1_avg=0.50535 acc_top5_avg=0.85951 lr=0.00100 gn=31.37740 time=56.10it/s +epoch=73 global_step=28650 loss=4.27317 loss_avg=5.12929 acc=0.60938 acc_top1_avg=0.50007 acc_top5_avg=0.85916 lr=0.00100 gn=26.74093 time=57.38it/s +epoch=73 global_step=28700 loss=5.22873 loss_avg=5.16358 acc=0.49219 acc_top1_avg=0.49582 acc_top5_avg=0.86156 lr=0.00100 gn=30.94221 time=57.72it/s +epoch=73 global_step=28750 loss=5.48567 loss_avg=5.19832 acc=0.46875 acc_top1_avg=0.49238 acc_top5_avg=0.86062 lr=0.00100 gn=34.81104 time=57.74it/s +epoch=73 global_step=28800 loss=4.93451 loss_avg=5.21316 acc=0.50781 acc_top1_avg=0.49000 acc_top5_avg=0.85858 lr=0.00100 gn=30.18450 time=62.55it/s +epoch=73 global_step=28850 loss=5.38111 loss_avg=5.22341 acc=0.45312 acc_top1_avg=0.48870 acc_top5_avg=0.85851 lr=0.00100 gn=23.10346 time=55.08it/s +epoch=73 global_step=28900 loss=5.54014 loss_avg=5.22500 acc=0.42188 acc_top1_avg=0.48877 acc_top5_avg=0.85861 lr=0.00100 gn=33.92961 time=54.75it/s +====================Eval==================== +epoch=73 global_step=28934 loss=5.22006 test_loss_avg=3.47680 acc=0.00000 test_acc_avg=0.20312 test_acc_top5_avg=0.83523 time=237.84it/s +epoch=73 global_step=28934 loss=6.80477 test_loss_avg=3.27675 acc=0.00000 test_acc_avg=0.32189 test_acc_top5_avg=0.86719 time=487.94it/s +curr_acc 0.3219 +BEST_ACC 0.3411 +curr_acc_top5 0.8672 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=4.80087 loss_avg=5.15853 acc=0.52344 acc_top1_avg=0.49414 acc_top5_avg=0.87061 lr=0.00100 gn=25.95029 time=63.36it/s +epoch=74 global_step=29000 loss=4.56629 loss_avg=5.18156 acc=0.57812 acc_top1_avg=0.49325 acc_top5_avg=0.86648 lr=0.00100 gn=29.95845 time=52.53it/s +epoch=74 global_step=29050 loss=4.91703 loss_avg=5.16620 acc=0.55469 acc_top1_avg=0.49569 acc_top5_avg=0.86335 lr=0.00100 gn=28.63735 time=60.04it/s +epoch=74 global_step=29100 loss=5.29225 loss_avg=5.18393 acc=0.48438 acc_top1_avg=0.49388 acc_top5_avg=0.86093 lr=0.00100 gn=29.30559 time=54.46it/s +epoch=74 global_step=29150 loss=4.86894 loss_avg=5.17999 acc=0.54688 acc_top1_avg=0.49468 acc_top5_avg=0.85865 lr=0.00100 gn=32.07017 time=47.80it/s +epoch=74 global_step=29200 loss=5.45281 loss_avg=5.18564 acc=0.47656 acc_top1_avg=0.49413 acc_top5_avg=0.85838 lr=0.00100 gn=31.78179 time=54.40it/s +epoch=74 global_step=29250 loss=4.81059 loss_avg=5.19529 acc=0.53906 acc_top1_avg=0.49295 acc_top5_avg=0.85883 lr=0.00100 gn=26.69464 time=56.58it/s +epoch=74 global_step=29300 loss=5.55602 loss_avg=5.20485 acc=0.45312 acc_top1_avg=0.49197 acc_top5_avg=0.85925 lr=0.00100 gn=31.39109 time=59.69it/s +====================Eval==================== +epoch=74 global_step=29325 loss=5.13564 test_loss_avg=5.14199 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.65820 time=240.33it/s +epoch=74 global_step=29325 loss=5.21390 test_loss_avg=3.50545 acc=0.00000 test_acc_avg=0.21875 test_acc_top5_avg=0.82914 time=229.17it/s +epoch=74 global_step=29325 loss=7.18594 test_loss_avg=3.27206 acc=0.00000 test_acc_avg=0.32209 test_acc_top5_avg=0.86758 time=497.78it/s +curr_acc 0.3221 +BEST_ACC 0.3411 +curr_acc_top5 0.8676 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=5.35814 loss_avg=5.20093 acc=0.46094 acc_top1_avg=0.49000 acc_top5_avg=0.85625 lr=0.00100 gn=30.93333 time=54.73it/s +epoch=75 global_step=29400 loss=4.70797 loss_avg=5.25289 acc=0.57031 acc_top1_avg=0.48385 acc_top5_avg=0.85417 lr=0.00100 gn=35.08321 time=59.98it/s +epoch=75 global_step=29450 loss=4.97170 loss_avg=5.19901 acc=0.54688 acc_top1_avg=0.49144 acc_top5_avg=0.85481 lr=0.00100 gn=31.02589 time=59.98it/s +epoch=75 global_step=29500 loss=5.40102 loss_avg=5.18670 acc=0.44531 acc_top1_avg=0.49241 acc_top5_avg=0.85571 lr=0.00100 gn=31.38086 time=55.48it/s +epoch=75 global_step=29550 loss=4.80734 loss_avg=5.19119 acc=0.53125 acc_top1_avg=0.49191 acc_top5_avg=0.85622 lr=0.00100 gn=31.93552 time=63.82it/s +epoch=75 global_step=29600 loss=4.89266 loss_avg=5.18944 acc=0.52344 acc_top1_avg=0.49295 acc_top5_avg=0.85750 lr=0.00100 gn=28.41822 time=55.03it/s +epoch=75 global_step=29650 loss=5.30451 loss_avg=5.20598 acc=0.46875 acc_top1_avg=0.49106 acc_top5_avg=0.85663 lr=0.00100 gn=31.39298 time=27.06it/s +epoch=75 global_step=29700 loss=4.70527 loss_avg=5.20862 acc=0.56250 acc_top1_avg=0.49069 acc_top5_avg=0.85621 lr=0.00100 gn=36.22404 time=60.49it/s +====================Eval==================== +epoch=75 global_step=29716 loss=4.23241 test_loss_avg=3.44229 acc=0.10156 test_acc_avg=0.23688 test_acc_top5_avg=0.79531 time=238.72it/s +epoch=75 global_step=29716 loss=6.82863 test_loss_avg=3.23920 acc=0.00000 test_acc_avg=0.32677 test_acc_top5_avg=0.85229 time=245.02it/s +epoch=75 global_step=29716 loss=6.97470 test_loss_avg=3.40924 acc=0.00000 test_acc_avg=0.31023 test_acc_top5_avg=0.85403 time=810.96it/s +curr_acc 0.3102 +BEST_ACC 0.3411 +curr_acc_top5 0.8540 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=5.24803 loss_avg=5.21963 acc=0.49219 acc_top1_avg=0.48736 acc_top5_avg=0.85432 lr=0.00100 gn=36.44374 time=55.02it/s +epoch=76 global_step=29800 loss=4.98586 loss_avg=5.20749 acc=0.50000 acc_top1_avg=0.49042 acc_top5_avg=0.85510 lr=0.00100 gn=29.82606 time=59.71it/s +epoch=76 global_step=29850 loss=5.34458 loss_avg=5.16749 acc=0.46875 acc_top1_avg=0.49504 acc_top5_avg=0.85646 lr=0.00100 gn=33.31322 time=54.62it/s +epoch=76 global_step=29900 loss=5.11606 loss_avg=5.13667 acc=0.48438 acc_top1_avg=0.49949 acc_top5_avg=0.85802 lr=0.00100 gn=31.19700 time=58.44it/s +epoch=76 global_step=29950 loss=5.18857 loss_avg=5.14392 acc=0.50000 acc_top1_avg=0.49820 acc_top5_avg=0.85971 lr=0.00100 gn=35.54219 time=58.89it/s +epoch=76 global_step=30000 loss=5.07451 loss_avg=5.16743 acc=0.52344 acc_top1_avg=0.49576 acc_top5_avg=0.85902 lr=0.00100 gn=37.40067 time=55.19it/s +epoch=76 global_step=30050 loss=5.50685 loss_avg=5.18020 acc=0.46094 acc_top1_avg=0.49380 acc_top5_avg=0.85802 lr=0.00100 gn=36.34814 time=59.13it/s +epoch=76 global_step=30100 loss=5.35695 loss_avg=5.19211 acc=0.47656 acc_top1_avg=0.49280 acc_top5_avg=0.85789 lr=0.00100 gn=30.79593 time=41.97it/s +====================Eval==================== +epoch=76 global_step=30107 loss=1.72774 test_loss_avg=3.50044 acc=0.49219 test_acc_avg=0.20228 test_acc_top5_avg=0.77497 time=227.05it/s +epoch=76 global_step=30107 loss=6.75005 test_loss_avg=3.39678 acc=0.00000 test_acc_avg=0.30657 test_acc_top5_avg=0.85057 time=508.52it/s +curr_acc 0.3066 +BEST_ACC 0.3411 +curr_acc_top5 0.8506 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=5.33634 loss_avg=5.08323 acc=0.46875 acc_top1_avg=0.50400 acc_top5_avg=0.85683 lr=0.00100 gn=34.38816 time=58.70it/s +epoch=77 global_step=30200 loss=5.27227 loss_avg=5.12429 acc=0.47656 acc_top1_avg=0.49908 acc_top5_avg=0.85879 lr=0.00100 gn=31.86454 time=53.06it/s +epoch=77 global_step=30250 loss=4.90596 loss_avg=5.12938 acc=0.53125 acc_top1_avg=0.49863 acc_top5_avg=0.85916 lr=0.00100 gn=32.96062 time=49.81it/s +epoch=77 global_step=30300 loss=5.83106 loss_avg=5.15813 acc=0.43750 acc_top1_avg=0.49486 acc_top5_avg=0.85682 lr=0.00100 gn=35.46577 time=61.32it/s +epoch=77 global_step=30350 loss=5.15756 loss_avg=5.16686 acc=0.50000 acc_top1_avg=0.49447 acc_top5_avg=0.85796 lr=0.00100 gn=28.98479 time=61.47it/s +epoch=77 global_step=30400 loss=5.57292 loss_avg=5.17267 acc=0.43750 acc_top1_avg=0.49443 acc_top5_avg=0.85634 lr=0.00100 gn=36.99741 time=63.45it/s +epoch=77 global_step=30450 loss=5.14405 loss_avg=5.17049 acc=0.50000 acc_top1_avg=0.49462 acc_top5_avg=0.85648 lr=0.00100 gn=34.77327 time=58.71it/s +====================Eval==================== +epoch=77 global_step=30498 loss=3.41601 test_loss_avg=3.32851 acc=0.14062 test_acc_avg=0.30653 test_acc_top5_avg=0.80285 time=232.62it/s +epoch=77 global_step=30498 loss=0.21402 test_loss_avg=3.03706 acc=0.91406 test_acc_avg=0.33244 test_acc_top5_avg=0.85203 time=239.44it/s +epoch=77 global_step=30498 loss=6.94665 test_loss_avg=3.30058 acc=0.00000 test_acc_avg=0.32170 test_acc_top5_avg=0.86007 time=792.57it/s +curr_acc 0.3217 +BEST_ACC 0.3411 +curr_acc_top5 0.8601 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=5.04527 loss_avg=4.99856 acc=0.49219 acc_top1_avg=0.49609 acc_top5_avg=0.83984 lr=0.00100 gn=32.46340 time=38.53it/s +epoch=78 global_step=30550 loss=5.10311 loss_avg=5.08566 acc=0.52344 acc_top1_avg=0.50391 acc_top5_avg=0.85322 lr=0.00100 gn=33.95936 time=57.08it/s +epoch=78 global_step=30600 loss=4.58895 loss_avg=5.12167 acc=0.58594 acc_top1_avg=0.50038 acc_top5_avg=0.85486 lr=0.00100 gn=35.07562 time=57.37it/s +epoch=78 global_step=30650 loss=5.59405 loss_avg=5.11914 acc=0.44531 acc_top1_avg=0.50093 acc_top5_avg=0.85752 lr=0.00100 gn=34.35024 time=54.99it/s +epoch=78 global_step=30700 loss=5.16801 loss_avg=5.14474 acc=0.51562 acc_top1_avg=0.49822 acc_top5_avg=0.85647 lr=0.00100 gn=35.65829 time=56.11it/s +epoch=78 global_step=30750 loss=5.12509 loss_avg=5.13673 acc=0.51562 acc_top1_avg=0.49922 acc_top5_avg=0.85801 lr=0.00100 gn=39.58521 time=51.31it/s +epoch=78 global_step=30800 loss=5.08975 loss_avg=5.14772 acc=0.50781 acc_top1_avg=0.49816 acc_top5_avg=0.85736 lr=0.00100 gn=31.41530 time=61.43it/s +epoch=78 global_step=30850 loss=5.42587 loss_avg=5.15387 acc=0.47656 acc_top1_avg=0.49740 acc_top5_avg=0.85811 lr=0.00100 gn=30.76070 time=61.59it/s +====================Eval==================== +epoch=78 global_step=30889 loss=4.15251 test_loss_avg=3.38914 acc=0.03125 test_acc_avg=0.19757 test_acc_top5_avg=0.76254 time=235.50it/s +epoch=78 global_step=30889 loss=7.27078 test_loss_avg=3.26470 acc=0.00000 test_acc_avg=0.30380 test_acc_top5_avg=0.84711 time=840.21it/s +curr_acc 0.3038 +BEST_ACC 0.3411 +curr_acc_top5 0.8471 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=5.48874 loss_avg=5.12135 acc=0.46875 acc_top1_avg=0.50852 acc_top5_avg=0.86719 lr=0.00100 gn=36.69156 time=59.42it/s +epoch=79 global_step=30950 loss=5.02722 loss_avg=5.06863 acc=0.51562 acc_top1_avg=0.50781 acc_top5_avg=0.86130 lr=0.00100 gn=31.10578 time=62.57it/s +epoch=79 global_step=31000 loss=5.26348 loss_avg=5.08769 acc=0.50781 acc_top1_avg=0.50633 acc_top5_avg=0.85839 lr=0.00100 gn=33.53952 time=57.50it/s +epoch=79 global_step=31050 loss=5.48397 loss_avg=5.10493 acc=0.45312 acc_top1_avg=0.50388 acc_top5_avg=0.85986 lr=0.00100 gn=37.90755 time=37.92it/s +epoch=79 global_step=31100 loss=5.40011 loss_avg=5.13589 acc=0.43750 acc_top1_avg=0.50048 acc_top5_avg=0.85930 lr=0.00100 gn=37.33917 time=59.31it/s +epoch=79 global_step=31150 loss=4.54738 loss_avg=5.14789 acc=0.57812 acc_top1_avg=0.49916 acc_top5_avg=0.85797 lr=0.00100 gn=32.86907 time=45.77it/s +epoch=79 global_step=31200 loss=5.10557 loss_avg=5.15422 acc=0.53906 acc_top1_avg=0.49839 acc_top5_avg=0.85852 lr=0.00100 gn=44.82366 time=63.27it/s +epoch=79 global_step=31250 loss=5.04706 loss_avg=5.15561 acc=0.51562 acc_top1_avg=0.49827 acc_top5_avg=0.85916 lr=0.00100 gn=31.37749 time=54.21it/s +====================Eval==================== +epoch=79 global_step=31280 loss=1.43060 test_loss_avg=4.73526 acc=0.49219 test_acc_avg=0.06424 test_acc_top5_avg=0.62847 time=240.25it/s +epoch=79 global_step=31280 loss=0.64143 test_loss_avg=3.31291 acc=0.79688 test_acc_avg=0.24166 test_acc_top5_avg=0.82084 time=203.16it/s +epoch=79 global_step=31280 loss=7.42367 test_loss_avg=3.26082 acc=0.00000 test_acc_avg=0.31161 test_acc_top5_avg=0.85403 time=503.88it/s +curr_acc 0.3116 +BEST_ACC 0.3411 +curr_acc_top5 0.8540 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=4.81392 loss_avg=4.97308 acc=0.54688 acc_top1_avg=0.51914 acc_top5_avg=0.86367 lr=0.00010 gn=35.12630 time=60.17it/s +epoch=80 global_step=31350 loss=5.36938 loss_avg=5.04833 acc=0.45312 acc_top1_avg=0.50614 acc_top5_avg=0.85871 lr=0.00010 gn=32.16255 time=62.07it/s +epoch=80 global_step=31400 loss=5.08136 loss_avg=5.01136 acc=0.50000 acc_top1_avg=0.51133 acc_top5_avg=0.86289 lr=0.00010 gn=29.78840 time=54.59it/s +epoch=80 global_step=31450 loss=4.97011 loss_avg=5.00770 acc=0.51562 acc_top1_avg=0.51222 acc_top5_avg=0.86039 lr=0.00010 gn=34.00518 time=53.52it/s +epoch=80 global_step=31500 loss=4.43690 loss_avg=4.97194 acc=0.55469 acc_top1_avg=0.51690 acc_top5_avg=0.86083 lr=0.00010 gn=30.92311 time=58.70it/s +epoch=80 global_step=31550 loss=5.19042 loss_avg=4.97587 acc=0.49219 acc_top1_avg=0.51652 acc_top5_avg=0.86053 lr=0.00010 gn=35.79241 time=54.28it/s +epoch=80 global_step=31600 loss=4.91281 loss_avg=4.97434 acc=0.51562 acc_top1_avg=0.51709 acc_top5_avg=0.86069 lr=0.00010 gn=32.14993 time=63.41it/s +epoch=80 global_step=31650 loss=5.54230 loss_avg=4.97469 acc=0.46875 acc_top1_avg=0.51689 acc_top5_avg=0.85980 lr=0.00010 gn=49.54173 time=53.76it/s +====================Eval==================== +epoch=80 global_step=31671 loss=3.32803 test_loss_avg=3.38853 acc=0.15625 test_acc_avg=0.20938 test_acc_top5_avg=0.82474 time=240.93it/s +epoch=80 global_step=31671 loss=7.24402 test_loss_avg=3.28530 acc=0.00000 test_acc_avg=0.31161 test_acc_top5_avg=0.85977 time=493.85it/s +curr_acc 0.3116 +BEST_ACC 0.3411 +curr_acc_top5 0.8598 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=4.92887 loss_avg=4.91266 acc=0.52344 acc_top1_avg=0.52452 acc_top5_avg=0.85264 lr=0.00010 gn=33.37168 time=58.16it/s +epoch=81 global_step=31750 loss=5.08131 loss_avg=4.95589 acc=0.50000 acc_top1_avg=0.52126 acc_top5_avg=0.85858 lr=0.00010 gn=27.70857 time=54.08it/s +epoch=81 global_step=31800 loss=5.51506 loss_avg=4.93298 acc=0.44531 acc_top1_avg=0.52380 acc_top5_avg=0.85950 lr=0.00010 gn=29.61873 time=63.80it/s +epoch=81 global_step=31850 loss=4.90695 loss_avg=4.90322 acc=0.53906 acc_top1_avg=0.52710 acc_top5_avg=0.85994 lr=0.00010 gn=39.61395 time=60.57it/s +epoch=81 global_step=31900 loss=5.01153 loss_avg=4.90455 acc=0.50781 acc_top1_avg=0.52664 acc_top5_avg=0.85890 lr=0.00010 gn=34.20877 time=55.88it/s +epoch=81 global_step=31950 loss=4.52371 loss_avg=4.91482 acc=0.57031 acc_top1_avg=0.52503 acc_top5_avg=0.85884 lr=0.00010 gn=39.84917 time=50.07it/s +epoch=81 global_step=32000 loss=4.79865 loss_avg=4.92441 acc=0.52344 acc_top1_avg=0.52386 acc_top5_avg=0.85800 lr=0.00010 gn=31.36837 time=55.14it/s +epoch=81 global_step=32050 loss=4.25511 loss_avg=4.92446 acc=0.60156 acc_top1_avg=0.52393 acc_top5_avg=0.85810 lr=0.00010 gn=36.16103 time=46.00it/s +====================Eval==================== +epoch=81 global_step=32062 loss=5.43297 test_loss_avg=5.43297 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.61719 time=210.49it/s +epoch=81 global_step=32062 loss=5.35423 test_loss_avg=3.46531 acc=0.00000 test_acc_avg=0.19960 test_acc_top5_avg=0.80898 time=238.02it/s +epoch=81 global_step=32062 loss=7.20578 test_loss_avg=3.29033 acc=0.00000 test_acc_avg=0.30479 test_acc_top5_avg=0.86254 time=490.45it/s +curr_acc 0.3048 +BEST_ACC 0.3411 +curr_acc_top5 0.8625 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=5.02257 loss_avg=5.01220 acc=0.50000 acc_top1_avg=0.51377 acc_top5_avg=0.85526 lr=0.00010 gn=26.84507 time=52.60it/s +epoch=82 global_step=32150 loss=4.87587 loss_avg=4.90363 acc=0.53125 acc_top1_avg=0.52566 acc_top5_avg=0.85813 lr=0.00010 gn=30.87472 time=55.94it/s +epoch=82 global_step=32200 loss=3.84417 loss_avg=4.86976 acc=0.66406 acc_top1_avg=0.52848 acc_top5_avg=0.86017 lr=0.00010 gn=38.23357 time=54.24it/s +epoch=82 global_step=32250 loss=5.12974 loss_avg=4.89473 acc=0.50000 acc_top1_avg=0.52589 acc_top5_avg=0.85834 lr=0.00010 gn=29.09951 time=62.62it/s +epoch=82 global_step=32300 loss=4.72408 loss_avg=4.88488 acc=0.54688 acc_top1_avg=0.52705 acc_top5_avg=0.85964 lr=0.00010 gn=34.34172 time=61.88it/s +epoch=82 global_step=32350 loss=5.75581 loss_avg=4.90004 acc=0.42188 acc_top1_avg=0.52528 acc_top5_avg=0.86003 lr=0.00010 gn=31.54524 time=59.08it/s +epoch=82 global_step=32400 loss=4.30130 loss_avg=4.89676 acc=0.57812 acc_top1_avg=0.52577 acc_top5_avg=0.85991 lr=0.00010 gn=37.80121 time=52.34it/s +epoch=82 global_step=32450 loss=4.56328 loss_avg=4.89985 acc=0.56250 acc_top1_avg=0.52545 acc_top5_avg=0.86014 lr=0.00010 gn=37.33553 time=57.22it/s +====================Eval==================== +epoch=82 global_step=32453 loss=3.05320 test_loss_avg=3.28875 acc=0.25781 test_acc_avg=0.23580 test_acc_top5_avg=0.79510 time=240.66it/s +epoch=82 global_step=32453 loss=6.47150 test_loss_avg=2.92224 acc=0.00000 test_acc_avg=0.34429 test_acc_top5_avg=0.85189 time=241.07it/s +epoch=82 global_step=32453 loss=7.29514 test_loss_avg=3.26197 acc=0.00000 test_acc_avg=0.31379 test_acc_top5_avg=0.85710 time=508.09it/s +curr_acc 0.3138 +BEST_ACC 0.3411 +curr_acc_top5 0.8571 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=4.40187 loss_avg=4.72003 acc=0.61719 acc_top1_avg=0.54737 acc_top5_avg=0.86935 lr=0.00010 gn=35.50570 time=60.94it/s +epoch=83 global_step=32550 loss=4.82296 loss_avg=4.81902 acc=0.53906 acc_top1_avg=0.53745 acc_top5_avg=0.86477 lr=0.00010 gn=33.34626 time=54.94it/s +epoch=83 global_step=32600 loss=5.37288 loss_avg=4.83995 acc=0.50000 acc_top1_avg=0.53375 acc_top5_avg=0.86299 lr=0.00010 gn=35.80421 time=56.90it/s +epoch=83 global_step=32650 loss=5.33782 loss_avg=4.83795 acc=0.45312 acc_top1_avg=0.53339 acc_top5_avg=0.86215 lr=0.00010 gn=40.97382 time=61.64it/s +epoch=83 global_step=32700 loss=5.26997 loss_avg=4.84454 acc=0.48438 acc_top1_avg=0.53242 acc_top5_avg=0.86099 lr=0.00010 gn=37.54827 time=48.20it/s +epoch=83 global_step=32750 loss=4.73074 loss_avg=4.85181 acc=0.53906 acc_top1_avg=0.53141 acc_top5_avg=0.86103 lr=0.00010 gn=40.94780 time=50.06it/s +epoch=83 global_step=32800 loss=4.41446 loss_avg=4.86933 acc=0.57812 acc_top1_avg=0.52938 acc_top5_avg=0.85980 lr=0.00010 gn=38.75117 time=52.18it/s +====================Eval==================== +epoch=83 global_step=32844 loss=1.94192 test_loss_avg=3.39583 acc=0.43750 test_acc_avg=0.20658 test_acc_top5_avg=0.76581 time=248.68it/s +epoch=83 global_step=32844 loss=7.24850 test_loss_avg=3.26821 acc=0.00000 test_acc_avg=0.31388 test_acc_top5_avg=0.85413 time=825.49it/s +curr_acc 0.3139 +BEST_ACC 0.3411 +curr_acc_top5 0.8541 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=5.29041 loss_avg=4.89269 acc=0.48438 acc_top1_avg=0.51953 acc_top5_avg=0.85286 lr=0.00010 gn=35.02338 time=55.51it/s +epoch=84 global_step=32900 loss=5.11817 loss_avg=4.92841 acc=0.48438 acc_top1_avg=0.52121 acc_top5_avg=0.85742 lr=0.00010 gn=35.99728 time=61.62it/s +epoch=84 global_step=32950 loss=4.48746 loss_avg=4.87078 acc=0.56250 acc_top1_avg=0.52882 acc_top5_avg=0.86240 lr=0.00010 gn=33.13436 time=53.55it/s +epoch=84 global_step=33000 loss=4.82021 loss_avg=4.84864 acc=0.50781 acc_top1_avg=0.53165 acc_top5_avg=0.86263 lr=0.00010 gn=29.94139 time=53.97it/s +epoch=84 global_step=33050 loss=5.19445 loss_avg=4.84004 acc=0.49219 acc_top1_avg=0.53144 acc_top5_avg=0.86180 lr=0.00010 gn=34.70880 time=58.65it/s +epoch=84 global_step=33100 loss=4.76086 loss_avg=4.84807 acc=0.53906 acc_top1_avg=0.53070 acc_top5_avg=0.86047 lr=0.00010 gn=36.96615 time=55.86it/s +epoch=84 global_step=33150 loss=6.03745 loss_avg=4.84348 acc=0.42188 acc_top1_avg=0.53125 acc_top5_avg=0.86114 lr=0.00010 gn=34.49751 time=60.04it/s +epoch=84 global_step=33200 loss=4.39418 loss_avg=4.84770 acc=0.57812 acc_top1_avg=0.53127 acc_top5_avg=0.86085 lr=0.00010 gn=40.96650 time=54.30it/s +====================Eval==================== +epoch=84 global_step=33235 loss=2.09823 test_loss_avg=3.80757 acc=0.37500 test_acc_avg=0.17411 test_acc_top5_avg=0.76228 time=106.45it/s +epoch=84 global_step=33235 loss=0.32051 test_loss_avg=3.12425 acc=0.90625 test_acc_avg=0.28357 test_acc_top5_avg=0.84448 time=240.83it/s +epoch=84 global_step=33235 loss=7.24259 test_loss_avg=3.28075 acc=0.00000 test_acc_avg=0.30489 test_acc_top5_avg=0.86284 time=827.12it/s +curr_acc 0.3049 +BEST_ACC 0.3411 +curr_acc_top5 0.8628 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=4.36393 loss_avg=4.79268 acc=0.58594 acc_top1_avg=0.53229 acc_top5_avg=0.85677 lr=0.00010 gn=37.62580 time=55.39it/s +epoch=85 global_step=33300 loss=4.29479 loss_avg=4.80390 acc=0.59375 acc_top1_avg=0.53534 acc_top5_avg=0.86046 lr=0.00010 gn=38.91262 time=56.22it/s +epoch=85 global_step=33350 loss=4.98552 loss_avg=4.82722 acc=0.50000 acc_top1_avg=0.53281 acc_top5_avg=0.85550 lr=0.00010 gn=31.97883 time=62.40it/s +epoch=85 global_step=33400 loss=5.10901 loss_avg=4.84962 acc=0.49219 acc_top1_avg=0.53059 acc_top5_avg=0.85573 lr=0.00010 gn=33.16851 time=54.79it/s +epoch=85 global_step=33450 loss=4.90111 loss_avg=4.86716 acc=0.53125 acc_top1_avg=0.52900 acc_top5_avg=0.85650 lr=0.00010 gn=43.80128 time=52.90it/s +epoch=85 global_step=33500 loss=4.95099 loss_avg=4.84713 acc=0.51562 acc_top1_avg=0.53172 acc_top5_avg=0.85710 lr=0.00010 gn=31.09320 time=51.85it/s +epoch=85 global_step=33550 loss=4.86643 loss_avg=4.84328 acc=0.52344 acc_top1_avg=0.53229 acc_top5_avg=0.85789 lr=0.00010 gn=30.75140 time=48.14it/s +epoch=85 global_step=33600 loss=4.88163 loss_avg=4.85205 acc=0.53125 acc_top1_avg=0.53129 acc_top5_avg=0.85747 lr=0.00010 gn=40.72668 time=63.72it/s +====================Eval==================== +epoch=85 global_step=33626 loss=4.41556 test_loss_avg=3.46166 acc=0.03125 test_acc_avg=0.18616 test_acc_top5_avg=0.79353 time=227.98it/s +epoch=85 global_step=33626 loss=7.36180 test_loss_avg=3.26893 acc=0.00000 test_acc_avg=0.31013 test_acc_top5_avg=0.85888 time=823.54it/s +curr_acc 0.3101 +BEST_ACC 0.3411 +curr_acc_top5 0.8589 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=4.92407 loss_avg=4.74622 acc=0.51562 acc_top1_avg=0.54199 acc_top5_avg=0.86491 lr=0.00010 gn=33.43177 time=60.36it/s +epoch=86 global_step=33700 loss=4.78296 loss_avg=4.75655 acc=0.52344 acc_top1_avg=0.54413 acc_top5_avg=0.86244 lr=0.00010 gn=45.12342 time=59.85it/s +epoch=86 global_step=33750 loss=4.93661 loss_avg=4.82085 acc=0.52344 acc_top1_avg=0.53604 acc_top5_avg=0.85982 lr=0.00010 gn=41.05849 time=58.37it/s +epoch=86 global_step=33800 loss=5.38576 loss_avg=4.83250 acc=0.46094 acc_top1_avg=0.53484 acc_top5_avg=0.85852 lr=0.00010 gn=36.51651 time=57.36it/s +epoch=86 global_step=33850 loss=5.23687 loss_avg=4.84819 acc=0.48438 acc_top1_avg=0.53244 acc_top5_avg=0.85624 lr=0.00010 gn=26.13507 time=56.90it/s +epoch=86 global_step=33900 loss=5.36549 loss_avg=4.85555 acc=0.45312 acc_top1_avg=0.53139 acc_top5_avg=0.85715 lr=0.00010 gn=33.75695 time=49.82it/s +epoch=86 global_step=33950 loss=4.55719 loss_avg=4.83632 acc=0.54688 acc_top1_avg=0.53388 acc_top5_avg=0.85819 lr=0.00010 gn=25.89875 time=55.64it/s +epoch=86 global_step=34000 loss=4.87642 loss_avg=4.82594 acc=0.53125 acc_top1_avg=0.53465 acc_top5_avg=0.85921 lr=0.00010 gn=31.51884 time=62.99it/s +====================Eval==================== +epoch=86 global_step=34017 loss=5.05729 test_loss_avg=5.23406 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.55208 time=242.80it/s +epoch=86 global_step=34017 loss=0.80774 test_loss_avg=3.48764 acc=0.73438 test_acc_avg=0.21401 test_acc_top5_avg=0.82115 time=235.48it/s +epoch=86 global_step=34017 loss=7.03783 test_loss_avg=3.25473 acc=0.00000 test_acc_avg=0.31309 test_acc_top5_avg=0.85928 time=822.57it/s +curr_acc 0.3131 +BEST_ACC 0.3411 +curr_acc_top5 0.8593 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=5.16143 loss_avg=4.83462 acc=0.50781 acc_top1_avg=0.53220 acc_top5_avg=0.85866 lr=0.00010 gn=40.46045 time=62.44it/s +epoch=87 global_step=34100 loss=4.97938 loss_avg=4.83717 acc=0.53125 acc_top1_avg=0.53445 acc_top5_avg=0.85966 lr=0.00010 gn=43.17358 time=59.52it/s +epoch=87 global_step=34150 loss=4.68705 loss_avg=4.80959 acc=0.55469 acc_top1_avg=0.53742 acc_top5_avg=0.86137 lr=0.00010 gn=38.23083 time=60.14it/s +epoch=87 global_step=34200 loss=5.13945 loss_avg=4.83534 acc=0.49219 acc_top1_avg=0.53437 acc_top5_avg=0.85993 lr=0.00010 gn=33.84552 time=51.97it/s +epoch=87 global_step=34250 loss=4.57644 loss_avg=4.83308 acc=0.55469 acc_top1_avg=0.53407 acc_top5_avg=0.86055 lr=0.00010 gn=30.68244 time=57.95it/s +epoch=87 global_step=34300 loss=5.38012 loss_avg=4.83507 acc=0.46875 acc_top1_avg=0.53409 acc_top5_avg=0.86053 lr=0.00010 gn=37.02970 time=56.12it/s +epoch=87 global_step=34350 loss=4.57172 loss_avg=4.82349 acc=0.56250 acc_top1_avg=0.53564 acc_top5_avg=0.86069 lr=0.00010 gn=31.30263 time=59.14it/s +epoch=87 global_step=34400 loss=5.04589 loss_avg=4.81887 acc=0.51562 acc_top1_avg=0.53568 acc_top5_avg=0.86052 lr=0.00010 gn=37.22975 time=59.66it/s +====================Eval==================== +epoch=87 global_step=34408 loss=3.35992 test_loss_avg=3.42594 acc=0.18750 test_acc_avg=0.20544 test_acc_top5_avg=0.81510 time=240.00it/s +epoch=87 global_step=34408 loss=6.30684 test_loss_avg=3.18604 acc=0.00000 test_acc_avg=0.31412 test_acc_top5_avg=0.85440 time=245.83it/s +epoch=87 global_step=34408 loss=7.04938 test_loss_avg=3.27597 acc=0.00000 test_acc_avg=0.30617 test_acc_top5_avg=0.85700 time=801.20it/s +curr_acc 0.3062 +BEST_ACC 0.3411 +curr_acc_top5 0.8570 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=4.33469 loss_avg=4.73986 acc=0.58594 acc_top1_avg=0.54408 acc_top5_avg=0.85938 lr=0.00010 gn=35.46755 time=54.20it/s +epoch=88 global_step=34500 loss=4.77712 loss_avg=4.78105 acc=0.52344 acc_top1_avg=0.53957 acc_top5_avg=0.86184 lr=0.00010 gn=33.65725 time=55.83it/s +epoch=88 global_step=34550 loss=4.58480 loss_avg=4.76303 acc=0.56250 acc_top1_avg=0.54165 acc_top5_avg=0.86141 lr=0.00010 gn=41.89467 time=60.01it/s +epoch=88 global_step=34600 loss=5.35121 loss_avg=4.79329 acc=0.46875 acc_top1_avg=0.53788 acc_top5_avg=0.86088 lr=0.00010 gn=41.08387 time=51.07it/s +epoch=88 global_step=34650 loss=4.32262 loss_avg=4.81213 acc=0.56250 acc_top1_avg=0.53499 acc_top5_avg=0.85960 lr=0.00010 gn=30.33922 time=61.68it/s +epoch=88 global_step=34700 loss=4.53444 loss_avg=4.80857 acc=0.55469 acc_top1_avg=0.53556 acc_top5_avg=0.85978 lr=0.00010 gn=33.66028 time=47.52it/s +epoch=88 global_step=34750 loss=4.42901 loss_avg=4.80319 acc=0.56250 acc_top1_avg=0.53621 acc_top5_avg=0.85912 lr=0.00010 gn=28.34371 time=55.05it/s +====================Eval==================== +epoch=88 global_step=34799 loss=5.15313 test_loss_avg=3.27963 acc=0.00000 test_acc_avg=0.22493 test_acc_top5_avg=0.79541 time=240.72it/s +epoch=88 global_step=34799 loss=7.09504 test_loss_avg=3.23695 acc=0.00000 test_acc_avg=0.31131 test_acc_top5_avg=0.85947 time=822.90it/s +curr_acc 0.3113 +BEST_ACC 0.3411 +curr_acc_top5 0.8595 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=5.21971 loss_avg=5.21971 acc=0.49219 acc_top1_avg=0.49219 acc_top5_avg=0.82812 lr=0.00010 gn=41.65485 time=29.72it/s +epoch=89 global_step=34850 loss=4.54658 loss_avg=4.79231 acc=0.55469 acc_top1_avg=0.53998 acc_top5_avg=0.85800 lr=0.00010 gn=36.66535 time=56.75it/s +epoch=89 global_step=34900 loss=4.71129 loss_avg=4.75547 acc=0.53906 acc_top1_avg=0.54401 acc_top5_avg=0.85744 lr=0.00010 gn=41.53011 time=54.41it/s +epoch=89 global_step=34950 loss=4.89773 loss_avg=4.75087 acc=0.53125 acc_top1_avg=0.54424 acc_top5_avg=0.85767 lr=0.00010 gn=45.43613 time=50.70it/s +epoch=89 global_step=35000 loss=4.86081 loss_avg=4.76523 acc=0.52344 acc_top1_avg=0.54233 acc_top5_avg=0.85887 lr=0.00010 gn=42.53878 time=58.49it/s +epoch=89 global_step=35050 loss=4.37103 loss_avg=4.76227 acc=0.60156 acc_top1_avg=0.54211 acc_top5_avg=0.85978 lr=0.00010 gn=42.72740 time=55.76it/s +epoch=89 global_step=35100 loss=4.92965 loss_avg=4.77072 acc=0.52344 acc_top1_avg=0.54122 acc_top5_avg=0.85958 lr=0.00010 gn=32.84646 time=55.16it/s +epoch=89 global_step=35150 loss=4.29991 loss_avg=4.78224 acc=0.57031 acc_top1_avg=0.53966 acc_top5_avg=0.85946 lr=0.00010 gn=34.93351 time=57.66it/s +====================Eval==================== +epoch=89 global_step=35190 loss=3.17618 test_loss_avg=3.40777 acc=0.17188 test_acc_avg=0.21834 test_acc_top5_avg=0.76480 time=242.80it/s +epoch=89 global_step=35190 loss=0.23863 test_loss_avg=2.95084 acc=0.96094 test_acc_avg=0.32960 test_acc_top5_avg=0.83832 time=224.93it/s +epoch=89 global_step=35190 loss=7.04712 test_loss_avg=3.29679 acc=0.00000 test_acc_avg=0.30380 test_acc_top5_avg=0.84691 time=826.14it/s +curr_acc 0.3038 +BEST_ACC 0.3411 +curr_acc_top5 0.8469 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=4.65462 loss_avg=5.11920 acc=0.53906 acc_top1_avg=0.50000 acc_top5_avg=0.84609 lr=0.00010 gn=36.94307 time=53.05it/s +epoch=90 global_step=35250 loss=4.59027 loss_avg=4.78621 acc=0.57031 acc_top1_avg=0.53646 acc_top5_avg=0.85156 lr=0.00010 gn=35.77062 time=53.94it/s +epoch=90 global_step=35300 loss=4.70034 loss_avg=4.77939 acc=0.54688 acc_top1_avg=0.53828 acc_top5_avg=0.85675 lr=0.00010 gn=40.35583 time=52.92it/s +epoch=90 global_step=35350 loss=4.69891 loss_avg=4.80696 acc=0.54688 acc_top1_avg=0.53623 acc_top5_avg=0.85640 lr=0.00010 gn=32.04974 time=56.46it/s +epoch=90 global_step=35400 loss=4.59113 loss_avg=4.79025 acc=0.56250 acc_top1_avg=0.53806 acc_top5_avg=0.85811 lr=0.00010 gn=34.99027 time=61.83it/s +epoch=90 global_step=35450 loss=4.92031 loss_avg=4.78119 acc=0.53906 acc_top1_avg=0.53912 acc_top5_avg=0.85965 lr=0.00010 gn=44.76376 time=54.76it/s +epoch=90 global_step=35500 loss=4.41630 loss_avg=4.79046 acc=0.58594 acc_top1_avg=0.53863 acc_top5_avg=0.85930 lr=0.00010 gn=33.25728 time=61.47it/s +epoch=90 global_step=35550 loss=4.78489 loss_avg=4.79611 acc=0.53906 acc_top1_avg=0.53774 acc_top5_avg=0.85968 lr=0.00010 gn=37.65790 time=54.33it/s +====================Eval==================== +epoch=90 global_step=35581 loss=1.98454 test_loss_avg=3.52897 acc=0.44531 test_acc_avg=0.18223 test_acc_top5_avg=0.76914 time=239.76it/s +epoch=90 global_step=35581 loss=6.73438 test_loss_avg=3.21914 acc=0.00000 test_acc_avg=0.31062 test_acc_top5_avg=0.85750 time=500.87it/s +curr_acc 0.3106 +BEST_ACC 0.3411 +curr_acc_top5 0.8575 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=4.08650 loss_avg=4.72608 acc=0.60938 acc_top1_avg=0.54605 acc_top5_avg=0.86020 lr=0.00010 gn=41.82990 time=61.39it/s +epoch=91 global_step=35650 loss=5.01728 loss_avg=4.77465 acc=0.51562 acc_top1_avg=0.53918 acc_top5_avg=0.85971 lr=0.00010 gn=43.90226 time=54.94it/s +epoch=91 global_step=35700 loss=5.27115 loss_avg=4.78002 acc=0.48438 acc_top1_avg=0.53762 acc_top5_avg=0.85997 lr=0.00010 gn=34.71748 time=61.55it/s +epoch=91 global_step=35750 loss=4.24497 loss_avg=4.74332 acc=0.59375 acc_top1_avg=0.54184 acc_top5_avg=0.86293 lr=0.00010 gn=37.78243 time=49.05it/s +epoch=91 global_step=35800 loss=4.75259 loss_avg=4.74880 acc=0.57031 acc_top1_avg=0.54160 acc_top5_avg=0.86284 lr=0.00010 gn=43.46343 time=50.04it/s +epoch=91 global_step=35850 loss=4.57598 loss_avg=4.74708 acc=0.55469 acc_top1_avg=0.54229 acc_top5_avg=0.86277 lr=0.00010 gn=38.84497 time=58.13it/s +epoch=91 global_step=35900 loss=5.02823 loss_avg=4.76130 acc=0.50781 acc_top1_avg=0.54097 acc_top5_avg=0.86175 lr=0.00010 gn=48.19124 time=59.60it/s +epoch=91 global_step=35950 loss=5.36607 loss_avg=4.78237 acc=0.43750 acc_top1_avg=0.53824 acc_top5_avg=0.86045 lr=0.00010 gn=33.60393 time=59.98it/s +====================Eval==================== +epoch=91 global_step=35972 loss=2.07190 test_loss_avg=4.11762 acc=0.43750 test_acc_avg=0.13352 test_acc_top5_avg=0.70312 time=242.14it/s +epoch=91 global_step=35972 loss=0.65876 test_loss_avg=3.21167 acc=0.80469 test_acc_avg=0.26217 test_acc_top5_avg=0.83299 time=238.86it/s +epoch=91 global_step=35972 loss=7.07586 test_loss_avg=3.23407 acc=0.00000 test_acc_avg=0.31171 test_acc_top5_avg=0.85809 time=498.31it/s +curr_acc 0.3117 +BEST_ACC 0.3411 +curr_acc_top5 0.8581 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=5.07964 loss_avg=4.83166 acc=0.51562 acc_top1_avg=0.52930 acc_top5_avg=0.85910 lr=0.00010 gn=43.06393 time=56.18it/s +epoch=92 global_step=36050 loss=4.23155 loss_avg=4.74754 acc=0.62500 acc_top1_avg=0.54257 acc_top5_avg=0.86348 lr=0.00010 gn=44.58918 time=54.67it/s +epoch=92 global_step=36100 loss=4.75056 loss_avg=4.76791 acc=0.54688 acc_top1_avg=0.54108 acc_top5_avg=0.86041 lr=0.00010 gn=35.64531 time=59.07it/s +epoch=92 global_step=36150 loss=4.55021 loss_avg=4.75466 acc=0.55469 acc_top1_avg=0.54279 acc_top5_avg=0.85995 lr=0.00010 gn=34.26423 time=43.69it/s +epoch=92 global_step=36200 loss=4.98353 loss_avg=4.76281 acc=0.50781 acc_top1_avg=0.54242 acc_top5_avg=0.85879 lr=0.00010 gn=40.95100 time=53.91it/s +epoch=92 global_step=36250 loss=4.93457 loss_avg=4.76813 acc=0.51562 acc_top1_avg=0.54204 acc_top5_avg=0.85862 lr=0.00010 gn=54.51843 time=54.07it/s +epoch=92 global_step=36300 loss=4.49949 loss_avg=4.76765 acc=0.55469 acc_top1_avg=0.54185 acc_top5_avg=0.85933 lr=0.00010 gn=37.49737 time=40.57it/s +epoch=92 global_step=36350 loss=5.05695 loss_avg=4.77113 acc=0.52344 acc_top1_avg=0.54119 acc_top5_avg=0.85931 lr=0.00010 gn=40.65232 time=58.19it/s +====================Eval==================== +epoch=92 global_step=36363 loss=4.11688 test_loss_avg=3.37711 acc=0.05469 test_acc_avg=0.20605 test_acc_top5_avg=0.82007 time=223.21it/s +epoch=92 global_step=36363 loss=6.88344 test_loss_avg=3.25224 acc=0.00000 test_acc_avg=0.30884 test_acc_top5_avg=0.85591 time=822.74it/s +curr_acc 0.3088 +BEST_ACC 0.3411 +curr_acc_top5 0.8559 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=4.73923 loss_avg=4.79223 acc=0.53906 acc_top1_avg=0.53463 acc_top5_avg=0.85621 lr=0.00010 gn=41.17405 time=57.20it/s +epoch=93 global_step=36450 loss=5.10819 loss_avg=4.73902 acc=0.52344 acc_top1_avg=0.54310 acc_top5_avg=0.85803 lr=0.00010 gn=39.26582 time=58.58it/s +epoch=93 global_step=36500 loss=4.92067 loss_avg=4.77177 acc=0.52344 acc_top1_avg=0.54094 acc_top5_avg=0.85806 lr=0.00010 gn=38.63087 time=61.29it/s +epoch=93 global_step=36550 loss=5.47456 loss_avg=4.78274 acc=0.46875 acc_top1_avg=0.53977 acc_top5_avg=0.86084 lr=0.00010 gn=36.49549 time=58.71it/s +epoch=93 global_step=36600 loss=5.15117 loss_avg=4.77103 acc=0.49219 acc_top1_avg=0.54094 acc_top5_avg=0.85934 lr=0.00010 gn=40.16009 time=52.58it/s +epoch=93 global_step=36650 loss=5.09248 loss_avg=4.76852 acc=0.50781 acc_top1_avg=0.54124 acc_top5_avg=0.86016 lr=0.00010 gn=35.86232 time=61.26it/s +epoch=93 global_step=36700 loss=4.60771 loss_avg=4.77602 acc=0.55469 acc_top1_avg=0.54006 acc_top5_avg=0.85900 lr=0.00010 gn=39.61595 time=53.28it/s +epoch=93 global_step=36750 loss=4.34641 loss_avg=4.76466 acc=0.57031 acc_top1_avg=0.54138 acc_top5_avg=0.85976 lr=0.00010 gn=37.54605 time=63.38it/s +====================Eval==================== +epoch=93 global_step=36754 loss=5.30260 test_loss_avg=5.25894 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.54688 time=153.86it/s +epoch=93 global_step=36754 loss=5.65625 test_loss_avg=3.53484 acc=0.00000 test_acc_avg=0.20371 test_acc_top5_avg=0.80911 time=242.67it/s +epoch=93 global_step=36754 loss=6.82409 test_loss_avg=3.26267 acc=0.00000 test_acc_avg=0.30924 test_acc_top5_avg=0.85552 time=802.74it/s +curr_acc 0.3092 +BEST_ACC 0.3411 +curr_acc_top5 0.8555 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=4.16942 loss_avg=4.64551 acc=0.61719 acc_top1_avg=0.55537 acc_top5_avg=0.86141 lr=0.00010 gn=37.67174 time=60.80it/s +epoch=94 global_step=36850 loss=4.71445 loss_avg=4.71397 acc=0.51562 acc_top1_avg=0.54712 acc_top5_avg=0.86051 lr=0.00010 gn=36.71244 time=46.58it/s +epoch=94 global_step=36900 loss=5.00419 loss_avg=4.70717 acc=0.50781 acc_top1_avg=0.54896 acc_top5_avg=0.86221 lr=0.00010 gn=40.09642 time=55.48it/s +epoch=94 global_step=36950 loss=5.35350 loss_avg=4.75468 acc=0.46875 acc_top1_avg=0.54337 acc_top5_avg=0.86041 lr=0.00010 gn=40.47831 time=53.54it/s +epoch=94 global_step=37000 loss=4.70887 loss_avg=4.75535 acc=0.56250 acc_top1_avg=0.54316 acc_top5_avg=0.86030 lr=0.00010 gn=37.79639 time=54.98it/s +epoch=94 global_step=37050 loss=5.00249 loss_avg=4.75521 acc=0.54688 acc_top1_avg=0.54302 acc_top5_avg=0.86072 lr=0.00010 gn=46.07262 time=62.30it/s +epoch=94 global_step=37100 loss=4.30908 loss_avg=4.76038 acc=0.60938 acc_top1_avg=0.54265 acc_top5_avg=0.86037 lr=0.00010 gn=38.18485 time=50.40it/s +====================Eval==================== +epoch=94 global_step=37145 loss=3.44794 test_loss_avg=3.31481 acc=0.13281 test_acc_avg=0.22461 test_acc_top5_avg=0.80208 time=236.79it/s +epoch=94 global_step=37145 loss=6.50494 test_loss_avg=3.04824 acc=0.00000 test_acc_avg=0.32580 test_acc_top5_avg=0.85389 time=242.49it/s +epoch=94 global_step=37145 loss=6.93992 test_loss_avg=3.26630 acc=0.00000 test_acc_avg=0.30518 test_acc_top5_avg=0.85750 time=711.14it/s +curr_acc 0.3052 +BEST_ACC 0.3411 +curr_acc_top5 0.8575 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=4.29773 loss_avg=4.74525 acc=0.57812 acc_top1_avg=0.54219 acc_top5_avg=0.87344 lr=0.00010 gn=32.07255 time=58.97it/s +epoch=95 global_step=37200 loss=4.45607 loss_avg=4.70704 acc=0.58594 acc_top1_avg=0.54744 acc_top5_avg=0.86420 lr=0.00010 gn=39.85130 time=58.07it/s +epoch=95 global_step=37250 loss=4.65107 loss_avg=4.71578 acc=0.54688 acc_top1_avg=0.54717 acc_top5_avg=0.85811 lr=0.00010 gn=39.94408 time=57.93it/s +epoch=95 global_step=37300 loss=5.34147 loss_avg=4.74196 acc=0.46875 acc_top1_avg=0.54430 acc_top5_avg=0.85958 lr=0.00010 gn=35.17133 time=52.90it/s +epoch=95 global_step=37350 loss=4.60259 loss_avg=4.75644 acc=0.55469 acc_top1_avg=0.54272 acc_top5_avg=0.86094 lr=0.00010 gn=29.85404 time=54.78it/s +epoch=95 global_step=37400 loss=5.42007 loss_avg=4.75531 acc=0.49219 acc_top1_avg=0.54311 acc_top5_avg=0.86127 lr=0.00010 gn=45.75439 time=61.03it/s +epoch=95 global_step=37450 loss=4.48559 loss_avg=4.74782 acc=0.57031 acc_top1_avg=0.54365 acc_top5_avg=0.86050 lr=0.00010 gn=44.52929 time=54.65it/s +epoch=95 global_step=37500 loss=5.05497 loss_avg=4.75390 acc=0.51562 acc_top1_avg=0.54278 acc_top5_avg=0.86001 lr=0.00010 gn=37.72485 time=59.11it/s +====================Eval==================== +epoch=95 global_step=37536 loss=1.70070 test_loss_avg=3.31559 acc=0.55469 test_acc_avg=0.22760 test_acc_top5_avg=0.77986 time=236.90it/s +epoch=95 global_step=37536 loss=6.79731 test_loss_avg=3.24178 acc=0.00000 test_acc_avg=0.31408 test_acc_top5_avg=0.85463 time=556.20it/s +curr_acc 0.3141 +BEST_ACC 0.3411 +curr_acc_top5 0.8546 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=4.77945 loss_avg=4.74621 acc=0.53906 acc_top1_avg=0.53906 acc_top5_avg=0.86663 lr=0.00010 gn=33.56995 time=54.59it/s +epoch=96 global_step=37600 loss=4.92594 loss_avg=4.78505 acc=0.49219 acc_top1_avg=0.53918 acc_top5_avg=0.85474 lr=0.00010 gn=41.02966 time=53.71it/s +epoch=96 global_step=37650 loss=3.54732 loss_avg=4.75433 acc=0.67188 acc_top1_avg=0.54132 acc_top5_avg=0.85787 lr=0.00010 gn=36.38345 time=63.90it/s +epoch=96 global_step=37700 loss=4.25318 loss_avg=4.75477 acc=0.60938 acc_top1_avg=0.54192 acc_top5_avg=0.85795 lr=0.00010 gn=44.18125 time=53.31it/s +epoch=96 global_step=37750 loss=4.75816 loss_avg=4.74998 acc=0.51562 acc_top1_avg=0.54249 acc_top5_avg=0.85908 lr=0.00010 gn=39.10045 time=55.18it/s +epoch=96 global_step=37800 loss=4.07404 loss_avg=4.74785 acc=0.63281 acc_top1_avg=0.54244 acc_top5_avg=0.85946 lr=0.00010 gn=35.59748 time=56.30it/s +epoch=96 global_step=37850 loss=4.40864 loss_avg=4.75143 acc=0.58594 acc_top1_avg=0.54210 acc_top5_avg=0.85952 lr=0.00010 gn=37.98156 time=60.03it/s +epoch=96 global_step=37900 loss=4.91766 loss_avg=4.74092 acc=0.51562 acc_top1_avg=0.54340 acc_top5_avg=0.86060 lr=0.00010 gn=35.98775 time=53.60it/s +====================Eval==================== +epoch=96 global_step=37927 loss=2.21642 test_loss_avg=3.40868 acc=0.37500 test_acc_avg=0.23779 test_acc_top5_avg=0.77295 time=98.67it/s +epoch=96 global_step=37927 loss=0.13490 test_loss_avg=2.98669 acc=0.95312 test_acc_avg=0.31179 test_acc_top5_avg=0.84541 time=242.42it/s +epoch=96 global_step=37927 loss=6.88923 test_loss_avg=3.21309 acc=0.00000 test_acc_avg=0.31220 test_acc_top5_avg=0.85839 time=851.46it/s +curr_acc 0.3122 +BEST_ACC 0.3411 +curr_acc_top5 0.8584 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=4.87576 loss_avg=4.65222 acc=0.52344 acc_top1_avg=0.54959 acc_top5_avg=0.86413 lr=0.00010 gn=38.81686 time=52.69it/s +epoch=97 global_step=38000 loss=4.83902 loss_avg=4.70969 acc=0.53906 acc_top1_avg=0.54463 acc_top5_avg=0.86280 lr=0.00010 gn=40.38995 time=63.05it/s +epoch=97 global_step=38050 loss=4.88131 loss_avg=4.71736 acc=0.52344 acc_top1_avg=0.54478 acc_top5_avg=0.86230 lr=0.00010 gn=43.11821 time=62.92it/s +epoch=97 global_step=38100 loss=4.66706 loss_avg=4.72968 acc=0.54688 acc_top1_avg=0.54426 acc_top5_avg=0.86222 lr=0.00010 gn=37.81672 time=55.69it/s +epoch=97 global_step=38150 loss=4.21070 loss_avg=4.71908 acc=0.60156 acc_top1_avg=0.54593 acc_top5_avg=0.86190 lr=0.00010 gn=40.32310 time=56.42it/s +epoch=97 global_step=38200 loss=4.87184 loss_avg=4.72404 acc=0.53125 acc_top1_avg=0.54533 acc_top5_avg=0.86129 lr=0.00010 gn=34.45970 time=45.93it/s +epoch=97 global_step=38250 loss=4.86959 loss_avg=4.72138 acc=0.53906 acc_top1_avg=0.54617 acc_top5_avg=0.86083 lr=0.00010 gn=48.26964 time=54.31it/s +epoch=97 global_step=38300 loss=5.29203 loss_avg=4.72413 acc=0.47656 acc_top1_avg=0.54543 acc_top5_avg=0.85969 lr=0.00010 gn=41.44958 time=63.93it/s +====================Eval==================== +epoch=97 global_step=38318 loss=4.51988 test_loss_avg=3.49580 acc=0.04688 test_acc_avg=0.18476 test_acc_top5_avg=0.77154 time=235.52it/s +epoch=97 global_step=38318 loss=6.93493 test_loss_avg=3.23358 acc=0.00000 test_acc_avg=0.30815 test_acc_top5_avg=0.85384 time=823.54it/s +curr_acc 0.3081 +BEST_ACC 0.3411 +curr_acc_top5 0.8538 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=4.98987 loss_avg=4.66596 acc=0.52344 acc_top1_avg=0.55249 acc_top5_avg=0.85938 lr=0.00010 gn=33.13814 time=51.36it/s +epoch=98 global_step=38400 loss=4.56270 loss_avg=4.71099 acc=0.57031 acc_top1_avg=0.54649 acc_top5_avg=0.85928 lr=0.00010 gn=43.52071 time=55.37it/s +epoch=98 global_step=38450 loss=4.79096 loss_avg=4.69392 acc=0.53906 acc_top1_avg=0.54753 acc_top5_avg=0.85825 lr=0.00010 gn=34.72810 time=59.91it/s +epoch=98 global_step=38500 loss=4.70735 loss_avg=4.71762 acc=0.53906 acc_top1_avg=0.54606 acc_top5_avg=0.85792 lr=0.00010 gn=44.47825 time=56.46it/s +epoch=98 global_step=38550 loss=5.21009 loss_avg=4.71375 acc=0.48438 acc_top1_avg=0.54694 acc_top5_avg=0.85712 lr=0.00010 gn=39.11499 time=59.65it/s +epoch=98 global_step=38600 loss=4.49364 loss_avg=4.70709 acc=0.57812 acc_top1_avg=0.54748 acc_top5_avg=0.85874 lr=0.00010 gn=43.18113 time=55.04it/s +epoch=98 global_step=38650 loss=4.43722 loss_avg=4.71548 acc=0.57812 acc_top1_avg=0.54615 acc_top5_avg=0.85865 lr=0.00010 gn=42.70106 time=55.55it/s +epoch=98 global_step=38700 loss=4.91199 loss_avg=4.72075 acc=0.53125 acc_top1_avg=0.54546 acc_top5_avg=0.85899 lr=0.00010 gn=41.81454 time=63.54it/s +====================Eval==================== +epoch=98 global_step=38709 loss=4.47711 test_loss_avg=4.97542 acc=0.09375 test_acc_avg=0.01172 test_acc_top5_avg=0.60254 time=240.46it/s +epoch=98 global_step=38709 loss=0.48663 test_loss_avg=3.31269 acc=0.84375 test_acc_avg=0.23505 test_acc_top5_avg=0.82934 time=233.93it/s +epoch=98 global_step=38709 loss=6.94841 test_loss_avg=3.19079 acc=0.00000 test_acc_avg=0.31329 test_acc_top5_avg=0.85997 time=487.48it/s +curr_acc 0.3133 +BEST_ACC 0.3411 +curr_acc_top5 0.8600 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=5.54069 loss_avg=4.66067 acc=0.45312 acc_top1_avg=0.55240 acc_top5_avg=0.86471 lr=0.00010 gn=38.76295 time=54.15it/s +epoch=99 global_step=38800 loss=5.08998 loss_avg=4.66111 acc=0.50781 acc_top1_avg=0.55323 acc_top5_avg=0.86118 lr=0.00010 gn=36.15793 time=57.88it/s +epoch=99 global_step=38850 loss=4.82075 loss_avg=4.69574 acc=0.55469 acc_top1_avg=0.54992 acc_top5_avg=0.86209 lr=0.00010 gn=46.54090 time=55.64it/s +epoch=99 global_step=38900 loss=4.49517 loss_avg=4.70199 acc=0.57031 acc_top1_avg=0.54835 acc_top5_avg=0.86191 lr=0.00010 gn=41.83314 time=60.96it/s +epoch=99 global_step=38950 loss=4.10031 loss_avg=4.67732 acc=0.60938 acc_top1_avg=0.55067 acc_top5_avg=0.86268 lr=0.00010 gn=35.72839 time=51.18it/s +epoch=99 global_step=39000 loss=4.74896 loss_avg=4.68917 acc=0.55469 acc_top1_avg=0.54937 acc_top5_avg=0.86045 lr=0.00010 gn=38.71130 time=61.05it/s +epoch=99 global_step=39050 loss=4.66175 loss_avg=4.70267 acc=0.54688 acc_top1_avg=0.54811 acc_top5_avg=0.85926 lr=0.00010 gn=42.67165 time=54.80it/s +epoch=99 global_step=39100 loss=4.41697 loss_avg=4.71317 acc=0.58750 acc_top1_avg=0.54710 acc_top5_avg=0.85922 lr=0.00010 gn=52.34475 time=85.30it/s +====================Eval==================== +epoch=99 global_step=39100 loss=3.21032 test_loss_avg=3.28627 acc=0.17969 test_acc_avg=0.22629 test_acc_top5_avg=0.82435 time=241.43it/s +epoch=99 global_step=39100 loss=7.02014 test_loss_avg=3.22706 acc=0.00000 test_acc_avg=0.31675 test_acc_top5_avg=0.86056 time=505.03it/s +epoch=99 global_step=39100 loss=7.02014 test_loss_avg=3.22706 acc=0.00000 test_acc_avg=0.31675 test_acc_top5_avg=0.86056 time=505.03it/s +curr_acc 0.3168 +BEST_ACC 0.3411 +curr_acc_top5 0.8606 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=5.21369 loss_avg=4.74313 acc=0.48438 acc_top1_avg=0.54281 acc_top5_avg=0.85328 lr=0.00010 gn=32.94824 time=54.14it/s +epoch=100 global_step=39200 loss=4.43201 loss_avg=4.69911 acc=0.57812 acc_top1_avg=0.54742 acc_top5_avg=0.86000 lr=0.00010 gn=40.50388 time=61.56it/s +epoch=100 global_step=39250 loss=4.92916 loss_avg=4.70322 acc=0.53125 acc_top1_avg=0.54688 acc_top5_avg=0.86005 lr=0.00010 gn=40.22387 time=55.14it/s +epoch=100 global_step=39300 loss=4.47498 loss_avg=4.69971 acc=0.57812 acc_top1_avg=0.54801 acc_top5_avg=0.85930 lr=0.00010 gn=49.14364 time=60.29it/s +epoch=100 global_step=39350 loss=4.72895 loss_avg=4.70506 acc=0.55469 acc_top1_avg=0.54775 acc_top5_avg=0.85956 lr=0.00010 gn=41.43960 time=57.18it/s +epoch=100 global_step=39400 loss=5.29152 loss_avg=4.70553 acc=0.49219 acc_top1_avg=0.54810 acc_top5_avg=0.85875 lr=0.00010 gn=45.11587 time=61.74it/s +epoch=100 global_step=39450 loss=4.91256 loss_avg=4.69983 acc=0.50781 acc_top1_avg=0.54853 acc_top5_avg=0.85991 lr=0.00010 gn=34.62625 time=56.64it/s +====================Eval==================== +epoch=100 global_step=39491 loss=5.40807 test_loss_avg=3.34197 acc=0.00000 test_acc_avg=0.21516 test_acc_top5_avg=0.79563 time=242.25it/s +epoch=100 global_step=39491 loss=6.95646 test_loss_avg=3.22275 acc=0.00000 test_acc_avg=0.30696 test_acc_top5_avg=0.85394 time=497.96it/s +curr_acc 0.3070 +BEST_ACC 0.3411 +curr_acc_top5 0.8539 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=4.99789 loss_avg=4.88125 acc=0.51562 acc_top1_avg=0.52604 acc_top5_avg=0.86806 lr=0.00010 gn=40.62114 time=55.23it/s +epoch=101 global_step=39550 loss=4.88137 loss_avg=4.77706 acc=0.53906 acc_top1_avg=0.53933 acc_top5_avg=0.86269 lr=0.00010 gn=45.97303 time=60.20it/s +epoch=101 global_step=39600 loss=4.73492 loss_avg=4.73538 acc=0.53906 acc_top1_avg=0.54472 acc_top5_avg=0.86067 lr=0.00010 gn=39.79110 time=54.07it/s +epoch=101 global_step=39650 loss=4.65638 loss_avg=4.70468 acc=0.56250 acc_top1_avg=0.54751 acc_top5_avg=0.86409 lr=0.00010 gn=46.50070 time=59.70it/s +epoch=101 global_step=39700 loss=5.34704 loss_avg=4.70714 acc=0.48438 acc_top1_avg=0.54736 acc_top5_avg=0.86266 lr=0.00010 gn=40.61434 time=39.96it/s +epoch=101 global_step=39750 loss=5.53514 loss_avg=4.69692 acc=0.45312 acc_top1_avg=0.54823 acc_top5_avg=0.86260 lr=0.00010 gn=41.16800 time=47.02it/s +epoch=101 global_step=39800 loss=5.07411 loss_avg=4.70381 acc=0.50000 acc_top1_avg=0.54763 acc_top5_avg=0.86125 lr=0.00010 gn=39.92293 time=59.33it/s +epoch=101 global_step=39850 loss=4.75195 loss_avg=4.70253 acc=0.54688 acc_top1_avg=0.54790 acc_top5_avg=0.86083 lr=0.00010 gn=37.05118 time=62.78it/s +====================Eval==================== +epoch=101 global_step=39882 loss=2.78921 test_loss_avg=3.39733 acc=0.25781 test_acc_avg=0.22582 test_acc_top5_avg=0.77493 time=247.13it/s +epoch=101 global_step=39882 loss=4.41487 test_loss_avg=2.94341 acc=0.29688 test_acc_avg=0.34056 test_acc_top5_avg=0.84716 time=241.20it/s +epoch=101 global_step=39882 loss=7.11258 test_loss_avg=3.31160 acc=0.00000 test_acc_avg=0.30607 test_acc_top5_avg=0.85225 time=839.36it/s +curr_acc 0.3061 +BEST_ACC 0.3411 +curr_acc_top5 0.8523 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=5.01697 loss_avg=4.63009 acc=0.51562 acc_top1_avg=0.55642 acc_top5_avg=0.86458 lr=0.00010 gn=44.25927 time=60.67it/s +epoch=102 global_step=39950 loss=4.24765 loss_avg=4.61509 acc=0.59375 acc_top1_avg=0.55699 acc_top5_avg=0.86638 lr=0.00010 gn=40.57418 time=52.42it/s +epoch=102 global_step=40000 loss=4.37026 loss_avg=4.60504 acc=0.59375 acc_top1_avg=0.55846 acc_top5_avg=0.86514 lr=0.00010 gn=41.92937 time=58.66it/s +epoch=102 global_step=40050 loss=4.42150 loss_avg=4.64991 acc=0.57031 acc_top1_avg=0.55339 acc_top5_avg=0.86319 lr=0.00010 gn=41.55599 time=54.34it/s +epoch=102 global_step=40100 loss=4.46215 loss_avg=4.64965 acc=0.56250 acc_top1_avg=0.55257 acc_top5_avg=0.86138 lr=0.00010 gn=31.89060 time=31.39it/s +epoch=102 global_step=40150 loss=4.74708 loss_avg=4.68296 acc=0.55469 acc_top1_avg=0.54967 acc_top5_avg=0.86005 lr=0.00010 gn=41.69735 time=49.94it/s +epoch=102 global_step=40200 loss=4.68208 loss_avg=4.68565 acc=0.57031 acc_top1_avg=0.54923 acc_top5_avg=0.85994 lr=0.00010 gn=40.23891 time=46.66it/s +epoch=102 global_step=40250 loss=5.42498 loss_avg=4.68487 acc=0.48438 acc_top1_avg=0.54978 acc_top5_avg=0.86126 lr=0.00010 gn=46.79295 time=60.37it/s +====================Eval==================== +epoch=102 global_step=40273 loss=1.68642 test_loss_avg=3.39044 acc=0.47656 test_acc_avg=0.20517 test_acc_top5_avg=0.77083 time=231.21it/s +epoch=102 global_step=40273 loss=6.91884 test_loss_avg=3.25096 acc=0.00000 test_acc_avg=0.30568 test_acc_top5_avg=0.85423 time=512.56it/s +curr_acc 0.3057 +BEST_ACC 0.3411 +curr_acc_top5 0.8542 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=4.41163 loss_avg=4.82756 acc=0.58594 acc_top1_avg=0.53328 acc_top5_avg=0.85503 lr=0.00010 gn=38.59305 time=58.48it/s +epoch=103 global_step=40350 loss=4.32818 loss_avg=4.70149 acc=0.60156 acc_top1_avg=0.54840 acc_top5_avg=0.85714 lr=0.00010 gn=41.01960 time=56.80it/s +epoch=103 global_step=40400 loss=4.37582 loss_avg=4.68884 acc=0.59375 acc_top1_avg=0.54891 acc_top5_avg=0.86048 lr=0.00010 gn=41.74757 time=51.38it/s +epoch=103 global_step=40450 loss=4.58262 loss_avg=4.69404 acc=0.54688 acc_top1_avg=0.54855 acc_top5_avg=0.85955 lr=0.00010 gn=37.95621 time=55.85it/s +epoch=103 global_step=40500 loss=3.98450 loss_avg=4.68469 acc=0.64062 acc_top1_avg=0.54942 acc_top5_avg=0.86024 lr=0.00010 gn=45.41631 time=56.36it/s +epoch=103 global_step=40550 loss=4.17721 loss_avg=4.67874 acc=0.60938 acc_top1_avg=0.55054 acc_top5_avg=0.86047 lr=0.00010 gn=37.47225 time=50.61it/s +epoch=103 global_step=40600 loss=4.54984 loss_avg=4.68300 acc=0.57031 acc_top1_avg=0.54996 acc_top5_avg=0.86002 lr=0.00010 gn=36.25074 time=60.34it/s +epoch=103 global_step=40650 loss=5.17104 loss_avg=4.69671 acc=0.48438 acc_top1_avg=0.54855 acc_top5_avg=0.85931 lr=0.00010 gn=36.26819 time=60.41it/s +====================Eval==================== +epoch=103 global_step=40664 loss=1.71961 test_loss_avg=3.69579 acc=0.56250 test_acc_avg=0.19291 test_acc_top5_avg=0.74219 time=233.72it/s +epoch=103 global_step=40664 loss=0.25357 test_loss_avg=3.13853 acc=0.89844 test_acc_avg=0.27641 test_acc_top5_avg=0.83346 time=233.77it/s +epoch=103 global_step=40664 loss=6.65403 test_loss_avg=3.20864 acc=0.00000 test_acc_avg=0.30726 test_acc_top5_avg=0.85295 time=510.57it/s +curr_acc 0.3073 +BEST_ACC 0.3411 +curr_acc_top5 0.8529 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=4.90757 loss_avg=4.69653 acc=0.53906 acc_top1_avg=0.55360 acc_top5_avg=0.85872 lr=0.00010 gn=46.09764 time=58.55it/s +epoch=104 global_step=40750 loss=4.32958 loss_avg=4.72617 acc=0.59375 acc_top1_avg=0.54688 acc_top5_avg=0.85956 lr=0.00010 gn=44.87974 time=60.25it/s +epoch=104 global_step=40800 loss=4.95298 loss_avg=4.68743 acc=0.52344 acc_top1_avg=0.55153 acc_top5_avg=0.86150 lr=0.00010 gn=34.82639 time=58.28it/s +epoch=104 global_step=40850 loss=4.74143 loss_avg=4.68596 acc=0.55469 acc_top1_avg=0.55137 acc_top5_avg=0.86085 lr=0.00010 gn=38.07094 time=57.10it/s +epoch=104 global_step=40900 loss=5.05738 loss_avg=4.68135 acc=0.51562 acc_top1_avg=0.55131 acc_top5_avg=0.86047 lr=0.00010 gn=39.52217 time=59.63it/s +epoch=104 global_step=40950 loss=5.08983 loss_avg=4.67460 acc=0.50000 acc_top1_avg=0.55207 acc_top5_avg=0.86071 lr=0.00010 gn=36.76795 time=55.96it/s +epoch=104 global_step=41000 loss=4.70002 loss_avg=4.67623 acc=0.56250 acc_top1_avg=0.55190 acc_top5_avg=0.85989 lr=0.00010 gn=41.76512 time=54.72it/s +epoch=104 global_step=41050 loss=4.25762 loss_avg=4.67939 acc=0.59375 acc_top1_avg=0.55169 acc_top5_avg=0.86016 lr=0.00010 gn=32.24014 time=55.95it/s +====================Eval==================== +epoch=104 global_step=41055 loss=4.59590 test_loss_avg=3.47479 acc=0.01562 test_acc_avg=0.19210 test_acc_top5_avg=0.79779 time=227.94it/s +epoch=104 global_step=41055 loss=6.75742 test_loss_avg=3.27874 acc=0.00000 test_acc_avg=0.30192 test_acc_top5_avg=0.85176 time=505.09it/s +curr_acc 0.3019 +BEST_ACC 0.3411 +curr_acc_top5 0.8518 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=4.50131 loss_avg=4.59145 acc=0.58594 acc_top1_avg=0.56007 acc_top5_avg=0.86042 lr=0.00010 gn=45.89355 time=60.01it/s +epoch=105 global_step=41150 loss=4.04591 loss_avg=4.62912 acc=0.62500 acc_top1_avg=0.55625 acc_top5_avg=0.85855 lr=0.00010 gn=38.02501 time=57.88it/s +epoch=105 global_step=41200 loss=4.53181 loss_avg=4.63604 acc=0.55469 acc_top1_avg=0.55609 acc_top5_avg=0.85695 lr=0.00010 gn=32.65054 time=57.30it/s +epoch=105 global_step=41250 loss=5.26362 loss_avg=4.64843 acc=0.46875 acc_top1_avg=0.55489 acc_top5_avg=0.85753 lr=0.00010 gn=40.71373 time=59.32it/s +epoch=105 global_step=41300 loss=5.03103 loss_avg=4.65236 acc=0.51562 acc_top1_avg=0.55443 acc_top5_avg=0.85906 lr=0.00010 gn=47.68324 time=58.07it/s +epoch=105 global_step=41350 loss=4.51684 loss_avg=4.65890 acc=0.57031 acc_top1_avg=0.55358 acc_top5_avg=0.85900 lr=0.00010 gn=41.22014 time=62.79it/s +epoch=105 global_step=41400 loss=4.62613 loss_avg=4.66357 acc=0.53906 acc_top1_avg=0.55290 acc_top5_avg=0.85910 lr=0.00010 gn=35.13550 time=53.99it/s +====================Eval==================== +epoch=105 global_step=41446 loss=5.11685 test_loss_avg=5.27253 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.58281 time=238.31it/s +epoch=105 global_step=41446 loss=3.74039 test_loss_avg=3.53798 acc=0.23438 test_acc_avg=0.20625 test_acc_top5_avg=0.82372 time=202.06it/s +epoch=105 global_step=41446 loss=6.90452 test_loss_avg=3.26067 acc=0.00000 test_acc_avg=0.30845 test_acc_top5_avg=0.86145 time=721.91it/s +curr_acc 0.3084 +BEST_ACC 0.3411 +curr_acc_top5 0.8615 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=4.82307 loss_avg=4.75718 acc=0.52344 acc_top1_avg=0.53906 acc_top5_avg=0.85742 lr=0.00010 gn=43.96722 time=36.54it/s +epoch=106 global_step=41500 loss=5.49263 loss_avg=4.66130 acc=0.46094 acc_top1_avg=0.55295 acc_top5_avg=0.85952 lr=0.00010 gn=34.31224 time=55.45it/s +epoch=106 global_step=41550 loss=4.48768 loss_avg=4.62925 acc=0.57031 acc_top1_avg=0.55529 acc_top5_avg=0.86005 lr=0.00010 gn=32.24410 time=59.01it/s +epoch=106 global_step=41600 loss=4.46586 loss_avg=4.62848 acc=0.57031 acc_top1_avg=0.55621 acc_top5_avg=0.86105 lr=0.00010 gn=44.35462 time=60.15it/s +epoch=106 global_step=41650 loss=4.34205 loss_avg=4.64388 acc=0.58594 acc_top1_avg=0.55411 acc_top5_avg=0.86006 lr=0.00010 gn=43.33058 time=53.40it/s +epoch=106 global_step=41700 loss=4.99953 loss_avg=4.65807 acc=0.52344 acc_top1_avg=0.55287 acc_top5_avg=0.86011 lr=0.00010 gn=41.38205 time=57.50it/s +epoch=106 global_step=41750 loss=4.60279 loss_avg=4.66615 acc=0.53906 acc_top1_avg=0.55183 acc_top5_avg=0.85932 lr=0.00010 gn=48.32423 time=56.46it/s +epoch=106 global_step=41800 loss=4.47533 loss_avg=4.66570 acc=0.57812 acc_top1_avg=0.55173 acc_top5_avg=0.86010 lr=0.00010 gn=40.27942 time=55.11it/s +====================Eval==================== +epoch=106 global_step=41837 loss=2.88431 test_loss_avg=3.27865 acc=0.22656 test_acc_avg=0.22837 test_acc_top5_avg=0.80349 time=239.44it/s +epoch=106 global_step=41837 loss=6.18523 test_loss_avg=3.10477 acc=0.00000 test_acc_avg=0.31980 test_acc_top5_avg=0.85341 time=243.76it/s +epoch=106 global_step=41837 loss=6.72391 test_loss_avg=3.22796 acc=0.00000 test_acc_avg=0.30765 test_acc_top5_avg=0.85502 time=500.04it/s +curr_acc 0.3077 +BEST_ACC 0.3411 +curr_acc_top5 0.8550 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=4.30510 loss_avg=4.67377 acc=0.60938 acc_top1_avg=0.54748 acc_top5_avg=0.85156 lr=0.00010 gn=41.55419 time=55.02it/s +epoch=107 global_step=41900 loss=5.44607 loss_avg=4.69207 acc=0.47656 acc_top1_avg=0.54774 acc_top5_avg=0.85851 lr=0.00010 gn=47.27753 time=59.29it/s +epoch=107 global_step=41950 loss=4.60396 loss_avg=4.65785 acc=0.54688 acc_top1_avg=0.55241 acc_top5_avg=0.86138 lr=0.00010 gn=44.91654 time=63.73it/s +epoch=107 global_step=42000 loss=4.47196 loss_avg=4.66201 acc=0.57031 acc_top1_avg=0.55196 acc_top5_avg=0.86076 lr=0.00010 gn=42.27980 time=57.72it/s +epoch=107 global_step=42050 loss=4.28171 loss_avg=4.67185 acc=0.57812 acc_top1_avg=0.55128 acc_top5_avg=0.86044 lr=0.00010 gn=40.72772 time=55.03it/s +epoch=107 global_step=42100 loss=4.44765 loss_avg=4.66827 acc=0.57031 acc_top1_avg=0.55267 acc_top5_avg=0.86139 lr=0.00010 gn=36.04266 time=53.21it/s +epoch=107 global_step=42150 loss=5.36321 loss_avg=4.66733 acc=0.46875 acc_top1_avg=0.55267 acc_top5_avg=0.86117 lr=0.00010 gn=40.29672 time=55.91it/s +epoch=107 global_step=42200 loss=4.79336 loss_avg=4.66586 acc=0.53906 acc_top1_avg=0.55284 acc_top5_avg=0.86062 lr=0.00010 gn=35.44364 time=53.56it/s +====================Eval==================== +epoch=107 global_step=42228 loss=2.19179 test_loss_avg=3.25810 acc=0.47656 test_acc_avg=0.23172 test_acc_top5_avg=0.77793 time=163.60it/s +epoch=107 global_step=42228 loss=6.67117 test_loss_avg=3.25600 acc=0.00000 test_acc_avg=0.30696 test_acc_top5_avg=0.84919 time=832.70it/s +curr_acc 0.3070 +BEST_ACC 0.3411 +curr_acc_top5 0.8492 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=5.27176 loss_avg=4.60711 acc=0.49219 acc_top1_avg=0.56108 acc_top5_avg=0.86648 lr=0.00010 gn=43.95494 time=57.62it/s +epoch=108 global_step=42300 loss=4.63489 loss_avg=4.66599 acc=0.55469 acc_top1_avg=0.55469 acc_top5_avg=0.86589 lr=0.00010 gn=43.70976 time=50.93it/s +epoch=108 global_step=42350 loss=4.74694 loss_avg=4.65658 acc=0.53125 acc_top1_avg=0.55405 acc_top5_avg=0.86386 lr=0.00010 gn=39.40258 time=63.77it/s +epoch=108 global_step=42400 loss=4.35508 loss_avg=4.66717 acc=0.57812 acc_top1_avg=0.55296 acc_top5_avg=0.86215 lr=0.00010 gn=48.65750 time=58.76it/s +epoch=108 global_step=42450 loss=4.87304 loss_avg=4.67793 acc=0.53125 acc_top1_avg=0.55131 acc_top5_avg=0.86113 lr=0.00010 gn=42.91114 time=56.50it/s +epoch=108 global_step=42500 loss=5.12826 loss_avg=4.66463 acc=0.51562 acc_top1_avg=0.55276 acc_top5_avg=0.86124 lr=0.00010 gn=42.66644 time=47.09it/s +epoch=108 global_step=42550 loss=4.28037 loss_avg=4.66060 acc=0.60938 acc_top1_avg=0.55343 acc_top5_avg=0.86076 lr=0.00010 gn=50.23089 time=42.59it/s +epoch=108 global_step=42600 loss=4.55247 loss_avg=4.65409 acc=0.58594 acc_top1_avg=0.55452 acc_top5_avg=0.86169 lr=0.00010 gn=45.56994 time=57.72it/s +====================Eval==================== +epoch=108 global_step=42619 loss=2.85279 test_loss_avg=3.47487 acc=0.28125 test_acc_avg=0.22830 test_acc_top5_avg=0.75304 time=246.00it/s +epoch=108 global_step=42619 loss=0.32180 test_loss_avg=3.00712 acc=0.92969 test_acc_avg=0.31905 test_acc_top5_avg=0.84030 time=236.82it/s +epoch=108 global_step=42619 loss=6.66186 test_loss_avg=3.28218 acc=0.00000 test_acc_avg=0.30241 test_acc_top5_avg=0.84929 time=497.37it/s +curr_acc 0.3024 +BEST_ACC 0.3411 +curr_acc_top5 0.8493 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=4.65107 loss_avg=4.56856 acc=0.54688 acc_top1_avg=0.56351 acc_top5_avg=0.86013 lr=0.00010 gn=46.69100 time=60.30it/s +epoch=109 global_step=42700 loss=5.04538 loss_avg=4.64203 acc=0.51562 acc_top1_avg=0.55536 acc_top5_avg=0.85600 lr=0.00010 gn=35.09363 time=53.43it/s +epoch=109 global_step=42750 loss=4.63697 loss_avg=4.63696 acc=0.53125 acc_top1_avg=0.55618 acc_top5_avg=0.85812 lr=0.00010 gn=36.78264 time=57.73it/s +epoch=109 global_step=42800 loss=4.48630 loss_avg=4.63433 acc=0.58594 acc_top1_avg=0.55637 acc_top5_avg=0.85868 lr=0.00010 gn=47.74898 time=54.74it/s +epoch=109 global_step=42850 loss=4.03517 loss_avg=4.63408 acc=0.62500 acc_top1_avg=0.55716 acc_top5_avg=0.85907 lr=0.00010 gn=43.20554 time=60.88it/s +epoch=109 global_step=42900 loss=4.89253 loss_avg=4.64322 acc=0.52344 acc_top1_avg=0.55555 acc_top5_avg=0.85921 lr=0.00010 gn=45.96725 time=38.33it/s +epoch=109 global_step=42950 loss=4.53469 loss_avg=4.65015 acc=0.57031 acc_top1_avg=0.55490 acc_top5_avg=0.86077 lr=0.00010 gn=40.97759 time=57.83it/s +epoch=109 global_step=43000 loss=4.69864 loss_avg=4.64963 acc=0.55469 acc_top1_avg=0.55497 acc_top5_avg=0.86067 lr=0.00010 gn=41.36694 time=62.74it/s +====================Eval==================== +epoch=109 global_step=43010 loss=4.33362 test_loss_avg=3.51822 acc=0.01562 test_acc_avg=0.18490 test_acc_top5_avg=0.75280 time=239.36it/s +epoch=109 global_step=43010 loss=6.68944 test_loss_avg=3.24108 acc=0.00000 test_acc_avg=0.30904 test_acc_top5_avg=0.84889 time=631.96it/s +curr_acc 0.3090 +BEST_ACC 0.3411 +curr_acc_top5 0.8489 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=5.13435 loss_avg=4.64525 acc=0.49219 acc_top1_avg=0.55371 acc_top5_avg=0.85391 lr=0.00010 gn=45.38358 time=52.98it/s +epoch=110 global_step=43100 loss=4.76582 loss_avg=4.62700 acc=0.54688 acc_top1_avg=0.55634 acc_top5_avg=0.85642 lr=0.00010 gn=46.39200 time=55.15it/s +epoch=110 global_step=43150 loss=4.86257 loss_avg=4.62029 acc=0.52344 acc_top1_avg=0.55781 acc_top5_avg=0.86032 lr=0.00010 gn=40.28784 time=53.96it/s +epoch=110 global_step=43200 loss=4.14518 loss_avg=4.61335 acc=0.60938 acc_top1_avg=0.55847 acc_top5_avg=0.86007 lr=0.00010 gn=39.18012 time=62.97it/s +epoch=110 global_step=43250 loss=5.43661 loss_avg=4.62211 acc=0.43750 acc_top1_avg=0.55726 acc_top5_avg=0.85964 lr=0.00010 gn=34.25517 time=55.44it/s +epoch=110 global_step=43300 loss=5.35830 loss_avg=4.62596 acc=0.46875 acc_top1_avg=0.55744 acc_top5_avg=0.85905 lr=0.00010 gn=38.96047 time=45.64it/s +epoch=110 global_step=43350 loss=4.73251 loss_avg=4.62319 acc=0.54688 acc_top1_avg=0.55726 acc_top5_avg=0.85910 lr=0.00010 gn=45.01460 time=54.71it/s +epoch=110 global_step=43400 loss=4.64815 loss_avg=4.62886 acc=0.57031 acc_top1_avg=0.55659 acc_top5_avg=0.85948 lr=0.00010 gn=44.03603 time=55.66it/s +====================Eval==================== +epoch=110 global_step=43401 loss=1.59798 test_loss_avg=4.31936 acc=0.53906 test_acc_avg=0.11016 test_acc_top5_avg=0.63516 time=242.70it/s +epoch=110 global_step=43401 loss=0.57764 test_loss_avg=3.26902 acc=0.80469 test_acc_avg=0.24753 test_acc_top5_avg=0.81875 time=240.18it/s +epoch=110 global_step=43401 loss=6.84328 test_loss_avg=3.22399 acc=0.00000 test_acc_avg=0.30538 test_acc_top5_avg=0.84771 time=500.04it/s +curr_acc 0.3054 +BEST_ACC 0.3411 +curr_acc_top5 0.8477 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=4.75511 loss_avg=4.64649 acc=0.54688 acc_top1_avg=0.55389 acc_top5_avg=0.85953 lr=0.00010 gn=40.56367 time=50.79it/s +epoch=111 global_step=43500 loss=4.70425 loss_avg=4.60716 acc=0.54688 acc_top1_avg=0.55911 acc_top5_avg=0.86269 lr=0.00010 gn=32.63307 time=52.00it/s +epoch=111 global_step=43550 loss=4.79338 loss_avg=4.60016 acc=0.53906 acc_top1_avg=0.56030 acc_top5_avg=0.86011 lr=0.00010 gn=42.21826 time=52.96it/s +epoch=111 global_step=43600 loss=4.67484 loss_avg=4.62327 acc=0.56250 acc_top1_avg=0.55787 acc_top5_avg=0.86079 lr=0.00010 gn=45.08034 time=58.44it/s +epoch=111 global_step=43650 loss=4.52396 loss_avg=4.62055 acc=0.58594 acc_top1_avg=0.55764 acc_top5_avg=0.86038 lr=0.00010 gn=46.39896 time=59.56it/s +epoch=111 global_step=43700 loss=4.30904 loss_avg=4.63012 acc=0.58594 acc_top1_avg=0.55654 acc_top5_avg=0.86081 lr=0.00010 gn=46.65957 time=60.57it/s +epoch=111 global_step=43750 loss=4.88137 loss_avg=4.62861 acc=0.51562 acc_top1_avg=0.55697 acc_top5_avg=0.86067 lr=0.00010 gn=43.58411 time=56.25it/s +====================Eval==================== +epoch=111 global_step=43792 loss=3.26319 test_loss_avg=3.22964 acc=0.19531 test_acc_avg=0.23185 test_acc_top5_avg=0.82434 time=240.71it/s +epoch=111 global_step=43792 loss=6.95681 test_loss_avg=3.20472 acc=0.00000 test_acc_avg=0.31260 test_acc_top5_avg=0.85255 time=506.25it/s +curr_acc 0.3126 +BEST_ACC 0.3411 +curr_acc_top5 0.8526 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=4.46915 loss_avg=4.69625 acc=0.57031 acc_top1_avg=0.55273 acc_top5_avg=0.86426 lr=0.00010 gn=39.97892 time=55.56it/s +epoch=112 global_step=43850 loss=4.83355 loss_avg=4.61872 acc=0.53906 acc_top1_avg=0.55873 acc_top5_avg=0.86018 lr=0.00010 gn=41.61926 time=56.09it/s +epoch=112 global_step=43900 loss=4.83476 loss_avg=4.64008 acc=0.53906 acc_top1_avg=0.55664 acc_top5_avg=0.85829 lr=0.00010 gn=44.93834 time=59.03it/s +epoch=112 global_step=43950 loss=4.57521 loss_avg=4.65575 acc=0.54688 acc_top1_avg=0.55454 acc_top5_avg=0.85844 lr=0.00010 gn=41.93271 time=52.02it/s +epoch=112 global_step=44000 loss=4.66631 loss_avg=4.64260 acc=0.55469 acc_top1_avg=0.55578 acc_top5_avg=0.85727 lr=0.00010 gn=44.66310 time=53.34it/s +epoch=112 global_step=44050 loss=4.85549 loss_avg=4.64322 acc=0.52344 acc_top1_avg=0.55557 acc_top5_avg=0.85862 lr=0.00010 gn=38.13434 time=55.32it/s +epoch=112 global_step=44100 loss=4.39414 loss_avg=4.62249 acc=0.59375 acc_top1_avg=0.55793 acc_top5_avg=0.85907 lr=0.00010 gn=48.81653 time=59.98it/s +epoch=112 global_step=44150 loss=4.27134 loss_avg=4.62706 acc=0.58594 acc_top1_avg=0.55713 acc_top5_avg=0.85907 lr=0.00010 gn=32.64388 time=49.21it/s +====================Eval==================== +epoch=112 global_step=44183 loss=4.98707 test_loss_avg=5.17834 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.53125 time=240.50it/s +epoch=112 global_step=44183 loss=5.42198 test_loss_avg=3.44983 acc=0.00000 test_acc_avg=0.21364 test_acc_top5_avg=0.80123 time=236.19it/s +epoch=112 global_step=44183 loss=6.74442 test_loss_avg=3.24720 acc=0.00000 test_acc_avg=0.30854 test_acc_top5_avg=0.85117 time=503.03it/s +curr_acc 0.3085 +BEST_ACC 0.3411 +curr_acc_top5 0.8512 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=5.07875 loss_avg=4.49568 acc=0.50781 acc_top1_avg=0.57169 acc_top5_avg=0.86397 lr=0.00010 gn=41.00407 time=57.23it/s +epoch=113 global_step=44250 loss=5.23144 loss_avg=4.62076 acc=0.50000 acc_top1_avg=0.55877 acc_top5_avg=0.86182 lr=0.00010 gn=48.79643 time=57.18it/s +epoch=113 global_step=44300 loss=4.77144 loss_avg=4.58975 acc=0.55469 acc_top1_avg=0.56183 acc_top5_avg=0.86251 lr=0.00010 gn=49.33029 time=55.08it/s +epoch=113 global_step=44350 loss=4.24953 loss_avg=4.56550 acc=0.60156 acc_top1_avg=0.56395 acc_top5_avg=0.86195 lr=0.00010 gn=46.29413 time=47.22it/s +epoch=113 global_step=44400 loss=3.89653 loss_avg=4.58496 acc=0.62500 acc_top1_avg=0.56135 acc_top5_avg=0.86186 lr=0.00010 gn=40.58019 time=60.69it/s +epoch=113 global_step=44450 loss=4.27763 loss_avg=4.59668 acc=0.60938 acc_top1_avg=0.56057 acc_top5_avg=0.86254 lr=0.00010 gn=47.95200 time=59.22it/s +epoch=113 global_step=44500 loss=4.26162 loss_avg=4.61537 acc=0.60938 acc_top1_avg=0.55841 acc_top5_avg=0.86159 lr=0.00010 gn=44.67210 time=52.48it/s +epoch=113 global_step=44550 loss=3.36128 loss_avg=4.62580 acc=0.72656 acc_top1_avg=0.55720 acc_top5_avg=0.86159 lr=0.00010 gn=46.34539 time=61.92it/s +====================Eval==================== +epoch=113 global_step=44574 loss=2.70462 test_loss_avg=3.30132 acc=0.27344 test_acc_avg=0.23132 test_acc_top5_avg=0.79042 time=237.38it/s +epoch=113 global_step=44574 loss=6.45186 test_loss_avg=2.96969 acc=0.00000 test_acc_avg=0.32909 test_acc_top5_avg=0.84835 time=242.74it/s +epoch=113 global_step=44574 loss=6.79630 test_loss_avg=3.23160 acc=0.00000 test_acc_avg=0.30409 test_acc_top5_avg=0.85225 time=820.00it/s +curr_acc 0.3041 +BEST_ACC 0.3411 +curr_acc_top5 0.8523 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=4.83832 loss_avg=4.60401 acc=0.53125 acc_top1_avg=0.55919 acc_top5_avg=0.86178 lr=0.00010 gn=42.93147 time=52.15it/s +epoch=114 global_step=44650 loss=4.97373 loss_avg=4.60236 acc=0.52344 acc_top1_avg=0.55849 acc_top5_avg=0.86051 lr=0.00010 gn=50.21246 time=62.23it/s +epoch=114 global_step=44700 loss=4.77945 loss_avg=4.58141 acc=0.55469 acc_top1_avg=0.56163 acc_top5_avg=0.86142 lr=0.00010 gn=54.08928 time=63.42it/s +epoch=114 global_step=44750 loss=4.93238 loss_avg=4.60344 acc=0.52344 acc_top1_avg=0.55886 acc_top5_avg=0.86213 lr=0.00010 gn=38.76456 time=53.83it/s +epoch=114 global_step=44800 loss=4.88028 loss_avg=4.60226 acc=0.54688 acc_top1_avg=0.55894 acc_top5_avg=0.86211 lr=0.00010 gn=42.76152 time=60.32it/s +epoch=114 global_step=44850 loss=4.13278 loss_avg=4.61358 acc=0.61719 acc_top1_avg=0.55738 acc_top5_avg=0.86073 lr=0.00010 gn=40.51140 time=56.51it/s +epoch=114 global_step=44900 loss=4.92066 loss_avg=4.63325 acc=0.51562 acc_top1_avg=0.55596 acc_top5_avg=0.86065 lr=0.00010 gn=42.82626 time=54.45it/s +epoch=114 global_step=44950 loss=3.84630 loss_avg=4.62864 acc=0.65625 acc_top1_avg=0.55670 acc_top5_avg=0.86085 lr=0.00010 gn=49.39873 time=53.15it/s +====================Eval==================== +epoch=114 global_step=44965 loss=1.91229 test_loss_avg=3.35192 acc=0.46094 test_acc_avg=0.21254 test_acc_top5_avg=0.76048 time=236.54it/s +epoch=114 global_step=44965 loss=6.85250 test_loss_avg=3.25865 acc=0.00000 test_acc_avg=0.30716 test_acc_top5_avg=0.84622 time=754.64it/s +curr_acc 0.3072 +BEST_ACC 0.3411 +curr_acc_top5 0.8462 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=4.49141 loss_avg=4.63920 acc=0.58594 acc_top1_avg=0.55402 acc_top5_avg=0.85179 lr=0.00010 gn=49.88668 time=57.78it/s +epoch=115 global_step=45050 loss=4.37853 loss_avg=4.64187 acc=0.61719 acc_top1_avg=0.55588 acc_top5_avg=0.85561 lr=0.00010 gn=51.82988 time=56.31it/s +epoch=115 global_step=45100 loss=5.00513 loss_avg=4.60263 acc=0.53906 acc_top1_avg=0.56030 acc_top5_avg=0.85677 lr=0.00010 gn=47.15998 time=53.88it/s +epoch=115 global_step=45150 loss=4.41035 loss_avg=4.58907 acc=0.58594 acc_top1_avg=0.56208 acc_top5_avg=0.85785 lr=0.00010 gn=36.56402 time=54.19it/s +epoch=115 global_step=45200 loss=4.98083 loss_avg=4.59618 acc=0.53906 acc_top1_avg=0.56110 acc_top5_avg=0.85868 lr=0.00010 gn=45.37715 time=58.89it/s +epoch=115 global_step=45250 loss=4.37170 loss_avg=4.59913 acc=0.57031 acc_top1_avg=0.56105 acc_top5_avg=0.85984 lr=0.00010 gn=42.03671 time=55.52it/s +epoch=115 global_step=45300 loss=4.69974 loss_avg=4.61750 acc=0.53906 acc_top1_avg=0.55875 acc_top5_avg=0.85989 lr=0.00010 gn=45.73551 time=54.02it/s +epoch=115 global_step=45350 loss=4.86134 loss_avg=4.61712 acc=0.53125 acc_top1_avg=0.55875 acc_top5_avg=0.86065 lr=0.00010 gn=37.06423 time=54.20it/s +====================Eval==================== +epoch=115 global_step=45356 loss=2.09296 test_loss_avg=3.71833 acc=0.42969 test_acc_avg=0.20208 test_acc_top5_avg=0.75365 time=238.80it/s +epoch=115 global_step=45356 loss=0.18947 test_loss_avg=3.12557 acc=0.92969 test_acc_avg=0.28714 test_acc_top5_avg=0.83462 time=228.83it/s +epoch=115 global_step=45356 loss=6.77284 test_loss_avg=3.28604 acc=0.00000 test_acc_avg=0.29974 test_acc_top5_avg=0.85087 time=568.64it/s +curr_acc 0.2997 +BEST_ACC 0.3411 +curr_acc_top5 0.8509 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=4.57080 loss_avg=4.61662 acc=0.57031 acc_top1_avg=0.55575 acc_top5_avg=0.86222 lr=0.00010 gn=44.45495 time=53.53it/s +epoch=116 global_step=45450 loss=4.89796 loss_avg=4.62700 acc=0.53906 acc_top1_avg=0.55618 acc_top5_avg=0.85888 lr=0.00010 gn=45.80340 time=57.00it/s +epoch=116 global_step=45500 loss=4.42784 loss_avg=4.61054 acc=0.57031 acc_top1_avg=0.55881 acc_top5_avg=0.86057 lr=0.00010 gn=39.34202 time=61.14it/s +epoch=116 global_step=45550 loss=4.18247 loss_avg=4.61996 acc=0.58594 acc_top1_avg=0.55763 acc_top5_avg=0.85958 lr=0.00010 gn=35.09812 time=55.51it/s +epoch=116 global_step=45600 loss=4.79823 loss_avg=4.61393 acc=0.52344 acc_top1_avg=0.55837 acc_top5_avg=0.85970 lr=0.00010 gn=39.89450 time=60.39it/s +epoch=116 global_step=45650 loss=4.41086 loss_avg=4.61039 acc=0.56250 acc_top1_avg=0.55849 acc_top5_avg=0.86023 lr=0.00010 gn=37.97489 time=62.27it/s +epoch=116 global_step=45700 loss=4.92254 loss_avg=4.60555 acc=0.53906 acc_top1_avg=0.55946 acc_top5_avg=0.86037 lr=0.00010 gn=52.50977 time=53.11it/s +====================Eval==================== +epoch=116 global_step=45747 loss=4.56530 test_loss_avg=3.40846 acc=0.03125 test_acc_avg=0.20161 test_acc_top5_avg=0.77951 time=165.72it/s +epoch=116 global_step=45747 loss=6.62161 test_loss_avg=3.20290 acc=0.00000 test_acc_avg=0.31072 test_acc_top5_avg=0.85077 time=499.14it/s +curr_acc 0.3107 +BEST_ACC 0.3411 +curr_acc_top5 0.8508 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=4.86097 loss_avg=4.57022 acc=0.54688 acc_top1_avg=0.55729 acc_top5_avg=0.82292 lr=0.00010 gn=44.53377 time=44.36it/s +epoch=117 global_step=45800 loss=4.41317 loss_avg=4.59724 acc=0.58594 acc_top1_avg=0.56221 acc_top5_avg=0.85230 lr=0.00010 gn=42.73711 time=57.99it/s +epoch=117 global_step=45850 loss=4.46899 loss_avg=4.59896 acc=0.57031 acc_top1_avg=0.55992 acc_top5_avg=0.85725 lr=0.00010 gn=37.30280 time=58.84it/s +epoch=117 global_step=45900 loss=4.32865 loss_avg=4.58276 acc=0.62500 acc_top1_avg=0.56265 acc_top5_avg=0.85835 lr=0.00010 gn=47.92424 time=62.41it/s +epoch=117 global_step=45950 loss=4.79777 loss_avg=4.59013 acc=0.53125 acc_top1_avg=0.56200 acc_top5_avg=0.85780 lr=0.00010 gn=39.02033 time=56.39it/s +epoch=117 global_step=46000 loss=4.99355 loss_avg=4.60247 acc=0.52344 acc_top1_avg=0.56040 acc_top5_avg=0.85866 lr=0.00010 gn=43.14694 time=62.33it/s +epoch=117 global_step=46050 loss=5.02371 loss_avg=4.60836 acc=0.50000 acc_top1_avg=0.55959 acc_top5_avg=0.85909 lr=0.00010 gn=43.27127 time=54.62it/s +epoch=117 global_step=46100 loss=4.91748 loss_avg=4.59786 acc=0.52344 acc_top1_avg=0.56124 acc_top5_avg=0.85980 lr=0.00010 gn=48.20473 time=52.04it/s +====================Eval==================== +epoch=117 global_step=46138 loss=5.03823 test_loss_avg=5.06319 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.52455 time=238.95it/s +epoch=117 global_step=46138 loss=0.67232 test_loss_avg=3.38641 acc=0.76562 test_acc_avg=0.22423 test_acc_top5_avg=0.81209 time=240.10it/s +epoch=117 global_step=46138 loss=6.77269 test_loss_avg=3.21468 acc=0.00000 test_acc_avg=0.31062 test_acc_top5_avg=0.84968 time=769.17it/s +curr_acc 0.3106 +BEST_ACC 0.3411 +curr_acc_top5 0.8497 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=4.38373 loss_avg=4.64102 acc=0.58594 acc_top1_avg=0.55534 acc_top5_avg=0.85091 lr=0.00010 gn=46.93558 time=55.44it/s +epoch=118 global_step=46200 loss=4.51187 loss_avg=4.55949 acc=0.57812 acc_top1_avg=0.56250 acc_top5_avg=0.85912 lr=0.00010 gn=44.83573 time=62.06it/s +epoch=118 global_step=46250 loss=4.92611 loss_avg=4.55926 acc=0.52344 acc_top1_avg=0.56299 acc_top5_avg=0.85986 lr=0.00010 gn=45.21123 time=60.83it/s +epoch=118 global_step=46300 loss=4.75111 loss_avg=4.57448 acc=0.53125 acc_top1_avg=0.56182 acc_top5_avg=0.85938 lr=0.00010 gn=47.73095 time=53.00it/s +epoch=118 global_step=46350 loss=4.81013 loss_avg=4.58271 acc=0.53906 acc_top1_avg=0.56180 acc_top5_avg=0.85982 lr=0.00010 gn=39.92647 time=54.31it/s +epoch=118 global_step=46400 loss=4.75482 loss_avg=4.59375 acc=0.54688 acc_top1_avg=0.56107 acc_top5_avg=0.86149 lr=0.00010 gn=39.32035 time=54.27it/s +epoch=118 global_step=46450 loss=4.52630 loss_avg=4.59593 acc=0.58594 acc_top1_avg=0.56110 acc_top5_avg=0.85950 lr=0.00010 gn=40.98041 time=63.85it/s +epoch=118 global_step=46500 loss=4.34380 loss_avg=4.59831 acc=0.58594 acc_top1_avg=0.56075 acc_top5_avg=0.85946 lr=0.00010 gn=45.63241 time=53.88it/s +====================Eval==================== +epoch=118 global_step=46529 loss=3.19208 test_loss_avg=3.26551 acc=0.21094 test_acc_avg=0.23270 test_acc_top5_avg=0.80831 time=245.86it/s +epoch=118 global_step=46529 loss=6.02445 test_loss_avg=3.16585 acc=0.00000 test_acc_avg=0.31510 test_acc_top5_avg=0.85206 time=261.08it/s +epoch=118 global_step=46529 loss=6.48206 test_loss_avg=3.20783 acc=0.00000 test_acc_avg=0.31112 test_acc_top5_avg=0.85394 time=857.91it/s +curr_acc 0.3111 +BEST_ACC 0.3411 +curr_acc_top5 0.8539 +BEST_ACC_top5 0.8744 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=5.26320 loss_avg=4.60252 acc=0.49219 acc_top1_avg=0.56138 acc_top5_avg=0.87128 lr=0.00010 gn=41.89239 time=63.73it/s +epoch=119 global_step=46600 loss=4.35858 loss_avg=4.54426 acc=0.57812 acc_top1_avg=0.56877 acc_top5_avg=0.86576 lr=0.00010 gn=43.10584 time=60.39it/s +epoch=119 global_step=46650 loss=3.90589 loss_avg=4.56609 acc=0.64062 acc_top1_avg=0.56528 acc_top5_avg=0.86073 lr=0.00010 gn=43.75392 time=61.99it/s +epoch=119 global_step=46700 loss=3.99894 loss_avg=4.57659 acc=0.61719 acc_top1_avg=0.56355 acc_top5_avg=0.86111 lr=0.00010 gn=44.36535 time=63.37it/s +epoch=119 global_step=46750 loss=4.31579 loss_avg=4.58576 acc=0.59375 acc_top1_avg=0.56254 acc_top5_avg=0.86100 lr=0.00010 gn=45.64247 time=60.30it/s +epoch=119 global_step=46800 loss=4.97204 loss_avg=4.58586 acc=0.53125 acc_top1_avg=0.56213 acc_top5_avg=0.86165 lr=0.00010 gn=49.15178 time=64.80it/s +epoch=119 global_step=46850 loss=4.75703 loss_avg=4.58695 acc=0.54688 acc_top1_avg=0.56199 acc_top5_avg=0.86093 lr=0.00010 gn=47.89681 time=55.86it/s +epoch=119 global_step=46900 loss=4.31946 loss_avg=4.58646 acc=0.60156 acc_top1_avg=0.56214 acc_top5_avg=0.86055 lr=0.00010 gn=44.55061 time=56.81it/s +====================Eval==================== +epoch=119 global_step=46920 loss=5.41431 test_loss_avg=3.33683 acc=0.00000 test_acc_avg=0.22481 test_acc_top5_avg=0.79305 time=242.36it/s +epoch=119 global_step=46920 loss=6.66564 test_loss_avg=3.24868 acc=0.00000 test_acc_avg=0.30864 test_acc_top5_avg=0.85443 time=899.68it/s +curr_acc 0.3086 +BEST_ACC 0.3411 +curr_acc_top5 0.8544 +BEST_ACC_top5 0.8744 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_6_2.log b/other_methods/sceloss/sceloss_results/out_6_2.log new file mode 100644 index 0000000..e04fabf --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_6_2.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.6__noise_amount__0.2.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=6.87908 loss_avg=7.47073 acc=0.32812 acc_top1_avg=0.24563 acc_top5_avg=0.70391 lr=0.01000 gn=7.81771 time=65.82it/s +epoch=0 global_step=100 loss=5.64437 loss_avg=6.96117 acc=0.44531 acc_top1_avg=0.29977 acc_top5_avg=0.75367 lr=0.01000 gn=8.82604 time=65.96it/s +epoch=0 global_step=150 loss=6.01716 loss_avg=6.67083 acc=0.40625 acc_top1_avg=0.32969 acc_top5_avg=0.77635 lr=0.01000 gn=8.13245 time=63.36it/s +epoch=0 global_step=200 loss=6.20065 loss_avg=6.50262 acc=0.38281 acc_top1_avg=0.34664 acc_top5_avg=0.78930 lr=0.01000 gn=6.41432 time=65.91it/s +epoch=0 global_step=250 loss=5.88028 loss_avg=6.37340 acc=0.40625 acc_top1_avg=0.35994 acc_top5_avg=0.79912 lr=0.01000 gn=6.64743 time=63.58it/s +epoch=0 global_step=300 loss=5.67455 loss_avg=6.24454 acc=0.45312 acc_top1_avg=0.37320 acc_top5_avg=0.81008 lr=0.01000 gn=7.27008 time=64.13it/s +epoch=0 global_step=350 loss=5.88411 loss_avg=6.13725 acc=0.42188 acc_top1_avg=0.38482 acc_top5_avg=0.81781 lr=0.01000 gn=6.16955 time=62.98it/s +====================Eval==================== +epoch=0 global_step=391 loss=0.89941 test_loss_avg=2.39425 acc=0.75000 test_acc_avg=0.44703 test_acc_top5_avg=0.87328 time=245.63it/s +epoch=0 global_step=391 loss=3.38503 test_loss_avg=2.06186 acc=0.18750 test_acc_avg=0.51592 test_acc_top5_avg=0.90615 time=35.00it/s +curr_acc 0.5159 +BEST_ACC 0.0000 +curr_acc_top5 0.9062 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=4.28377 loss_avg=5.16862 acc=0.58594 acc_top1_avg=0.48872 acc_top5_avg=0.88455 lr=0.01000 gn=5.50184 time=54.83it/s +epoch=1 global_step=450 loss=5.26109 loss_avg=5.20662 acc=0.49219 acc_top1_avg=0.48570 acc_top5_avg=0.87672 lr=0.01000 gn=6.62559 time=63.63it/s +epoch=1 global_step=500 loss=4.61199 loss_avg=5.19109 acc=0.56250 acc_top1_avg=0.48731 acc_top5_avg=0.87672 lr=0.01000 gn=6.98805 time=63.55it/s +epoch=1 global_step=550 loss=5.43511 loss_avg=5.13657 acc=0.45312 acc_top1_avg=0.49288 acc_top5_avg=0.87947 lr=0.01000 gn=6.45771 time=62.29it/s +epoch=1 global_step=600 loss=4.48881 loss_avg=5.07754 acc=0.57031 acc_top1_avg=0.49940 acc_top5_avg=0.88094 lr=0.01000 gn=6.39389 time=65.29it/s +epoch=1 global_step=650 loss=4.54567 loss_avg=5.02829 acc=0.55469 acc_top1_avg=0.50374 acc_top5_avg=0.88308 lr=0.01000 gn=5.61732 time=64.68it/s +epoch=1 global_step=700 loss=4.77784 loss_avg=4.99265 acc=0.50781 acc_top1_avg=0.50743 acc_top5_avg=0.88519 lr=0.01000 gn=5.17203 time=59.65it/s +epoch=1 global_step=750 loss=4.95867 loss_avg=4.95715 acc=0.50781 acc_top1_avg=0.51084 acc_top5_avg=0.88671 lr=0.01000 gn=6.53167 time=62.95it/s +====================Eval==================== +epoch=1 global_step=782 loss=2.83406 test_loss_avg=2.11542 acc=0.32031 test_acc_avg=0.53385 test_acc_top5_avg=0.94940 time=256.94it/s +epoch=1 global_step=782 loss=1.38787 test_loss_avg=2.19848 acc=0.66406 test_acc_avg=0.55755 test_acc_top5_avg=0.93860 time=246.81it/s +epoch=1 global_step=782 loss=1.23762 test_loss_avg=2.19965 acc=0.43750 test_acc_avg=0.55083 test_acc_top5_avg=0.93542 time=731.73it/s +curr_acc 0.5508 +BEST_ACC 0.5159 +curr_acc_top5 0.9354 +BEST_ACC_top5 0.9062 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=4.86855 loss_avg=4.62227 acc=0.53125 acc_top1_avg=0.54818 acc_top5_avg=0.90451 lr=0.01000 gn=6.52589 time=57.83it/s +epoch=2 global_step=850 loss=4.12626 loss_avg=4.58934 acc=0.57812 acc_top1_avg=0.55101 acc_top5_avg=0.90499 lr=0.01000 gn=5.39182 time=60.54it/s +epoch=2 global_step=900 loss=4.41553 loss_avg=4.55808 acc=0.56250 acc_top1_avg=0.55277 acc_top5_avg=0.90632 lr=0.01000 gn=6.36063 time=58.70it/s +epoch=2 global_step=950 loss=4.39570 loss_avg=4.54365 acc=0.56250 acc_top1_avg=0.55352 acc_top5_avg=0.90574 lr=0.01000 gn=5.48217 time=64.91it/s +epoch=2 global_step=1000 loss=5.32481 loss_avg=4.54047 acc=0.48438 acc_top1_avg=0.55297 acc_top5_avg=0.90736 lr=0.01000 gn=6.13045 time=58.12it/s +epoch=2 global_step=1050 loss=3.54436 loss_avg=4.54428 acc=0.64844 acc_top1_avg=0.55247 acc_top5_avg=0.90613 lr=0.01000 gn=6.77433 time=64.19it/s +epoch=2 global_step=1100 loss=4.31796 loss_avg=4.51938 acc=0.58594 acc_top1_avg=0.55508 acc_top5_avg=0.90682 lr=0.01000 gn=5.77849 time=60.52it/s +epoch=2 global_step=1150 loss=4.80136 loss_avg=4.49751 acc=0.50781 acc_top1_avg=0.55728 acc_top5_avg=0.90852 lr=0.01000 gn=6.32742 time=55.63it/s +====================Eval==================== +epoch=2 global_step=1173 loss=3.57061 test_loss_avg=1.86424 acc=0.27344 test_acc_avg=0.54743 test_acc_top5_avg=0.93527 time=219.00it/s +epoch=2 global_step=1173 loss=1.84423 test_loss_avg=1.61368 acc=0.50000 test_acc_avg=0.61788 test_acc_top5_avg=0.93295 time=466.09it/s +curr_acc 0.6179 +BEST_ACC 0.5508 +curr_acc_top5 0.9330 +BEST_ACC_top5 0.9354 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=3.78783 loss_avg=4.18298 acc=0.62500 acc_top1_avg=0.58362 acc_top5_avg=0.92361 lr=0.01000 gn=6.10623 time=54.66it/s +epoch=3 global_step=1250 loss=4.35994 loss_avg=4.23196 acc=0.57812 acc_top1_avg=0.58198 acc_top5_avg=0.91985 lr=0.01000 gn=6.59597 time=51.96it/s +epoch=3 global_step=1300 loss=4.67135 loss_avg=4.28322 acc=0.53125 acc_top1_avg=0.57745 acc_top5_avg=0.91837 lr=0.01000 gn=7.33279 time=62.53it/s +epoch=3 global_step=1350 loss=3.59008 loss_avg=4.28006 acc=0.64844 acc_top1_avg=0.57817 acc_top5_avg=0.91812 lr=0.01000 gn=5.23356 time=61.41it/s +epoch=3 global_step=1400 loss=4.52702 loss_avg=4.28321 acc=0.53125 acc_top1_avg=0.57792 acc_top5_avg=0.91864 lr=0.01000 gn=4.85510 time=58.55it/s +epoch=3 global_step=1450 loss=4.65086 loss_avg=4.26763 acc=0.54688 acc_top1_avg=0.58016 acc_top5_avg=0.91829 lr=0.01000 gn=6.35812 time=61.26it/s +epoch=3 global_step=1500 loss=3.55282 loss_avg=4.26751 acc=0.67188 acc_top1_avg=0.58037 acc_top5_avg=0.91848 lr=0.01000 gn=6.12699 time=55.43it/s +epoch=3 global_step=1550 loss=4.39555 loss_avg=4.24757 acc=0.54688 acc_top1_avg=0.58273 acc_top5_avg=0.91852 lr=0.01000 gn=5.36315 time=64.21it/s +====================Eval==================== +epoch=3 global_step=1564 loss=0.99939 test_loss_avg=1.29551 acc=0.76562 test_acc_avg=0.68209 test_acc_top5_avg=0.96154 time=199.38it/s +epoch=3 global_step=1564 loss=0.81532 test_loss_avg=1.41166 acc=0.79688 test_acc_avg=0.67436 test_acc_top5_avg=0.96726 time=240.22it/s +epoch=3 global_step=1564 loss=0.14253 test_loss_avg=1.20032 acc=1.00000 test_acc_avg=0.72004 test_acc_top5_avg=0.97271 time=602.37it/s +curr_acc 0.7200 +BEST_ACC 0.6179 +curr_acc_top5 0.9727 +BEST_ACC_top5 0.9354 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.62014 loss_avg=4.08397 acc=0.55469 acc_top1_avg=0.60286 acc_top5_avg=0.92253 lr=0.01000 gn=7.13076 time=63.17it/s +epoch=4 global_step=1650 loss=4.34113 loss_avg=4.11322 acc=0.57812 acc_top1_avg=0.59784 acc_top5_avg=0.92369 lr=0.01000 gn=6.81891 time=60.30it/s +epoch=4 global_step=1700 loss=3.83906 loss_avg=4.12139 acc=0.63281 acc_top1_avg=0.59743 acc_top5_avg=0.92245 lr=0.01000 gn=5.74721 time=58.15it/s +epoch=4 global_step=1750 loss=4.56967 loss_avg=4.10599 acc=0.53906 acc_top1_avg=0.59913 acc_top5_avg=0.92335 lr=0.01000 gn=6.84618 time=63.93it/s +epoch=4 global_step=1800 loss=3.81394 loss_avg=4.10164 acc=0.65625 acc_top1_avg=0.59928 acc_top5_avg=0.92399 lr=0.01000 gn=5.49541 time=55.24it/s +epoch=4 global_step=1850 loss=3.32913 loss_avg=4.09459 acc=0.67969 acc_top1_avg=0.59984 acc_top5_avg=0.92502 lr=0.01000 gn=6.31950 time=52.29it/s +epoch=4 global_step=1900 loss=3.97374 loss_avg=4.07840 acc=0.60156 acc_top1_avg=0.60135 acc_top5_avg=0.92508 lr=0.01000 gn=5.23023 time=56.82it/s +epoch=4 global_step=1950 loss=3.61288 loss_avg=4.06518 acc=0.64062 acc_top1_avg=0.60290 acc_top5_avg=0.92469 lr=0.01000 gn=5.74790 time=62.05it/s +====================Eval==================== +epoch=4 global_step=1955 loss=1.10185 test_loss_avg=1.21641 acc=0.72656 test_acc_avg=0.67992 test_acc_top5_avg=0.96691 time=237.14it/s +epoch=4 global_step=1955 loss=0.15993 test_loss_avg=0.95259 acc=0.93750 test_acc_avg=0.74575 test_acc_top5_avg=0.96865 time=651.59it/s +curr_acc 0.7457 +BEST_ACC 0.7200 +curr_acc_top5 0.9687 +BEST_ACC_top5 0.9727 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=3.62036 loss_avg=4.00358 acc=0.66406 acc_top1_avg=0.61042 acc_top5_avg=0.92691 lr=0.01000 gn=6.86242 time=55.14it/s +epoch=5 global_step=2050 loss=3.94186 loss_avg=4.01304 acc=0.64062 acc_top1_avg=0.60666 acc_top5_avg=0.92541 lr=0.01000 gn=6.89307 time=60.43it/s +epoch=5 global_step=2100 loss=3.73441 loss_avg=4.02399 acc=0.66406 acc_top1_avg=0.60571 acc_top5_avg=0.92726 lr=0.01000 gn=5.76438 time=56.62it/s +epoch=5 global_step=2150 loss=3.25870 loss_avg=3.95684 acc=0.71094 acc_top1_avg=0.61374 acc_top5_avg=0.92865 lr=0.01000 gn=7.66801 time=54.24it/s +epoch=5 global_step=2200 loss=3.85779 loss_avg=3.96674 acc=0.63281 acc_top1_avg=0.61234 acc_top5_avg=0.92765 lr=0.01000 gn=5.74029 time=55.06it/s +epoch=5 global_step=2250 loss=4.18968 loss_avg=3.95372 acc=0.59375 acc_top1_avg=0.61382 acc_top5_avg=0.92762 lr=0.01000 gn=6.71903 time=53.91it/s +epoch=5 global_step=2300 loss=3.67735 loss_avg=3.94303 acc=0.64062 acc_top1_avg=0.61501 acc_top5_avg=0.92722 lr=0.01000 gn=7.48439 time=62.31it/s +====================Eval==================== +epoch=5 global_step=2346 loss=0.75087 test_loss_avg=1.02454 acc=0.76562 test_acc_avg=0.72969 test_acc_top5_avg=0.97188 time=238.11it/s +epoch=5 global_step=2346 loss=0.32995 test_loss_avg=1.07599 acc=0.89062 test_acc_avg=0.72372 test_acc_top5_avg=0.97301 time=40.33it/s +epoch=5 global_step=2346 loss=0.15199 test_loss_avg=0.89787 acc=0.93750 test_acc_avg=0.76661 test_acc_top5_avg=0.97725 time=638.11it/s +curr_acc 0.7666 +BEST_ACC 0.7457 +curr_acc_top5 0.9773 +BEST_ACC_top5 0.9727 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=4.12574 loss_avg=3.96583 acc=0.58594 acc_top1_avg=0.60547 acc_top5_avg=0.92773 lr=0.01000 gn=5.92028 time=61.37it/s +epoch=6 global_step=2400 loss=3.94142 loss_avg=3.88912 acc=0.62500 acc_top1_avg=0.62211 acc_top5_avg=0.92824 lr=0.01000 gn=6.69817 time=56.91it/s +epoch=6 global_step=2450 loss=3.55746 loss_avg=3.87941 acc=0.65625 acc_top1_avg=0.62350 acc_top5_avg=0.92781 lr=0.01000 gn=6.75466 time=52.01it/s +epoch=6 global_step=2500 loss=3.31548 loss_avg=3.85661 acc=0.67969 acc_top1_avg=0.62515 acc_top5_avg=0.93065 lr=0.01000 gn=6.67944 time=53.94it/s +epoch=6 global_step=2550 loss=3.74208 loss_avg=3.87358 acc=0.63281 acc_top1_avg=0.62263 acc_top5_avg=0.93107 lr=0.01000 gn=7.40404 time=58.39it/s +epoch=6 global_step=2600 loss=4.01774 loss_avg=3.86335 acc=0.62500 acc_top1_avg=0.62438 acc_top5_avg=0.93147 lr=0.01000 gn=7.72295 time=60.80it/s +epoch=6 global_step=2650 loss=4.57575 loss_avg=3.85571 acc=0.54688 acc_top1_avg=0.62487 acc_top5_avg=0.93167 lr=0.01000 gn=7.01866 time=55.03it/s +epoch=6 global_step=2700 loss=3.97826 loss_avg=3.86082 acc=0.60156 acc_top1_avg=0.62396 acc_top5_avg=0.93200 lr=0.01000 gn=6.46060 time=56.57it/s +====================Eval==================== +epoch=6 global_step=2737 loss=1.81810 test_loss_avg=1.30640 acc=0.57812 test_acc_avg=0.68720 test_acc_top5_avg=0.95553 time=242.91it/s +epoch=6 global_step=2737 loss=0.48425 test_loss_avg=0.98426 acc=0.85156 test_acc_avg=0.75072 test_acc_top5_avg=0.96875 time=234.95it/s +epoch=6 global_step=2737 loss=0.31727 test_loss_avg=0.96510 acc=0.87500 test_acc_avg=0.75494 test_acc_top5_avg=0.96994 time=481.00it/s +curr_acc 0.7549 +BEST_ACC 0.7666 +curr_acc_top5 0.9699 +BEST_ACC_top5 0.9773 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=3.83810 loss_avg=3.83436 acc=0.64062 acc_top1_avg=0.63281 acc_top5_avg=0.93570 lr=0.01000 gn=7.59433 time=62.14it/s +epoch=7 global_step=2800 loss=3.92660 loss_avg=3.75304 acc=0.60938 acc_top1_avg=0.63591 acc_top5_avg=0.93490 lr=0.01000 gn=7.63641 time=60.29it/s +epoch=7 global_step=2850 loss=2.98396 loss_avg=3.77536 acc=0.73438 acc_top1_avg=0.63420 acc_top5_avg=0.93363 lr=0.01000 gn=6.98728 time=54.41it/s +epoch=7 global_step=2900 loss=3.88843 loss_avg=3.77620 acc=0.61719 acc_top1_avg=0.63435 acc_top5_avg=0.93261 lr=0.01000 gn=5.66412 time=55.68it/s +epoch=7 global_step=2950 loss=2.92241 loss_avg=3.77207 acc=0.74219 acc_top1_avg=0.63417 acc_top5_avg=0.93332 lr=0.01000 gn=7.44576 time=59.10it/s +epoch=7 global_step=3000 loss=3.94180 loss_avg=3.78536 acc=0.62500 acc_top1_avg=0.63246 acc_top5_avg=0.93224 lr=0.01000 gn=7.98359 time=55.75it/s +epoch=7 global_step=3050 loss=3.46935 loss_avg=3.77914 acc=0.65625 acc_top1_avg=0.63321 acc_top5_avg=0.93281 lr=0.01000 gn=6.57693 time=57.59it/s +epoch=7 global_step=3100 loss=3.62781 loss_avg=3.76512 acc=0.62500 acc_top1_avg=0.63492 acc_top5_avg=0.93371 lr=0.01000 gn=7.33194 time=58.15it/s +====================Eval==================== +epoch=7 global_step=3128 loss=1.41723 test_loss_avg=1.35444 acc=0.66406 test_acc_avg=0.69481 test_acc_top5_avg=0.95578 time=242.29it/s +epoch=7 global_step=3128 loss=0.84668 test_loss_avg=1.10172 acc=0.75000 test_acc_avg=0.74496 test_acc_top5_avg=0.96529 time=493.91it/s +curr_acc 0.7450 +BEST_ACC 0.7666 +curr_acc_top5 0.9653 +BEST_ACC_top5 0.9773 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=2.76868 loss_avg=3.59711 acc=0.75000 acc_top1_avg=0.64773 acc_top5_avg=0.93999 lr=0.01000 gn=6.16402 time=55.28it/s +epoch=8 global_step=3200 loss=4.08064 loss_avg=3.75868 acc=0.61719 acc_top1_avg=0.63422 acc_top5_avg=0.93696 lr=0.01000 gn=7.95255 time=59.53it/s +epoch=8 global_step=3250 loss=4.81414 loss_avg=3.72658 acc=0.51562 acc_top1_avg=0.63800 acc_top5_avg=0.93923 lr=0.01000 gn=8.46658 time=60.48it/s +epoch=8 global_step=3300 loss=3.81587 loss_avg=3.72723 acc=0.63281 acc_top1_avg=0.63894 acc_top5_avg=0.93786 lr=0.01000 gn=8.33769 time=59.12it/s +epoch=8 global_step=3350 loss=3.71461 loss_avg=3.74480 acc=0.66406 acc_top1_avg=0.63682 acc_top5_avg=0.93571 lr=0.01000 gn=8.42720 time=50.73it/s +epoch=8 global_step=3400 loss=3.96406 loss_avg=3.74559 acc=0.59375 acc_top1_avg=0.63660 acc_top5_avg=0.93491 lr=0.01000 gn=6.93294 time=54.48it/s +epoch=8 global_step=3450 loss=3.42323 loss_avg=3.72356 acc=0.67188 acc_top1_avg=0.63878 acc_top5_avg=0.93566 lr=0.01000 gn=5.70278 time=63.03it/s +epoch=8 global_step=3500 loss=2.77541 loss_avg=3.71442 acc=0.73438 acc_top1_avg=0.63987 acc_top5_avg=0.93584 lr=0.01000 gn=6.43395 time=54.25it/s +====================Eval==================== +epoch=8 global_step=3519 loss=1.05782 test_loss_avg=0.79497 acc=0.71094 test_acc_avg=0.78516 test_acc_top5_avg=0.97569 time=235.19it/s +epoch=8 global_step=3519 loss=0.94776 test_loss_avg=0.99614 acc=0.71875 test_acc_avg=0.74609 test_acc_top5_avg=0.96978 time=228.10it/s +epoch=8 global_step=3519 loss=0.19395 test_loss_avg=0.92552 acc=0.87500 test_acc_avg=0.76108 test_acc_top5_avg=0.97251 time=451.05it/s +curr_acc 0.7611 +BEST_ACC 0.7666 +curr_acc_top5 0.9725 +BEST_ACC_top5 0.9773 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=3.67833 loss_avg=3.66062 acc=0.65625 acc_top1_avg=0.64390 acc_top5_avg=0.94355 lr=0.01000 gn=7.24662 time=51.80it/s +epoch=9 global_step=3600 loss=4.08567 loss_avg=3.67717 acc=0.59375 acc_top1_avg=0.64333 acc_top5_avg=0.93866 lr=0.01000 gn=7.45694 time=54.29it/s +epoch=9 global_step=3650 loss=3.70895 loss_avg=3.69172 acc=0.61719 acc_top1_avg=0.64200 acc_top5_avg=0.93756 lr=0.01000 gn=7.36297 time=52.22it/s +epoch=9 global_step=3700 loss=4.10645 loss_avg=3.67229 acc=0.58594 acc_top1_avg=0.64347 acc_top5_avg=0.93888 lr=0.01000 gn=7.32020 time=52.70it/s +epoch=9 global_step=3750 loss=3.91580 loss_avg=3.68257 acc=0.63281 acc_top1_avg=0.64221 acc_top5_avg=0.93912 lr=0.01000 gn=9.37989 time=56.09it/s +epoch=9 global_step=3800 loss=3.72696 loss_avg=3.67435 acc=0.63281 acc_top1_avg=0.64327 acc_top5_avg=0.93858 lr=0.01000 gn=9.83394 time=54.23it/s +epoch=9 global_step=3850 loss=3.73646 loss_avg=3.67906 acc=0.63281 acc_top1_avg=0.64310 acc_top5_avg=0.93795 lr=0.01000 gn=7.05439 time=63.71it/s +epoch=9 global_step=3900 loss=3.75963 loss_avg=3.69056 acc=0.63281 acc_top1_avg=0.64200 acc_top5_avg=0.93789 lr=0.01000 gn=8.40017 time=56.50it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.58787 test_loss_avg=1.09543 acc=0.85938 test_acc_avg=0.72256 test_acc_top5_avg=0.97496 time=76.43it/s +epoch=9 global_step=3910 loss=0.45078 test_loss_avg=1.00993 acc=0.87500 test_acc_avg=0.74120 test_acc_top5_avg=0.97419 time=644.88it/s +curr_acc 0.7412 +BEST_ACC 0.7666 +curr_acc_top5 0.9742 +BEST_ACC_top5 0.9773 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=4.02947 loss_avg=3.64305 acc=0.61719 acc_top1_avg=0.64570 acc_top5_avg=0.94512 lr=0.01000 gn=7.46448 time=58.38it/s +epoch=10 global_step=4000 loss=3.27679 loss_avg=3.62217 acc=0.70312 acc_top1_avg=0.64852 acc_top5_avg=0.94158 lr=0.01000 gn=7.76035 time=56.50it/s +epoch=10 global_step=4050 loss=3.62283 loss_avg=3.66644 acc=0.64844 acc_top1_avg=0.64520 acc_top5_avg=0.93934 lr=0.01000 gn=8.12165 time=54.95it/s +epoch=10 global_step=4100 loss=3.68396 loss_avg=3.65849 acc=0.66406 acc_top1_avg=0.64589 acc_top5_avg=0.93902 lr=0.01000 gn=9.76911 time=54.99it/s +epoch=10 global_step=4150 loss=3.86385 loss_avg=3.65719 acc=0.61719 acc_top1_avg=0.64609 acc_top5_avg=0.93955 lr=0.01000 gn=7.15038 time=63.09it/s +epoch=10 global_step=4200 loss=3.63488 loss_avg=3.67359 acc=0.64844 acc_top1_avg=0.64402 acc_top5_avg=0.93882 lr=0.01000 gn=9.40452 time=58.21it/s +epoch=10 global_step=4250 loss=3.29993 loss_avg=3.67565 acc=0.67969 acc_top1_avg=0.64407 acc_top5_avg=0.93886 lr=0.01000 gn=6.26249 time=63.88it/s +epoch=10 global_step=4300 loss=3.48830 loss_avg=3.66510 acc=0.68750 acc_top1_avg=0.64509 acc_top5_avg=0.93886 lr=0.01000 gn=8.77071 time=60.63it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.30570 test_loss_avg=0.55666 acc=0.89062 test_acc_avg=0.83437 test_acc_top5_avg=0.99297 time=245.31it/s +epoch=10 global_step=4301 loss=1.11550 test_loss_avg=0.98452 acc=0.70312 test_acc_avg=0.75182 test_acc_top5_avg=0.97344 time=219.22it/s +epoch=10 global_step=4301 loss=0.49884 test_loss_avg=0.87851 acc=0.68750 test_acc_avg=0.77354 test_acc_top5_avg=0.97745 time=707.78it/s +curr_acc 0.7735 +BEST_ACC 0.7666 +curr_acc_top5 0.9775 +BEST_ACC_top5 0.9773 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=3.57475 loss_avg=3.63416 acc=0.64844 acc_top1_avg=0.64796 acc_top5_avg=0.93750 lr=0.01000 gn=8.54319 time=50.89it/s +epoch=11 global_step=4400 loss=3.74864 loss_avg=3.63937 acc=0.65625 acc_top1_avg=0.64781 acc_top5_avg=0.93845 lr=0.01000 gn=8.44703 time=55.04it/s +epoch=11 global_step=4450 loss=3.45762 loss_avg=3.61247 acc=0.67188 acc_top1_avg=0.65221 acc_top5_avg=0.94017 lr=0.01000 gn=7.82582 time=54.31it/s +epoch=11 global_step=4500 loss=3.30876 loss_avg=3.61227 acc=0.67969 acc_top1_avg=0.65170 acc_top5_avg=0.93923 lr=0.01000 gn=7.78543 time=54.17it/s +epoch=11 global_step=4550 loss=3.96380 loss_avg=3.62246 acc=0.60938 acc_top1_avg=0.65010 acc_top5_avg=0.93913 lr=0.01000 gn=8.35457 time=52.04it/s +epoch=11 global_step=4600 loss=4.19085 loss_avg=3.62620 acc=0.57812 acc_top1_avg=0.64938 acc_top5_avg=0.93938 lr=0.01000 gn=8.42692 time=54.47it/s +epoch=11 global_step=4650 loss=3.52694 loss_avg=3.61814 acc=0.65625 acc_top1_avg=0.65023 acc_top5_avg=0.93998 lr=0.01000 gn=8.33691 time=53.99it/s +====================Eval==================== +epoch=11 global_step=4692 loss=1.36016 test_loss_avg=1.14043 acc=0.64844 test_acc_avg=0.72455 test_acc_top5_avg=0.96573 time=248.15it/s +epoch=11 global_step=4692 loss=0.07854 test_loss_avg=1.03327 acc=1.00000 test_acc_avg=0.74555 test_acc_top5_avg=0.96568 time=450.61it/s +curr_acc 0.7455 +BEST_ACC 0.7735 +curr_acc_top5 0.9657 +BEST_ACC_top5 0.9775 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=4.29110 loss_avg=3.51794 acc=0.58594 acc_top1_avg=0.67090 acc_top5_avg=0.94043 lr=0.01000 gn=7.28625 time=52.68it/s +epoch=12 global_step=4750 loss=4.16222 loss_avg=3.55703 acc=0.60938 acc_top1_avg=0.66002 acc_top5_avg=0.94356 lr=0.01000 gn=11.02501 time=54.04it/s +epoch=12 global_step=4800 loss=3.64506 loss_avg=3.59414 acc=0.64062 acc_top1_avg=0.65444 acc_top5_avg=0.94242 lr=0.01000 gn=8.08096 time=56.01it/s +epoch=12 global_step=4850 loss=3.87494 loss_avg=3.59788 acc=0.60938 acc_top1_avg=0.65259 acc_top5_avg=0.94190 lr=0.01000 gn=7.93949 time=52.20it/s +epoch=12 global_step=4900 loss=3.34150 loss_avg=3.60867 acc=0.67969 acc_top1_avg=0.65114 acc_top5_avg=0.94193 lr=0.01000 gn=7.23507 time=56.81it/s +epoch=12 global_step=4950 loss=3.73254 loss_avg=3.62594 acc=0.64062 acc_top1_avg=0.64938 acc_top5_avg=0.94168 lr=0.01000 gn=9.54112 time=58.56it/s +epoch=12 global_step=5000 loss=3.47102 loss_avg=3.62074 acc=0.67188 acc_top1_avg=0.65009 acc_top5_avg=0.94212 lr=0.01000 gn=8.12758 time=57.51it/s +epoch=12 global_step=5050 loss=3.21688 loss_avg=3.61116 acc=0.71094 acc_top1_avg=0.65106 acc_top5_avg=0.94186 lr=0.01000 gn=7.53349 time=53.52it/s +====================Eval==================== +epoch=12 global_step=5083 loss=1.22205 test_loss_avg=1.31242 acc=0.71875 test_acc_avg=0.69141 test_acc_top5_avg=0.98828 time=229.17it/s +epoch=12 global_step=5083 loss=0.13454 test_loss_avg=1.11390 acc=0.95312 test_acc_avg=0.73017 test_acc_top5_avg=0.97010 time=208.69it/s +epoch=12 global_step=5083 loss=0.42599 test_loss_avg=0.96492 acc=0.87500 test_acc_avg=0.76197 test_acc_top5_avg=0.97280 time=842.06it/s +curr_acc 0.7620 +BEST_ACC 0.7735 +curr_acc_top5 0.9728 +BEST_ACC_top5 0.9775 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=3.81502 loss_avg=3.56195 acc=0.63281 acc_top1_avg=0.65763 acc_top5_avg=0.93888 lr=0.01000 gn=10.08154 time=54.97it/s +epoch=13 global_step=5150 loss=3.17115 loss_avg=3.59465 acc=0.70312 acc_top1_avg=0.65275 acc_top5_avg=0.94298 lr=0.01000 gn=8.69338 time=48.97it/s +epoch=13 global_step=5200 loss=3.30823 loss_avg=3.52741 acc=0.68750 acc_top1_avg=0.65859 acc_top5_avg=0.94251 lr=0.01000 gn=9.21301 time=55.27it/s +epoch=13 global_step=5250 loss=3.29901 loss_avg=3.53252 acc=0.67969 acc_top1_avg=0.65854 acc_top5_avg=0.94241 lr=0.01000 gn=7.19377 time=57.59it/s +epoch=13 global_step=5300 loss=3.52473 loss_avg=3.52924 acc=0.67188 acc_top1_avg=0.65870 acc_top5_avg=0.94272 lr=0.01000 gn=8.99009 time=51.49it/s +epoch=13 global_step=5350 loss=3.85254 loss_avg=3.53584 acc=0.64844 acc_top1_avg=0.65827 acc_top5_avg=0.94256 lr=0.01000 gn=10.22814 time=42.02it/s +epoch=13 global_step=5400 loss=3.32520 loss_avg=3.55571 acc=0.67188 acc_top1_avg=0.65669 acc_top5_avg=0.94201 lr=0.01000 gn=8.21033 time=61.76it/s +epoch=13 global_step=5450 loss=3.78541 loss_avg=3.57430 acc=0.61719 acc_top1_avg=0.65487 acc_top5_avg=0.94174 lr=0.01000 gn=9.48471 time=56.54it/s +====================Eval==================== +epoch=13 global_step=5474 loss=2.16449 test_loss_avg=1.31357 acc=0.46094 test_acc_avg=0.68784 test_acc_top5_avg=0.95720 time=240.72it/s +epoch=13 global_step=5474 loss=0.47671 test_loss_avg=0.95393 acc=0.89062 test_acc_avg=0.76124 test_acc_top5_avg=0.96854 time=246.30it/s +epoch=13 global_step=5474 loss=0.21373 test_loss_avg=0.91012 acc=0.93750 test_acc_avg=0.77146 test_acc_top5_avg=0.97063 time=512.94it/s +curr_acc 0.7715 +BEST_ACC 0.7735 +curr_acc_top5 0.9706 +BEST_ACC_top5 0.9775 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=3.50695 loss_avg=3.52685 acc=0.64844 acc_top1_avg=0.66076 acc_top5_avg=0.94832 lr=0.01000 gn=8.55571 time=55.57it/s +epoch=14 global_step=5550 loss=3.56964 loss_avg=3.53504 acc=0.67188 acc_top1_avg=0.65995 acc_top5_avg=0.94223 lr=0.01000 gn=10.41763 time=53.31it/s +epoch=14 global_step=5600 loss=2.74442 loss_avg=3.51546 acc=0.73438 acc_top1_avg=0.66071 acc_top5_avg=0.94178 lr=0.01000 gn=8.04640 time=62.20it/s +epoch=14 global_step=5650 loss=3.37307 loss_avg=3.51373 acc=0.69531 acc_top1_avg=0.66202 acc_top5_avg=0.94136 lr=0.01000 gn=10.00913 time=51.06it/s +epoch=14 global_step=5700 loss=3.83189 loss_avg=3.52143 acc=0.64062 acc_top1_avg=0.66099 acc_top5_avg=0.94248 lr=0.01000 gn=9.15954 time=58.42it/s +epoch=14 global_step=5750 loss=2.94576 loss_avg=3.54108 acc=0.71875 acc_top1_avg=0.65854 acc_top5_avg=0.94226 lr=0.01000 gn=9.58706 time=62.75it/s +epoch=14 global_step=5800 loss=3.79270 loss_avg=3.54486 acc=0.62500 acc_top1_avg=0.65822 acc_top5_avg=0.94268 lr=0.01000 gn=10.88142 time=53.92it/s +epoch=14 global_step=5850 loss=3.04393 loss_avg=3.54843 acc=0.71875 acc_top1_avg=0.65770 acc_top5_avg=0.94236 lr=0.01000 gn=8.85152 time=62.59it/s +====================Eval==================== +epoch=14 global_step=5865 loss=3.09908 test_loss_avg=0.89719 acc=0.40625 test_acc_avg=0.77344 test_acc_top5_avg=0.98065 time=233.21it/s +epoch=14 global_step=5865 loss=1.50337 test_loss_avg=0.98594 acc=0.62500 test_acc_avg=0.75267 test_acc_top5_avg=0.97182 time=506.93it/s +curr_acc 0.7527 +BEST_ACC 0.7735 +curr_acc_top5 0.9718 +BEST_ACC_top5 0.9775 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=3.62930 loss_avg=3.49580 acc=0.63281 acc_top1_avg=0.66250 acc_top5_avg=0.94978 lr=0.01000 gn=9.59571 time=44.85it/s +epoch=15 global_step=5950 loss=3.52030 loss_avg=3.53229 acc=0.67188 acc_top1_avg=0.65901 acc_top5_avg=0.94375 lr=0.01000 gn=11.73771 time=52.09it/s +epoch=15 global_step=6000 loss=3.38906 loss_avg=3.55959 acc=0.67969 acc_top1_avg=0.65567 acc_top5_avg=0.94375 lr=0.01000 gn=9.30604 time=58.57it/s +epoch=15 global_step=6050 loss=3.31729 loss_avg=3.56063 acc=0.68750 acc_top1_avg=0.65608 acc_top5_avg=0.94413 lr=0.01000 gn=8.78842 time=59.06it/s +epoch=15 global_step=6100 loss=3.31886 loss_avg=3.56605 acc=0.69531 acc_top1_avg=0.65545 acc_top5_avg=0.94372 lr=0.01000 gn=9.72232 time=55.89it/s +epoch=15 global_step=6150 loss=2.82042 loss_avg=3.55488 acc=0.72656 acc_top1_avg=0.65691 acc_top5_avg=0.94378 lr=0.01000 gn=10.90659 time=54.17it/s +epoch=15 global_step=6200 loss=3.74842 loss_avg=3.55026 acc=0.64844 acc_top1_avg=0.65718 acc_top5_avg=0.94289 lr=0.01000 gn=11.03192 time=52.16it/s +epoch=15 global_step=6250 loss=3.95699 loss_avg=3.54361 acc=0.61719 acc_top1_avg=0.65787 acc_top5_avg=0.94247 lr=0.01000 gn=8.74896 time=60.23it/s +====================Eval==================== +epoch=15 global_step=6256 loss=0.85440 test_loss_avg=1.00946 acc=0.72656 test_acc_avg=0.71042 test_acc_top5_avg=0.96823 time=243.56it/s +epoch=15 global_step=6256 loss=0.40984 test_loss_avg=0.91494 acc=0.88281 test_acc_avg=0.75577 test_acc_top5_avg=0.97993 time=239.81it/s +epoch=15 global_step=6256 loss=0.03713 test_loss_avg=0.82159 acc=1.00000 test_acc_avg=0.77789 test_acc_top5_avg=0.98210 time=582.62it/s +curr_acc 0.7779 +BEST_ACC 0.7735 +curr_acc_top5 0.9821 +BEST_ACC_top5 0.9775 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=3.96728 loss_avg=3.59068 acc=0.63281 acc_top1_avg=0.65536 acc_top5_avg=0.93981 lr=0.01000 gn=10.69806 time=52.77it/s +epoch=16 global_step=6350 loss=3.55163 loss_avg=3.50930 acc=0.67188 acc_top1_avg=0.66298 acc_top5_avg=0.94091 lr=0.01000 gn=10.31453 time=51.90it/s +epoch=16 global_step=6400 loss=3.89795 loss_avg=3.52510 acc=0.62500 acc_top1_avg=0.66178 acc_top5_avg=0.94097 lr=0.01000 gn=12.51586 time=58.45it/s +epoch=16 global_step=6450 loss=3.57069 loss_avg=3.52612 acc=0.64844 acc_top1_avg=0.66144 acc_top5_avg=0.94165 lr=0.01000 gn=8.54028 time=60.89it/s +epoch=16 global_step=6500 loss=3.20307 loss_avg=3.53162 acc=0.69531 acc_top1_avg=0.66000 acc_top5_avg=0.94246 lr=0.01000 gn=9.04498 time=54.87it/s +epoch=16 global_step=6550 loss=4.22924 loss_avg=3.52766 acc=0.58594 acc_top1_avg=0.66048 acc_top5_avg=0.94266 lr=0.01000 gn=9.64171 time=58.16it/s +epoch=16 global_step=6600 loss=4.01936 loss_avg=3.52259 acc=0.60938 acc_top1_avg=0.66070 acc_top5_avg=0.94345 lr=0.01000 gn=8.29759 time=52.46it/s +====================Eval==================== +epoch=16 global_step=6647 loss=1.49658 test_loss_avg=0.90284 acc=0.62500 test_acc_avg=0.74653 test_acc_top5_avg=0.97418 time=241.44it/s +epoch=16 global_step=6647 loss=0.80632 test_loss_avg=0.86235 acc=0.87500 test_acc_avg=0.76424 test_acc_top5_avg=0.97231 time=840.37it/s +curr_acc 0.7642 +BEST_ACC 0.7779 +curr_acc_top5 0.9723 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=3.19334 loss_avg=3.22721 acc=0.70312 acc_top1_avg=0.69531 acc_top5_avg=0.96094 lr=0.01000 gn=12.51537 time=50.10it/s +epoch=17 global_step=6700 loss=3.17515 loss_avg=3.49124 acc=0.68750 acc_top1_avg=0.66126 acc_top5_avg=0.94782 lr=0.01000 gn=8.91114 time=57.59it/s +epoch=17 global_step=6750 loss=3.59483 loss_avg=3.50072 acc=0.66406 acc_top1_avg=0.66217 acc_top5_avg=0.94622 lr=0.01000 gn=10.01755 time=51.87it/s +epoch=17 global_step=6800 loss=3.24832 loss_avg=3.47777 acc=0.67969 acc_top1_avg=0.66457 acc_top5_avg=0.94552 lr=0.01000 gn=9.15942 time=57.46it/s +epoch=17 global_step=6850 loss=2.89375 loss_avg=3.47605 acc=0.72656 acc_top1_avg=0.66537 acc_top5_avg=0.94508 lr=0.01000 gn=9.32194 time=53.26it/s +epoch=17 global_step=6900 loss=3.68008 loss_avg=3.49012 acc=0.65625 acc_top1_avg=0.66375 acc_top5_avg=0.94463 lr=0.01000 gn=8.88998 time=55.15it/s +epoch=17 global_step=6950 loss=3.73361 loss_avg=3.50758 acc=0.64844 acc_top1_avg=0.66223 acc_top5_avg=0.94397 lr=0.01000 gn=8.95513 time=49.70it/s +epoch=17 global_step=7000 loss=3.87153 loss_avg=3.51203 acc=0.64844 acc_top1_avg=0.66169 acc_top5_avg=0.94412 lr=0.01000 gn=11.36598 time=62.96it/s +====================Eval==================== +epoch=17 global_step=7038 loss=1.01499 test_loss_avg=1.05673 acc=0.74219 test_acc_avg=0.71540 test_acc_top5_avg=0.97545 time=239.43it/s +epoch=17 global_step=7038 loss=0.37529 test_loss_avg=1.47604 acc=0.90625 test_acc_avg=0.65721 test_acc_top5_avg=0.95450 time=235.04it/s +epoch=17 global_step=7038 loss=1.30827 test_loss_avg=1.30216 acc=0.62500 test_acc_avg=0.69027 test_acc_top5_avg=0.95827 time=509.39it/s +curr_acc 0.6903 +BEST_ACC 0.7779 +curr_acc_top5 0.9583 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=3.66269 loss_avg=3.45729 acc=0.63281 acc_top1_avg=0.67383 acc_top5_avg=0.94987 lr=0.01000 gn=11.74840 time=59.49it/s +epoch=18 global_step=7100 loss=3.92533 loss_avg=3.50941 acc=0.62500 acc_top1_avg=0.66280 acc_top5_avg=0.94317 lr=0.01000 gn=8.43436 time=56.43it/s +epoch=18 global_step=7150 loss=3.24929 loss_avg=3.50897 acc=0.69531 acc_top1_avg=0.66267 acc_top5_avg=0.94266 lr=0.01000 gn=9.67546 time=54.55it/s +epoch=18 global_step=7200 loss=3.92154 loss_avg=3.51711 acc=0.60938 acc_top1_avg=0.66223 acc_top5_avg=0.94194 lr=0.01000 gn=9.30166 time=60.61it/s +epoch=18 global_step=7250 loss=3.38414 loss_avg=3.52585 acc=0.67969 acc_top1_avg=0.66104 acc_top5_avg=0.94321 lr=0.01000 gn=8.72141 time=62.94it/s +epoch=18 global_step=7300 loss=3.52975 loss_avg=3.52096 acc=0.64844 acc_top1_avg=0.66150 acc_top5_avg=0.94400 lr=0.01000 gn=8.90086 time=53.70it/s +epoch=18 global_step=7350 loss=3.69967 loss_avg=3.52255 acc=0.64062 acc_top1_avg=0.66113 acc_top5_avg=0.94436 lr=0.01000 gn=8.09944 time=59.44it/s +epoch=18 global_step=7400 loss=3.56749 loss_avg=3.51451 acc=0.67188 acc_top1_avg=0.66175 acc_top5_avg=0.94555 lr=0.01000 gn=11.63617 time=62.32it/s +====================Eval==================== +epoch=18 global_step=7429 loss=2.01242 test_loss_avg=0.86943 acc=0.53125 test_acc_avg=0.76172 test_acc_top5_avg=0.97377 time=240.86it/s +epoch=18 global_step=7429 loss=0.47506 test_loss_avg=0.77772 acc=0.89062 test_acc_avg=0.78956 test_acc_top5_avg=0.97546 time=248.99it/s +epoch=18 global_step=7429 loss=0.10899 test_loss_avg=0.76926 acc=0.93750 test_acc_avg=0.79144 test_acc_top5_avg=0.97577 time=522.00it/s +curr_acc 0.7914 +BEST_ACC 0.7779 +curr_acc_top5 0.9758 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=3.19332 loss_avg=3.21712 acc=0.68750 acc_top1_avg=0.69345 acc_top5_avg=0.94271 lr=0.01000 gn=6.58071 time=51.92it/s +epoch=19 global_step=7500 loss=3.28563 loss_avg=3.35989 acc=0.68750 acc_top1_avg=0.67848 acc_top5_avg=0.94366 lr=0.01000 gn=10.70408 time=63.67it/s +epoch=19 global_step=7550 loss=2.99519 loss_avg=3.45723 acc=0.71094 acc_top1_avg=0.66813 acc_top5_avg=0.94396 lr=0.01000 gn=8.73372 time=59.96it/s +epoch=19 global_step=7600 loss=3.84714 loss_avg=3.46723 acc=0.62500 acc_top1_avg=0.66630 acc_top5_avg=0.94380 lr=0.01000 gn=9.02469 time=54.46it/s +epoch=19 global_step=7650 loss=3.38941 loss_avg=3.47770 acc=0.65625 acc_top1_avg=0.66523 acc_top5_avg=0.94439 lr=0.01000 gn=10.11835 time=56.22it/s +epoch=19 global_step=7700 loss=3.53590 loss_avg=3.47918 acc=0.66406 acc_top1_avg=0.66501 acc_top5_avg=0.94384 lr=0.01000 gn=9.24880 time=52.09it/s +epoch=19 global_step=7750 loss=3.83407 loss_avg=3.49309 acc=0.61719 acc_top1_avg=0.66319 acc_top5_avg=0.94470 lr=0.01000 gn=9.57191 time=56.32it/s +epoch=19 global_step=7800 loss=3.82008 loss_avg=3.49846 acc=0.62500 acc_top1_avg=0.66276 acc_top5_avg=0.94502 lr=0.01000 gn=8.35033 time=53.77it/s +====================Eval==================== +epoch=19 global_step=7820 loss=0.57616 test_loss_avg=1.05487 acc=0.84375 test_acc_avg=0.72592 test_acc_top5_avg=0.96205 time=253.13it/s +epoch=19 global_step=7820 loss=0.86989 test_loss_avg=0.93153 acc=0.75000 test_acc_avg=0.75623 test_acc_top5_avg=0.96974 time=757.23it/s +curr_acc 0.7562 +BEST_ACC 0.7914 +curr_acc_top5 0.9697 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=3.89373 loss_avg=3.42012 acc=0.63281 acc_top1_avg=0.67214 acc_top5_avg=0.94427 lr=0.01000 gn=8.43852 time=60.42it/s +epoch=20 global_step=7900 loss=3.77722 loss_avg=3.48423 acc=0.62500 acc_top1_avg=0.66367 acc_top5_avg=0.94307 lr=0.01000 gn=8.14661 time=62.32it/s +epoch=20 global_step=7950 loss=2.58577 loss_avg=3.45144 acc=0.75781 acc_top1_avg=0.66653 acc_top5_avg=0.94447 lr=0.01000 gn=7.40312 time=49.58it/s +epoch=20 global_step=8000 loss=3.43095 loss_avg=3.44013 acc=0.67188 acc_top1_avg=0.66780 acc_top5_avg=0.94557 lr=0.01000 gn=8.34865 time=56.34it/s +epoch=20 global_step=8050 loss=3.15406 loss_avg=3.45049 acc=0.70312 acc_top1_avg=0.66647 acc_top5_avg=0.94616 lr=0.01000 gn=9.05128 time=53.80it/s +epoch=20 global_step=8100 loss=3.34840 loss_avg=3.46689 acc=0.67188 acc_top1_avg=0.66493 acc_top5_avg=0.94590 lr=0.01000 gn=7.43590 time=51.76it/s +epoch=20 global_step=8150 loss=3.75756 loss_avg=3.47740 acc=0.63281 acc_top1_avg=0.66335 acc_top5_avg=0.94607 lr=0.01000 gn=7.93658 time=53.54it/s +epoch=20 global_step=8200 loss=3.86755 loss_avg=3.49496 acc=0.62500 acc_top1_avg=0.66176 acc_top5_avg=0.94599 lr=0.01000 gn=9.39493 time=55.33it/s +====================Eval==================== +epoch=20 global_step=8211 loss=2.12591 test_loss_avg=0.95485 acc=0.49219 test_acc_avg=0.75820 test_acc_top5_avg=0.97383 time=243.32it/s +epoch=20 global_step=8211 loss=0.09983 test_loss_avg=1.28889 acc=0.96875 test_acc_avg=0.70737 test_acc_top5_avg=0.95033 time=237.15it/s +epoch=20 global_step=8211 loss=1.75183 test_loss_avg=1.30486 acc=0.43750 test_acc_avg=0.69788 test_acc_top5_avg=0.95402 time=382.17it/s +curr_acc 0.6979 +BEST_ACC 0.7914 +curr_acc_top5 0.9540 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=4.05314 loss_avg=3.45281 acc=0.60156 acc_top1_avg=0.67167 acc_top5_avg=0.94852 lr=0.01000 gn=11.11462 time=50.96it/s +epoch=21 global_step=8300 loss=3.67561 loss_avg=3.49038 acc=0.63281 acc_top1_avg=0.66608 acc_top5_avg=0.94733 lr=0.01000 gn=9.06285 time=53.07it/s +epoch=21 global_step=8350 loss=3.08847 loss_avg=3.49157 acc=0.71094 acc_top1_avg=0.66552 acc_top5_avg=0.94542 lr=0.01000 gn=8.19474 time=55.44it/s +epoch=21 global_step=8400 loss=3.34213 loss_avg=3.47143 acc=0.67188 acc_top1_avg=0.66683 acc_top5_avg=0.94527 lr=0.01000 gn=8.35548 time=58.69it/s +epoch=21 global_step=8450 loss=3.72864 loss_avg=3.49461 acc=0.64062 acc_top1_avg=0.66410 acc_top5_avg=0.94502 lr=0.01000 gn=9.47846 time=58.02it/s +epoch=21 global_step=8500 loss=3.94976 loss_avg=3.50234 acc=0.62500 acc_top1_avg=0.66306 acc_top5_avg=0.94456 lr=0.01000 gn=10.64171 time=59.96it/s +epoch=21 global_step=8550 loss=3.82074 loss_avg=3.49725 acc=0.60938 acc_top1_avg=0.66342 acc_top5_avg=0.94444 lr=0.01000 gn=11.33570 time=54.63it/s +epoch=21 global_step=8600 loss=3.78963 loss_avg=3.49423 acc=0.63281 acc_top1_avg=0.66360 acc_top5_avg=0.94493 lr=0.01000 gn=8.65632 time=62.93it/s +====================Eval==================== +epoch=21 global_step=8602 loss=1.71693 test_loss_avg=1.91750 acc=0.67969 test_acc_avg=0.61147 test_acc_top5_avg=0.96494 time=235.82it/s +epoch=21 global_step=8602 loss=1.11081 test_loss_avg=1.32930 acc=0.56250 test_acc_avg=0.71776 test_acc_top5_avg=0.97636 time=714.05it/s +curr_acc 0.7178 +BEST_ACC 0.7914 +curr_acc_top5 0.9764 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=3.20305 loss_avg=3.38333 acc=0.69531 acc_top1_avg=0.67301 acc_top5_avg=0.94710 lr=0.01000 gn=9.16831 time=52.98it/s +epoch=22 global_step=8700 loss=3.60026 loss_avg=3.39767 acc=0.64062 acc_top1_avg=0.67068 acc_top5_avg=0.94635 lr=0.01000 gn=11.44035 time=54.69it/s +epoch=22 global_step=8750 loss=3.74093 loss_avg=3.40452 acc=0.64844 acc_top1_avg=0.67050 acc_top5_avg=0.94705 lr=0.01000 gn=7.27714 time=53.47it/s +epoch=22 global_step=8800 loss=3.32495 loss_avg=3.41081 acc=0.66406 acc_top1_avg=0.67053 acc_top5_avg=0.94721 lr=0.01000 gn=8.05328 time=54.02it/s +epoch=22 global_step=8850 loss=4.07030 loss_avg=3.43495 acc=0.58594 acc_top1_avg=0.66803 acc_top5_avg=0.94642 lr=0.01000 gn=9.59697 time=60.91it/s +epoch=22 global_step=8900 loss=3.26766 loss_avg=3.46590 acc=0.67188 acc_top1_avg=0.66495 acc_top5_avg=0.94505 lr=0.01000 gn=10.63844 time=59.58it/s +epoch=22 global_step=8950 loss=3.44457 loss_avg=3.45706 acc=0.65625 acc_top1_avg=0.66586 acc_top5_avg=0.94520 lr=0.01000 gn=9.46552 time=49.96it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.26570 test_loss_avg=1.30818 acc=0.92188 test_acc_avg=0.70443 test_acc_top5_avg=0.98242 time=140.79it/s +epoch=22 global_step=8993 loss=1.45240 test_loss_avg=1.31506 acc=0.66406 test_acc_avg=0.69115 test_acc_top5_avg=0.96736 time=243.18it/s +epoch=22 global_step=8993 loss=0.84754 test_loss_avg=1.10993 acc=0.87500 test_acc_avg=0.73794 test_acc_top5_avg=0.97379 time=495.96it/s +curr_acc 0.7379 +BEST_ACC 0.7914 +curr_acc_top5 0.9738 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=3.32057 loss_avg=3.34888 acc=0.67188 acc_top1_avg=0.69196 acc_top5_avg=0.94643 lr=0.01000 gn=9.34745 time=47.45it/s +epoch=23 global_step=9050 loss=3.75541 loss_avg=3.43792 acc=0.62500 acc_top1_avg=0.67023 acc_top5_avg=0.94833 lr=0.01000 gn=9.01092 time=54.06it/s +epoch=23 global_step=9100 loss=3.09108 loss_avg=3.50575 acc=0.70312 acc_top1_avg=0.66195 acc_top5_avg=0.94582 lr=0.01000 gn=8.59901 time=39.23it/s +epoch=23 global_step=9150 loss=3.29068 loss_avg=3.51792 acc=0.67188 acc_top1_avg=0.65988 acc_top5_avg=0.94621 lr=0.01000 gn=9.95036 time=59.27it/s +epoch=23 global_step=9200 loss=4.37785 loss_avg=3.50323 acc=0.57031 acc_top1_avg=0.66108 acc_top5_avg=0.94671 lr=0.01000 gn=9.52229 time=55.50it/s +epoch=23 global_step=9250 loss=3.83789 loss_avg=3.50836 acc=0.61719 acc_top1_avg=0.66060 acc_top5_avg=0.94513 lr=0.01000 gn=9.57521 time=61.24it/s +epoch=23 global_step=9300 loss=2.82810 loss_avg=3.49960 acc=0.74219 acc_top1_avg=0.66180 acc_top5_avg=0.94526 lr=0.01000 gn=10.33701 time=43.78it/s +epoch=23 global_step=9350 loss=3.51490 loss_avg=3.50111 acc=0.67188 acc_top1_avg=0.66150 acc_top5_avg=0.94522 lr=0.01000 gn=10.35569 time=55.49it/s +====================Eval==================== +epoch=23 global_step=9384 loss=1.27589 test_loss_avg=1.02840 acc=0.67188 test_acc_avg=0.72917 test_acc_top5_avg=0.96662 time=241.54it/s +epoch=23 global_step=9384 loss=0.22630 test_loss_avg=0.84637 acc=0.87500 test_acc_avg=0.77304 test_acc_top5_avg=0.96905 time=612.49it/s +curr_acc 0.7730 +BEST_ACC 0.7914 +curr_acc_top5 0.9690 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=3.30822 loss_avg=3.56098 acc=0.67188 acc_top1_avg=0.65430 acc_top5_avg=0.93994 lr=0.01000 gn=10.67333 time=59.93it/s +epoch=24 global_step=9450 loss=4.04406 loss_avg=3.43570 acc=0.62500 acc_top1_avg=0.66761 acc_top5_avg=0.94425 lr=0.01000 gn=11.82415 time=53.93it/s +epoch=24 global_step=9500 loss=3.73079 loss_avg=3.41706 acc=0.64844 acc_top1_avg=0.66925 acc_top5_avg=0.94531 lr=0.01000 gn=10.10221 time=58.41it/s +epoch=24 global_step=9550 loss=4.20557 loss_avg=3.44826 acc=0.58594 acc_top1_avg=0.66665 acc_top5_avg=0.94437 lr=0.01000 gn=8.41922 time=55.30it/s +epoch=24 global_step=9600 loss=3.54469 loss_avg=3.47867 acc=0.67188 acc_top1_avg=0.66359 acc_top5_avg=0.94452 lr=0.01000 gn=8.96932 time=62.48it/s +epoch=24 global_step=9650 loss=4.01250 loss_avg=3.48914 acc=0.60156 acc_top1_avg=0.66274 acc_top5_avg=0.94481 lr=0.01000 gn=11.51952 time=54.93it/s +epoch=24 global_step=9700 loss=3.68488 loss_avg=3.47031 acc=0.65625 acc_top1_avg=0.66485 acc_top5_avg=0.94549 lr=0.01000 gn=10.10926 time=48.87it/s +epoch=24 global_step=9750 loss=3.01760 loss_avg=3.47880 acc=0.70312 acc_top1_avg=0.66421 acc_top5_avg=0.94553 lr=0.01000 gn=10.23825 time=62.69it/s +====================Eval==================== +epoch=24 global_step=9775 loss=0.99625 test_loss_avg=0.97970 acc=0.73438 test_acc_avg=0.73828 test_acc_top5_avg=0.98633 time=242.92it/s +epoch=24 global_step=9775 loss=1.86615 test_loss_avg=1.29459 acc=0.56250 test_acc_avg=0.66739 test_acc_top5_avg=0.95255 time=233.60it/s +epoch=24 global_step=9775 loss=0.19250 test_loss_avg=1.07720 acc=0.93750 test_acc_avg=0.71845 test_acc_top5_avg=0.96331 time=514.58it/s +curr_acc 0.7185 +BEST_ACC 0.7914 +curr_acc_top5 0.9633 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=3.41451 loss_avg=3.32589 acc=0.66406 acc_top1_avg=0.68219 acc_top5_avg=0.94875 lr=0.01000 gn=9.49409 time=57.73it/s +epoch=25 global_step=9850 loss=3.55940 loss_avg=3.47208 acc=0.66406 acc_top1_avg=0.66417 acc_top5_avg=0.94573 lr=0.01000 gn=8.89987 time=54.64it/s +epoch=25 global_step=9900 loss=3.22226 loss_avg=3.45137 acc=0.70312 acc_top1_avg=0.66750 acc_top5_avg=0.94675 lr=0.01000 gn=8.79536 time=51.13it/s +epoch=25 global_step=9950 loss=3.00662 loss_avg=3.44618 acc=0.74219 acc_top1_avg=0.66871 acc_top5_avg=0.94621 lr=0.01000 gn=10.33672 time=60.41it/s +epoch=25 global_step=10000 loss=3.06753 loss_avg=3.44278 acc=0.73438 acc_top1_avg=0.66906 acc_top5_avg=0.94639 lr=0.01000 gn=11.53933 time=54.43it/s +epoch=25 global_step=10050 loss=3.82351 loss_avg=3.45478 acc=0.63281 acc_top1_avg=0.66798 acc_top5_avg=0.94597 lr=0.01000 gn=7.71107 time=56.61it/s +epoch=25 global_step=10100 loss=3.98106 loss_avg=3.45700 acc=0.63281 acc_top1_avg=0.66769 acc_top5_avg=0.94603 lr=0.01000 gn=11.81211 time=39.99it/s +epoch=25 global_step=10150 loss=3.59812 loss_avg=3.46537 acc=0.66406 acc_top1_avg=0.66677 acc_top5_avg=0.94610 lr=0.01000 gn=8.98739 time=55.06it/s +====================Eval==================== +epoch=25 global_step=10166 loss=1.87305 test_loss_avg=0.87521 acc=0.57031 test_acc_avg=0.75906 test_acc_top5_avg=0.97844 time=243.64it/s +epoch=25 global_step=10166 loss=0.44996 test_loss_avg=0.84332 acc=0.85938 test_acc_avg=0.77885 test_acc_top5_avg=0.97208 time=242.24it/s +epoch=25 global_step=10166 loss=0.70822 test_loss_avg=0.83214 acc=0.81250 test_acc_avg=0.78165 test_acc_top5_avg=0.97330 time=677.81it/s +curr_acc 0.7816 +BEST_ACC 0.7914 +curr_acc_top5 0.9733 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=3.59152 loss_avg=3.44215 acc=0.66406 acc_top1_avg=0.66590 acc_top5_avg=0.94830 lr=0.01000 gn=10.05973 time=54.82it/s +epoch=26 global_step=10250 loss=3.49586 loss_avg=3.38711 acc=0.65625 acc_top1_avg=0.67429 acc_top5_avg=0.94829 lr=0.01000 gn=9.73769 time=53.22it/s +epoch=26 global_step=10300 loss=3.08022 loss_avg=3.43094 acc=0.71094 acc_top1_avg=0.66849 acc_top5_avg=0.94619 lr=0.01000 gn=10.81134 time=59.22it/s +epoch=26 global_step=10350 loss=3.66235 loss_avg=3.42933 acc=0.64844 acc_top1_avg=0.66890 acc_top5_avg=0.94544 lr=0.01000 gn=10.43530 time=61.90it/s +epoch=26 global_step=10400 loss=4.04031 loss_avg=3.44375 acc=0.59375 acc_top1_avg=0.66743 acc_top5_avg=0.94551 lr=0.01000 gn=8.38766 time=55.65it/s +epoch=26 global_step=10450 loss=3.57015 loss_avg=3.46382 acc=0.62500 acc_top1_avg=0.66563 acc_top5_avg=0.94460 lr=0.01000 gn=10.17600 time=62.25it/s +epoch=26 global_step=10500 loss=3.53480 loss_avg=3.46774 acc=0.64844 acc_top1_avg=0.66497 acc_top5_avg=0.94456 lr=0.01000 gn=8.91576 time=53.94it/s +epoch=26 global_step=10550 loss=3.51839 loss_avg=3.46597 acc=0.67188 acc_top1_avg=0.66526 acc_top5_avg=0.94482 lr=0.01000 gn=8.81580 time=54.78it/s +====================Eval==================== +epoch=26 global_step=10557 loss=3.49117 test_loss_avg=1.37709 acc=0.35938 test_acc_avg=0.68393 test_acc_top5_avg=0.97164 time=196.91it/s +epoch=26 global_step=10557 loss=0.78593 test_loss_avg=1.26432 acc=0.68750 test_acc_avg=0.69739 test_acc_top5_avg=0.96282 time=499.08it/s +curr_acc 0.6974 +BEST_ACC 0.7914 +curr_acc_top5 0.9628 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=2.60580 loss_avg=3.46996 acc=0.75781 acc_top1_avg=0.66170 acc_top5_avg=0.94531 lr=0.01000 gn=10.84574 time=54.78it/s +epoch=27 global_step=10650 loss=3.42872 loss_avg=3.41751 acc=0.68750 acc_top1_avg=0.66927 acc_top5_avg=0.94934 lr=0.01000 gn=11.06509 time=57.15it/s +epoch=27 global_step=10700 loss=3.51404 loss_avg=3.39700 acc=0.62500 acc_top1_avg=0.67269 acc_top5_avg=0.94925 lr=0.01000 gn=9.04923 time=52.88it/s +epoch=27 global_step=10750 loss=3.08394 loss_avg=3.41117 acc=0.69531 acc_top1_avg=0.67111 acc_top5_avg=0.94798 lr=0.01000 gn=8.84866 time=52.97it/s +epoch=27 global_step=10800 loss=3.77457 loss_avg=3.43001 acc=0.62500 acc_top1_avg=0.66911 acc_top5_avg=0.94763 lr=0.01000 gn=11.65417 time=55.65it/s +epoch=27 global_step=10850 loss=3.85979 loss_avg=3.44204 acc=0.61719 acc_top1_avg=0.66790 acc_top5_avg=0.94737 lr=0.01000 gn=9.37992 time=62.81it/s +epoch=27 global_step=10900 loss=3.60599 loss_avg=3.45356 acc=0.66406 acc_top1_avg=0.66673 acc_top5_avg=0.94711 lr=0.01000 gn=7.27003 time=54.38it/s +====================Eval==================== +epoch=27 global_step=10948 loss=1.08023 test_loss_avg=0.73753 acc=0.72656 test_acc_avg=0.79320 test_acc_top5_avg=0.99357 time=233.86it/s +epoch=27 global_step=10948 loss=0.19853 test_loss_avg=1.06548 acc=0.92969 test_acc_avg=0.73659 test_acc_top5_avg=0.96828 time=242.17it/s +epoch=27 global_step=10948 loss=0.35319 test_loss_avg=1.04546 acc=0.75000 test_acc_avg=0.73774 test_acc_top5_avg=0.96974 time=832.53it/s +curr_acc 0.7377 +BEST_ACC 0.7914 +curr_acc_top5 0.9697 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=3.78061 loss_avg=3.55198 acc=0.63281 acc_top1_avg=0.66016 acc_top5_avg=0.96094 lr=0.01000 gn=9.47985 time=55.19it/s +epoch=28 global_step=11000 loss=3.38050 loss_avg=3.45119 acc=0.67188 acc_top1_avg=0.67067 acc_top5_avg=0.94636 lr=0.01000 gn=11.32440 time=42.81it/s +epoch=28 global_step=11050 loss=3.40030 loss_avg=3.47798 acc=0.66406 acc_top1_avg=0.66697 acc_top5_avg=0.94508 lr=0.01000 gn=10.87956 time=50.90it/s +epoch=28 global_step=11100 loss=3.69668 loss_avg=3.47476 acc=0.64844 acc_top1_avg=0.66643 acc_top5_avg=0.94660 lr=0.01000 gn=12.19688 time=59.03it/s +epoch=28 global_step=11150 loss=3.29170 loss_avg=3.44105 acc=0.70312 acc_top1_avg=0.66963 acc_top5_avg=0.94667 lr=0.01000 gn=10.74887 time=57.28it/s +epoch=28 global_step=11200 loss=3.57460 loss_avg=3.43811 acc=0.65625 acc_top1_avg=0.66983 acc_top5_avg=0.94590 lr=0.01000 gn=10.42329 time=55.28it/s +epoch=28 global_step=11250 loss=3.29897 loss_avg=3.45046 acc=0.68750 acc_top1_avg=0.66825 acc_top5_avg=0.94477 lr=0.01000 gn=8.30881 time=62.74it/s +epoch=28 global_step=11300 loss=4.01180 loss_avg=3.44043 acc=0.58594 acc_top1_avg=0.66954 acc_top5_avg=0.94531 lr=0.01000 gn=9.34095 time=56.34it/s +====================Eval==================== +epoch=28 global_step=11339 loss=1.42490 test_loss_avg=1.36434 acc=0.67188 test_acc_avg=0.69305 test_acc_top5_avg=0.96669 time=238.86it/s +epoch=28 global_step=11339 loss=0.32334 test_loss_avg=1.23742 acc=0.81250 test_acc_avg=0.71855 test_acc_top5_avg=0.97330 time=843.75it/s +curr_acc 0.7186 +BEST_ACC 0.7914 +curr_acc_top5 0.9733 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=3.35292 loss_avg=3.24736 acc=0.64844 acc_top1_avg=0.68821 acc_top5_avg=0.95099 lr=0.01000 gn=8.75144 time=60.03it/s +epoch=29 global_step=11400 loss=3.42259 loss_avg=3.46467 acc=0.67188 acc_top1_avg=0.66496 acc_top5_avg=0.95082 lr=0.01000 gn=11.48846 time=33.94it/s +epoch=29 global_step=11450 loss=3.91275 loss_avg=3.45361 acc=0.60938 acc_top1_avg=0.66610 acc_top5_avg=0.94742 lr=0.01000 gn=8.25990 time=56.47it/s +epoch=29 global_step=11500 loss=3.05460 loss_avg=3.42788 acc=0.71875 acc_top1_avg=0.67071 acc_top5_avg=0.94687 lr=0.01000 gn=11.14265 time=56.95it/s +epoch=29 global_step=11550 loss=3.16230 loss_avg=3.42845 acc=0.68750 acc_top1_avg=0.67095 acc_top5_avg=0.94609 lr=0.01000 gn=8.67722 time=53.61it/s +epoch=29 global_step=11600 loss=3.30926 loss_avg=3.43385 acc=0.67188 acc_top1_avg=0.66972 acc_top5_avg=0.94582 lr=0.01000 gn=11.29226 time=53.81it/s +epoch=29 global_step=11650 loss=3.30783 loss_avg=3.43345 acc=0.66406 acc_top1_avg=0.66956 acc_top5_avg=0.94644 lr=0.01000 gn=10.88315 time=51.95it/s +epoch=29 global_step=11700 loss=3.54322 loss_avg=3.43340 acc=0.65625 acc_top1_avg=0.66978 acc_top5_avg=0.94622 lr=0.01000 gn=11.24461 time=18.32it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.42906 test_loss_avg=0.87410 acc=0.88281 test_acc_avg=0.77083 test_acc_top5_avg=0.96441 time=241.89it/s +epoch=29 global_step=11730 loss=0.34057 test_loss_avg=0.99181 acc=0.89844 test_acc_avg=0.75689 test_acc_top5_avg=0.97113 time=242.33it/s +epoch=29 global_step=11730 loss=0.07200 test_loss_avg=0.86967 acc=0.93750 test_acc_avg=0.78174 test_acc_top5_avg=0.97528 time=838.36it/s +curr_acc 0.7817 +BEST_ACC 0.7914 +curr_acc_top5 0.9753 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=3.06303 loss_avg=3.37965 acc=0.71094 acc_top1_avg=0.67852 acc_top5_avg=0.95000 lr=0.01000 gn=8.52507 time=59.03it/s +epoch=30 global_step=11800 loss=3.03564 loss_avg=3.34654 acc=0.71875 acc_top1_avg=0.68092 acc_top5_avg=0.95056 lr=0.01000 gn=10.90496 time=42.45it/s +epoch=30 global_step=11850 loss=3.51279 loss_avg=3.37100 acc=0.67188 acc_top1_avg=0.67747 acc_top5_avg=0.94902 lr=0.01000 gn=10.26733 time=55.20it/s +epoch=30 global_step=11900 loss=2.99572 loss_avg=3.39037 acc=0.71094 acc_top1_avg=0.67509 acc_top5_avg=0.94692 lr=0.01000 gn=8.83421 time=57.38it/s +epoch=30 global_step=11950 loss=3.46911 loss_avg=3.39054 acc=0.65625 acc_top1_avg=0.67482 acc_top5_avg=0.94730 lr=0.01000 gn=7.20637 time=53.58it/s +epoch=30 global_step=12000 loss=3.19274 loss_avg=3.43004 acc=0.67969 acc_top1_avg=0.67028 acc_top5_avg=0.94667 lr=0.01000 gn=10.04908 time=51.83it/s +epoch=30 global_step=12050 loss=3.07107 loss_avg=3.42901 acc=0.69531 acc_top1_avg=0.67065 acc_top5_avg=0.94688 lr=0.01000 gn=6.80172 time=50.63it/s +epoch=30 global_step=12100 loss=4.32543 loss_avg=3.43193 acc=0.58594 acc_top1_avg=0.67042 acc_top5_avg=0.94685 lr=0.01000 gn=9.86584 time=55.72it/s +====================Eval==================== +epoch=30 global_step=12121 loss=1.76209 test_loss_avg=1.04203 acc=0.59375 test_acc_avg=0.73672 test_acc_top5_avg=0.97630 time=252.46it/s +epoch=30 global_step=12121 loss=0.21480 test_loss_avg=0.88872 acc=0.81250 test_acc_avg=0.77571 test_acc_top5_avg=0.97627 time=846.31it/s +curr_acc 0.7757 +BEST_ACC 0.7914 +curr_acc_top5 0.9763 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=3.55610 loss_avg=3.48642 acc=0.63281 acc_top1_avg=0.66083 acc_top5_avg=0.94639 lr=0.01000 gn=10.14384 time=60.93it/s +epoch=31 global_step=12200 loss=3.78685 loss_avg=3.43712 acc=0.62500 acc_top1_avg=0.66663 acc_top5_avg=0.94709 lr=0.01000 gn=8.85528 time=48.08it/s +epoch=31 global_step=12250 loss=4.03953 loss_avg=3.45871 acc=0.60938 acc_top1_avg=0.66418 acc_top5_avg=0.94683 lr=0.01000 gn=10.74611 time=34.37it/s +epoch=31 global_step=12300 loss=3.19821 loss_avg=3.45908 acc=0.71875 acc_top1_avg=0.66454 acc_top5_avg=0.94797 lr=0.01000 gn=12.14323 time=55.73it/s +epoch=31 global_step=12350 loss=2.98208 loss_avg=3.44770 acc=0.71875 acc_top1_avg=0.66659 acc_top5_avg=0.94719 lr=0.01000 gn=9.31619 time=54.17it/s +epoch=31 global_step=12400 loss=2.83786 loss_avg=3.44205 acc=0.75781 acc_top1_avg=0.66759 acc_top5_avg=0.94778 lr=0.01000 gn=9.02006 time=59.43it/s +epoch=31 global_step=12450 loss=3.54037 loss_avg=3.43224 acc=0.67969 acc_top1_avg=0.66876 acc_top5_avg=0.94809 lr=0.01000 gn=11.96896 time=61.00it/s +epoch=31 global_step=12500 loss=4.11307 loss_avg=3.43779 acc=0.60156 acc_top1_avg=0.66858 acc_top5_avg=0.94807 lr=0.01000 gn=12.01884 time=55.18it/s +====================Eval==================== +epoch=31 global_step=12512 loss=1.09028 test_loss_avg=1.09028 acc=0.69531 test_acc_avg=0.69531 test_acc_top5_avg=0.97656 time=212.11it/s +epoch=31 global_step=12512 loss=0.08574 test_loss_avg=1.68087 acc=0.97656 test_acc_avg=0.67969 test_acc_top5_avg=0.94225 time=238.96it/s +epoch=31 global_step=12512 loss=1.10144 test_loss_avg=1.39197 acc=0.81250 test_acc_avg=0.71875 test_acc_top5_avg=0.95362 time=843.92it/s +curr_acc 0.7188 +BEST_ACC 0.7914 +curr_acc_top5 0.9536 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=3.57586 loss_avg=3.48054 acc=0.64844 acc_top1_avg=0.66735 acc_top5_avg=0.94901 lr=0.01000 gn=11.48472 time=63.05it/s +epoch=32 global_step=12600 loss=3.60602 loss_avg=3.42867 acc=0.66406 acc_top1_avg=0.67019 acc_top5_avg=0.94700 lr=0.01000 gn=9.69278 time=54.95it/s +epoch=32 global_step=12650 loss=3.28463 loss_avg=3.41556 acc=0.67969 acc_top1_avg=0.67142 acc_top5_avg=0.94718 lr=0.01000 gn=8.90457 time=59.19it/s +epoch=32 global_step=12700 loss=2.86950 loss_avg=3.42183 acc=0.72656 acc_top1_avg=0.67063 acc_top5_avg=0.94602 lr=0.01000 gn=11.79823 time=55.09it/s +epoch=32 global_step=12750 loss=3.91944 loss_avg=3.44399 acc=0.63281 acc_top1_avg=0.66807 acc_top5_avg=0.94610 lr=0.01000 gn=8.88871 time=59.86it/s +epoch=32 global_step=12800 loss=3.27255 loss_avg=3.44318 acc=0.70312 acc_top1_avg=0.66813 acc_top5_avg=0.94588 lr=0.01000 gn=10.03988 time=37.56it/s +epoch=32 global_step=12850 loss=3.34650 loss_avg=3.43512 acc=0.67969 acc_top1_avg=0.66947 acc_top5_avg=0.94647 lr=0.01000 gn=10.06427 time=51.65it/s +epoch=32 global_step=12900 loss=3.28281 loss_avg=3.43055 acc=0.69531 acc_top1_avg=0.67024 acc_top5_avg=0.94666 lr=0.01000 gn=10.26921 time=55.62it/s +====================Eval==================== +epoch=32 global_step=12903 loss=1.01598 test_loss_avg=1.24184 acc=0.74219 test_acc_avg=0.69567 test_acc_top5_avg=0.95881 time=89.97it/s +epoch=32 global_step=12903 loss=0.52800 test_loss_avg=1.04568 acc=0.83594 test_acc_avg=0.75022 test_acc_top5_avg=0.96886 time=245.68it/s +epoch=32 global_step=12903 loss=0.02375 test_loss_avg=0.97509 acc=1.00000 test_acc_avg=0.76572 test_acc_top5_avg=0.97122 time=597.65it/s +curr_acc 0.7657 +BEST_ACC 0.7914 +curr_acc_top5 0.9712 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=3.83345 loss_avg=3.49277 acc=0.60938 acc_top1_avg=0.66539 acc_top5_avg=0.94731 lr=0.01000 gn=8.54360 time=56.18it/s +epoch=33 global_step=13000 loss=3.49107 loss_avg=3.44264 acc=0.64844 acc_top1_avg=0.66881 acc_top5_avg=0.95079 lr=0.01000 gn=10.29902 time=56.50it/s +epoch=33 global_step=13050 loss=3.55270 loss_avg=3.41747 acc=0.65625 acc_top1_avg=0.67124 acc_top5_avg=0.94978 lr=0.01000 gn=12.35258 time=57.36it/s +epoch=33 global_step=13100 loss=3.48770 loss_avg=3.43788 acc=0.66406 acc_top1_avg=0.66882 acc_top5_avg=0.94952 lr=0.01000 gn=7.80406 time=55.41it/s +epoch=33 global_step=13150 loss=3.31806 loss_avg=3.44695 acc=0.69531 acc_top1_avg=0.66789 acc_top5_avg=0.94939 lr=0.01000 gn=8.99296 time=53.98it/s +epoch=33 global_step=13200 loss=3.61066 loss_avg=3.43800 acc=0.66406 acc_top1_avg=0.66903 acc_top5_avg=0.94921 lr=0.01000 gn=11.28550 time=52.94it/s +epoch=33 global_step=13250 loss=4.02134 loss_avg=3.43536 acc=0.60156 acc_top1_avg=0.66929 acc_top5_avg=0.94855 lr=0.01000 gn=8.45058 time=54.43it/s +====================Eval==================== +epoch=33 global_step=13294 loss=0.77899 test_loss_avg=1.22105 acc=0.80469 test_acc_avg=0.71021 test_acc_top5_avg=0.95749 time=242.22it/s +epoch=33 global_step=13294 loss=0.29196 test_loss_avg=0.86153 acc=0.93750 test_acc_avg=0.78788 test_acc_top5_avg=0.97300 time=839.70it/s +curr_acc 0.7879 +BEST_ACC 0.7914 +curr_acc_top5 0.9730 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=3.54081 loss_avg=3.43298 acc=0.66406 acc_top1_avg=0.67578 acc_top5_avg=0.94271 lr=0.01000 gn=8.20790 time=56.54it/s +epoch=34 global_step=13350 loss=2.69461 loss_avg=3.33405 acc=0.76562 acc_top1_avg=0.68150 acc_top5_avg=0.94461 lr=0.01000 gn=11.02755 time=60.95it/s +epoch=34 global_step=13400 loss=3.42051 loss_avg=3.38824 acc=0.67969 acc_top1_avg=0.67622 acc_top5_avg=0.94716 lr=0.01000 gn=10.22177 time=59.90it/s +epoch=34 global_step=13450 loss=3.05392 loss_avg=3.39122 acc=0.70312 acc_top1_avg=0.67563 acc_top5_avg=0.94792 lr=0.01000 gn=8.79078 time=59.29it/s +epoch=34 global_step=13500 loss=3.55952 loss_avg=3.40798 acc=0.66406 acc_top1_avg=0.67362 acc_top5_avg=0.94789 lr=0.01000 gn=9.02535 time=53.96it/s +epoch=34 global_step=13550 loss=2.42773 loss_avg=3.41201 acc=0.77344 acc_top1_avg=0.67288 acc_top5_avg=0.94803 lr=0.01000 gn=7.25788 time=56.40it/s +epoch=34 global_step=13600 loss=3.83650 loss_avg=3.39775 acc=0.63281 acc_top1_avg=0.67407 acc_top5_avg=0.94771 lr=0.01000 gn=9.61123 time=62.14it/s +epoch=34 global_step=13650 loss=3.28845 loss_avg=3.40389 acc=0.67188 acc_top1_avg=0.67319 acc_top5_avg=0.94764 lr=0.01000 gn=11.24336 time=58.02it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.27713 test_loss_avg=0.85133 acc=0.89844 test_acc_avg=0.77065 test_acc_top5_avg=0.98326 time=236.85it/s +epoch=34 global_step=13685 loss=0.86921 test_loss_avg=0.94785 acc=0.77344 test_acc_avg=0.75793 test_acc_top5_avg=0.97400 time=237.18it/s +epoch=34 global_step=13685 loss=0.54413 test_loss_avg=0.90443 acc=0.87500 test_acc_avg=0.76622 test_acc_top5_avg=0.97636 time=367.73it/s +curr_acc 0.7662 +BEST_ACC 0.7914 +curr_acc_top5 0.9764 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=3.89872 loss_avg=3.38099 acc=0.62500 acc_top1_avg=0.67656 acc_top5_avg=0.95469 lr=0.01000 gn=9.79742 time=56.08it/s +epoch=35 global_step=13750 loss=2.98072 loss_avg=3.39268 acc=0.71875 acc_top1_avg=0.67644 acc_top5_avg=0.94916 lr=0.01000 gn=10.17909 time=56.10it/s +epoch=35 global_step=13800 loss=3.11333 loss_avg=3.38273 acc=0.71094 acc_top1_avg=0.67745 acc_top5_avg=0.94823 lr=0.01000 gn=9.41941 time=51.78it/s +epoch=35 global_step=13850 loss=3.96522 loss_avg=3.38590 acc=0.61719 acc_top1_avg=0.67604 acc_top5_avg=0.94863 lr=0.01000 gn=9.98127 time=54.90it/s +epoch=35 global_step=13900 loss=3.26969 loss_avg=3.38632 acc=0.68750 acc_top1_avg=0.67533 acc_top5_avg=0.94916 lr=0.01000 gn=9.43328 time=51.83it/s +epoch=35 global_step=13950 loss=2.74259 loss_avg=3.39787 acc=0.75781 acc_top1_avg=0.67409 acc_top5_avg=0.94847 lr=0.01000 gn=8.69011 time=55.27it/s +epoch=35 global_step=14000 loss=3.05252 loss_avg=3.39862 acc=0.70312 acc_top1_avg=0.67383 acc_top5_avg=0.94839 lr=0.01000 gn=8.78083 time=56.37it/s +epoch=35 global_step=14050 loss=3.59179 loss_avg=3.41331 acc=0.65625 acc_top1_avg=0.67260 acc_top5_avg=0.94799 lr=0.01000 gn=9.99897 time=53.44it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.55250 test_loss_avg=0.78319 acc=0.85938 test_acc_avg=0.78862 test_acc_top5_avg=0.98237 time=240.07it/s +epoch=35 global_step=14076 loss=0.76785 test_loss_avg=0.83659 acc=0.75000 test_acc_avg=0.77393 test_acc_top5_avg=0.98200 time=849.39it/s +curr_acc 0.7739 +BEST_ACC 0.7914 +curr_acc_top5 0.9820 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=3.56809 loss_avg=3.28905 acc=0.67188 acc_top1_avg=0.68099 acc_top5_avg=0.95182 lr=0.01000 gn=11.75916 time=63.76it/s +epoch=36 global_step=14150 loss=3.35923 loss_avg=3.39352 acc=0.67969 acc_top1_avg=0.67367 acc_top5_avg=0.94880 lr=0.01000 gn=8.19510 time=54.97it/s +epoch=36 global_step=14200 loss=3.51517 loss_avg=3.38381 acc=0.67969 acc_top1_avg=0.67509 acc_top5_avg=0.94979 lr=0.01000 gn=8.42642 time=63.22it/s +epoch=36 global_step=14250 loss=3.33048 loss_avg=3.41501 acc=0.67969 acc_top1_avg=0.67210 acc_top5_avg=0.94805 lr=0.01000 gn=10.82998 time=55.09it/s +epoch=36 global_step=14300 loss=3.37429 loss_avg=3.41562 acc=0.67969 acc_top1_avg=0.67194 acc_top5_avg=0.94890 lr=0.01000 gn=10.32581 time=55.05it/s +epoch=36 global_step=14350 loss=2.75658 loss_avg=3.42346 acc=0.76562 acc_top1_avg=0.67165 acc_top5_avg=0.94822 lr=0.01000 gn=8.38618 time=60.31it/s +epoch=36 global_step=14400 loss=3.21484 loss_avg=3.42068 acc=0.70312 acc_top1_avg=0.67134 acc_top5_avg=0.94809 lr=0.01000 gn=9.09587 time=55.49it/s +epoch=36 global_step=14450 loss=2.64612 loss_avg=3.40513 acc=0.72656 acc_top1_avg=0.67294 acc_top5_avg=0.94834 lr=0.01000 gn=10.73595 time=57.83it/s +====================Eval==================== +epoch=36 global_step=14467 loss=0.82629 test_loss_avg=0.77854 acc=0.75000 test_acc_avg=0.77995 test_acc_top5_avg=0.98177 time=239.31it/s +epoch=36 global_step=14467 loss=1.40986 test_loss_avg=0.78441 acc=0.66406 test_acc_avg=0.79869 test_acc_top5_avg=0.98451 time=241.41it/s +epoch=36 global_step=14467 loss=0.74894 test_loss_avg=0.82316 acc=0.68750 test_acc_avg=0.78590 test_acc_top5_avg=0.97913 time=491.02it/s +curr_acc 0.7859 +BEST_ACC 0.7914 +curr_acc_top5 0.9791 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=3.45993 loss_avg=3.34804 acc=0.65625 acc_top1_avg=0.67661 acc_top5_avg=0.94697 lr=0.01000 gn=8.15592 time=54.00it/s +epoch=37 global_step=14550 loss=3.44888 loss_avg=3.33479 acc=0.67188 acc_top1_avg=0.67875 acc_top5_avg=0.94955 lr=0.01000 gn=10.17836 time=54.28it/s +epoch=37 global_step=14600 loss=2.79025 loss_avg=3.33165 acc=0.73438 acc_top1_avg=0.67975 acc_top5_avg=0.94860 lr=0.01000 gn=8.64389 time=53.27it/s +epoch=37 global_step=14650 loss=2.74137 loss_avg=3.34793 acc=0.75781 acc_top1_avg=0.67798 acc_top5_avg=0.94962 lr=0.01000 gn=10.04309 time=56.18it/s +epoch=37 global_step=14700 loss=3.41628 loss_avg=3.38373 acc=0.67188 acc_top1_avg=0.67345 acc_top5_avg=0.95004 lr=0.01000 gn=11.58550 time=47.21it/s +epoch=37 global_step=14750 loss=3.46904 loss_avg=3.40115 acc=0.67188 acc_top1_avg=0.67207 acc_top5_avg=0.94943 lr=0.01000 gn=7.29225 time=49.34it/s +epoch=37 global_step=14800 loss=3.65497 loss_avg=3.41503 acc=0.64062 acc_top1_avg=0.67049 acc_top5_avg=0.94806 lr=0.01000 gn=8.65948 time=55.04it/s +epoch=37 global_step=14850 loss=3.55451 loss_avg=3.41970 acc=0.65625 acc_top1_avg=0.66984 acc_top5_avg=0.94803 lr=0.01000 gn=8.50340 time=59.64it/s +====================Eval==================== +epoch=37 global_step=14858 loss=0.82294 test_loss_avg=0.93620 acc=0.77344 test_acc_avg=0.75463 test_acc_top5_avg=0.97859 time=227.94it/s +epoch=37 global_step=14858 loss=0.63381 test_loss_avg=0.84187 acc=0.82031 test_acc_avg=0.78632 test_acc_top5_avg=0.97971 time=258.54it/s +epoch=37 global_step=14858 loss=0.09973 test_loss_avg=0.82838 acc=0.93750 test_acc_avg=0.78877 test_acc_top5_avg=0.98022 time=855.28it/s +curr_acc 0.7888 +BEST_ACC 0.7914 +curr_acc_top5 0.9802 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=3.65118 loss_avg=3.31641 acc=0.63281 acc_top1_avg=0.68304 acc_top5_avg=0.94661 lr=0.01000 gn=9.74306 time=51.63it/s +epoch=38 global_step=14950 loss=3.11041 loss_avg=3.36544 acc=0.71875 acc_top1_avg=0.67782 acc_top5_avg=0.94565 lr=0.01000 gn=8.12448 time=54.54it/s +epoch=38 global_step=15000 loss=3.31266 loss_avg=3.35944 acc=0.69531 acc_top1_avg=0.67809 acc_top5_avg=0.94630 lr=0.01000 gn=9.73126 time=49.86it/s +epoch=38 global_step=15050 loss=2.83842 loss_avg=3.37096 acc=0.72656 acc_top1_avg=0.67647 acc_top5_avg=0.94706 lr=0.01000 gn=8.36760 time=51.05it/s +epoch=38 global_step=15100 loss=3.83973 loss_avg=3.37458 acc=0.65625 acc_top1_avg=0.67614 acc_top5_avg=0.94735 lr=0.01000 gn=12.07023 time=56.84it/s +epoch=38 global_step=15150 loss=3.90522 loss_avg=3.38548 acc=0.60156 acc_top1_avg=0.67471 acc_top5_avg=0.94780 lr=0.01000 gn=12.36020 time=58.92it/s +epoch=38 global_step=15200 loss=3.65788 loss_avg=3.40106 acc=0.66406 acc_top1_avg=0.67331 acc_top5_avg=0.94808 lr=0.01000 gn=11.66068 time=53.61it/s +====================Eval==================== +epoch=38 global_step=15249 loss=0.51494 test_loss_avg=1.06967 acc=0.86719 test_acc_avg=0.70801 test_acc_top5_avg=0.97331 time=243.49it/s +epoch=38 global_step=15249 loss=0.48333 test_loss_avg=0.80679 acc=0.87500 test_acc_avg=0.77611 test_acc_top5_avg=0.98091 time=839.36it/s +curr_acc 0.7761 +BEST_ACC 0.7914 +curr_acc_top5 0.9809 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=4.23574 loss_avg=4.23574 acc=0.57031 acc_top1_avg=0.57031 acc_top5_avg=0.92969 lr=0.01000 gn=10.36357 time=48.75it/s +epoch=39 global_step=15300 loss=3.37199 loss_avg=3.38887 acc=0.66406 acc_top1_avg=0.67494 acc_top5_avg=0.94256 lr=0.01000 gn=9.06273 time=53.28it/s +epoch=39 global_step=15350 loss=3.61882 loss_avg=3.35989 acc=0.64062 acc_top1_avg=0.67713 acc_top5_avg=0.94601 lr=0.01000 gn=10.20562 time=53.96it/s +epoch=39 global_step=15400 loss=3.29904 loss_avg=3.37902 acc=0.70312 acc_top1_avg=0.67607 acc_top5_avg=0.94640 lr=0.01000 gn=11.48015 time=56.68it/s +epoch=39 global_step=15450 loss=3.33061 loss_avg=3.37716 acc=0.67188 acc_top1_avg=0.67588 acc_top5_avg=0.94733 lr=0.01000 gn=8.66750 time=57.25it/s +epoch=39 global_step=15500 loss=3.57535 loss_avg=3.40020 acc=0.66406 acc_top1_avg=0.67334 acc_top5_avg=0.94687 lr=0.01000 gn=10.05591 time=48.84it/s +epoch=39 global_step=15550 loss=3.75752 loss_avg=3.40496 acc=0.60938 acc_top1_avg=0.67258 acc_top5_avg=0.94760 lr=0.01000 gn=8.21695 time=60.47it/s +epoch=39 global_step=15600 loss=3.70999 loss_avg=3.40958 acc=0.64062 acc_top1_avg=0.67216 acc_top5_avg=0.94781 lr=0.01000 gn=9.16711 time=55.70it/s +====================Eval==================== +epoch=39 global_step=15640 loss=1.25266 test_loss_avg=0.53160 acc=0.68750 test_acc_avg=0.85609 test_acc_top5_avg=0.99301 time=243.33it/s +epoch=39 global_step=15640 loss=0.78864 test_loss_avg=0.76399 acc=0.78125 test_acc_avg=0.79529 test_acc_top5_avg=0.98064 time=248.15it/s +epoch=39 global_step=15640 loss=0.04373 test_loss_avg=0.71615 acc=1.00000 test_acc_avg=0.80756 test_acc_top5_avg=0.98250 time=801.82it/s +curr_acc 0.8076 +BEST_ACC 0.7914 +curr_acc_top5 0.9825 +BEST_ACC_top5 0.9821 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=2.96839 loss_avg=3.24926 acc=0.74219 acc_top1_avg=0.69219 acc_top5_avg=0.96250 lr=0.00100 gn=9.57179 time=54.88it/s +epoch=40 global_step=15700 loss=2.61441 loss_avg=3.16903 acc=0.75000 acc_top1_avg=0.69792 acc_top5_avg=0.95638 lr=0.00100 gn=6.27644 time=55.92it/s +epoch=40 global_step=15750 loss=2.78678 loss_avg=3.07091 acc=0.73438 acc_top1_avg=0.70739 acc_top5_avg=0.95781 lr=0.00100 gn=10.29446 time=52.18it/s +epoch=40 global_step=15800 loss=2.82333 loss_avg=3.03957 acc=0.71875 acc_top1_avg=0.70977 acc_top5_avg=0.95869 lr=0.00100 gn=8.64570 time=60.12it/s +epoch=40 global_step=15850 loss=3.18086 loss_avg=3.02860 acc=0.70312 acc_top1_avg=0.71153 acc_top5_avg=0.95822 lr=0.00100 gn=8.36842 time=54.94it/s +epoch=40 global_step=15900 loss=3.19798 loss_avg=3.01726 acc=0.69531 acc_top1_avg=0.71235 acc_top5_avg=0.95847 lr=0.00100 gn=7.49021 time=48.82it/s +epoch=40 global_step=15950 loss=2.95727 loss_avg=3.01780 acc=0.71094 acc_top1_avg=0.71222 acc_top5_avg=0.95827 lr=0.00100 gn=8.73331 time=54.73it/s +epoch=40 global_step=16000 loss=3.04299 loss_avg=3.00090 acc=0.69531 acc_top1_avg=0.71376 acc_top5_avg=0.95881 lr=0.00100 gn=9.18078 time=54.98it/s +====================Eval==================== +epoch=40 global_step=16031 loss=0.74312 test_loss_avg=0.46758 acc=0.80469 test_acc_avg=0.86621 test_acc_top5_avg=0.99316 time=234.07it/s +epoch=40 global_step=16031 loss=0.06966 test_loss_avg=0.42193 acc=1.00000 test_acc_avg=0.87896 test_acc_top5_avg=0.99308 time=824.19it/s +curr_acc 0.8790 +BEST_ACC 0.8076 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9825 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=3.23808 loss_avg=2.85848 acc=0.68750 acc_top1_avg=0.72245 acc_top5_avg=0.95765 lr=0.00100 gn=8.00568 time=60.66it/s +epoch=41 global_step=16100 loss=2.36584 loss_avg=2.82809 acc=0.78125 acc_top1_avg=0.73041 acc_top5_avg=0.96105 lr=0.00100 gn=9.19649 time=57.19it/s +epoch=41 global_step=16150 loss=2.73665 loss_avg=2.81972 acc=0.73438 acc_top1_avg=0.73057 acc_top5_avg=0.96205 lr=0.00100 gn=8.73497 time=57.17it/s +epoch=41 global_step=16200 loss=2.57282 loss_avg=2.82897 acc=0.76562 acc_top1_avg=0.73054 acc_top5_avg=0.96223 lr=0.00100 gn=10.98818 time=53.93it/s +epoch=41 global_step=16250 loss=2.96437 loss_avg=2.83029 acc=0.71875 acc_top1_avg=0.72974 acc_top5_avg=0.96286 lr=0.00100 gn=7.80091 time=60.03it/s +epoch=41 global_step=16300 loss=3.51765 loss_avg=2.85080 acc=0.67188 acc_top1_avg=0.72755 acc_top5_avg=0.96195 lr=0.00100 gn=10.97513 time=55.32it/s +epoch=41 global_step=16350 loss=2.50389 loss_avg=2.84641 acc=0.77344 acc_top1_avg=0.72808 acc_top5_avg=0.96182 lr=0.00100 gn=8.81984 time=52.93it/s +epoch=41 global_step=16400 loss=2.88702 loss_avg=2.84141 acc=0.72656 acc_top1_avg=0.72840 acc_top5_avg=0.96233 lr=0.00100 gn=9.46876 time=62.54it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.11612 test_loss_avg=0.52242 acc=0.96094 test_acc_avg=0.84233 test_acc_top5_avg=0.99077 time=228.20it/s +epoch=41 global_step=16422 loss=0.14466 test_loss_avg=0.48262 acc=0.95312 test_acc_avg=0.86488 test_acc_top5_avg=0.99193 time=240.28it/s +epoch=41 global_step=16422 loss=0.03618 test_loss_avg=0.41774 acc=1.00000 test_acc_avg=0.88113 test_acc_top5_avg=0.99268 time=841.05it/s +curr_acc 0.8811 +BEST_ACC 0.8790 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9931 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=2.67892 loss_avg=2.77701 acc=0.76562 acc_top1_avg=0.73410 acc_top5_avg=0.96345 lr=0.00100 gn=7.47803 time=63.40it/s +epoch=42 global_step=16500 loss=2.74735 loss_avg=2.75231 acc=0.74219 acc_top1_avg=0.73728 acc_top5_avg=0.96494 lr=0.00100 gn=8.88493 time=56.59it/s +epoch=42 global_step=16550 loss=3.02034 loss_avg=2.71352 acc=0.70312 acc_top1_avg=0.74109 acc_top5_avg=0.96497 lr=0.00100 gn=10.68143 time=62.49it/s +epoch=42 global_step=16600 loss=2.79978 loss_avg=2.75436 acc=0.71094 acc_top1_avg=0.73648 acc_top5_avg=0.96379 lr=0.00100 gn=8.03761 time=61.46it/s +epoch=42 global_step=16650 loss=2.56646 loss_avg=2.74492 acc=0.75781 acc_top1_avg=0.73790 acc_top5_avg=0.96358 lr=0.00100 gn=8.61964 time=53.89it/s +epoch=42 global_step=16700 loss=3.57799 loss_avg=2.76555 acc=0.64062 acc_top1_avg=0.73620 acc_top5_avg=0.96335 lr=0.00100 gn=12.00110 time=52.34it/s +epoch=42 global_step=16750 loss=3.10477 loss_avg=2.76686 acc=0.70312 acc_top1_avg=0.73604 acc_top5_avg=0.96339 lr=0.00100 gn=6.42217 time=53.28it/s +epoch=42 global_step=16800 loss=3.02291 loss_avg=2.76700 acc=0.70312 acc_top1_avg=0.73613 acc_top5_avg=0.96317 lr=0.00100 gn=10.01895 time=61.34it/s +====================Eval==================== +epoch=42 global_step=16813 loss=0.33787 test_loss_avg=0.55509 acc=0.87500 test_acc_avg=0.84717 test_acc_top5_avg=0.99121 time=222.20it/s +epoch=42 global_step=16813 loss=0.05581 test_loss_avg=0.40119 acc=1.00000 test_acc_avg=0.88766 test_acc_top5_avg=0.99367 time=792.13it/s +curr_acc 0.8877 +BEST_ACC 0.8811 +curr_acc_top5 0.9937 +BEST_ACC_top5 0.9931 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=3.14742 loss_avg=2.70352 acc=0.67969 acc_top1_avg=0.74219 acc_top5_avg=0.96706 lr=0.00100 gn=8.44113 time=44.04it/s +epoch=43 global_step=16900 loss=2.79427 loss_avg=2.76384 acc=0.73438 acc_top1_avg=0.73689 acc_top5_avg=0.96498 lr=0.00100 gn=10.67212 time=50.71it/s +epoch=43 global_step=16950 loss=2.72648 loss_avg=2.77205 acc=0.73438 acc_top1_avg=0.73557 acc_top5_avg=0.96253 lr=0.00100 gn=9.41185 time=56.64it/s +epoch=43 global_step=17000 loss=3.16181 loss_avg=2.77038 acc=0.68750 acc_top1_avg=0.73534 acc_top5_avg=0.96369 lr=0.00100 gn=7.81289 time=60.56it/s +epoch=43 global_step=17050 loss=2.82629 loss_avg=2.75229 acc=0.74219 acc_top1_avg=0.73761 acc_top5_avg=0.96328 lr=0.00100 gn=12.11675 time=57.04it/s +epoch=43 global_step=17100 loss=2.50019 loss_avg=2.73360 acc=0.76562 acc_top1_avg=0.73968 acc_top5_avg=0.96352 lr=0.00100 gn=7.26513 time=52.86it/s +epoch=43 global_step=17150 loss=2.88992 loss_avg=2.72675 acc=0.72656 acc_top1_avg=0.74066 acc_top5_avg=0.96312 lr=0.00100 gn=10.12092 time=51.35it/s +epoch=43 global_step=17200 loss=3.24426 loss_avg=2.72594 acc=0.68750 acc_top1_avg=0.74083 acc_top5_avg=0.96388 lr=0.00100 gn=10.22015 time=63.94it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.34530 test_loss_avg=0.45212 acc=0.90625 test_acc_avg=0.85938 test_acc_top5_avg=0.98958 time=242.17it/s +epoch=43 global_step=17204 loss=0.23488 test_loss_avg=0.45178 acc=0.93750 test_acc_avg=0.87117 test_acc_top5_avg=0.99101 time=243.23it/s +epoch=43 global_step=17204 loss=0.16421 test_loss_avg=0.39058 acc=0.93750 test_acc_avg=0.88687 test_acc_top5_avg=0.99298 time=841.89it/s +curr_acc 0.8869 +BEST_ACC 0.8877 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=3.08126 loss_avg=2.70729 acc=0.70312 acc_top1_avg=0.74338 acc_top5_avg=0.96179 lr=0.00100 gn=11.41541 time=56.27it/s +epoch=44 global_step=17300 loss=2.54797 loss_avg=2.69770 acc=0.75000 acc_top1_avg=0.74447 acc_top5_avg=0.96346 lr=0.00100 gn=8.78801 time=58.07it/s +epoch=44 global_step=17350 loss=2.13908 loss_avg=2.69650 acc=0.78906 acc_top1_avg=0.74460 acc_top5_avg=0.96260 lr=0.00100 gn=10.29394 time=53.85it/s +epoch=44 global_step=17400 loss=2.49332 loss_avg=2.67363 acc=0.77344 acc_top1_avg=0.74693 acc_top5_avg=0.96377 lr=0.00100 gn=11.60555 time=59.04it/s +epoch=44 global_step=17450 loss=3.04307 loss_avg=2.69149 acc=0.70312 acc_top1_avg=0.74473 acc_top5_avg=0.96376 lr=0.00100 gn=8.66872 time=52.91it/s +epoch=44 global_step=17500 loss=2.58269 loss_avg=2.68969 acc=0.73438 acc_top1_avg=0.74462 acc_top5_avg=0.96389 lr=0.00100 gn=10.60469 time=59.11it/s +epoch=44 global_step=17550 loss=2.62458 loss_avg=2.68000 acc=0.75781 acc_top1_avg=0.74503 acc_top5_avg=0.96399 lr=0.00100 gn=11.09493 time=50.70it/s +====================Eval==================== +epoch=44 global_step=17595 loss=0.93422 test_loss_avg=0.46461 acc=0.75000 test_acc_avg=0.86523 test_acc_top5_avg=0.99316 time=244.03it/s +epoch=44 global_step=17595 loss=0.18719 test_loss_avg=0.40065 acc=0.92969 test_acc_avg=0.88524 test_acc_top5_avg=0.99314 time=245.37it/s +epoch=44 global_step=17595 loss=0.04011 test_loss_avg=0.38649 acc=1.00000 test_acc_avg=0.88924 test_acc_top5_avg=0.99347 time=847.51it/s +curr_acc 0.8892 +BEST_ACC 0.8877 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=2.40270 loss_avg=2.49373 acc=0.76562 acc_top1_avg=0.75625 acc_top5_avg=0.97031 lr=0.00100 gn=7.84115 time=51.77it/s +epoch=45 global_step=17650 loss=2.03795 loss_avg=2.66238 acc=0.81250 acc_top1_avg=0.74645 acc_top5_avg=0.96293 lr=0.00100 gn=9.23158 time=59.13it/s +epoch=45 global_step=17700 loss=3.00781 loss_avg=2.67732 acc=0.70312 acc_top1_avg=0.74516 acc_top5_avg=0.96302 lr=0.00100 gn=10.63631 time=62.50it/s +epoch=45 global_step=17750 loss=2.78968 loss_avg=2.65322 acc=0.74219 acc_top1_avg=0.74778 acc_top5_avg=0.96462 lr=0.00100 gn=10.11270 time=55.13it/s +epoch=45 global_step=17800 loss=3.07693 loss_avg=2.63309 acc=0.70312 acc_top1_avg=0.75004 acc_top5_avg=0.96559 lr=0.00100 gn=9.45613 time=48.34it/s +epoch=45 global_step=17850 loss=3.04425 loss_avg=2.65356 acc=0.70312 acc_top1_avg=0.74776 acc_top5_avg=0.96406 lr=0.00100 gn=9.89326 time=62.89it/s +epoch=45 global_step=17900 loss=2.16827 loss_avg=2.66308 acc=0.80469 acc_top1_avg=0.74698 acc_top5_avg=0.96440 lr=0.00100 gn=9.90342 time=60.16it/s +epoch=45 global_step=17950 loss=2.42903 loss_avg=2.65143 acc=0.75781 acc_top1_avg=0.74828 acc_top5_avg=0.96466 lr=0.00100 gn=7.89135 time=55.51it/s +====================Eval==================== +epoch=45 global_step=17986 loss=0.57168 test_loss_avg=0.49100 acc=0.85938 test_acc_avg=0.85990 test_acc_top5_avg=0.99132 time=241.84it/s +epoch=45 global_step=17986 loss=0.04923 test_loss_avg=0.37982 acc=1.00000 test_acc_avg=0.89102 test_acc_top5_avg=0.99328 time=842.91it/s +curr_acc 0.8910 +BEST_ACC 0.8892 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=3.50566 loss_avg=2.79060 acc=0.64844 acc_top1_avg=0.72824 acc_top5_avg=0.95926 lr=0.00100 gn=8.34935 time=57.42it/s +epoch=46 global_step=18050 loss=2.82669 loss_avg=2.64032 acc=0.72656 acc_top1_avg=0.74841 acc_top5_avg=0.96484 lr=0.00100 gn=12.49140 time=54.02it/s +epoch=46 global_step=18100 loss=2.69641 loss_avg=2.60078 acc=0.74219 acc_top1_avg=0.75233 acc_top5_avg=0.96663 lr=0.00100 gn=10.86097 time=63.45it/s +epoch=46 global_step=18150 loss=2.31375 loss_avg=2.59605 acc=0.78125 acc_top1_avg=0.75305 acc_top5_avg=0.96661 lr=0.00100 gn=10.17012 time=60.14it/s +epoch=46 global_step=18200 loss=2.89747 loss_avg=2.60562 acc=0.71875 acc_top1_avg=0.75179 acc_top5_avg=0.96638 lr=0.00100 gn=9.13638 time=52.58it/s +epoch=46 global_step=18250 loss=2.99667 loss_avg=2.60710 acc=0.70312 acc_top1_avg=0.75130 acc_top5_avg=0.96626 lr=0.00100 gn=11.72517 time=52.54it/s +epoch=46 global_step=18300 loss=2.44433 loss_avg=2.61610 acc=0.75000 acc_top1_avg=0.75035 acc_top5_avg=0.96591 lr=0.00100 gn=10.49484 time=54.79it/s +epoch=46 global_step=18350 loss=2.33260 loss_avg=2.60934 acc=0.78125 acc_top1_avg=0.75120 acc_top5_avg=0.96639 lr=0.00100 gn=11.55022 time=55.16it/s +====================Eval==================== +epoch=46 global_step=18377 loss=0.57380 test_loss_avg=0.33356 acc=0.85938 test_acc_avg=0.89648 test_acc_top5_avg=0.99512 time=240.02it/s +epoch=46 global_step=18377 loss=0.11311 test_loss_avg=0.41434 acc=0.96094 test_acc_avg=0.88554 test_acc_top5_avg=0.99302 time=249.84it/s +epoch=46 global_step=18377 loss=0.14087 test_loss_avg=0.39149 acc=0.87500 test_acc_avg=0.89062 test_acc_top5_avg=0.99357 time=508.65it/s +curr_acc 0.8906 +BEST_ACC 0.8910 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=2.34075 loss_avg=2.55736 acc=0.78125 acc_top1_avg=0.75645 acc_top5_avg=0.96705 lr=0.00100 gn=8.54903 time=54.12it/s +epoch=47 global_step=18450 loss=3.18251 loss_avg=2.52225 acc=0.69531 acc_top1_avg=0.76027 acc_top5_avg=0.96757 lr=0.00100 gn=11.03411 time=60.99it/s +epoch=47 global_step=18500 loss=3.02530 loss_avg=2.55320 acc=0.71094 acc_top1_avg=0.75743 acc_top5_avg=0.96570 lr=0.00100 gn=12.67021 time=59.06it/s +epoch=47 global_step=18550 loss=2.42185 loss_avg=2.55662 acc=0.77344 acc_top1_avg=0.75727 acc_top5_avg=0.96563 lr=0.00100 gn=11.36040 time=54.18it/s +epoch=47 global_step=18600 loss=2.90769 loss_avg=2.58156 acc=0.70312 acc_top1_avg=0.75473 acc_top5_avg=0.96532 lr=0.00100 gn=11.79861 time=55.17it/s +epoch=47 global_step=18650 loss=2.74987 loss_avg=2.59470 acc=0.74219 acc_top1_avg=0.75295 acc_top5_avg=0.96497 lr=0.00100 gn=10.23981 time=55.33it/s +epoch=47 global_step=18700 loss=2.28784 loss_avg=2.59799 acc=0.78125 acc_top1_avg=0.75242 acc_top5_avg=0.96515 lr=0.00100 gn=12.85471 time=52.97it/s +epoch=47 global_step=18750 loss=1.89170 loss_avg=2.59613 acc=0.82812 acc_top1_avg=0.75276 acc_top5_avg=0.96565 lr=0.00100 gn=9.59016 time=57.56it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.19968 test_loss_avg=0.44150 acc=0.93750 test_acc_avg=0.87563 test_acc_top5_avg=0.99198 time=242.78it/s +epoch=47 global_step=18768 loss=0.04447 test_loss_avg=0.38243 acc=1.00000 test_acc_avg=0.89389 test_acc_top5_avg=0.99347 time=736.88it/s +curr_acc 0.8939 +BEST_ACC 0.8910 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=1.72119 loss_avg=2.46551 acc=0.84375 acc_top1_avg=0.76978 acc_top5_avg=0.96802 lr=0.00100 gn=9.95542 time=58.35it/s +epoch=48 global_step=18850 loss=2.51966 loss_avg=2.56853 acc=0.75781 acc_top1_avg=0.75705 acc_top5_avg=0.96665 lr=0.00100 gn=16.03463 time=56.51it/s +epoch=48 global_step=18900 loss=2.74019 loss_avg=2.56737 acc=0.73438 acc_top1_avg=0.75616 acc_top5_avg=0.96638 lr=0.00100 gn=9.78480 time=57.12it/s +epoch=48 global_step=18950 loss=2.62277 loss_avg=2.55561 acc=0.76562 acc_top1_avg=0.75717 acc_top5_avg=0.96600 lr=0.00100 gn=12.38481 time=58.56it/s +epoch=48 global_step=19000 loss=2.30052 loss_avg=2.56199 acc=0.78125 acc_top1_avg=0.75650 acc_top5_avg=0.96599 lr=0.00100 gn=9.75298 time=52.62it/s +epoch=48 global_step=19050 loss=2.96418 loss_avg=2.57711 acc=0.70312 acc_top1_avg=0.75499 acc_top5_avg=0.96595 lr=0.00100 gn=9.61733 time=57.28it/s +epoch=48 global_step=19100 loss=2.50130 loss_avg=2.56547 acc=0.76562 acc_top1_avg=0.75649 acc_top5_avg=0.96593 lr=0.00100 gn=11.14609 time=54.99it/s +epoch=48 global_step=19150 loss=2.59320 loss_avg=2.56710 acc=0.76562 acc_top1_avg=0.75646 acc_top5_avg=0.96572 lr=0.00100 gn=12.04553 time=62.75it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.25740 test_loss_avg=0.48219 acc=0.90625 test_acc_avg=0.85742 test_acc_top5_avg=0.99316 time=240.82it/s +epoch=48 global_step=19159 loss=0.20857 test_loss_avg=0.44026 acc=0.92969 test_acc_avg=0.87810 test_acc_top5_avg=0.99300 time=233.48it/s +epoch=48 global_step=19159 loss=0.05764 test_loss_avg=0.38660 acc=1.00000 test_acc_avg=0.89231 test_acc_top5_avg=0.99417 time=844.60it/s +curr_acc 0.8923 +BEST_ACC 0.8939 +curr_acc_top5 0.9942 +BEST_ACC_top5 0.9937 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=2.35749 loss_avg=2.66563 acc=0.78125 acc_top1_avg=0.74600 acc_top5_avg=0.96341 lr=0.00100 gn=12.68997 time=56.71it/s +epoch=49 global_step=19250 loss=2.65429 loss_avg=2.58188 acc=0.75000 acc_top1_avg=0.75361 acc_top5_avg=0.96437 lr=0.00100 gn=10.76159 time=55.80it/s +epoch=49 global_step=19300 loss=3.16686 loss_avg=2.57188 acc=0.67969 acc_top1_avg=0.75532 acc_top5_avg=0.96354 lr=0.00100 gn=9.74099 time=56.47it/s +epoch=49 global_step=19350 loss=2.23587 loss_avg=2.56305 acc=0.78906 acc_top1_avg=0.75744 acc_top5_avg=0.96413 lr=0.00100 gn=11.39336 time=55.90it/s +epoch=49 global_step=19400 loss=2.47778 loss_avg=2.55187 acc=0.75781 acc_top1_avg=0.75814 acc_top5_avg=0.96489 lr=0.00100 gn=11.92055 time=56.48it/s +epoch=49 global_step=19450 loss=2.38150 loss_avg=2.54700 acc=0.77344 acc_top1_avg=0.75843 acc_top5_avg=0.96534 lr=0.00100 gn=8.41772 time=62.66it/s +epoch=49 global_step=19500 loss=2.84280 loss_avg=2.54805 acc=0.71875 acc_top1_avg=0.75839 acc_top5_avg=0.96566 lr=0.00100 gn=12.53273 time=55.34it/s +epoch=49 global_step=19550 loss=3.68219 loss_avg=2.54848 acc=0.63750 acc_top1_avg=0.75864 acc_top5_avg=0.96586 lr=0.00100 gn=14.04916 time=67.69it/s +====================Eval==================== +epoch=49 global_step=19550 loss=0.44404 test_loss_avg=0.44917 acc=0.85156 test_acc_avg=0.87177 test_acc_top5_avg=0.99380 time=240.07it/s +epoch=49 global_step=19550 loss=0.17990 test_loss_avg=0.37940 acc=0.87500 test_acc_avg=0.89320 test_acc_top5_avg=0.99387 time=859.31it/s +epoch=49 global_step=19550 loss=0.17990 test_loss_avg=0.37940 acc=0.87500 test_acc_avg=0.89320 test_acc_top5_avg=0.99387 time=859.31it/s +curr_acc 0.8932 +BEST_ACC 0.8939 +curr_acc_top5 0.9939 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=2.54822 loss_avg=2.59222 acc=0.74219 acc_top1_avg=0.75313 acc_top5_avg=0.96437 lr=0.00100 gn=11.07442 time=58.36it/s +epoch=50 global_step=19650 loss=2.49648 loss_avg=2.54605 acc=0.76562 acc_top1_avg=0.75758 acc_top5_avg=0.96656 lr=0.00100 gn=10.57737 time=57.39it/s +epoch=50 global_step=19700 loss=1.66137 loss_avg=2.51523 acc=0.85156 acc_top1_avg=0.76052 acc_top5_avg=0.96589 lr=0.00100 gn=9.67778 time=54.70it/s +epoch=50 global_step=19750 loss=2.52385 loss_avg=2.51758 acc=0.75000 acc_top1_avg=0.76059 acc_top5_avg=0.96562 lr=0.00100 gn=13.22947 time=59.88it/s +epoch=50 global_step=19800 loss=2.77130 loss_avg=2.51601 acc=0.74219 acc_top1_avg=0.76144 acc_top5_avg=0.96578 lr=0.00100 gn=11.94352 time=52.80it/s +epoch=50 global_step=19850 loss=2.82666 loss_avg=2.51646 acc=0.72656 acc_top1_avg=0.76146 acc_top5_avg=0.96680 lr=0.00100 gn=10.21639 time=56.78it/s +epoch=50 global_step=19900 loss=2.48267 loss_avg=2.51313 acc=0.76562 acc_top1_avg=0.76181 acc_top5_avg=0.96694 lr=0.00100 gn=13.18497 time=50.64it/s +====================Eval==================== +epoch=50 global_step=19941 loss=0.22945 test_loss_avg=0.48912 acc=0.92969 test_acc_avg=0.86141 test_acc_top5_avg=0.99125 time=238.22it/s +epoch=50 global_step=19941 loss=0.36390 test_loss_avg=0.39607 acc=0.87500 test_acc_avg=0.88657 test_acc_top5_avg=0.99357 time=802.74it/s +curr_acc 0.8866 +BEST_ACC 0.8939 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=2.57486 loss_avg=2.51253 acc=0.74219 acc_top1_avg=0.76042 acc_top5_avg=0.95052 lr=0.00100 gn=9.65119 time=52.03it/s +epoch=51 global_step=20000 loss=2.44111 loss_avg=2.45745 acc=0.76562 acc_top1_avg=0.76841 acc_top5_avg=0.96253 lr=0.00100 gn=10.23055 time=54.50it/s +epoch=51 global_step=20050 loss=2.71239 loss_avg=2.44941 acc=0.73438 acc_top1_avg=0.76864 acc_top5_avg=0.96617 lr=0.00100 gn=10.62903 time=53.91it/s +epoch=51 global_step=20100 loss=2.51346 loss_avg=2.49137 acc=0.76562 acc_top1_avg=0.76430 acc_top5_avg=0.96728 lr=0.00100 gn=11.08621 time=55.68it/s +epoch=51 global_step=20150 loss=2.18191 loss_avg=2.48582 acc=0.78906 acc_top1_avg=0.76518 acc_top5_avg=0.96681 lr=0.00100 gn=14.05813 time=59.42it/s +epoch=51 global_step=20200 loss=2.20609 loss_avg=2.49247 acc=0.78906 acc_top1_avg=0.76406 acc_top5_avg=0.96682 lr=0.00100 gn=9.87827 time=53.05it/s +epoch=51 global_step=20250 loss=1.79366 loss_avg=2.49123 acc=0.84375 acc_top1_avg=0.76436 acc_top5_avg=0.96647 lr=0.00100 gn=16.92061 time=60.76it/s +epoch=51 global_step=20300 loss=2.35149 loss_avg=2.48499 acc=0.78125 acc_top1_avg=0.76502 acc_top5_avg=0.96668 lr=0.00100 gn=10.82794 time=59.20it/s +====================Eval==================== +epoch=51 global_step=20332 loss=0.51413 test_loss_avg=0.36174 acc=0.85156 test_acc_avg=0.89174 test_acc_top5_avg=0.99554 time=216.10it/s +epoch=51 global_step=20332 loss=0.22910 test_loss_avg=0.39274 acc=0.93750 test_acc_avg=0.88765 test_acc_top5_avg=0.99351 time=240.11it/s +epoch=51 global_step=20332 loss=0.14876 test_loss_avg=0.38486 acc=0.93750 test_acc_avg=0.89033 test_acc_top5_avg=0.99357 time=859.66it/s +curr_acc 0.8903 +BEST_ACC 0.8939 +curr_acc_top5 0.9936 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=2.45899 loss_avg=2.43112 acc=0.76562 acc_top1_avg=0.76823 acc_top5_avg=0.96398 lr=0.00100 gn=11.69848 time=49.58it/s +epoch=52 global_step=20400 loss=2.82457 loss_avg=2.47968 acc=0.73438 acc_top1_avg=0.76540 acc_top5_avg=0.96461 lr=0.00100 gn=13.23403 time=54.51it/s +epoch=52 global_step=20450 loss=2.34223 loss_avg=2.46689 acc=0.77344 acc_top1_avg=0.76662 acc_top5_avg=0.96398 lr=0.00100 gn=12.37233 time=54.01it/s +epoch=52 global_step=20500 loss=2.68923 loss_avg=2.45853 acc=0.74219 acc_top1_avg=0.76818 acc_top5_avg=0.96568 lr=0.00100 gn=11.53825 time=51.83it/s +epoch=52 global_step=20550 loss=2.94613 loss_avg=2.45209 acc=0.71094 acc_top1_avg=0.76878 acc_top5_avg=0.96567 lr=0.00100 gn=13.79384 time=53.48it/s +epoch=52 global_step=20600 loss=2.50507 loss_avg=2.45634 acc=0.75000 acc_top1_avg=0.76834 acc_top5_avg=0.96621 lr=0.00100 gn=12.89091 time=53.81it/s +epoch=52 global_step=20650 loss=2.57397 loss_avg=2.46628 acc=0.75000 acc_top1_avg=0.76725 acc_top5_avg=0.96681 lr=0.00100 gn=10.62633 time=33.33it/s +epoch=52 global_step=20700 loss=2.45129 loss_avg=2.46309 acc=0.76562 acc_top1_avg=0.76764 acc_top5_avg=0.96703 lr=0.00100 gn=10.24994 time=56.02it/s +====================Eval==================== +epoch=52 global_step=20723 loss=0.46772 test_loss_avg=0.45684 acc=0.86719 test_acc_avg=0.87016 test_acc_top5_avg=0.99107 time=231.41it/s +epoch=52 global_step=20723 loss=0.33743 test_loss_avg=0.37393 acc=0.87500 test_acc_avg=0.89171 test_acc_top5_avg=0.99288 time=513.19it/s +curr_acc 0.8917 +BEST_ACC 0.8939 +curr_acc_top5 0.9929 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=2.39730 loss_avg=2.34358 acc=0.77344 acc_top1_avg=0.78096 acc_top5_avg=0.97251 lr=0.00100 gn=12.63630 time=58.77it/s +epoch=53 global_step=20800 loss=2.52370 loss_avg=2.42998 acc=0.75000 acc_top1_avg=0.77141 acc_top5_avg=0.96794 lr=0.00100 gn=12.30841 time=63.31it/s +epoch=53 global_step=20850 loss=1.77194 loss_avg=2.44092 acc=0.83594 acc_top1_avg=0.76969 acc_top5_avg=0.96740 lr=0.00100 gn=13.48755 time=59.90it/s +epoch=53 global_step=20900 loss=2.40181 loss_avg=2.46178 acc=0.76562 acc_top1_avg=0.76735 acc_top5_avg=0.96751 lr=0.00100 gn=12.95275 time=54.99it/s +epoch=53 global_step=20950 loss=2.50360 loss_avg=2.47572 acc=0.75781 acc_top1_avg=0.76597 acc_top5_avg=0.96730 lr=0.00100 gn=11.77506 time=46.54it/s +epoch=53 global_step=21000 loss=3.08340 loss_avg=2.44906 acc=0.69531 acc_top1_avg=0.76895 acc_top5_avg=0.96751 lr=0.00100 gn=15.44446 time=57.38it/s +epoch=53 global_step=21050 loss=2.53158 loss_avg=2.44968 acc=0.75000 acc_top1_avg=0.76892 acc_top5_avg=0.96729 lr=0.00100 gn=13.95232 time=46.11it/s +epoch=53 global_step=21100 loss=2.36335 loss_avg=2.45691 acc=0.79688 acc_top1_avg=0.76836 acc_top5_avg=0.96693 lr=0.00100 gn=15.88041 time=60.28it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.05636 test_loss_avg=0.41241 acc=0.98438 test_acc_avg=0.86779 test_acc_top5_avg=0.99399 time=232.59it/s +epoch=53 global_step=21114 loss=0.16117 test_loss_avg=0.41510 acc=0.95312 test_acc_avg=0.87810 test_acc_top5_avg=0.99182 time=235.37it/s +epoch=53 global_step=21114 loss=0.39709 test_loss_avg=0.39008 acc=0.87500 test_acc_avg=0.88627 test_acc_top5_avg=0.99278 time=858.08it/s +curr_acc 0.8863 +BEST_ACC 0.8939 +curr_acc_top5 0.9928 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=2.74086 loss_avg=2.45711 acc=0.73438 acc_top1_avg=0.76584 acc_top5_avg=0.96940 lr=0.00100 gn=13.75961 time=54.28it/s +epoch=54 global_step=21200 loss=2.29966 loss_avg=2.45110 acc=0.78125 acc_top1_avg=0.77062 acc_top5_avg=0.96802 lr=0.00100 gn=10.49091 time=55.88it/s +epoch=54 global_step=21250 loss=2.89718 loss_avg=2.42558 acc=0.72656 acc_top1_avg=0.77258 acc_top5_avg=0.96829 lr=0.00100 gn=10.43297 time=55.56it/s +epoch=54 global_step=21300 loss=2.99245 loss_avg=2.44607 acc=0.71875 acc_top1_avg=0.77062 acc_top5_avg=0.96778 lr=0.00100 gn=16.07980 time=53.37it/s +epoch=54 global_step=21350 loss=2.97061 loss_avg=2.44309 acc=0.71094 acc_top1_avg=0.77079 acc_top5_avg=0.96792 lr=0.00100 gn=14.62551 time=62.70it/s +epoch=54 global_step=21400 loss=1.98078 loss_avg=2.43366 acc=0.82031 acc_top1_avg=0.77128 acc_top5_avg=0.96788 lr=0.00100 gn=13.43033 time=52.53it/s +epoch=54 global_step=21450 loss=2.71939 loss_avg=2.42798 acc=0.75000 acc_top1_avg=0.77181 acc_top5_avg=0.96789 lr=0.00100 gn=11.77723 time=54.96it/s +epoch=54 global_step=21500 loss=1.97334 loss_avg=2.43805 acc=0.82812 acc_top1_avg=0.77064 acc_top5_avg=0.96741 lr=0.00100 gn=13.81442 time=56.04it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.23825 test_loss_avg=0.49809 acc=0.93750 test_acc_avg=0.86052 test_acc_top5_avg=0.99150 time=232.91it/s +epoch=54 global_step=21505 loss=0.14514 test_loss_avg=0.40178 acc=0.87500 test_acc_avg=0.88538 test_acc_top5_avg=0.99328 time=640.25it/s +curr_acc 0.8854 +BEST_ACC 0.8939 +curr_acc_top5 0.9933 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=2.26623 loss_avg=2.40211 acc=0.78125 acc_top1_avg=0.77187 acc_top5_avg=0.96927 lr=0.00100 gn=11.32717 time=62.27it/s +epoch=55 global_step=21600 loss=3.11898 loss_avg=2.39266 acc=0.70312 acc_top1_avg=0.77393 acc_top5_avg=0.96924 lr=0.00100 gn=14.28837 time=51.42it/s +epoch=55 global_step=21650 loss=2.84244 loss_avg=2.40447 acc=0.71875 acc_top1_avg=0.77236 acc_top5_avg=0.96794 lr=0.00100 gn=18.12271 time=54.52it/s +epoch=55 global_step=21700 loss=3.34530 loss_avg=2.41874 acc=0.68750 acc_top1_avg=0.77175 acc_top5_avg=0.96811 lr=0.00100 gn=18.96548 time=55.40it/s +epoch=55 global_step=21750 loss=2.23530 loss_avg=2.42667 acc=0.78906 acc_top1_avg=0.77105 acc_top5_avg=0.96716 lr=0.00100 gn=16.42225 time=64.08it/s +epoch=55 global_step=21800 loss=2.06326 loss_avg=2.41681 acc=0.81250 acc_top1_avg=0.77227 acc_top5_avg=0.96716 lr=0.00100 gn=12.79112 time=54.37it/s +epoch=55 global_step=21850 loss=2.44042 loss_avg=2.41442 acc=0.76562 acc_top1_avg=0.77271 acc_top5_avg=0.96716 lr=0.00100 gn=13.73906 time=62.60it/s +====================Eval==================== +epoch=55 global_step=21896 loss=0.40070 test_loss_avg=0.55595 acc=0.86719 test_acc_avg=0.82969 test_acc_top5_avg=0.99531 time=242.91it/s +epoch=55 global_step=21896 loss=0.20509 test_loss_avg=0.47550 acc=0.91406 test_acc_avg=0.86520 test_acc_top5_avg=0.99134 time=233.87it/s +epoch=55 global_step=21896 loss=0.16733 test_loss_avg=0.39760 acc=0.87500 test_acc_avg=0.88578 test_acc_top5_avg=0.99268 time=853.37it/s +curr_acc 0.8858 +BEST_ACC 0.8939 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=3.01598 loss_avg=2.61838 acc=0.72656 acc_top1_avg=0.75781 acc_top5_avg=0.96289 lr=0.00100 gn=14.95463 time=53.64it/s +epoch=56 global_step=21950 loss=2.37180 loss_avg=2.37925 acc=0.75781 acc_top1_avg=0.77778 acc_top5_avg=0.96846 lr=0.00100 gn=15.48695 time=63.22it/s +epoch=56 global_step=22000 loss=2.52152 loss_avg=2.38734 acc=0.75781 acc_top1_avg=0.77622 acc_top5_avg=0.96717 lr=0.00100 gn=12.39149 time=55.22it/s +epoch=56 global_step=22050 loss=2.27480 loss_avg=2.39388 acc=0.79688 acc_top1_avg=0.77587 acc_top5_avg=0.96733 lr=0.00100 gn=16.03069 time=63.74it/s +epoch=56 global_step=22100 loss=3.26088 loss_avg=2.39799 acc=0.68750 acc_top1_avg=0.77554 acc_top5_avg=0.96760 lr=0.00100 gn=16.41481 time=53.80it/s +epoch=56 global_step=22150 loss=2.05126 loss_avg=2.38226 acc=0.80469 acc_top1_avg=0.77688 acc_top5_avg=0.96869 lr=0.00100 gn=17.47914 time=55.49it/s +epoch=56 global_step=22200 loss=2.67082 loss_avg=2.38908 acc=0.75000 acc_top1_avg=0.77642 acc_top5_avg=0.96808 lr=0.00100 gn=12.39942 time=50.90it/s +epoch=56 global_step=22250 loss=2.59690 loss_avg=2.39101 acc=0.76562 acc_top1_avg=0.77611 acc_top5_avg=0.96798 lr=0.00100 gn=15.80891 time=61.99it/s +====================Eval==================== +epoch=56 global_step=22287 loss=0.56307 test_loss_avg=0.47863 acc=0.82031 test_acc_avg=0.85877 test_acc_top5_avg=0.99069 time=240.07it/s +epoch=56 global_step=22287 loss=0.31646 test_loss_avg=0.40188 acc=0.91406 test_acc_avg=0.88477 test_acc_top5_avg=0.99239 time=245.15it/s +epoch=56 global_step=22287 loss=0.44704 test_loss_avg=0.40030 acc=0.87500 test_acc_avg=0.88528 test_acc_top5_avg=0.99268 time=588.10it/s +curr_acc 0.8853 +BEST_ACC 0.8939 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=3.07687 loss_avg=2.30454 acc=0.69531 acc_top1_avg=0.78486 acc_top5_avg=0.96695 lr=0.00100 gn=13.33226 time=54.62it/s +epoch=57 global_step=22350 loss=2.17151 loss_avg=2.34971 acc=0.79688 acc_top1_avg=0.77976 acc_top5_avg=0.96763 lr=0.00100 gn=17.94166 time=56.30it/s +epoch=57 global_step=22400 loss=2.07792 loss_avg=2.36271 acc=0.79688 acc_top1_avg=0.77800 acc_top5_avg=0.96910 lr=0.00100 gn=16.63343 time=49.30it/s +epoch=57 global_step=22450 loss=2.54015 loss_avg=2.33695 acc=0.76562 acc_top1_avg=0.78183 acc_top5_avg=0.96865 lr=0.00100 gn=19.72372 time=63.71it/s +epoch=57 global_step=22500 loss=2.24278 loss_avg=2.35240 acc=0.79688 acc_top1_avg=0.77975 acc_top5_avg=0.96794 lr=0.00100 gn=16.58789 time=61.14it/s +epoch=57 global_step=22550 loss=3.21640 loss_avg=2.36471 acc=0.68750 acc_top1_avg=0.77875 acc_top5_avg=0.96771 lr=0.00100 gn=20.71899 time=44.98it/s +epoch=57 global_step=22600 loss=2.25947 loss_avg=2.36937 acc=0.79688 acc_top1_avg=0.77833 acc_top5_avg=0.96780 lr=0.00100 gn=15.73188 time=31.51it/s +epoch=57 global_step=22650 loss=2.15789 loss_avg=2.37428 acc=0.81250 acc_top1_avg=0.77813 acc_top5_avg=0.96731 lr=0.00100 gn=16.45999 time=53.98it/s +====================Eval==================== +epoch=57 global_step=22678 loss=0.31706 test_loss_avg=0.50930 acc=0.89062 test_acc_avg=0.85805 test_acc_top5_avg=0.99136 time=227.32it/s +epoch=57 global_step=22678 loss=0.30905 test_loss_avg=0.39035 acc=0.93750 test_acc_avg=0.89102 test_acc_top5_avg=0.99298 time=434.10it/s +curr_acc 0.8910 +BEST_ACC 0.8939 +curr_acc_top5 0.9930 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=2.81292 loss_avg=2.32627 acc=0.73438 acc_top1_avg=0.78196 acc_top5_avg=0.96875 lr=0.00100 gn=13.86587 time=59.21it/s +epoch=58 global_step=22750 loss=2.41686 loss_avg=2.41834 acc=0.77344 acc_top1_avg=0.77355 acc_top5_avg=0.96832 lr=0.00100 gn=14.44852 time=52.27it/s +epoch=58 global_step=22800 loss=2.38964 loss_avg=2.38097 acc=0.78125 acc_top1_avg=0.77683 acc_top5_avg=0.96862 lr=0.00100 gn=16.70522 time=54.08it/s +epoch=58 global_step=22850 loss=2.26775 loss_avg=2.38435 acc=0.78125 acc_top1_avg=0.77693 acc_top5_avg=0.96784 lr=0.00100 gn=12.87732 time=54.10it/s +epoch=58 global_step=22900 loss=2.42157 loss_avg=2.39236 acc=0.76562 acc_top1_avg=0.77594 acc_top5_avg=0.96794 lr=0.00100 gn=17.15556 time=58.30it/s +epoch=58 global_step=22950 loss=2.26339 loss_avg=2.38571 acc=0.80469 acc_top1_avg=0.77648 acc_top5_avg=0.96774 lr=0.00100 gn=21.83837 time=50.54it/s +epoch=58 global_step=23000 loss=2.45379 loss_avg=2.38030 acc=0.76562 acc_top1_avg=0.77698 acc_top5_avg=0.96681 lr=0.00100 gn=18.50648 time=54.33it/s +epoch=58 global_step=23050 loss=2.93356 loss_avg=2.37606 acc=0.72656 acc_top1_avg=0.77757 acc_top5_avg=0.96743 lr=0.00100 gn=16.20734 time=56.30it/s +====================Eval==================== +epoch=58 global_step=23069 loss=0.61305 test_loss_avg=0.44669 acc=0.83594 test_acc_avg=0.86589 test_acc_top5_avg=0.98785 time=234.76it/s +epoch=58 global_step=23069 loss=0.28883 test_loss_avg=0.41598 acc=0.91406 test_acc_avg=0.88166 test_acc_top5_avg=0.99219 time=242.14it/s +epoch=58 global_step=23069 loss=0.26543 test_loss_avg=0.39096 acc=0.93750 test_acc_avg=0.88894 test_acc_top5_avg=0.99258 time=452.61it/s +curr_acc 0.8889 +BEST_ACC 0.8939 +curr_acc_top5 0.9926 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=2.14574 loss_avg=2.38118 acc=0.79688 acc_top1_avg=0.77747 acc_top5_avg=0.96724 lr=0.00100 gn=15.06939 time=62.48it/s +epoch=59 global_step=23150 loss=2.77383 loss_avg=2.35167 acc=0.73438 acc_top1_avg=0.77961 acc_top5_avg=0.96769 lr=0.00100 gn=17.15945 time=42.50it/s +epoch=59 global_step=23200 loss=2.33166 loss_avg=2.35029 acc=0.78906 acc_top1_avg=0.78006 acc_top5_avg=0.96774 lr=0.00100 gn=17.49084 time=60.82it/s +epoch=59 global_step=23250 loss=2.42887 loss_avg=2.34744 acc=0.76562 acc_top1_avg=0.78121 acc_top5_avg=0.96806 lr=0.00100 gn=12.79266 time=59.21it/s +epoch=59 global_step=23300 loss=2.45320 loss_avg=2.36757 acc=0.78906 acc_top1_avg=0.77844 acc_top5_avg=0.96770 lr=0.00100 gn=18.00075 time=55.47it/s +epoch=59 global_step=23350 loss=2.62004 loss_avg=2.36288 acc=0.75000 acc_top1_avg=0.77886 acc_top5_avg=0.96786 lr=0.00100 gn=16.63305 time=56.44it/s +epoch=59 global_step=23400 loss=2.32038 loss_avg=2.36225 acc=0.78906 acc_top1_avg=0.77884 acc_top5_avg=0.96790 lr=0.00100 gn=14.06882 time=51.38it/s +epoch=59 global_step=23450 loss=2.03691 loss_avg=2.35893 acc=0.80469 acc_top1_avg=0.77922 acc_top5_avg=0.96756 lr=0.00100 gn=14.80035 time=63.96it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.11514 test_loss_avg=0.46039 acc=0.96094 test_acc_avg=0.85998 test_acc_top5_avg=0.99219 time=240.03it/s +epoch=59 global_step=23460 loss=0.26452 test_loss_avg=0.40735 acc=0.87500 test_acc_avg=0.87846 test_acc_top5_avg=0.99318 time=503.03it/s +curr_acc 0.8785 +BEST_ACC 0.8939 +curr_acc_top5 0.9932 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=2.19333 loss_avg=2.32217 acc=0.81250 acc_top1_avg=0.78535 acc_top5_avg=0.96680 lr=0.00100 gn=17.22222 time=63.84it/s +epoch=60 global_step=23550 loss=1.73921 loss_avg=2.33880 acc=0.85156 acc_top1_avg=0.78299 acc_top5_avg=0.96632 lr=0.00100 gn=14.75250 time=55.29it/s +epoch=60 global_step=23600 loss=2.49917 loss_avg=2.34154 acc=0.78125 acc_top1_avg=0.78326 acc_top5_avg=0.96685 lr=0.00100 gn=16.47173 time=54.97it/s +epoch=60 global_step=23650 loss=2.16659 loss_avg=2.33479 acc=0.81250 acc_top1_avg=0.78372 acc_top5_avg=0.96739 lr=0.00100 gn=14.63014 time=56.06it/s +epoch=60 global_step=23700 loss=1.72190 loss_avg=2.32634 acc=0.83594 acc_top1_avg=0.78447 acc_top5_avg=0.96673 lr=0.00100 gn=17.93758 time=62.50it/s +epoch=60 global_step=23750 loss=2.21795 loss_avg=2.32923 acc=0.79688 acc_top1_avg=0.78389 acc_top5_avg=0.96697 lr=0.00100 gn=16.13917 time=54.10it/s +epoch=60 global_step=23800 loss=2.49996 loss_avg=2.34024 acc=0.75781 acc_top1_avg=0.78254 acc_top5_avg=0.96744 lr=0.00100 gn=20.34103 time=60.69it/s +epoch=60 global_step=23850 loss=2.57439 loss_avg=2.34263 acc=0.75000 acc_top1_avg=0.78241 acc_top5_avg=0.96747 lr=0.00100 gn=17.55028 time=62.30it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.17262 test_loss_avg=0.57339 acc=0.94531 test_acc_avg=0.83359 test_acc_top5_avg=0.99141 time=240.00it/s +epoch=60 global_step=23851 loss=0.21642 test_loss_avg=0.46068 acc=0.93750 test_acc_avg=0.86797 test_acc_top5_avg=0.99245 time=243.16it/s +epoch=60 global_step=23851 loss=0.16910 test_loss_avg=0.40166 acc=0.93750 test_acc_avg=0.88439 test_acc_top5_avg=0.99347 time=532.27it/s +curr_acc 0.8844 +BEST_ACC 0.8939 +curr_acc_top5 0.9935 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=2.88230 loss_avg=2.26246 acc=0.71875 acc_top1_avg=0.79082 acc_top5_avg=0.97337 lr=0.00100 gn=19.45909 time=55.55it/s +epoch=61 global_step=23950 loss=2.43050 loss_avg=2.29851 acc=0.78906 acc_top1_avg=0.78733 acc_top5_avg=0.96978 lr=0.00100 gn=19.59709 time=54.59it/s +epoch=61 global_step=24000 loss=2.30000 loss_avg=2.30719 acc=0.79688 acc_top1_avg=0.78602 acc_top5_avg=0.96906 lr=0.00100 gn=15.75531 time=54.04it/s +epoch=61 global_step=24050 loss=2.06975 loss_avg=2.30121 acc=0.81250 acc_top1_avg=0.78647 acc_top5_avg=0.96875 lr=0.00100 gn=15.44725 time=56.04it/s +epoch=61 global_step=24100 loss=2.25401 loss_avg=2.31307 acc=0.78125 acc_top1_avg=0.78505 acc_top5_avg=0.96913 lr=0.00100 gn=14.83398 time=60.36it/s +epoch=61 global_step=24150 loss=2.52162 loss_avg=2.31611 acc=0.76562 acc_top1_avg=0.78488 acc_top5_avg=0.96846 lr=0.00100 gn=17.60325 time=61.44it/s +epoch=61 global_step=24200 loss=2.38441 loss_avg=2.31653 acc=0.78125 acc_top1_avg=0.78472 acc_top5_avg=0.96808 lr=0.00100 gn=17.85631 time=53.71it/s +====================Eval==================== +epoch=61 global_step=24242 loss=0.49221 test_loss_avg=0.48647 acc=0.85938 test_acc_avg=0.85585 test_acc_top5_avg=0.99294 time=241.44it/s +epoch=61 global_step=24242 loss=0.20471 test_loss_avg=0.39057 acc=0.87500 test_acc_avg=0.88627 test_acc_top5_avg=0.99426 time=859.31it/s +curr_acc 0.8863 +BEST_ACC 0.8939 +curr_acc_top5 0.9943 +BEST_ACC_top5 0.9942 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=2.25780 loss_avg=2.29890 acc=0.78906 acc_top1_avg=0.78906 acc_top5_avg=0.97266 lr=0.00100 gn=19.97933 time=51.50it/s +epoch=62 global_step=24300 loss=3.24081 loss_avg=2.29867 acc=0.69531 acc_top1_avg=0.78731 acc_top5_avg=0.96727 lr=0.00100 gn=20.64016 time=52.37it/s +epoch=62 global_step=24350 loss=2.12606 loss_avg=2.28030 acc=0.80469 acc_top1_avg=0.78935 acc_top5_avg=0.96861 lr=0.00100 gn=19.93929 time=56.85it/s +epoch=62 global_step=24400 loss=2.02549 loss_avg=2.28101 acc=0.81250 acc_top1_avg=0.78872 acc_top5_avg=0.96860 lr=0.00100 gn=15.38657 time=51.42it/s +epoch=62 global_step=24450 loss=1.69751 loss_avg=2.28691 acc=0.85938 acc_top1_avg=0.78816 acc_top5_avg=0.96909 lr=0.00100 gn=18.61709 time=59.98it/s +epoch=62 global_step=24500 loss=1.75590 loss_avg=2.29709 acc=0.84375 acc_top1_avg=0.78740 acc_top5_avg=0.96872 lr=0.00100 gn=16.95093 time=64.06it/s +epoch=62 global_step=24550 loss=2.50681 loss_avg=2.31168 acc=0.76562 acc_top1_avg=0.78589 acc_top5_avg=0.96784 lr=0.00100 gn=22.54550 time=62.85it/s +epoch=62 global_step=24600 loss=2.32454 loss_avg=2.32357 acc=0.78906 acc_top1_avg=0.78481 acc_top5_avg=0.96794 lr=0.00100 gn=18.33553 time=48.54it/s +====================Eval==================== +epoch=62 global_step=24633 loss=0.95241 test_loss_avg=0.81911 acc=0.73438 test_acc_avg=0.76953 test_acc_top5_avg=0.98438 time=239.74it/s +epoch=62 global_step=24633 loss=0.08506 test_loss_avg=0.51625 acc=0.96875 test_acc_avg=0.85096 test_acc_top5_avg=0.98828 time=242.04it/s +epoch=62 global_step=24633 loss=0.45761 test_loss_avg=0.42455 acc=0.93750 test_acc_avg=0.87816 test_acc_top5_avg=0.99150 time=509.51it/s +curr_acc 0.8782 +BEST_ACC 0.8939 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=1.80677 loss_avg=2.28525 acc=0.84375 acc_top1_avg=0.78952 acc_top5_avg=0.96415 lr=0.00100 gn=18.60679 time=54.32it/s +epoch=63 global_step=24700 loss=2.30409 loss_avg=2.26169 acc=0.78906 acc_top1_avg=0.79081 acc_top5_avg=0.96980 lr=0.00100 gn=19.77415 time=50.42it/s +epoch=63 global_step=24750 loss=2.16732 loss_avg=2.26637 acc=0.80469 acc_top1_avg=0.79033 acc_top5_avg=0.96828 lr=0.00100 gn=20.53969 time=63.01it/s +epoch=63 global_step=24800 loss=2.32145 loss_avg=2.28087 acc=0.78125 acc_top1_avg=0.78962 acc_top5_avg=0.96814 lr=0.00100 gn=18.18657 time=54.83it/s +epoch=63 global_step=24850 loss=2.27577 loss_avg=2.29487 acc=0.78906 acc_top1_avg=0.78820 acc_top5_avg=0.96792 lr=0.00100 gn=16.67123 time=53.34it/s +epoch=63 global_step=24900 loss=2.47708 loss_avg=2.29973 acc=0.77344 acc_top1_avg=0.78807 acc_top5_avg=0.96849 lr=0.00100 gn=25.18636 time=60.21it/s +epoch=63 global_step=24950 loss=2.94289 loss_avg=2.31291 acc=0.70312 acc_top1_avg=0.78650 acc_top5_avg=0.96826 lr=0.00100 gn=24.42136 time=62.12it/s +epoch=63 global_step=25000 loss=1.99627 loss_avg=2.30585 acc=0.83594 acc_top1_avg=0.78738 acc_top5_avg=0.96837 lr=0.00100 gn=19.77373 time=54.38it/s +====================Eval==================== +epoch=63 global_step=25024 loss=0.92164 test_loss_avg=0.57467 acc=0.73438 test_acc_avg=0.82745 test_acc_top5_avg=0.98913 time=229.60it/s +epoch=63 global_step=25024 loss=0.38480 test_loss_avg=0.43954 acc=0.91406 test_acc_avg=0.87361 test_acc_top5_avg=0.99165 time=257.15it/s +epoch=63 global_step=25024 loss=0.44258 test_loss_avg=0.43295 acc=0.93750 test_acc_avg=0.87688 test_acc_top5_avg=0.99209 time=842.06it/s +curr_acc 0.8769 +BEST_ACC 0.8939 +curr_acc_top5 0.9921 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=2.47435 loss_avg=2.24187 acc=0.78906 acc_top1_avg=0.79237 acc_top5_avg=0.97055 lr=0.00100 gn=27.44947 time=55.84it/s +epoch=64 global_step=25100 loss=1.79392 loss_avg=2.24063 acc=0.84375 acc_top1_avg=0.79379 acc_top5_avg=0.97050 lr=0.00100 gn=19.07342 time=41.19it/s +epoch=64 global_step=25150 loss=1.92870 loss_avg=2.23774 acc=0.83594 acc_top1_avg=0.79458 acc_top5_avg=0.97111 lr=0.00100 gn=17.53738 time=64.03it/s +epoch=64 global_step=25200 loss=2.53181 loss_avg=2.27696 acc=0.76562 acc_top1_avg=0.79039 acc_top5_avg=0.97008 lr=0.00100 gn=19.75017 time=48.48it/s +epoch=64 global_step=25250 loss=2.20756 loss_avg=2.29479 acc=0.78906 acc_top1_avg=0.78820 acc_top5_avg=0.96968 lr=0.00100 gn=18.97041 time=24.06it/s +epoch=64 global_step=25300 loss=2.15258 loss_avg=2.27636 acc=0.80469 acc_top1_avg=0.79002 acc_top5_avg=0.96912 lr=0.00100 gn=15.01168 time=56.06it/s +epoch=64 global_step=25350 loss=2.15835 loss_avg=2.28212 acc=0.80469 acc_top1_avg=0.78966 acc_top5_avg=0.96880 lr=0.00100 gn=19.27733 time=55.29it/s +epoch=64 global_step=25400 loss=1.81680 loss_avg=2.28527 acc=0.82812 acc_top1_avg=0.78937 acc_top5_avg=0.96829 lr=0.00100 gn=18.58153 time=57.72it/s +====================Eval==================== +epoch=64 global_step=25415 loss=0.95795 test_loss_avg=0.50732 acc=0.76562 test_acc_avg=0.85032 test_acc_top5_avg=0.99201 time=235.56it/s +epoch=64 global_step=25415 loss=0.23031 test_loss_avg=0.40884 acc=0.93750 test_acc_avg=0.88182 test_acc_top5_avg=0.99308 time=500.93it/s +curr_acc 0.8818 +BEST_ACC 0.8939 +curr_acc_top5 0.9931 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=2.06195 loss_avg=2.23186 acc=0.81250 acc_top1_avg=0.79576 acc_top5_avg=0.97232 lr=0.00100 gn=18.14393 time=55.57it/s +epoch=65 global_step=25500 loss=1.55890 loss_avg=2.18876 acc=0.87500 acc_top1_avg=0.80037 acc_top5_avg=0.97031 lr=0.00100 gn=22.22043 time=59.61it/s +epoch=65 global_step=25550 loss=2.68191 loss_avg=2.22565 acc=0.77344 acc_top1_avg=0.79716 acc_top5_avg=0.96794 lr=0.00100 gn=23.97531 time=55.92it/s +epoch=65 global_step=25600 loss=2.09077 loss_avg=2.25159 acc=0.80469 acc_top1_avg=0.79438 acc_top5_avg=0.96719 lr=0.00100 gn=22.73494 time=63.11it/s +epoch=65 global_step=25650 loss=1.62709 loss_avg=2.25384 acc=0.84375 acc_top1_avg=0.79372 acc_top5_avg=0.96785 lr=0.00100 gn=20.70377 time=59.94it/s +epoch=65 global_step=25700 loss=1.94627 loss_avg=2.25713 acc=0.80469 acc_top1_avg=0.79309 acc_top5_avg=0.96831 lr=0.00100 gn=20.99833 time=54.90it/s +epoch=65 global_step=25750 loss=2.76062 loss_avg=2.26672 acc=0.73438 acc_top1_avg=0.79205 acc_top5_avg=0.96828 lr=0.00100 gn=22.15837 time=63.92it/s +epoch=65 global_step=25800 loss=2.81415 loss_avg=2.26980 acc=0.74219 acc_top1_avg=0.79170 acc_top5_avg=0.96822 lr=0.00100 gn=27.46851 time=56.08it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.14711 test_loss_avg=0.37717 acc=0.97656 test_acc_avg=0.88385 test_acc_top5_avg=0.99427 time=224.84it/s +epoch=65 global_step=25806 loss=0.13469 test_loss_avg=0.41931 acc=0.95312 test_acc_avg=0.87704 test_acc_top5_avg=0.99351 time=242.54it/s +epoch=65 global_step=25806 loss=0.27681 test_loss_avg=0.38911 acc=0.93750 test_acc_avg=0.88617 test_acc_top5_avg=0.99407 time=821.29it/s +curr_acc 0.8862 +BEST_ACC 0.8939 +curr_acc_top5 0.9941 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=2.02263 loss_avg=2.21961 acc=0.81250 acc_top1_avg=0.79439 acc_top5_avg=0.96609 lr=0.00100 gn=18.49664 time=63.18it/s +epoch=66 global_step=25900 loss=2.13055 loss_avg=2.22483 acc=0.80469 acc_top1_avg=0.79530 acc_top5_avg=0.96692 lr=0.00100 gn=19.64930 time=54.24it/s +epoch=66 global_step=25950 loss=2.29789 loss_avg=2.23812 acc=0.78125 acc_top1_avg=0.79335 acc_top5_avg=0.96853 lr=0.00100 gn=17.66274 time=54.74it/s +epoch=66 global_step=26000 loss=2.17973 loss_avg=2.24020 acc=0.80469 acc_top1_avg=0.79357 acc_top5_avg=0.96742 lr=0.00100 gn=22.66562 time=60.78it/s +epoch=66 global_step=26050 loss=2.31851 loss_avg=2.23646 acc=0.78125 acc_top1_avg=0.79457 acc_top5_avg=0.96763 lr=0.00100 gn=20.48021 time=54.11it/s +epoch=66 global_step=26100 loss=2.36009 loss_avg=2.24863 acc=0.78125 acc_top1_avg=0.79339 acc_top5_avg=0.96787 lr=0.00100 gn=23.01910 time=54.54it/s +epoch=66 global_step=26150 loss=1.70116 loss_avg=2.25988 acc=0.85156 acc_top1_avg=0.79222 acc_top5_avg=0.96791 lr=0.00100 gn=17.82630 time=30.51it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.21413 test_loss_avg=0.45798 acc=0.92188 test_acc_avg=0.86654 test_acc_top5_avg=0.99154 time=235.90it/s +epoch=66 global_step=26197 loss=0.26588 test_loss_avg=0.40516 acc=0.93750 test_acc_avg=0.88558 test_acc_top5_avg=0.99258 time=515.52it/s +curr_acc 0.8856 +BEST_ACC 0.8939 +curr_acc_top5 0.9926 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=1.90608 loss_avg=2.32785 acc=0.84375 acc_top1_avg=0.78646 acc_top5_avg=0.96354 lr=0.00100 gn=22.01683 time=51.12it/s +epoch=67 global_step=26250 loss=2.38232 loss_avg=2.22637 acc=0.78125 acc_top1_avg=0.79761 acc_top5_avg=0.96713 lr=0.00100 gn=16.76725 time=56.93it/s +epoch=67 global_step=26300 loss=2.62381 loss_avg=2.23034 acc=0.75000 acc_top1_avg=0.79612 acc_top5_avg=0.96625 lr=0.00100 gn=24.50609 time=57.55it/s +epoch=67 global_step=26350 loss=2.00155 loss_avg=2.24164 acc=0.82031 acc_top1_avg=0.79468 acc_top5_avg=0.96569 lr=0.00100 gn=22.62078 time=46.63it/s +epoch=67 global_step=26400 loss=2.45176 loss_avg=2.24010 acc=0.76562 acc_top1_avg=0.79433 acc_top5_avg=0.96686 lr=0.00100 gn=20.58401 time=50.45it/s +epoch=67 global_step=26450 loss=1.87312 loss_avg=2.23599 acc=0.83594 acc_top1_avg=0.79515 acc_top5_avg=0.96671 lr=0.00100 gn=16.68477 time=55.43it/s +epoch=67 global_step=26500 loss=2.75460 loss_avg=2.23352 acc=0.72656 acc_top1_avg=0.79543 acc_top5_avg=0.96705 lr=0.00100 gn=19.71787 time=56.40it/s +epoch=67 global_step=26550 loss=1.76110 loss_avg=2.24397 acc=0.83594 acc_top1_avg=0.79413 acc_top5_avg=0.96744 lr=0.00100 gn=23.93274 time=51.96it/s +====================Eval==================== +epoch=67 global_step=26588 loss=0.88961 test_loss_avg=0.82316 acc=0.77344 test_acc_avg=0.75893 test_acc_top5_avg=0.98103 time=236.89it/s +epoch=67 global_step=26588 loss=0.32400 test_loss_avg=0.50486 acc=0.91406 test_acc_avg=0.85252 test_acc_top5_avg=0.98945 time=191.94it/s +epoch=67 global_step=26588 loss=0.25229 test_loss_avg=0.42363 acc=0.93750 test_acc_avg=0.87549 test_acc_top5_avg=0.99179 time=433.83it/s +curr_acc 0.8755 +BEST_ACC 0.8939 +curr_acc_top5 0.9918 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=2.05796 loss_avg=2.24042 acc=0.81250 acc_top1_avg=0.79557 acc_top5_avg=0.97526 lr=0.00100 gn=20.39486 time=54.23it/s +epoch=68 global_step=26650 loss=2.16170 loss_avg=2.23249 acc=0.81250 acc_top1_avg=0.79587 acc_top5_avg=0.96837 lr=0.00100 gn=25.24917 time=55.07it/s +epoch=68 global_step=26700 loss=2.26228 loss_avg=2.21881 acc=0.78125 acc_top1_avg=0.79715 acc_top5_avg=0.96966 lr=0.00100 gn=19.10974 time=52.85it/s +epoch=68 global_step=26750 loss=1.74419 loss_avg=2.20292 acc=0.85938 acc_top1_avg=0.79784 acc_top5_avg=0.96947 lr=0.00100 gn=22.19236 time=53.52it/s +epoch=68 global_step=26800 loss=2.52513 loss_avg=2.21041 acc=0.77344 acc_top1_avg=0.79761 acc_top5_avg=0.96893 lr=0.00100 gn=23.82192 time=58.83it/s +epoch=68 global_step=26850 loss=2.77231 loss_avg=2.21852 acc=0.73438 acc_top1_avg=0.79702 acc_top5_avg=0.96839 lr=0.00100 gn=18.01236 time=54.85it/s +epoch=68 global_step=26900 loss=2.67274 loss_avg=2.23229 acc=0.74219 acc_top1_avg=0.79522 acc_top5_avg=0.96777 lr=0.00100 gn=14.78293 time=60.20it/s +epoch=68 global_step=26950 loss=2.55021 loss_avg=2.22977 acc=0.77344 acc_top1_avg=0.79573 acc_top5_avg=0.96791 lr=0.00100 gn=23.41704 time=56.42it/s +====================Eval==================== +epoch=68 global_step=26979 loss=0.42149 test_loss_avg=0.53285 acc=0.89844 test_acc_avg=0.83622 test_acc_top5_avg=0.99051 time=240.04it/s +epoch=68 global_step=26979 loss=0.29102 test_loss_avg=0.43275 acc=0.92969 test_acc_avg=0.87480 test_acc_top5_avg=0.99259 time=258.17it/s +epoch=68 global_step=26979 loss=0.25422 test_loss_avg=0.43049 acc=0.93750 test_acc_avg=0.87559 test_acc_top5_avg=0.99268 time=820.48it/s +curr_acc 0.8756 +BEST_ACC 0.8939 +curr_acc_top5 0.9927 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=2.57605 loss_avg=2.24427 acc=0.75000 acc_top1_avg=0.79576 acc_top5_avg=0.96540 lr=0.00100 gn=22.49585 time=48.09it/s +epoch=69 global_step=27050 loss=2.34256 loss_avg=2.20402 acc=0.77344 acc_top1_avg=0.79897 acc_top5_avg=0.96754 lr=0.00100 gn=16.88030 time=61.32it/s +epoch=69 global_step=27100 loss=1.66516 loss_avg=2.21800 acc=0.86719 acc_top1_avg=0.79778 acc_top5_avg=0.96888 lr=0.00100 gn=23.58387 time=58.73it/s +epoch=69 global_step=27150 loss=2.15927 loss_avg=2.23123 acc=0.81250 acc_top1_avg=0.79674 acc_top5_avg=0.96820 lr=0.00100 gn=23.71637 time=54.81it/s +epoch=69 global_step=27200 loss=2.02954 loss_avg=2.22326 acc=0.82031 acc_top1_avg=0.79730 acc_top5_avg=0.96889 lr=0.00100 gn=23.61864 time=47.56it/s +epoch=69 global_step=27250 loss=2.21934 loss_avg=2.21299 acc=0.82031 acc_top1_avg=0.79829 acc_top5_avg=0.96852 lr=0.00100 gn=24.69258 time=52.82it/s +epoch=69 global_step=27300 loss=2.38632 loss_avg=2.21800 acc=0.77344 acc_top1_avg=0.79797 acc_top5_avg=0.96853 lr=0.00100 gn=23.56085 time=60.53it/s +epoch=69 global_step=27350 loss=2.20304 loss_avg=2.22096 acc=0.78125 acc_top1_avg=0.79730 acc_top5_avg=0.96883 lr=0.00100 gn=29.06974 time=58.88it/s +====================Eval==================== +epoch=69 global_step=27370 loss=0.14772 test_loss_avg=0.55574 acc=0.96094 test_acc_avg=0.83689 test_acc_top5_avg=0.99027 time=241.19it/s +epoch=69 global_step=27370 loss=0.22910 test_loss_avg=0.44181 acc=0.93750 test_acc_avg=0.86996 test_acc_top5_avg=0.99229 time=720.55it/s +curr_acc 0.8700 +BEST_ACC 0.8939 +curr_acc_top5 0.9923 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=1.78608 loss_avg=2.17857 acc=0.87500 acc_top1_avg=0.80391 acc_top5_avg=0.96745 lr=0.00100 gn=27.43238 time=56.24it/s +epoch=70 global_step=27450 loss=2.38270 loss_avg=2.14191 acc=0.77344 acc_top1_avg=0.80674 acc_top5_avg=0.96631 lr=0.00100 gn=18.74464 time=54.69it/s +epoch=70 global_step=27500 loss=1.92524 loss_avg=2.15165 acc=0.82031 acc_top1_avg=0.80559 acc_top5_avg=0.96815 lr=0.00100 gn=23.36995 time=50.18it/s +epoch=70 global_step=27550 loss=2.47412 loss_avg=2.16945 acc=0.78125 acc_top1_avg=0.80369 acc_top5_avg=0.96797 lr=0.00100 gn=28.68940 time=52.34it/s +epoch=70 global_step=27600 loss=2.03452 loss_avg=2.17732 acc=0.80469 acc_top1_avg=0.80214 acc_top5_avg=0.96844 lr=0.00100 gn=24.76048 time=51.60it/s +epoch=70 global_step=27650 loss=2.18417 loss_avg=2.19484 acc=0.79688 acc_top1_avg=0.80022 acc_top5_avg=0.96794 lr=0.00100 gn=26.91599 time=48.73it/s +epoch=70 global_step=27700 loss=2.46877 loss_avg=2.20830 acc=0.77344 acc_top1_avg=0.79912 acc_top5_avg=0.96778 lr=0.00100 gn=21.32117 time=58.02it/s +epoch=70 global_step=27750 loss=2.19121 loss_avg=2.21292 acc=0.80469 acc_top1_avg=0.79903 acc_top5_avg=0.96760 lr=0.00100 gn=23.35132 time=55.21it/s +====================Eval==================== +epoch=70 global_step=27761 loss=1.07641 test_loss_avg=0.60098 acc=0.71875 test_acc_avg=0.80937 test_acc_top5_avg=0.98867 time=235.23it/s +epoch=70 global_step=27761 loss=0.12623 test_loss_avg=0.47849 acc=0.96875 test_acc_avg=0.86060 test_acc_top5_avg=0.99185 time=248.88it/s +epoch=70 global_step=27761 loss=0.13300 test_loss_avg=0.44826 acc=0.93750 test_acc_avg=0.86936 test_acc_top5_avg=0.99219 time=701.86it/s +curr_acc 0.8694 +BEST_ACC 0.8939 +curr_acc_top5 0.9922 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=2.40699 loss_avg=2.09315 acc=0.77344 acc_top1_avg=0.80950 acc_top5_avg=0.96955 lr=0.00100 gn=23.10023 time=54.62it/s +epoch=71 global_step=27850 loss=2.34996 loss_avg=2.15386 acc=0.78906 acc_top1_avg=0.80320 acc_top5_avg=0.96998 lr=0.00100 gn=25.15622 time=61.03it/s +epoch=71 global_step=27900 loss=2.48016 loss_avg=2.17667 acc=0.78906 acc_top1_avg=0.80199 acc_top5_avg=0.97072 lr=0.00100 gn=23.89904 time=49.64it/s +epoch=71 global_step=27950 loss=2.18436 loss_avg=2.17797 acc=0.78125 acc_top1_avg=0.80217 acc_top5_avg=0.96974 lr=0.00100 gn=23.45104 time=55.52it/s +epoch=71 global_step=28000 loss=2.65353 loss_avg=2.19371 acc=0.75781 acc_top1_avg=0.80060 acc_top5_avg=0.96960 lr=0.00100 gn=22.55892 time=62.64it/s +epoch=71 global_step=28050 loss=2.02081 loss_avg=2.19319 acc=0.82031 acc_top1_avg=0.80058 acc_top5_avg=0.96951 lr=0.00100 gn=21.56718 time=60.88it/s +epoch=71 global_step=28100 loss=3.03176 loss_avg=2.21080 acc=0.71094 acc_top1_avg=0.79913 acc_top5_avg=0.96893 lr=0.00100 gn=19.97389 time=57.62it/s +epoch=71 global_step=28150 loss=2.41248 loss_avg=2.21350 acc=0.78906 acc_top1_avg=0.79910 acc_top5_avg=0.96919 lr=0.00100 gn=29.07583 time=61.22it/s +====================Eval==================== +epoch=71 global_step=28152 loss=0.64809 test_loss_avg=0.68138 acc=0.81250 test_acc_avg=0.79230 test_acc_top5_avg=0.98761 time=246.01it/s +epoch=71 global_step=28152 loss=0.52599 test_loss_avg=0.50592 acc=0.93750 test_acc_avg=0.84968 test_acc_top5_avg=0.99051 time=525.27it/s +curr_acc 0.8497 +BEST_ACC 0.8939 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=2.41089 loss_avg=2.20214 acc=0.78125 acc_top1_avg=0.80029 acc_top5_avg=0.97038 lr=0.00100 gn=21.46487 time=62.98it/s +epoch=72 global_step=28250 loss=1.81490 loss_avg=2.17064 acc=0.84375 acc_top1_avg=0.80317 acc_top5_avg=0.97066 lr=0.00100 gn=24.55280 time=53.36it/s +epoch=72 global_step=28300 loss=2.04510 loss_avg=2.16537 acc=0.82031 acc_top1_avg=0.80326 acc_top5_avg=0.96954 lr=0.00100 gn=20.41518 time=56.36it/s +epoch=72 global_step=28350 loss=2.22108 loss_avg=2.17626 acc=0.79688 acc_top1_avg=0.80264 acc_top5_avg=0.96938 lr=0.00100 gn=24.87156 time=54.36it/s +epoch=72 global_step=28400 loss=2.20605 loss_avg=2.18361 acc=0.80469 acc_top1_avg=0.80207 acc_top5_avg=0.96935 lr=0.00100 gn=21.88609 time=58.83it/s +epoch=72 global_step=28450 loss=2.21126 loss_avg=2.18702 acc=0.80469 acc_top1_avg=0.80159 acc_top5_avg=0.96862 lr=0.00100 gn=24.80240 time=59.45it/s +epoch=72 global_step=28500 loss=1.98779 loss_avg=2.19188 acc=0.83594 acc_top1_avg=0.80128 acc_top5_avg=0.96859 lr=0.00100 gn=26.75841 time=52.58it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.07612 test_loss_avg=0.64784 acc=0.97656 test_acc_avg=0.78776 test_acc_top5_avg=0.98763 time=233.25it/s +epoch=72 global_step=28543 loss=0.30035 test_loss_avg=0.49809 acc=0.89062 test_acc_avg=0.84627 test_acc_top5_avg=0.99005 time=241.14it/s +epoch=72 global_step=28543 loss=0.41129 test_loss_avg=0.46063 acc=0.93750 test_acc_avg=0.86027 test_acc_top5_avg=0.99120 time=780.63it/s +curr_acc 0.8603 +BEST_ACC 0.8939 +curr_acc_top5 0.9912 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=2.63103 loss_avg=1.97639 acc=0.76562 acc_top1_avg=0.82478 acc_top5_avg=0.97098 lr=0.00100 gn=27.36307 time=50.31it/s +epoch=73 global_step=28600 loss=2.08965 loss_avg=2.17082 acc=0.81250 acc_top1_avg=0.80674 acc_top5_avg=0.97122 lr=0.00100 gn=23.30282 time=56.98it/s +epoch=73 global_step=28650 loss=1.70255 loss_avg=2.17064 acc=0.85156 acc_top1_avg=0.80564 acc_top5_avg=0.96955 lr=0.00100 gn=25.99443 time=60.73it/s +epoch=73 global_step=28700 loss=2.59116 loss_avg=2.15996 acc=0.75000 acc_top1_avg=0.80623 acc_top5_avg=0.96950 lr=0.00100 gn=24.04614 time=57.54it/s +epoch=73 global_step=28750 loss=1.91971 loss_avg=2.16974 acc=0.83594 acc_top1_avg=0.80491 acc_top5_avg=0.96909 lr=0.00100 gn=21.13759 time=55.75it/s +epoch=73 global_step=28800 loss=2.46048 loss_avg=2.17152 acc=0.76562 acc_top1_avg=0.80438 acc_top5_avg=0.96927 lr=0.00100 gn=21.82733 time=60.56it/s +epoch=73 global_step=28850 loss=2.57419 loss_avg=2.17444 acc=0.75000 acc_top1_avg=0.80415 acc_top5_avg=0.96852 lr=0.00100 gn=26.09835 time=51.09it/s +epoch=73 global_step=28900 loss=2.40075 loss_avg=2.19007 acc=0.78125 acc_top1_avg=0.80224 acc_top5_avg=0.96822 lr=0.00100 gn=21.73959 time=62.59it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.37171 test_loss_avg=0.58160 acc=0.89062 test_acc_avg=0.82292 test_acc_top5_avg=0.98887 time=236.49it/s +epoch=73 global_step=28934 loss=0.22482 test_loss_avg=0.41547 acc=0.93750 test_acc_avg=0.87638 test_acc_top5_avg=0.99209 time=552.10it/s +curr_acc 0.8764 +BEST_ACC 0.8939 +curr_acc_top5 0.9921 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=2.16816 loss_avg=2.04989 acc=0.80469 acc_top1_avg=0.81836 acc_top5_avg=0.96777 lr=0.00100 gn=20.78172 time=52.68it/s +epoch=74 global_step=29000 loss=2.06271 loss_avg=2.13401 acc=0.81250 acc_top1_avg=0.80990 acc_top5_avg=0.96875 lr=0.00100 gn=24.31018 time=44.68it/s +epoch=74 global_step=29050 loss=2.23015 loss_avg=2.12741 acc=0.80469 acc_top1_avg=0.80954 acc_top5_avg=0.96868 lr=0.00100 gn=27.60891 time=54.81it/s +epoch=74 global_step=29100 loss=2.29614 loss_avg=2.15590 acc=0.79688 acc_top1_avg=0.80648 acc_top5_avg=0.96908 lr=0.00100 gn=24.13243 time=58.43it/s +epoch=74 global_step=29150 loss=1.56503 loss_avg=2.16395 acc=0.86719 acc_top1_avg=0.80548 acc_top5_avg=0.96893 lr=0.00100 gn=20.34304 time=51.41it/s +epoch=74 global_step=29200 loss=1.91300 loss_avg=2.17131 acc=0.82812 acc_top1_avg=0.80486 acc_top5_avg=0.96904 lr=0.00100 gn=25.43960 time=60.65it/s +epoch=74 global_step=29250 loss=2.30550 loss_avg=2.17713 acc=0.78125 acc_top1_avg=0.80417 acc_top5_avg=0.96897 lr=0.00100 gn=23.06527 time=49.64it/s +epoch=74 global_step=29300 loss=2.00505 loss_avg=2.18308 acc=0.81250 acc_top1_avg=0.80366 acc_top5_avg=0.96807 lr=0.00100 gn=26.50016 time=56.65it/s +====================Eval==================== +epoch=74 global_step=29325 loss=1.48397 test_loss_avg=1.51920 acc=0.54688 test_acc_avg=0.54297 test_acc_top5_avg=0.94336 time=236.45it/s +epoch=74 global_step=29325 loss=0.22805 test_loss_avg=0.61568 acc=0.96094 test_acc_avg=0.81351 test_acc_top5_avg=0.98669 time=241.97it/s +epoch=74 global_step=29325 loss=0.23271 test_loss_avg=0.49740 acc=0.93750 test_acc_avg=0.85107 test_acc_top5_avg=0.98952 time=537.52it/s +curr_acc 0.8511 +BEST_ACC 0.8939 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=2.21034 loss_avg=2.09839 acc=0.78906 acc_top1_avg=0.81125 acc_top5_avg=0.97062 lr=0.00100 gn=20.02495 time=54.34it/s +epoch=75 global_step=29400 loss=2.25049 loss_avg=2.10480 acc=0.80469 acc_top1_avg=0.81104 acc_top5_avg=0.97115 lr=0.00100 gn=24.99923 time=58.14it/s +epoch=75 global_step=29450 loss=2.04858 loss_avg=2.12284 acc=0.81250 acc_top1_avg=0.80875 acc_top5_avg=0.97137 lr=0.00100 gn=19.00441 time=59.28it/s +epoch=75 global_step=29500 loss=2.50621 loss_avg=2.16884 acc=0.77344 acc_top1_avg=0.80433 acc_top5_avg=0.97022 lr=0.00100 gn=30.64140 time=55.08it/s +epoch=75 global_step=29550 loss=2.32510 loss_avg=2.16742 acc=0.79688 acc_top1_avg=0.80510 acc_top5_avg=0.96899 lr=0.00100 gn=31.30922 time=50.60it/s +epoch=75 global_step=29600 loss=2.49835 loss_avg=2.16945 acc=0.75781 acc_top1_avg=0.80511 acc_top5_avg=0.96929 lr=0.00100 gn=26.65407 time=56.81it/s +epoch=75 global_step=29650 loss=2.15450 loss_avg=2.17010 acc=0.79688 acc_top1_avg=0.80476 acc_top5_avg=0.96899 lr=0.00100 gn=26.77353 time=54.71it/s +epoch=75 global_step=29700 loss=1.92506 loss_avg=2.17178 acc=0.84375 acc_top1_avg=0.80442 acc_top5_avg=0.96890 lr=0.00100 gn=26.60708 time=60.09it/s +====================Eval==================== +epoch=75 global_step=29716 loss=0.54358 test_loss_avg=0.64365 acc=0.85156 test_acc_avg=0.79906 test_acc_top5_avg=0.98813 time=237.77it/s +epoch=75 global_step=29716 loss=0.24899 test_loss_avg=0.44972 acc=0.92969 test_acc_avg=0.86490 test_acc_top5_avg=0.99146 time=252.18it/s +epoch=75 global_step=29716 loss=0.36643 test_loss_avg=0.44162 acc=0.93750 test_acc_avg=0.86788 test_acc_top5_avg=0.99169 time=794.23it/s +curr_acc 0.8679 +BEST_ACC 0.8939 +curr_acc_top5 0.9917 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=2.55128 loss_avg=2.03319 acc=0.76562 acc_top1_avg=0.81893 acc_top5_avg=0.97013 lr=0.00100 gn=26.05745 time=54.58it/s +epoch=76 global_step=29800 loss=2.36448 loss_avg=2.09198 acc=0.78906 acc_top1_avg=0.81185 acc_top5_avg=0.96847 lr=0.00100 gn=29.03349 time=54.44it/s +epoch=76 global_step=29850 loss=2.66164 loss_avg=2.16412 acc=0.74219 acc_top1_avg=0.80521 acc_top5_avg=0.96735 lr=0.00100 gn=19.56737 time=54.38it/s +epoch=76 global_step=29900 loss=1.93452 loss_avg=2.17406 acc=0.82812 acc_top1_avg=0.80380 acc_top5_avg=0.96654 lr=0.00100 gn=17.71338 time=56.07it/s +epoch=76 global_step=29950 loss=2.36482 loss_avg=2.17652 acc=0.77344 acc_top1_avg=0.80352 acc_top5_avg=0.96698 lr=0.00100 gn=28.40973 time=61.62it/s +epoch=76 global_step=30000 loss=2.42579 loss_avg=2.16681 acc=0.77344 acc_top1_avg=0.80472 acc_top5_avg=0.96768 lr=0.00100 gn=25.76225 time=63.29it/s +epoch=76 global_step=30050 loss=2.15159 loss_avg=2.16405 acc=0.78906 acc_top1_avg=0.80516 acc_top5_avg=0.96732 lr=0.00100 gn=21.75558 time=59.91it/s +epoch=76 global_step=30100 loss=1.82156 loss_avg=2.16817 acc=0.83594 acc_top1_avg=0.80475 acc_top5_avg=0.96729 lr=0.00100 gn=26.20569 time=61.30it/s +====================Eval==================== +epoch=76 global_step=30107 loss=0.81761 test_loss_avg=0.65986 acc=0.79688 test_acc_avg=0.80418 test_acc_top5_avg=0.98590 time=244.34it/s +epoch=76 global_step=30107 loss=0.27074 test_loss_avg=0.47096 acc=0.93750 test_acc_avg=0.86145 test_acc_top5_avg=0.99031 time=460.00it/s +curr_acc 0.8615 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=2.66735 loss_avg=2.12262 acc=0.75000 acc_top1_avg=0.81086 acc_top5_avg=0.97057 lr=0.00100 gn=24.83743 time=53.71it/s +epoch=77 global_step=30200 loss=2.56311 loss_avg=2.13662 acc=0.75781 acc_top1_avg=0.80906 acc_top5_avg=0.96993 lr=0.00100 gn=24.99121 time=58.22it/s +epoch=77 global_step=30250 loss=2.40808 loss_avg=2.12947 acc=0.78125 acc_top1_avg=0.80878 acc_top5_avg=0.97023 lr=0.00100 gn=26.05617 time=55.17it/s +epoch=77 global_step=30300 loss=2.11961 loss_avg=2.14697 acc=0.80469 acc_top1_avg=0.80752 acc_top5_avg=0.96915 lr=0.00100 gn=23.37556 time=48.84it/s +epoch=77 global_step=30350 loss=2.16869 loss_avg=2.14618 acc=0.81250 acc_top1_avg=0.80787 acc_top5_avg=0.96965 lr=0.00100 gn=28.03385 time=59.58it/s +epoch=77 global_step=30400 loss=1.89020 loss_avg=2.13098 acc=0.84375 acc_top1_avg=0.80962 acc_top5_avg=0.96966 lr=0.00100 gn=26.71427 time=54.81it/s +epoch=77 global_step=30450 loss=1.54317 loss_avg=2.13073 acc=0.86719 acc_top1_avg=0.80970 acc_top5_avg=0.96923 lr=0.00100 gn=21.26287 time=58.65it/s +====================Eval==================== +epoch=77 global_step=30498 loss=0.76382 test_loss_avg=0.47979 acc=0.70312 test_acc_avg=0.83686 test_acc_top5_avg=0.99449 time=240.13it/s +epoch=77 global_step=30498 loss=0.21226 test_loss_avg=0.49646 acc=0.96875 test_acc_avg=0.85238 test_acc_top5_avg=0.99044 time=70.31it/s +epoch=77 global_step=30498 loss=0.09923 test_loss_avg=0.46103 acc=0.93750 test_acc_avg=0.86353 test_acc_top5_avg=0.99130 time=400.99it/s +curr_acc 0.8635 +BEST_ACC 0.8939 +curr_acc_top5 0.9913 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=2.32185 loss_avg=2.18116 acc=0.78125 acc_top1_avg=0.79297 acc_top5_avg=0.97656 lr=0.00100 gn=23.15495 time=49.76it/s +epoch=78 global_step=30550 loss=2.19571 loss_avg=2.14764 acc=0.82812 acc_top1_avg=0.80709 acc_top5_avg=0.96484 lr=0.00100 gn=33.40788 time=60.41it/s +epoch=78 global_step=30600 loss=2.68810 loss_avg=2.15919 acc=0.75000 acc_top1_avg=0.80446 acc_top5_avg=0.96722 lr=0.00100 gn=32.31716 time=62.40it/s +epoch=78 global_step=30650 loss=2.47715 loss_avg=2.15784 acc=0.78125 acc_top1_avg=0.80556 acc_top5_avg=0.96757 lr=0.00100 gn=28.10034 time=48.71it/s +epoch=78 global_step=30700 loss=2.06484 loss_avg=2.15287 acc=0.82812 acc_top1_avg=0.80670 acc_top5_avg=0.96813 lr=0.00100 gn=25.53386 time=60.37it/s +epoch=78 global_step=30750 loss=2.00731 loss_avg=2.16279 acc=0.82031 acc_top1_avg=0.80580 acc_top5_avg=0.96782 lr=0.00100 gn=28.08232 time=62.96it/s +epoch=78 global_step=30800 loss=2.07439 loss_avg=2.15362 acc=0.81250 acc_top1_avg=0.80720 acc_top5_avg=0.96847 lr=0.00100 gn=33.27250 time=52.71it/s +epoch=78 global_step=30850 loss=2.04015 loss_avg=2.14814 acc=0.82031 acc_top1_avg=0.80759 acc_top5_avg=0.96857 lr=0.00100 gn=31.61181 time=62.63it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.24170 test_loss_avg=0.65637 acc=0.93750 test_acc_avg=0.80016 test_acc_top5_avg=0.98294 time=252.65it/s +epoch=78 global_step=30889 loss=0.34271 test_loss_avg=0.47576 acc=0.93750 test_acc_avg=0.85819 test_acc_top5_avg=0.98813 time=675.96it/s +curr_acc 0.8582 +BEST_ACC 0.8939 +curr_acc_top5 0.9881 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=2.12210 loss_avg=2.19135 acc=0.80469 acc_top1_avg=0.80540 acc_top5_avg=0.96804 lr=0.00100 gn=28.87260 time=48.55it/s +epoch=79 global_step=30950 loss=2.19441 loss_avg=2.16406 acc=0.81250 acc_top1_avg=0.80802 acc_top5_avg=0.96811 lr=0.00100 gn=29.59374 time=55.51it/s +epoch=79 global_step=31000 loss=1.74771 loss_avg=2.12105 acc=0.85938 acc_top1_avg=0.81229 acc_top5_avg=0.96952 lr=0.00100 gn=27.64367 time=54.96it/s +epoch=79 global_step=31050 loss=1.61501 loss_avg=2.12915 acc=0.86719 acc_top1_avg=0.81104 acc_top5_avg=0.96817 lr=0.00100 gn=30.02319 time=52.75it/s +epoch=79 global_step=31100 loss=2.24696 loss_avg=2.12447 acc=0.80469 acc_top1_avg=0.81120 acc_top5_avg=0.96864 lr=0.00100 gn=25.32550 time=52.83it/s +epoch=79 global_step=31150 loss=1.91518 loss_avg=2.12767 acc=0.84375 acc_top1_avg=0.81094 acc_top5_avg=0.96836 lr=0.00100 gn=25.34506 time=63.72it/s +epoch=79 global_step=31200 loss=2.57771 loss_avg=2.11888 acc=0.75781 acc_top1_avg=0.81210 acc_top5_avg=0.96840 lr=0.00100 gn=29.20307 time=56.13it/s +epoch=79 global_step=31250 loss=2.85693 loss_avg=2.12544 acc=0.71875 acc_top1_avg=0.81137 acc_top5_avg=0.96869 lr=0.00100 gn=30.78218 time=55.00it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.09406 test_loss_avg=1.02176 acc=0.96875 test_acc_avg=0.67969 test_acc_top5_avg=0.96701 time=240.40it/s +epoch=79 global_step=31280 loss=0.39058 test_loss_avg=0.60887 acc=0.89062 test_acc_avg=0.81713 test_acc_top5_avg=0.98438 time=227.42it/s +epoch=79 global_step=31280 loss=0.32011 test_loss_avg=0.52729 acc=0.93750 test_acc_avg=0.84365 test_acc_top5_avg=0.98705 time=512.00it/s +curr_acc 0.8437 +BEST_ACC 0.8939 +curr_acc_top5 0.9870 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=2.01978 loss_avg=2.04189 acc=0.83594 acc_top1_avg=0.82031 acc_top5_avg=0.96914 lr=0.00010 gn=31.31728 time=50.77it/s +epoch=80 global_step=31350 loss=2.01087 loss_avg=1.98890 acc=0.80469 acc_top1_avg=0.82645 acc_top5_avg=0.97143 lr=0.00010 gn=25.78861 time=55.01it/s +epoch=80 global_step=31400 loss=2.38273 loss_avg=2.00308 acc=0.78125 acc_top1_avg=0.82357 acc_top5_avg=0.97012 lr=0.00010 gn=28.77959 time=52.17it/s +epoch=80 global_step=31450 loss=2.35827 loss_avg=2.01643 acc=0.78125 acc_top1_avg=0.82155 acc_top5_avg=0.96930 lr=0.00010 gn=28.71056 time=59.31it/s +epoch=80 global_step=31500 loss=2.37661 loss_avg=2.01141 acc=0.77344 acc_top1_avg=0.82269 acc_top5_avg=0.96918 lr=0.00010 gn=28.88910 time=57.23it/s +epoch=80 global_step=31550 loss=2.23312 loss_avg=2.00783 acc=0.80469 acc_top1_avg=0.82286 acc_top5_avg=0.96962 lr=0.00010 gn=22.39578 time=60.41it/s +epoch=80 global_step=31600 loss=2.92638 loss_avg=2.00975 acc=0.72656 acc_top1_avg=0.82253 acc_top5_avg=0.96970 lr=0.00010 gn=31.32620 time=63.25it/s +epoch=80 global_step=31650 loss=2.20873 loss_avg=2.01911 acc=0.82031 acc_top1_avg=0.82160 acc_top5_avg=0.96930 lr=0.00010 gn=24.83168 time=60.33it/s +====================Eval==================== +epoch=80 global_step=31671 loss=0.51210 test_loss_avg=0.64561 acc=0.82031 test_acc_avg=0.79583 test_acc_top5_avg=0.98490 time=226.38it/s +epoch=80 global_step=31671 loss=0.15965 test_loss_avg=0.43454 acc=0.93750 test_acc_avg=0.86748 test_acc_top5_avg=0.99061 time=869.11it/s +curr_acc 0.8675 +BEST_ACC 0.8939 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=2.03059 loss_avg=1.94366 acc=0.81250 acc_top1_avg=0.82732 acc_top5_avg=0.97548 lr=0.00010 gn=21.03091 time=50.65it/s +epoch=81 global_step=31750 loss=1.87576 loss_avg=1.92567 acc=0.82812 acc_top1_avg=0.83040 acc_top5_avg=0.97201 lr=0.00010 gn=27.20583 time=29.62it/s +epoch=81 global_step=31800 loss=1.36680 loss_avg=1.93779 acc=0.88281 acc_top1_avg=0.82867 acc_top5_avg=0.97069 lr=0.00010 gn=25.17736 time=57.28it/s +epoch=81 global_step=31850 loss=2.56498 loss_avg=1.97194 acc=0.78125 acc_top1_avg=0.82559 acc_top5_avg=0.96919 lr=0.00010 gn=29.04195 time=63.99it/s +epoch=81 global_step=31900 loss=2.20834 loss_avg=1.97294 acc=0.80469 acc_top1_avg=0.82543 acc_top5_avg=0.96916 lr=0.00010 gn=26.21606 time=49.83it/s +epoch=81 global_step=31950 loss=1.67521 loss_avg=1.96358 acc=0.86719 acc_top1_avg=0.82619 acc_top5_avg=0.96923 lr=0.00010 gn=25.51852 time=58.15it/s +epoch=81 global_step=32000 loss=1.91292 loss_avg=1.96892 acc=0.82812 acc_top1_avg=0.82549 acc_top5_avg=0.96913 lr=0.00010 gn=25.66265 time=50.26it/s +epoch=81 global_step=32050 loss=2.42107 loss_avg=1.96962 acc=0.78125 acc_top1_avg=0.82540 acc_top5_avg=0.96922 lr=0.00010 gn=32.91223 time=58.91it/s +====================Eval==================== +epoch=81 global_step=32062 loss=1.00150 test_loss_avg=1.00150 acc=0.71875 test_acc_avg=0.71875 test_acc_top5_avg=0.97656 time=203.78it/s +epoch=81 global_step=32062 loss=0.08732 test_loss_avg=0.57186 acc=0.98438 test_acc_avg=0.82215 test_acc_top5_avg=0.98729 time=236.55it/s +epoch=81 global_step=32062 loss=0.14887 test_loss_avg=0.44537 acc=0.93750 test_acc_avg=0.86303 test_acc_top5_avg=0.99070 time=862.49it/s +curr_acc 0.8630 +BEST_ACC 0.8939 +curr_acc_top5 0.9907 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=2.72732 loss_avg=1.86485 acc=0.75000 acc_top1_avg=0.83388 acc_top5_avg=0.96937 lr=0.00010 gn=27.98121 time=52.70it/s +epoch=82 global_step=32150 loss=2.16707 loss_avg=1.88157 acc=0.79688 acc_top1_avg=0.83256 acc_top5_avg=0.96857 lr=0.00010 gn=30.85975 time=57.55it/s +epoch=82 global_step=32200 loss=1.79925 loss_avg=1.90736 acc=0.85156 acc_top1_avg=0.83101 acc_top5_avg=0.96818 lr=0.00010 gn=22.52103 time=63.61it/s +epoch=82 global_step=32250 loss=1.62996 loss_avg=1.91310 acc=0.87500 acc_top1_avg=0.83091 acc_top5_avg=0.96917 lr=0.00010 gn=30.88671 time=55.51it/s +epoch=82 global_step=32300 loss=2.26888 loss_avg=1.92548 acc=0.80469 acc_top1_avg=0.82960 acc_top5_avg=0.96931 lr=0.00010 gn=25.66772 time=54.00it/s +epoch=82 global_step=32350 loss=2.04658 loss_avg=1.93525 acc=0.82031 acc_top1_avg=0.82821 acc_top5_avg=0.96921 lr=0.00010 gn=27.52287 time=51.08it/s +epoch=82 global_step=32400 loss=2.10115 loss_avg=1.94592 acc=0.79688 acc_top1_avg=0.82736 acc_top5_avg=0.96893 lr=0.00010 gn=29.78573 time=53.08it/s +epoch=82 global_step=32450 loss=1.80703 loss_avg=1.94562 acc=0.83594 acc_top1_avg=0.82730 acc_top5_avg=0.96903 lr=0.00010 gn=22.63888 time=63.48it/s +====================Eval==================== +epoch=82 global_step=32453 loss=1.09534 test_loss_avg=0.70277 acc=0.64062 test_acc_avg=0.77486 test_acc_top5_avg=0.98509 time=236.34it/s +epoch=82 global_step=32453 loss=0.28496 test_loss_avg=0.46389 acc=0.91406 test_acc_avg=0.85666 test_acc_top5_avg=0.98991 time=255.72it/s +epoch=82 global_step=32453 loss=0.16898 test_loss_avg=0.44282 acc=0.93750 test_acc_avg=0.86333 test_acc_top5_avg=0.99041 time=862.67it/s +curr_acc 0.8633 +BEST_ACC 0.8939 +curr_acc_top5 0.9904 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=2.08100 loss_avg=1.96548 acc=0.81250 acc_top1_avg=0.82563 acc_top5_avg=0.96576 lr=0.00010 gn=28.34740 time=62.71it/s +epoch=83 global_step=32550 loss=1.88864 loss_avg=1.93360 acc=0.83594 acc_top1_avg=0.82917 acc_top5_avg=0.96682 lr=0.00010 gn=21.84082 time=55.65it/s +epoch=83 global_step=32600 loss=2.01189 loss_avg=1.95064 acc=0.82812 acc_top1_avg=0.82706 acc_top5_avg=0.96763 lr=0.00010 gn=32.20847 time=60.68it/s +epoch=83 global_step=32650 loss=2.15115 loss_avg=1.96031 acc=0.80469 acc_top1_avg=0.82551 acc_top5_avg=0.96780 lr=0.00010 gn=28.03568 time=49.46it/s +epoch=83 global_step=32700 loss=1.49468 loss_avg=1.95296 acc=0.88281 acc_top1_avg=0.82620 acc_top5_avg=0.96809 lr=0.00010 gn=24.23602 time=51.82it/s +epoch=83 global_step=32750 loss=2.18331 loss_avg=1.95218 acc=0.79688 acc_top1_avg=0.82652 acc_top5_avg=0.96891 lr=0.00010 gn=34.50673 time=57.94it/s +epoch=83 global_step=32800 loss=1.80453 loss_avg=1.94855 acc=0.84375 acc_top1_avg=0.82680 acc_top5_avg=0.96880 lr=0.00010 gn=25.07632 time=45.17it/s +====================Eval==================== +epoch=83 global_step=32844 loss=0.71927 test_loss_avg=0.57533 acc=0.78125 test_acc_avg=0.81668 test_acc_top5_avg=0.98728 time=233.41it/s +epoch=83 global_step=32844 loss=0.16996 test_loss_avg=0.43283 acc=0.93750 test_acc_avg=0.86650 test_acc_top5_avg=0.99100 time=866.05it/s +curr_acc 0.8665 +BEST_ACC 0.8939 +curr_acc_top5 0.9910 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=2.22409 loss_avg=1.93239 acc=0.79688 acc_top1_avg=0.82812 acc_top5_avg=0.96354 lr=0.00010 gn=18.42125 time=51.00it/s +epoch=84 global_step=32900 loss=1.35331 loss_avg=1.87496 acc=0.89844 acc_top1_avg=0.83426 acc_top5_avg=0.96917 lr=0.00010 gn=28.20304 time=63.67it/s +epoch=84 global_step=32950 loss=1.67761 loss_avg=1.87986 acc=0.85156 acc_top1_avg=0.83432 acc_top5_avg=0.96794 lr=0.00010 gn=19.87553 time=55.77it/s +epoch=84 global_step=33000 loss=1.61132 loss_avg=1.88816 acc=0.86719 acc_top1_avg=0.83343 acc_top5_avg=0.96935 lr=0.00010 gn=30.69676 time=58.42it/s +epoch=84 global_step=33050 loss=1.78629 loss_avg=1.89153 acc=0.84375 acc_top1_avg=0.83309 acc_top5_avg=0.96902 lr=0.00010 gn=23.38884 time=61.50it/s +epoch=84 global_step=33100 loss=1.30969 loss_avg=1.92911 acc=0.90625 acc_top1_avg=0.82965 acc_top5_avg=0.96851 lr=0.00010 gn=21.26154 time=53.10it/s +epoch=84 global_step=33150 loss=1.75869 loss_avg=1.92723 acc=0.83594 acc_top1_avg=0.82955 acc_top5_avg=0.96847 lr=0.00010 gn=19.88593 time=34.56it/s +epoch=84 global_step=33200 loss=2.09462 loss_avg=1.92553 acc=0.82812 acc_top1_avg=0.82960 acc_top5_avg=0.96915 lr=0.00010 gn=33.47440 time=55.28it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.13638 test_loss_avg=0.61720 acc=0.96094 test_acc_avg=0.79743 test_acc_top5_avg=0.98996 time=218.69it/s +epoch=84 global_step=33235 loss=0.28576 test_loss_avg=0.51222 acc=0.92969 test_acc_avg=0.84131 test_acc_top5_avg=0.98926 time=240.24it/s +epoch=84 global_step=33235 loss=0.18503 test_loss_avg=0.45780 acc=0.93750 test_acc_avg=0.85967 test_acc_top5_avg=0.99061 time=508.28it/s +curr_acc 0.8597 +BEST_ACC 0.8939 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=1.98125 loss_avg=1.93047 acc=0.82812 acc_top1_avg=0.82708 acc_top5_avg=0.96875 lr=0.00010 gn=23.90033 time=55.75it/s +epoch=85 global_step=33300 loss=1.93665 loss_avg=1.94046 acc=0.81250 acc_top1_avg=0.82740 acc_top5_avg=0.96791 lr=0.00010 gn=24.18067 time=59.09it/s +epoch=85 global_step=33350 loss=1.73344 loss_avg=1.93833 acc=0.85156 acc_top1_avg=0.82792 acc_top5_avg=0.96793 lr=0.00010 gn=20.96830 time=57.87it/s +epoch=85 global_step=33400 loss=1.68905 loss_avg=1.92167 acc=0.85156 acc_top1_avg=0.82888 acc_top5_avg=0.96889 lr=0.00010 gn=30.39102 time=53.49it/s +epoch=85 global_step=33450 loss=1.96922 loss_avg=1.93500 acc=0.83594 acc_top1_avg=0.82751 acc_top5_avg=0.96940 lr=0.00010 gn=26.37496 time=54.26it/s +epoch=85 global_step=33500 loss=2.33496 loss_avg=1.94535 acc=0.77344 acc_top1_avg=0.82630 acc_top5_avg=0.96904 lr=0.00010 gn=26.05689 time=50.70it/s +epoch=85 global_step=33550 loss=1.69693 loss_avg=1.93429 acc=0.85938 acc_top1_avg=0.82755 acc_top5_avg=0.96915 lr=0.00010 gn=31.46946 time=42.83it/s +epoch=85 global_step=33600 loss=1.34206 loss_avg=1.91679 acc=0.88281 acc_top1_avg=0.82928 acc_top5_avg=0.96933 lr=0.00010 gn=17.45465 time=51.17it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.15897 test_loss_avg=0.60217 acc=0.93750 test_acc_avg=0.80625 test_acc_top5_avg=0.98750 time=239.50it/s +epoch=85 global_step=33626 loss=0.19795 test_loss_avg=0.43848 acc=0.93750 test_acc_avg=0.86481 test_acc_top5_avg=0.99110 time=545.57it/s +curr_acc 0.8648 +BEST_ACC 0.8939 +curr_acc_top5 0.9911 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=1.55094 loss_avg=1.90180 acc=0.89062 acc_top1_avg=0.83333 acc_top5_avg=0.97070 lr=0.00010 gn=27.28349 time=53.89it/s +epoch=86 global_step=33700 loss=2.18657 loss_avg=1.92266 acc=0.78906 acc_top1_avg=0.82981 acc_top5_avg=0.96970 lr=0.00010 gn=26.99964 time=55.02it/s +epoch=86 global_step=33750 loss=1.90331 loss_avg=1.87803 acc=0.82812 acc_top1_avg=0.83461 acc_top5_avg=0.96995 lr=0.00010 gn=28.55881 time=57.64it/s +epoch=86 global_step=33800 loss=1.72713 loss_avg=1.89594 acc=0.84375 acc_top1_avg=0.83239 acc_top5_avg=0.96929 lr=0.00010 gn=15.18731 time=48.00it/s +epoch=86 global_step=33850 loss=2.12459 loss_avg=1.89210 acc=0.79688 acc_top1_avg=0.83255 acc_top5_avg=0.96913 lr=0.00010 gn=26.36559 time=55.20it/s +epoch=86 global_step=33900 loss=1.83319 loss_avg=1.89322 acc=0.82812 acc_top1_avg=0.83246 acc_top5_avg=0.96906 lr=0.00010 gn=29.57989 time=54.61it/s +epoch=86 global_step=33950 loss=2.63655 loss_avg=1.90548 acc=0.75781 acc_top1_avg=0.83092 acc_top5_avg=0.96921 lr=0.00010 gn=33.69508 time=55.74it/s +epoch=86 global_step=34000 loss=1.74146 loss_avg=1.90776 acc=0.85156 acc_top1_avg=0.83072 acc_top5_avg=0.96954 lr=0.00010 gn=29.80586 time=46.34it/s +====================Eval==================== +epoch=86 global_step=34017 loss=0.92163 test_loss_avg=0.93851 acc=0.71094 test_acc_avg=0.70312 test_acc_top5_avg=0.98307 time=175.27it/s +epoch=86 global_step=34017 loss=0.28538 test_loss_avg=0.51818 acc=0.92188 test_acc_avg=0.84235 test_acc_top5_avg=0.98940 time=236.29it/s +epoch=86 global_step=34017 loss=0.16040 test_loss_avg=0.43377 acc=0.93750 test_acc_avg=0.86986 test_acc_top5_avg=0.99140 time=859.14it/s +curr_acc 0.8699 +BEST_ACC 0.8939 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=1.29008 loss_avg=1.88587 acc=0.89844 acc_top1_avg=0.83073 acc_top5_avg=0.96851 lr=0.00010 gn=24.17952 time=53.47it/s +epoch=87 global_step=34100 loss=1.46885 loss_avg=1.87855 acc=0.87500 acc_top1_avg=0.83227 acc_top5_avg=0.96856 lr=0.00010 gn=24.18553 time=59.93it/s +epoch=87 global_step=34150 loss=1.82834 loss_avg=1.88032 acc=0.83594 acc_top1_avg=0.83371 acc_top5_avg=0.96893 lr=0.00010 gn=30.78953 time=57.89it/s +epoch=87 global_step=34200 loss=1.83925 loss_avg=1.90489 acc=0.84375 acc_top1_avg=0.83081 acc_top5_avg=0.96790 lr=0.00010 gn=27.34816 time=55.22it/s +epoch=87 global_step=34250 loss=1.68163 loss_avg=1.89715 acc=0.85156 acc_top1_avg=0.83138 acc_top5_avg=0.96905 lr=0.00010 gn=23.30180 time=57.89it/s +epoch=87 global_step=34300 loss=1.88656 loss_avg=1.90725 acc=0.83594 acc_top1_avg=0.83025 acc_top5_avg=0.96941 lr=0.00010 gn=29.15368 time=44.23it/s +epoch=87 global_step=34350 loss=1.94072 loss_avg=1.90620 acc=0.82031 acc_top1_avg=0.83026 acc_top5_avg=0.96901 lr=0.00010 gn=27.30187 time=61.93it/s +epoch=87 global_step=34400 loss=2.34459 loss_avg=1.90249 acc=0.77344 acc_top1_avg=0.83096 acc_top5_avg=0.96891 lr=0.00010 gn=26.17175 time=60.52it/s +====================Eval==================== +epoch=87 global_step=34408 loss=0.56056 test_loss_avg=0.70017 acc=0.82812 test_acc_avg=0.77778 test_acc_top5_avg=0.98669 time=238.34it/s +epoch=87 global_step=34408 loss=0.34376 test_loss_avg=0.45605 acc=0.89062 test_acc_avg=0.86313 test_acc_top5_avg=0.99117 time=244.61it/s +epoch=87 global_step=34408 loss=0.20683 test_loss_avg=0.45013 acc=0.93750 test_acc_avg=0.86472 test_acc_top5_avg=0.99140 time=506.93it/s +curr_acc 0.8647 +BEST_ACC 0.8939 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=2.54922 loss_avg=1.92260 acc=0.76562 acc_top1_avg=0.82850 acc_top5_avg=0.97042 lr=0.00010 gn=27.40329 time=58.63it/s +epoch=88 global_step=34500 loss=1.91567 loss_avg=1.89525 acc=0.82812 acc_top1_avg=0.83246 acc_top5_avg=0.97002 lr=0.00010 gn=25.27322 time=56.40it/s +epoch=88 global_step=34550 loss=1.70041 loss_avg=1.92253 acc=0.85938 acc_top1_avg=0.82978 acc_top5_avg=0.96892 lr=0.00010 gn=25.00066 time=52.15it/s +epoch=88 global_step=34600 loss=1.93674 loss_avg=1.92588 acc=0.82031 acc_top1_avg=0.82865 acc_top5_avg=0.96932 lr=0.00010 gn=21.77449 time=57.42it/s +epoch=88 global_step=34650 loss=1.79294 loss_avg=1.91830 acc=0.84375 acc_top1_avg=0.82967 acc_top5_avg=0.96978 lr=0.00010 gn=22.35953 time=57.61it/s +epoch=88 global_step=34700 loss=2.03634 loss_avg=1.91100 acc=0.81250 acc_top1_avg=0.83043 acc_top5_avg=0.97025 lr=0.00010 gn=32.13514 time=55.38it/s +epoch=88 global_step=34750 loss=1.83785 loss_avg=1.89810 acc=0.84375 acc_top1_avg=0.83169 acc_top5_avg=0.97044 lr=0.00010 gn=24.93962 time=56.53it/s +====================Eval==================== +epoch=88 global_step=34799 loss=0.29522 test_loss_avg=0.58585 acc=0.92969 test_acc_avg=0.81722 test_acc_top5_avg=0.98584 time=243.04it/s +epoch=88 global_step=34799 loss=0.18871 test_loss_avg=0.44122 acc=0.93750 test_acc_avg=0.86481 test_acc_top5_avg=0.99051 time=833.03it/s +curr_acc 0.8648 +BEST_ACC 0.8939 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=2.09111 loss_avg=2.09111 acc=0.80469 acc_top1_avg=0.80469 acc_top5_avg=0.96094 lr=0.00010 gn=29.54255 time=50.12it/s +epoch=89 global_step=34850 loss=1.20739 loss_avg=1.86348 acc=0.90625 acc_top1_avg=0.83395 acc_top5_avg=0.96998 lr=0.00010 gn=28.07037 time=54.91it/s +epoch=89 global_step=34900 loss=1.53530 loss_avg=1.84924 acc=0.87500 acc_top1_avg=0.83594 acc_top5_avg=0.96906 lr=0.00010 gn=27.93652 time=55.17it/s +epoch=89 global_step=34950 loss=1.94880 loss_avg=1.87347 acc=0.81250 acc_top1_avg=0.83392 acc_top5_avg=0.96844 lr=0.00010 gn=18.92709 time=58.10it/s +epoch=89 global_step=35000 loss=2.56742 loss_avg=1.87214 acc=0.75781 acc_top1_avg=0.83423 acc_top5_avg=0.96887 lr=0.00010 gn=31.81119 time=53.06it/s +epoch=89 global_step=35050 loss=2.30022 loss_avg=1.86742 acc=0.78125 acc_top1_avg=0.83438 acc_top5_avg=0.96919 lr=0.00010 gn=26.33399 time=53.16it/s +epoch=89 global_step=35100 loss=2.08088 loss_avg=1.87456 acc=0.82031 acc_top1_avg=0.83391 acc_top5_avg=0.96981 lr=0.00010 gn=29.43318 time=63.81it/s +epoch=89 global_step=35150 loss=2.03471 loss_avg=1.88230 acc=0.82812 acc_top1_avg=0.83313 acc_top5_avg=0.96957 lr=0.00010 gn=31.73208 time=60.53it/s +====================Eval==================== +epoch=89 global_step=35190 loss=1.10817 test_loss_avg=0.65169 acc=0.67188 test_acc_avg=0.79359 test_acc_top5_avg=0.98766 time=232.80it/s +epoch=89 global_step=35190 loss=0.22222 test_loss_avg=0.49040 acc=0.92969 test_acc_avg=0.85168 test_acc_top5_avg=0.98981 time=245.42it/s +epoch=89 global_step=35190 loss=0.19130 test_loss_avg=0.45847 acc=0.93750 test_acc_avg=0.86185 test_acc_top5_avg=0.99061 time=627.23it/s +curr_acc 0.8618 +BEST_ACC 0.8939 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=1.59465 loss_avg=1.78627 acc=0.85938 acc_top1_avg=0.84531 acc_top5_avg=0.97188 lr=0.00010 gn=18.27338 time=57.18it/s +epoch=90 global_step=35250 loss=2.15355 loss_avg=1.83090 acc=0.79688 acc_top1_avg=0.83971 acc_top5_avg=0.97031 lr=0.00010 gn=26.89856 time=57.84it/s +epoch=90 global_step=35300 loss=1.75119 loss_avg=1.85098 acc=0.84375 acc_top1_avg=0.83679 acc_top5_avg=0.97053 lr=0.00010 gn=25.90435 time=57.35it/s +epoch=90 global_step=35350 loss=2.58598 loss_avg=1.87547 acc=0.75000 acc_top1_avg=0.83428 acc_top5_avg=0.96973 lr=0.00010 gn=23.36751 time=52.05it/s +epoch=90 global_step=35400 loss=1.56469 loss_avg=1.85759 acc=0.85938 acc_top1_avg=0.83646 acc_top5_avg=0.97020 lr=0.00010 gn=28.27079 time=63.37it/s +epoch=90 global_step=35450 loss=1.61763 loss_avg=1.88010 acc=0.85938 acc_top1_avg=0.83395 acc_top5_avg=0.96989 lr=0.00010 gn=26.07631 time=49.83it/s +epoch=90 global_step=35500 loss=2.09584 loss_avg=1.87656 acc=0.80469 acc_top1_avg=0.83420 acc_top5_avg=0.96973 lr=0.00010 gn=22.24360 time=63.09it/s +epoch=90 global_step=35550 loss=2.08344 loss_avg=1.87174 acc=0.80469 acc_top1_avg=0.83453 acc_top5_avg=0.96994 lr=0.00010 gn=28.32127 time=32.05it/s +====================Eval==================== +epoch=90 global_step=35581 loss=0.71298 test_loss_avg=0.63687 acc=0.78906 test_acc_avg=0.79609 test_acc_top5_avg=0.98633 time=125.02it/s +epoch=90 global_step=35581 loss=0.22611 test_loss_avg=0.47556 acc=0.93750 test_acc_avg=0.85295 test_acc_top5_avg=0.99031 time=846.65it/s +curr_acc 0.8529 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=1.60933 loss_avg=1.84882 acc=0.86719 acc_top1_avg=0.83964 acc_top5_avg=0.96628 lr=0.00010 gn=30.09152 time=50.85it/s +epoch=91 global_step=35650 loss=2.03333 loss_avg=1.90791 acc=0.81250 acc_top1_avg=0.82982 acc_top5_avg=0.96807 lr=0.00010 gn=26.52433 time=52.48it/s +epoch=91 global_step=35700 loss=1.16810 loss_avg=1.87453 acc=0.89844 acc_top1_avg=0.83397 acc_top5_avg=0.96987 lr=0.00010 gn=21.64867 time=52.77it/s +epoch=91 global_step=35750 loss=2.08375 loss_avg=1.87858 acc=0.81250 acc_top1_avg=0.83330 acc_top5_avg=0.96954 lr=0.00010 gn=22.68110 time=55.41it/s +epoch=91 global_step=35800 loss=1.40441 loss_avg=1.87439 acc=0.89062 acc_top1_avg=0.83398 acc_top5_avg=0.96950 lr=0.00010 gn=23.54376 time=60.02it/s +epoch=91 global_step=35850 loss=2.19002 loss_avg=1.88012 acc=0.79688 acc_top1_avg=0.83327 acc_top5_avg=0.96921 lr=0.00010 gn=27.36926 time=42.85it/s +epoch=91 global_step=35900 loss=1.99371 loss_avg=1.87045 acc=0.81250 acc_top1_avg=0.83491 acc_top5_avg=0.96919 lr=0.00010 gn=18.68584 time=55.34it/s +epoch=91 global_step=35950 loss=2.04311 loss_avg=1.87214 acc=0.82812 acc_top1_avg=0.83465 acc_top5_avg=0.96871 lr=0.00010 gn=36.64289 time=63.74it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.14424 test_loss_avg=0.71975 acc=0.95312 test_acc_avg=0.77344 test_acc_top5_avg=0.98580 time=243.83it/s +epoch=91 global_step=35972 loss=0.21571 test_loss_avg=0.52062 acc=0.92969 test_acc_avg=0.84042 test_acc_top5_avg=0.98873 time=226.78it/s +epoch=91 global_step=35972 loss=0.17258 test_loss_avg=0.45365 acc=0.93750 test_acc_avg=0.86284 test_acc_top5_avg=0.99031 time=546.77it/s +curr_acc 0.8628 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=1.86637 loss_avg=1.84539 acc=0.83594 acc_top1_avg=0.83845 acc_top5_avg=0.96317 lr=0.00010 gn=25.91262 time=54.11it/s +epoch=92 global_step=36050 loss=2.39411 loss_avg=1.88354 acc=0.78906 acc_top1_avg=0.83494 acc_top5_avg=0.96785 lr=0.00010 gn=35.29578 time=58.52it/s +epoch=92 global_step=36100 loss=1.91053 loss_avg=1.85620 acc=0.82812 acc_top1_avg=0.83722 acc_top5_avg=0.96930 lr=0.00010 gn=23.97855 time=51.68it/s +epoch=92 global_step=36150 loss=2.13979 loss_avg=1.84165 acc=0.78906 acc_top1_avg=0.83822 acc_top5_avg=0.96989 lr=0.00010 gn=28.24217 time=52.40it/s +epoch=92 global_step=36200 loss=2.21263 loss_avg=1.85669 acc=0.78906 acc_top1_avg=0.83662 acc_top5_avg=0.96944 lr=0.00010 gn=25.27631 time=60.15it/s +epoch=92 global_step=36250 loss=1.77333 loss_avg=1.85603 acc=0.84375 acc_top1_avg=0.83639 acc_top5_avg=0.96917 lr=0.00010 gn=28.01126 time=42.12it/s +epoch=92 global_step=36300 loss=1.31640 loss_avg=1.86198 acc=0.89062 acc_top1_avg=0.83565 acc_top5_avg=0.96930 lr=0.00010 gn=22.78286 time=56.22it/s +epoch=92 global_step=36350 loss=1.55787 loss_avg=1.87045 acc=0.88281 acc_top1_avg=0.83472 acc_top5_avg=0.96941 lr=0.00010 gn=32.74762 time=55.23it/s +====================Eval==================== +epoch=92 global_step=36363 loss=0.21573 test_loss_avg=0.64914 acc=0.91406 test_acc_avg=0.79419 test_acc_top5_avg=0.98853 time=91.68it/s +epoch=92 global_step=36363 loss=0.23433 test_loss_avg=0.45365 acc=0.93750 test_acc_avg=0.86195 test_acc_top5_avg=0.99159 time=834.69it/s +curr_acc 0.8619 +BEST_ACC 0.8939 +curr_acc_top5 0.9916 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=1.77412 loss_avg=1.87294 acc=0.84375 acc_top1_avg=0.83699 acc_top5_avg=0.96938 lr=0.00010 gn=21.60355 time=62.72it/s +epoch=93 global_step=36450 loss=1.83399 loss_avg=1.83180 acc=0.82812 acc_top1_avg=0.83908 acc_top5_avg=0.96893 lr=0.00010 gn=24.58826 time=54.74it/s +epoch=93 global_step=36500 loss=1.55334 loss_avg=1.84812 acc=0.85938 acc_top1_avg=0.83731 acc_top5_avg=0.96869 lr=0.00010 gn=15.77821 time=55.38it/s +epoch=93 global_step=36550 loss=1.71706 loss_avg=1.87074 acc=0.85156 acc_top1_avg=0.83477 acc_top5_avg=0.96842 lr=0.00010 gn=20.81381 time=52.98it/s +epoch=93 global_step=36600 loss=1.45485 loss_avg=1.85816 acc=0.87500 acc_top1_avg=0.83637 acc_top5_avg=0.96888 lr=0.00010 gn=28.65471 time=53.33it/s +epoch=93 global_step=36650 loss=1.76502 loss_avg=1.86221 acc=0.83594 acc_top1_avg=0.83558 acc_top5_avg=0.96919 lr=0.00010 gn=34.94869 time=54.99it/s +epoch=93 global_step=36700 loss=1.92161 loss_avg=1.85799 acc=0.84375 acc_top1_avg=0.83598 acc_top5_avg=0.96961 lr=0.00010 gn=33.89653 time=52.90it/s +epoch=93 global_step=36750 loss=1.86224 loss_avg=1.86066 acc=0.84375 acc_top1_avg=0.83568 acc_top5_avg=0.96980 lr=0.00010 gn=30.55197 time=55.19it/s +====================Eval==================== +epoch=93 global_step=36754 loss=1.01286 test_loss_avg=1.07400 acc=0.68750 test_acc_avg=0.67969 test_acc_top5_avg=0.97656 time=245.34it/s +epoch=93 global_step=36754 loss=0.14090 test_loss_avg=0.56232 acc=0.96875 test_acc_avg=0.82989 test_acc_top5_avg=0.98894 time=226.41it/s +epoch=93 global_step=36754 loss=0.19839 test_loss_avg=0.45506 acc=0.93750 test_acc_avg=0.86353 test_acc_top5_avg=0.99150 time=693.16it/s +curr_acc 0.8635 +BEST_ACC 0.8939 +curr_acc_top5 0.9915 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=1.44457 loss_avg=1.83851 acc=0.88281 acc_top1_avg=0.83713 acc_top5_avg=0.97028 lr=0.00010 gn=28.26206 time=54.76it/s +epoch=94 global_step=36850 loss=1.86623 loss_avg=1.86804 acc=0.82812 acc_top1_avg=0.83447 acc_top5_avg=0.96965 lr=0.00010 gn=25.36046 time=58.60it/s +epoch=94 global_step=36900 loss=1.59676 loss_avg=1.85736 acc=0.85156 acc_top1_avg=0.83572 acc_top5_avg=0.97084 lr=0.00010 gn=25.20112 time=56.88it/s +epoch=94 global_step=36950 loss=1.90039 loss_avg=1.85684 acc=0.82031 acc_top1_avg=0.83554 acc_top5_avg=0.97082 lr=0.00010 gn=23.15360 time=61.85it/s +epoch=94 global_step=37000 loss=1.81032 loss_avg=1.85203 acc=0.84375 acc_top1_avg=0.83626 acc_top5_avg=0.96970 lr=0.00010 gn=28.64347 time=52.86it/s +epoch=94 global_step=37050 loss=1.92288 loss_avg=1.84210 acc=0.82812 acc_top1_avg=0.83723 acc_top5_avg=0.96954 lr=0.00010 gn=27.10058 time=55.06it/s +epoch=94 global_step=37100 loss=2.01596 loss_avg=1.85170 acc=0.82031 acc_top1_avg=0.83587 acc_top5_avg=0.96934 lr=0.00010 gn=27.84663 time=55.30it/s +====================Eval==================== +epoch=94 global_step=37145 loss=0.88471 test_loss_avg=0.77699 acc=0.72656 test_acc_avg=0.75358 test_acc_top5_avg=0.98730 time=60.27it/s +epoch=94 global_step=37145 loss=0.14087 test_loss_avg=0.48716 acc=0.95312 test_acc_avg=0.85135 test_acc_top5_avg=0.99103 time=239.67it/s +epoch=94 global_step=37145 loss=0.18591 test_loss_avg=0.46982 acc=0.93750 test_acc_avg=0.85651 test_acc_top5_avg=0.99140 time=504.43it/s +curr_acc 0.8565 +BEST_ACC 0.8939 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=1.85514 loss_avg=2.00671 acc=0.83594 acc_top1_avg=0.81406 acc_top5_avg=0.97500 lr=0.00010 gn=25.45526 time=58.11it/s +epoch=95 global_step=37200 loss=2.00136 loss_avg=1.89231 acc=0.82031 acc_top1_avg=0.83310 acc_top5_avg=0.97017 lr=0.00010 gn=26.59112 time=57.99it/s +epoch=95 global_step=37250 loss=1.87520 loss_avg=1.84503 acc=0.83594 acc_top1_avg=0.83720 acc_top5_avg=0.97068 lr=0.00010 gn=20.01186 time=52.68it/s +epoch=95 global_step=37300 loss=1.59734 loss_avg=1.85834 acc=0.86719 acc_top1_avg=0.83523 acc_top5_avg=0.96971 lr=0.00010 gn=33.58745 time=54.96it/s +epoch=95 global_step=37350 loss=1.70413 loss_avg=1.87080 acc=0.83594 acc_top1_avg=0.83380 acc_top5_avg=0.96982 lr=0.00010 gn=23.52383 time=56.27it/s +epoch=95 global_step=37400 loss=1.95277 loss_avg=1.86911 acc=0.82812 acc_top1_avg=0.83456 acc_top5_avg=0.97010 lr=0.00010 gn=30.35427 time=54.78it/s +epoch=95 global_step=37450 loss=2.06590 loss_avg=1.85675 acc=0.81250 acc_top1_avg=0.83553 acc_top5_avg=0.97008 lr=0.00010 gn=34.30431 time=53.52it/s +epoch=95 global_step=37500 loss=1.95978 loss_avg=1.85811 acc=0.81250 acc_top1_avg=0.83532 acc_top5_avg=0.96994 lr=0.00010 gn=28.42898 time=51.00it/s +====================Eval==================== +epoch=95 global_step=37536 loss=0.60510 test_loss_avg=0.64776 acc=0.84375 test_acc_avg=0.79809 test_acc_top5_avg=0.98628 time=233.21it/s +epoch=95 global_step=37536 loss=0.17434 test_loss_avg=0.46815 acc=0.93750 test_acc_avg=0.85621 test_acc_top5_avg=0.99070 time=587.03it/s +curr_acc 0.8562 +BEST_ACC 0.8939 +curr_acc_top5 0.9907 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=1.56605 loss_avg=1.80446 acc=0.86719 acc_top1_avg=0.84152 acc_top5_avg=0.97154 lr=0.00010 gn=28.33454 time=50.98it/s +epoch=96 global_step=37600 loss=1.96799 loss_avg=1.77522 acc=0.84375 acc_top1_avg=0.84521 acc_top5_avg=0.97144 lr=0.00010 gn=37.00662 time=56.66it/s +epoch=96 global_step=37650 loss=2.04171 loss_avg=1.81833 acc=0.82812 acc_top1_avg=0.84039 acc_top5_avg=0.97183 lr=0.00010 gn=32.60013 time=63.29it/s +epoch=96 global_step=37700 loss=1.60667 loss_avg=1.83567 acc=0.85938 acc_top1_avg=0.83894 acc_top5_avg=0.97032 lr=0.00010 gn=29.91056 time=57.50it/s +epoch=96 global_step=37750 loss=1.51792 loss_avg=1.83891 acc=0.87500 acc_top1_avg=0.83875 acc_top5_avg=0.97025 lr=0.00010 gn=22.07618 time=54.01it/s +epoch=96 global_step=37800 loss=2.10519 loss_avg=1.84286 acc=0.80469 acc_top1_avg=0.83786 acc_top5_avg=0.97020 lr=0.00010 gn=31.03493 time=53.70it/s +epoch=96 global_step=37850 loss=1.59370 loss_avg=1.84212 acc=0.85938 acc_top1_avg=0.83780 acc_top5_avg=0.97032 lr=0.00010 gn=33.66439 time=56.60it/s +epoch=96 global_step=37900 loss=1.49293 loss_avg=1.84539 acc=0.87500 acc_top1_avg=0.83755 acc_top5_avg=0.97019 lr=0.00010 gn=26.13850 time=63.83it/s +====================Eval==================== +epoch=96 global_step=37927 loss=0.68483 test_loss_avg=0.66383 acc=0.83594 test_acc_avg=0.80078 test_acc_top5_avg=0.99072 time=241.40it/s +epoch=96 global_step=37927 loss=0.12054 test_loss_avg=0.54193 acc=0.95312 test_acc_avg=0.83653 test_acc_top5_avg=0.99053 time=221.50it/s +epoch=96 global_step=37927 loss=0.19851 test_loss_avg=0.49059 acc=0.93750 test_acc_avg=0.85275 test_acc_top5_avg=0.99140 time=843.75it/s +curr_acc 0.8527 +BEST_ACC 0.8939 +curr_acc_top5 0.9914 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=2.44911 loss_avg=1.93697 acc=0.77344 acc_top1_avg=0.82643 acc_top5_avg=0.96433 lr=0.00010 gn=24.68591 time=57.95it/s +epoch=97 global_step=38000 loss=1.88156 loss_avg=1.89878 acc=0.85938 acc_top1_avg=0.83080 acc_top5_avg=0.96533 lr=0.00010 gn=32.96417 time=63.26it/s +epoch=97 global_step=38050 loss=1.97549 loss_avg=1.87473 acc=0.82031 acc_top1_avg=0.83371 acc_top5_avg=0.96697 lr=0.00010 gn=25.69871 time=59.25it/s +epoch=97 global_step=38100 loss=1.59171 loss_avg=1.86104 acc=0.88281 acc_top1_avg=0.83512 acc_top5_avg=0.96861 lr=0.00010 gn=33.69482 time=51.67it/s +epoch=97 global_step=38150 loss=1.81830 loss_avg=1.86156 acc=0.83594 acc_top1_avg=0.83499 acc_top5_avg=0.96854 lr=0.00010 gn=21.28433 time=59.41it/s +epoch=97 global_step=38200 loss=1.97433 loss_avg=1.84626 acc=0.82812 acc_top1_avg=0.83662 acc_top5_avg=0.96906 lr=0.00010 gn=24.72186 time=54.32it/s +epoch=97 global_step=38250 loss=1.48828 loss_avg=1.84395 acc=0.87500 acc_top1_avg=0.83671 acc_top5_avg=0.96955 lr=0.00010 gn=27.63786 time=59.56it/s +epoch=97 global_step=38300 loss=1.60858 loss_avg=1.84609 acc=0.86719 acc_top1_avg=0.83625 acc_top5_avg=0.96967 lr=0.00010 gn=31.17551 time=57.52it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.24638 test_loss_avg=0.65287 acc=0.90625 test_acc_avg=0.79413 test_acc_top5_avg=0.98775 time=230.30it/s +epoch=97 global_step=38318 loss=0.18560 test_loss_avg=0.47592 acc=0.93750 test_acc_avg=0.85552 test_acc_top5_avg=0.99090 time=469.95it/s +curr_acc 0.8555 +BEST_ACC 0.8939 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=1.60411 loss_avg=1.81781 acc=0.86719 acc_top1_avg=0.84058 acc_top5_avg=0.96777 lr=0.00010 gn=35.88011 time=55.36it/s +epoch=98 global_step=38400 loss=1.65801 loss_avg=1.81664 acc=0.85156 acc_top1_avg=0.84108 acc_top5_avg=0.96932 lr=0.00010 gn=23.58964 time=53.73it/s +epoch=98 global_step=38450 loss=1.52571 loss_avg=1.80852 acc=0.86719 acc_top1_avg=0.84144 acc_top5_avg=0.96887 lr=0.00010 gn=28.31536 time=58.01it/s +epoch=98 global_step=38500 loss=1.24066 loss_avg=1.82461 acc=0.92188 acc_top1_avg=0.84049 acc_top5_avg=0.96858 lr=0.00010 gn=34.58846 time=54.31it/s +epoch=98 global_step=38550 loss=1.68676 loss_avg=1.82329 acc=0.85938 acc_top1_avg=0.84052 acc_top5_avg=0.96888 lr=0.00010 gn=29.34453 time=62.80it/s +epoch=98 global_step=38600 loss=1.39659 loss_avg=1.83050 acc=0.87500 acc_top1_avg=0.83951 acc_top5_avg=0.96930 lr=0.00010 gn=26.00042 time=53.59it/s +epoch=98 global_step=38650 loss=1.96222 loss_avg=1.84499 acc=0.82812 acc_top1_avg=0.83761 acc_top5_avg=0.96920 lr=0.00010 gn=34.58817 time=63.58it/s +epoch=98 global_step=38700 loss=1.23085 loss_avg=1.84179 acc=0.89844 acc_top1_avg=0.83774 acc_top5_avg=0.96928 lr=0.00010 gn=29.85249 time=60.25it/s +====================Eval==================== +epoch=98 global_step=38709 loss=0.70075 test_loss_avg=1.06996 acc=0.76562 test_acc_avg=0.66113 test_acc_top5_avg=0.98145 time=224.44it/s +epoch=98 global_step=38709 loss=0.22174 test_loss_avg=0.54448 acc=0.92969 test_acc_avg=0.83419 test_acc_top5_avg=0.98895 time=243.04it/s +epoch=98 global_step=38709 loss=0.21559 test_loss_avg=0.46109 acc=0.93750 test_acc_avg=0.86155 test_acc_top5_avg=0.99100 time=468.74it/s +curr_acc 0.8616 +BEST_ACC 0.8939 +curr_acc_top5 0.9910 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=1.99307 loss_avg=1.79531 acc=0.82031 acc_top1_avg=0.83975 acc_top5_avg=0.96913 lr=0.00010 gn=32.74953 time=55.16it/s +epoch=99 global_step=38800 loss=1.29374 loss_avg=1.77016 acc=0.89062 acc_top1_avg=0.84384 acc_top5_avg=0.97227 lr=0.00010 gn=28.44705 time=55.35it/s +epoch=99 global_step=38850 loss=1.61802 loss_avg=1.76620 acc=0.85938 acc_top1_avg=0.84541 acc_top5_avg=0.97169 lr=0.00010 gn=25.68921 time=56.14it/s +epoch=99 global_step=38900 loss=2.37526 loss_avg=1.79839 acc=0.78906 acc_top1_avg=0.84228 acc_top5_avg=0.96990 lr=0.00010 gn=34.18787 time=55.51it/s +epoch=99 global_step=38950 loss=1.88603 loss_avg=1.80957 acc=0.83594 acc_top1_avg=0.84099 acc_top5_avg=0.96950 lr=0.00010 gn=22.59061 time=60.10it/s +epoch=99 global_step=39000 loss=2.43777 loss_avg=1.81961 acc=0.78125 acc_top1_avg=0.83986 acc_top5_avg=0.96939 lr=0.00010 gn=37.29025 time=42.88it/s +epoch=99 global_step=39050 loss=1.63502 loss_avg=1.83037 acc=0.85156 acc_top1_avg=0.83848 acc_top5_avg=0.96976 lr=0.00010 gn=23.33414 time=48.48it/s +epoch=99 global_step=39100 loss=2.02155 loss_avg=1.83436 acc=0.82500 acc_top1_avg=0.83813 acc_top5_avg=0.96955 lr=0.00010 gn=43.02453 time=70.84it/s +====================Eval==================== +epoch=99 global_step=39100 loss=0.42138 test_loss_avg=0.75044 acc=0.87500 test_acc_avg=0.76886 test_acc_top5_avg=0.98518 time=232.76it/s +epoch=99 global_step=39100 loss=0.18925 test_loss_avg=0.47298 acc=0.93750 test_acc_avg=0.85759 test_acc_top5_avg=0.99090 time=809.40it/s +epoch=99 global_step=39100 loss=0.18925 test_loss_avg=0.47298 acc=0.93750 test_acc_avg=0.85759 test_acc_top5_avg=0.99090 time=809.40it/s +curr_acc 0.8576 +BEST_ACC 0.8939 +curr_acc_top5 0.9909 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=1.73717 loss_avg=1.77598 acc=0.85156 acc_top1_avg=0.84391 acc_top5_avg=0.97141 lr=0.00010 gn=29.16556 time=49.23it/s +epoch=100 global_step=39200 loss=1.53314 loss_avg=1.79385 acc=0.86719 acc_top1_avg=0.84211 acc_top5_avg=0.97195 lr=0.00010 gn=32.31316 time=53.92it/s +epoch=100 global_step=39250 loss=2.25607 loss_avg=1.81744 acc=0.79688 acc_top1_avg=0.84026 acc_top5_avg=0.96969 lr=0.00010 gn=35.87362 time=58.16it/s +epoch=100 global_step=39300 loss=1.67581 loss_avg=1.80102 acc=0.85156 acc_top1_avg=0.84168 acc_top5_avg=0.97008 lr=0.00010 gn=27.06551 time=54.61it/s +epoch=100 global_step=39350 loss=1.93697 loss_avg=1.81190 acc=0.83594 acc_top1_avg=0.84066 acc_top5_avg=0.96962 lr=0.00010 gn=27.75775 time=57.34it/s +epoch=100 global_step=39400 loss=2.12074 loss_avg=1.82203 acc=0.81250 acc_top1_avg=0.83961 acc_top5_avg=0.96974 lr=0.00010 gn=38.49059 time=56.55it/s +epoch=100 global_step=39450 loss=2.04703 loss_avg=1.82281 acc=0.80469 acc_top1_avg=0.83987 acc_top5_avg=0.96958 lr=0.00010 gn=21.23155 time=63.50it/s +====================Eval==================== +epoch=100 global_step=39491 loss=0.19781 test_loss_avg=0.63811 acc=0.92188 test_acc_avg=0.80578 test_acc_top5_avg=0.98656 time=242.77it/s +epoch=100 global_step=39491 loss=0.23644 test_loss_avg=0.48316 acc=0.93750 test_acc_avg=0.85423 test_acc_top5_avg=0.99061 time=492.69it/s +curr_acc 0.8542 +BEST_ACC 0.8939 +curr_acc_top5 0.9906 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=1.84358 loss_avg=1.71854 acc=0.84375 acc_top1_avg=0.85156 acc_top5_avg=0.97917 lr=0.00010 gn=35.88909 time=59.57it/s +epoch=101 global_step=39550 loss=2.05759 loss_avg=1.79145 acc=0.79688 acc_top1_avg=0.84296 acc_top5_avg=0.96849 lr=0.00010 gn=29.05360 time=51.74it/s +epoch=101 global_step=39600 loss=1.99264 loss_avg=1.78359 acc=0.82031 acc_top1_avg=0.84339 acc_top5_avg=0.96753 lr=0.00010 gn=26.75696 time=58.82it/s +epoch=101 global_step=39650 loss=1.88371 loss_avg=1.77769 acc=0.82812 acc_top1_avg=0.84350 acc_top5_avg=0.96836 lr=0.00010 gn=22.79444 time=55.41it/s +epoch=101 global_step=39700 loss=1.71682 loss_avg=1.79485 acc=0.85938 acc_top1_avg=0.84184 acc_top5_avg=0.96871 lr=0.00010 gn=31.76539 time=55.24it/s +epoch=101 global_step=39750 loss=2.43573 loss_avg=1.81136 acc=0.77344 acc_top1_avg=0.84031 acc_top5_avg=0.96917 lr=0.00010 gn=29.21945 time=49.83it/s +epoch=101 global_step=39800 loss=2.01519 loss_avg=1.81808 acc=0.81250 acc_top1_avg=0.83965 acc_top5_avg=0.96905 lr=0.00010 gn=26.90340 time=60.81it/s +epoch=101 global_step=39850 loss=1.54359 loss_avg=1.82099 acc=0.88281 acc_top1_avg=0.83920 acc_top5_avg=0.96934 lr=0.00010 gn=35.64565 time=56.21it/s +====================Eval==================== +epoch=101 global_step=39882 loss=1.19603 test_loss_avg=0.73061 acc=0.62500 test_acc_avg=0.77232 test_acc_top5_avg=0.98624 time=232.85it/s +epoch=101 global_step=39882 loss=0.20110 test_loss_avg=0.49335 acc=0.95312 test_acc_avg=0.85057 test_acc_top5_avg=0.98977 time=234.34it/s +epoch=101 global_step=39882 loss=0.27608 test_loss_avg=0.47230 acc=0.93750 test_acc_avg=0.85789 test_acc_top5_avg=0.99031 time=577.57it/s +curr_acc 0.8579 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=1.66437 loss_avg=1.98432 acc=0.85938 acc_top1_avg=0.82335 acc_top5_avg=0.96962 lr=0.00010 gn=25.51333 time=54.18it/s +epoch=102 global_step=39950 loss=1.86414 loss_avg=1.80563 acc=0.82812 acc_top1_avg=0.84111 acc_top5_avg=0.97128 lr=0.00010 gn=26.22463 time=52.28it/s +epoch=102 global_step=40000 loss=2.19305 loss_avg=1.78071 acc=0.79688 acc_top1_avg=0.84415 acc_top5_avg=0.97080 lr=0.00010 gn=30.07172 time=54.72it/s +epoch=102 global_step=40050 loss=2.34055 loss_avg=1.79801 acc=0.78125 acc_top1_avg=0.84208 acc_top5_avg=0.96973 lr=0.00010 gn=24.64451 time=52.10it/s +epoch=102 global_step=40100 loss=2.44714 loss_avg=1.81517 acc=0.79688 acc_top1_avg=0.84067 acc_top5_avg=0.97004 lr=0.00010 gn=36.47446 time=64.06it/s +epoch=102 global_step=40150 loss=1.74357 loss_avg=1.81207 acc=0.85156 acc_top1_avg=0.84098 acc_top5_avg=0.97003 lr=0.00010 gn=30.80332 time=48.76it/s +epoch=102 global_step=40200 loss=2.11509 loss_avg=1.80408 acc=0.82031 acc_top1_avg=0.84188 acc_top5_avg=0.97045 lr=0.00010 gn=37.20274 time=57.38it/s +epoch=102 global_step=40250 loss=1.68832 loss_avg=1.81201 acc=0.83594 acc_top1_avg=0.84101 acc_top5_avg=0.96983 lr=0.00010 gn=28.95435 time=54.37it/s +====================Eval==================== +epoch=102 global_step=40273 loss=0.71902 test_loss_avg=0.63404 acc=0.80469 test_acc_avg=0.80525 test_acc_top5_avg=0.98754 time=238.26it/s +epoch=102 global_step=40273 loss=0.27084 test_loss_avg=0.47912 acc=0.93750 test_acc_avg=0.85641 test_acc_top5_avg=0.99051 time=502.61it/s +curr_acc 0.8564 +BEST_ACC 0.8939 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=1.75878 loss_avg=1.77356 acc=0.83594 acc_top1_avg=0.84346 acc_top5_avg=0.97338 lr=0.00010 gn=21.52432 time=58.04it/s +epoch=103 global_step=40350 loss=1.49827 loss_avg=1.76482 acc=0.86719 acc_top1_avg=0.84578 acc_top5_avg=0.97129 lr=0.00010 gn=23.83093 time=63.73it/s +epoch=103 global_step=40400 loss=1.81313 loss_avg=1.80567 acc=0.84375 acc_top1_avg=0.84129 acc_top5_avg=0.96930 lr=0.00010 gn=26.13657 time=55.65it/s +epoch=103 global_step=40450 loss=1.31553 loss_avg=1.79662 acc=0.89062 acc_top1_avg=0.84260 acc_top5_avg=0.96985 lr=0.00010 gn=30.41288 time=52.61it/s +epoch=103 global_step=40500 loss=1.11391 loss_avg=1.79815 acc=0.92188 acc_top1_avg=0.84258 acc_top5_avg=0.96985 lr=0.00010 gn=27.73928 time=53.76it/s +epoch=103 global_step=40550 loss=2.54113 loss_avg=1.79932 acc=0.78125 acc_top1_avg=0.84251 acc_top5_avg=0.96982 lr=0.00010 gn=39.21806 time=58.82it/s +epoch=103 global_step=40600 loss=2.15158 loss_avg=1.80175 acc=0.81250 acc_top1_avg=0.84215 acc_top5_avg=0.96983 lr=0.00010 gn=31.18087 time=63.17it/s +epoch=103 global_step=40650 loss=1.83741 loss_avg=1.81751 acc=0.84375 acc_top1_avg=0.84068 acc_top5_avg=0.96970 lr=0.00010 gn=31.66511 time=57.41it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.13889 test_loss_avg=0.80033 acc=0.96094 test_acc_avg=0.75841 test_acc_top5_avg=0.98377 time=234.96it/s +epoch=103 global_step=40664 loss=0.16300 test_loss_avg=0.55132 acc=0.95312 test_acc_avg=0.83358 test_acc_top5_avg=0.98859 time=241.12it/s +epoch=103 global_step=40664 loss=0.22778 test_loss_avg=0.48333 acc=0.93750 test_acc_avg=0.85542 test_acc_top5_avg=0.99011 time=724.53it/s +curr_acc 0.8554 +BEST_ACC 0.8939 +curr_acc_top5 0.9901 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=1.93125 loss_avg=1.83060 acc=0.82812 acc_top1_avg=0.84006 acc_top5_avg=0.96788 lr=0.00010 gn=20.64579 time=58.52it/s +epoch=104 global_step=40750 loss=1.94387 loss_avg=1.76392 acc=0.82031 acc_top1_avg=0.84648 acc_top5_avg=0.97011 lr=0.00010 gn=21.91950 time=51.58it/s +epoch=104 global_step=40800 loss=1.66176 loss_avg=1.75296 acc=0.85938 acc_top1_avg=0.84743 acc_top5_avg=0.97070 lr=0.00010 gn=31.88103 time=56.53it/s +epoch=104 global_step=40850 loss=1.92517 loss_avg=1.76488 acc=0.82812 acc_top1_avg=0.84623 acc_top5_avg=0.97060 lr=0.00010 gn=24.28995 time=61.05it/s +epoch=104 global_step=40900 loss=1.97501 loss_avg=1.78949 acc=0.82031 acc_top1_avg=0.84365 acc_top5_avg=0.97037 lr=0.00010 gn=32.14204 time=56.45it/s +epoch=104 global_step=40950 loss=1.40439 loss_avg=1.80126 acc=0.88281 acc_top1_avg=0.84238 acc_top5_avg=0.97012 lr=0.00010 gn=42.96396 time=50.60it/s +epoch=104 global_step=41000 loss=1.39847 loss_avg=1.81296 acc=0.89062 acc_top1_avg=0.84122 acc_top5_avg=0.96989 lr=0.00010 gn=34.06496 time=42.88it/s +epoch=104 global_step=41050 loss=1.49608 loss_avg=1.81674 acc=0.87500 acc_top1_avg=0.84061 acc_top5_avg=0.96956 lr=0.00010 gn=23.54714 time=63.56it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.38320 test_loss_avg=0.70219 acc=0.90625 test_acc_avg=0.78585 test_acc_top5_avg=0.98667 time=233.16it/s +epoch=104 global_step=41055 loss=0.28414 test_loss_avg=0.48347 acc=0.93750 test_acc_avg=0.85522 test_acc_top5_avg=0.99051 time=475.54it/s +curr_acc 0.8552 +BEST_ACC 0.8939 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=1.61320 loss_avg=1.85987 acc=0.85938 acc_top1_avg=0.83490 acc_top5_avg=0.96806 lr=0.00010 gn=22.40750 time=59.59it/s +epoch=105 global_step=41150 loss=1.99835 loss_avg=1.81155 acc=0.82031 acc_top1_avg=0.84038 acc_top5_avg=0.96900 lr=0.00010 gn=31.24108 time=60.27it/s +epoch=105 global_step=41200 loss=1.85338 loss_avg=1.80363 acc=0.84375 acc_top1_avg=0.84159 acc_top5_avg=0.97053 lr=0.00010 gn=29.47849 time=54.19it/s +epoch=105 global_step=41250 loss=2.24510 loss_avg=1.80574 acc=0.78906 acc_top1_avg=0.84111 acc_top5_avg=0.97079 lr=0.00010 gn=29.68250 time=54.73it/s +epoch=105 global_step=41300 loss=1.86724 loss_avg=1.81426 acc=0.83594 acc_top1_avg=0.84056 acc_top5_avg=0.97015 lr=0.00010 gn=30.38043 time=58.26it/s +epoch=105 global_step=41350 loss=1.85648 loss_avg=1.81737 acc=0.83594 acc_top1_avg=0.84015 acc_top5_avg=0.96984 lr=0.00010 gn=19.35108 time=53.19it/s +epoch=105 global_step=41400 loss=1.66467 loss_avg=1.81087 acc=0.85156 acc_top1_avg=0.84087 acc_top5_avg=0.97000 lr=0.00010 gn=30.75598 time=60.29it/s +====================Eval==================== +epoch=105 global_step=41446 loss=0.87736 test_loss_avg=1.15876 acc=0.68750 test_acc_avg=0.64375 test_acc_top5_avg=0.97188 time=243.56it/s +epoch=105 global_step=41446 loss=0.26729 test_loss_avg=0.60280 acc=0.91406 test_acc_avg=0.81520 test_acc_top5_avg=0.98693 time=247.17it/s +epoch=105 global_step=41446 loss=0.19866 test_loss_avg=0.48517 acc=0.93750 test_acc_avg=0.85216 test_acc_top5_avg=0.98981 time=637.82it/s +curr_acc 0.8522 +BEST_ACC 0.8939 +curr_acc_top5 0.9898 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=2.11803 loss_avg=1.82907 acc=0.81250 acc_top1_avg=0.84180 acc_top5_avg=0.97852 lr=0.00010 gn=32.30930 time=57.71it/s +epoch=106 global_step=41500 loss=1.99412 loss_avg=1.80349 acc=0.82031 acc_top1_avg=0.84129 acc_top5_avg=0.97323 lr=0.00010 gn=22.72920 time=60.30it/s +epoch=106 global_step=41550 loss=2.61680 loss_avg=1.79358 acc=0.75000 acc_top1_avg=0.84165 acc_top5_avg=0.97236 lr=0.00010 gn=37.79440 time=55.46it/s +epoch=106 global_step=41600 loss=1.62905 loss_avg=1.82364 acc=0.85156 acc_top1_avg=0.83868 acc_top5_avg=0.97139 lr=0.00010 gn=21.66921 time=47.24it/s +epoch=106 global_step=41650 loss=1.90643 loss_avg=1.80859 acc=0.82031 acc_top1_avg=0.84053 acc_top5_avg=0.97097 lr=0.00010 gn=27.92572 time=58.61it/s +epoch=106 global_step=41700 loss=1.70174 loss_avg=1.79275 acc=0.84375 acc_top1_avg=0.84258 acc_top5_avg=0.97063 lr=0.00010 gn=25.97918 time=59.70it/s +epoch=106 global_step=41750 loss=2.25406 loss_avg=1.79341 acc=0.79688 acc_top1_avg=0.84254 acc_top5_avg=0.97027 lr=0.00010 gn=33.22784 time=55.24it/s +epoch=106 global_step=41800 loss=1.69816 loss_avg=1.79298 acc=0.85156 acc_top1_avg=0.84273 acc_top5_avg=0.97043 lr=0.00010 gn=29.91394 time=47.38it/s +====================Eval==================== +epoch=106 global_step=41837 loss=0.73388 test_loss_avg=0.82719 acc=0.76562 test_acc_avg=0.75090 test_acc_top5_avg=0.98317 time=241.40it/s +epoch=106 global_step=41837 loss=0.21316 test_loss_avg=0.50617 acc=0.94531 test_acc_avg=0.85023 test_acc_top5_avg=0.98982 time=259.32it/s +epoch=106 global_step=41837 loss=0.20756 test_loss_avg=0.49624 acc=0.93750 test_acc_avg=0.85305 test_acc_top5_avg=0.99021 time=862.49it/s +curr_acc 0.8530 +BEST_ACC 0.8939 +curr_acc_top5 0.9902 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=1.36742 loss_avg=1.71532 acc=0.89062 acc_top1_avg=0.85156 acc_top5_avg=0.97897 lr=0.00010 gn=29.69397 time=53.21it/s +epoch=107 global_step=41900 loss=1.75530 loss_avg=1.81362 acc=0.84375 acc_top1_avg=0.84077 acc_top5_avg=0.96850 lr=0.00010 gn=30.43061 time=54.75it/s +epoch=107 global_step=41950 loss=1.39305 loss_avg=1.81587 acc=0.88281 acc_top1_avg=0.84112 acc_top5_avg=0.96875 lr=0.00010 gn=23.04018 time=63.08it/s +epoch=107 global_step=42000 loss=1.54515 loss_avg=1.79096 acc=0.86719 acc_top1_avg=0.84346 acc_top5_avg=0.96937 lr=0.00010 gn=26.58478 time=62.74it/s +epoch=107 global_step=42050 loss=2.26114 loss_avg=1.78429 acc=0.78906 acc_top1_avg=0.84393 acc_top5_avg=0.96937 lr=0.00010 gn=37.92043 time=54.52it/s +epoch=107 global_step=42100 loss=1.93607 loss_avg=1.77806 acc=0.82031 acc_top1_avg=0.84437 acc_top5_avg=0.96940 lr=0.00010 gn=30.48782 time=47.57it/s +epoch=107 global_step=42150 loss=1.54655 loss_avg=1.78822 acc=0.88281 acc_top1_avg=0.84320 acc_top5_avg=0.96895 lr=0.00010 gn=28.72368 time=62.75it/s +epoch=107 global_step=42200 loss=1.58839 loss_avg=1.79268 acc=0.85938 acc_top1_avg=0.84274 acc_top5_avg=0.96890 lr=0.00010 gn=20.11958 time=54.95it/s +====================Eval==================== +epoch=107 global_step=42228 loss=0.65545 test_loss_avg=0.65821 acc=0.76562 test_acc_avg=0.80086 test_acc_top5_avg=0.98554 time=242.71it/s +epoch=107 global_step=42228 loss=0.23946 test_loss_avg=0.48191 acc=0.93750 test_acc_avg=0.85581 test_acc_top5_avg=0.99031 time=491.08it/s +curr_acc 0.8558 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=1.80761 loss_avg=1.73446 acc=0.83594 acc_top1_avg=0.84943 acc_top5_avg=0.96662 lr=0.00010 gn=21.76153 time=50.21it/s +epoch=108 global_step=42300 loss=1.78400 loss_avg=1.76313 acc=0.84375 acc_top1_avg=0.84614 acc_top5_avg=0.96756 lr=0.00010 gn=35.78619 time=57.50it/s +epoch=108 global_step=42350 loss=1.69668 loss_avg=1.76818 acc=0.83594 acc_top1_avg=0.84497 acc_top5_avg=0.96849 lr=0.00010 gn=26.81397 time=52.44it/s +epoch=108 global_step=42400 loss=2.38275 loss_avg=1.78877 acc=0.78125 acc_top1_avg=0.84275 acc_top5_avg=0.96843 lr=0.00010 gn=33.54438 time=57.11it/s +epoch=108 global_step=42450 loss=1.88787 loss_avg=1.80631 acc=0.82812 acc_top1_avg=0.84111 acc_top5_avg=0.96850 lr=0.00010 gn=30.13306 time=50.70it/s +epoch=108 global_step=42500 loss=1.74637 loss_avg=1.79162 acc=0.85156 acc_top1_avg=0.84283 acc_top5_avg=0.96878 lr=0.00010 gn=28.70566 time=46.67it/s +epoch=108 global_step=42550 loss=1.54812 loss_avg=1.79508 acc=0.86719 acc_top1_avg=0.84254 acc_top5_avg=0.96924 lr=0.00010 gn=21.49725 time=56.31it/s +epoch=108 global_step=42600 loss=1.95258 loss_avg=1.80023 acc=0.82812 acc_top1_avg=0.84213 acc_top5_avg=0.96913 lr=0.00010 gn=33.60627 time=62.08it/s +====================Eval==================== +epoch=108 global_step=42619 loss=0.88261 test_loss_avg=0.81305 acc=0.71875 test_acc_avg=0.75694 test_acc_top5_avg=0.98611 time=231.61it/s +epoch=108 global_step=42619 loss=0.29408 test_loss_avg=0.54756 acc=0.92188 test_acc_avg=0.83640 test_acc_top5_avg=0.98920 time=243.91it/s +epoch=108 global_step=42619 loss=0.18150 test_loss_avg=0.50193 acc=0.93750 test_acc_avg=0.85077 test_acc_top5_avg=0.99001 time=486.63it/s +curr_acc 0.8508 +BEST_ACC 0.8939 +curr_acc_top5 0.9900 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=1.39601 loss_avg=1.78996 acc=0.88281 acc_top1_avg=0.84400 acc_top5_avg=0.96547 lr=0.00010 gn=28.55862 time=57.08it/s +epoch=109 global_step=42700 loss=1.21669 loss_avg=1.76878 acc=0.89844 acc_top1_avg=0.84606 acc_top5_avg=0.96779 lr=0.00010 gn=24.03909 time=62.62it/s +epoch=109 global_step=42750 loss=1.87812 loss_avg=1.77802 acc=0.83594 acc_top1_avg=0.84566 acc_top5_avg=0.96881 lr=0.00010 gn=30.99211 time=50.90it/s +epoch=109 global_step=42800 loss=1.21496 loss_avg=1.77786 acc=0.91406 acc_top1_avg=0.84483 acc_top5_avg=0.96888 lr=0.00010 gn=43.13797 time=56.83it/s +epoch=109 global_step=42850 loss=2.05185 loss_avg=1.78504 acc=0.82812 acc_top1_avg=0.84399 acc_top5_avg=0.96899 lr=0.00010 gn=26.60138 time=58.59it/s +epoch=109 global_step=42900 loss=1.73841 loss_avg=1.78585 acc=0.85938 acc_top1_avg=0.84386 acc_top5_avg=0.96931 lr=0.00010 gn=30.99581 time=51.25it/s +epoch=109 global_step=42950 loss=1.81556 loss_avg=1.79241 acc=0.85156 acc_top1_avg=0.84304 acc_top5_avg=0.96962 lr=0.00010 gn=32.79813 time=56.97it/s +epoch=109 global_step=43000 loss=2.14404 loss_avg=1.79709 acc=0.79688 acc_top1_avg=0.84246 acc_top5_avg=0.96973 lr=0.00010 gn=20.49957 time=56.64it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.16590 test_loss_avg=0.68588 acc=0.93750 test_acc_avg=0.79067 test_acc_top5_avg=0.98658 time=124.66it/s +epoch=109 global_step=43010 loss=0.19463 test_loss_avg=0.50887 acc=0.93750 test_acc_avg=0.84899 test_acc_top5_avg=0.99031 time=498.43it/s +curr_acc 0.8490 +BEST_ACC 0.8939 +curr_acc_top5 0.9903 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=1.91250 loss_avg=1.80776 acc=0.82812 acc_top1_avg=0.84023 acc_top5_avg=0.96484 lr=0.00010 gn=23.08701 time=62.06it/s +epoch=110 global_step=43100 loss=1.66607 loss_avg=1.76673 acc=0.85938 acc_top1_avg=0.84540 acc_top5_avg=0.96814 lr=0.00010 gn=26.06849 time=56.46it/s +epoch=110 global_step=43150 loss=1.61559 loss_avg=1.75969 acc=0.85938 acc_top1_avg=0.84654 acc_top5_avg=0.96897 lr=0.00010 gn=25.34124 time=60.29it/s +epoch=110 global_step=43200 loss=1.79198 loss_avg=1.78474 acc=0.85156 acc_top1_avg=0.84387 acc_top5_avg=0.96822 lr=0.00010 gn=43.23454 time=56.09it/s +epoch=110 global_step=43250 loss=1.97665 loss_avg=1.78417 acc=0.82812 acc_top1_avg=0.84375 acc_top5_avg=0.96797 lr=0.00010 gn=25.39412 time=53.33it/s +epoch=110 global_step=43300 loss=1.78746 loss_avg=1.77742 acc=0.85156 acc_top1_avg=0.84440 acc_top5_avg=0.96915 lr=0.00010 gn=29.19012 time=56.41it/s +epoch=110 global_step=43350 loss=1.83239 loss_avg=1.78677 acc=0.82812 acc_top1_avg=0.84331 acc_top5_avg=0.96967 lr=0.00010 gn=36.92842 time=57.32it/s +epoch=110 global_step=43400 loss=2.29811 loss_avg=1.78446 acc=0.81250 acc_top1_avg=0.84377 acc_top5_avg=0.96949 lr=0.00010 gn=37.92096 time=53.97it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.09761 test_loss_avg=0.88008 acc=0.96875 test_acc_avg=0.73125 test_acc_top5_avg=0.98281 time=236.91it/s +epoch=110 global_step=43401 loss=0.14983 test_loss_avg=0.55953 acc=0.95312 test_acc_avg=0.83255 test_acc_top5_avg=0.98932 time=234.78it/s +epoch=110 global_step=43401 loss=0.25011 test_loss_avg=0.48305 acc=0.93750 test_acc_avg=0.85601 test_acc_top5_avg=0.99080 time=856.16it/s +curr_acc 0.8560 +BEST_ACC 0.8939 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=1.76998 loss_avg=1.73844 acc=0.85156 acc_top1_avg=0.84933 acc_top5_avg=0.97162 lr=0.00010 gn=31.68769 time=58.79it/s +epoch=111 global_step=43500 loss=2.08646 loss_avg=1.76834 acc=0.81250 acc_top1_avg=0.84391 acc_top5_avg=0.97009 lr=0.00010 gn=28.22783 time=56.39it/s +epoch=111 global_step=43550 loss=1.45238 loss_avg=1.78394 acc=0.88281 acc_top1_avg=0.84212 acc_top5_avg=0.97043 lr=0.00010 gn=34.97442 time=56.06it/s +epoch=111 global_step=43600 loss=2.14707 loss_avg=1.78973 acc=0.79688 acc_top1_avg=0.84171 acc_top5_avg=0.97052 lr=0.00010 gn=26.71293 time=56.35it/s +epoch=111 global_step=43650 loss=1.39723 loss_avg=1.79578 acc=0.87500 acc_top1_avg=0.84152 acc_top5_avg=0.97057 lr=0.00010 gn=19.26117 time=62.51it/s +epoch=111 global_step=43700 loss=1.75749 loss_avg=1.78620 acc=0.85156 acc_top1_avg=0.84265 acc_top5_avg=0.97016 lr=0.00010 gn=33.68673 time=58.24it/s +epoch=111 global_step=43750 loss=1.65646 loss_avg=1.78195 acc=0.85938 acc_top1_avg=0.84341 acc_top5_avg=0.96998 lr=0.00010 gn=27.92547 time=58.20it/s +====================Eval==================== +epoch=111 global_step=43792 loss=0.66428 test_loss_avg=0.79670 acc=0.78125 test_acc_avg=0.76008 test_acc_top5_avg=0.98463 time=51.35it/s +epoch=111 global_step=43792 loss=0.20548 test_loss_avg=0.49859 acc=0.93750 test_acc_avg=0.85206 test_acc_top5_avg=0.98952 time=827.77it/s +curr_acc 0.8521 +BEST_ACC 0.8939 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=1.70687 loss_avg=1.62272 acc=0.84375 acc_top1_avg=0.86523 acc_top5_avg=0.98047 lr=0.00010 gn=25.55290 time=35.86it/s +epoch=112 global_step=43850 loss=1.96367 loss_avg=1.75784 acc=0.82031 acc_top1_avg=0.84564 acc_top5_avg=0.96942 lr=0.00010 gn=23.23767 time=53.47it/s +epoch=112 global_step=43900 loss=2.25387 loss_avg=1.78854 acc=0.78906 acc_top1_avg=0.84361 acc_top5_avg=0.96810 lr=0.00010 gn=31.25796 time=52.26it/s +epoch=112 global_step=43950 loss=1.88519 loss_avg=1.78123 acc=0.82812 acc_top1_avg=0.84434 acc_top5_avg=0.96806 lr=0.00010 gn=24.08795 time=55.42it/s +epoch=112 global_step=44000 loss=1.85953 loss_avg=1.77021 acc=0.82812 acc_top1_avg=0.84537 acc_top5_avg=0.96901 lr=0.00010 gn=24.66319 time=59.25it/s +epoch=112 global_step=44050 loss=1.95722 loss_avg=1.77485 acc=0.82812 acc_top1_avg=0.84517 acc_top5_avg=0.96899 lr=0.00010 gn=25.66382 time=54.33it/s +epoch=112 global_step=44100 loss=2.04829 loss_avg=1.78411 acc=0.81250 acc_top1_avg=0.84443 acc_top5_avg=0.96938 lr=0.00010 gn=35.17719 time=53.96it/s +epoch=112 global_step=44150 loss=1.65922 loss_avg=1.78412 acc=0.85938 acc_top1_avg=0.84427 acc_top5_avg=0.96908 lr=0.00010 gn=27.17718 time=49.15it/s +====================Eval==================== +epoch=112 global_step=44183 loss=1.33713 test_loss_avg=1.28443 acc=0.59375 test_acc_avg=0.61719 test_acc_top5_avg=0.96875 time=227.40it/s +epoch=112 global_step=44183 loss=0.19954 test_loss_avg=0.59349 acc=0.92969 test_acc_avg=0.82347 test_acc_top5_avg=0.98708 time=241.79it/s +epoch=112 global_step=44183 loss=0.19593 test_loss_avg=0.47026 acc=0.93750 test_acc_avg=0.86116 test_acc_top5_avg=0.99051 time=549.93it/s +curr_acc 0.8612 +BEST_ACC 0.8939 +curr_acc_top5 0.9905 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=1.69833 loss_avg=1.75867 acc=0.85156 acc_top1_avg=0.84789 acc_top5_avg=0.96645 lr=0.00010 gn=27.86548 time=53.02it/s +epoch=113 global_step=44250 loss=2.00998 loss_avg=1.80159 acc=0.82031 acc_top1_avg=0.84305 acc_top5_avg=0.96770 lr=0.00010 gn=27.72217 time=55.01it/s +epoch=113 global_step=44300 loss=2.04374 loss_avg=1.81005 acc=0.81250 acc_top1_avg=0.84135 acc_top5_avg=0.96902 lr=0.00010 gn=29.93228 time=53.16it/s +epoch=113 global_step=44350 loss=1.97435 loss_avg=1.77321 acc=0.80469 acc_top1_avg=0.84525 acc_top5_avg=0.97006 lr=0.00010 gn=30.88727 time=55.27it/s +epoch=113 global_step=44400 loss=1.62664 loss_avg=1.77329 acc=0.87500 acc_top1_avg=0.84569 acc_top5_avg=0.97019 lr=0.00010 gn=31.35083 time=58.10it/s +epoch=113 global_step=44450 loss=1.95799 loss_avg=1.77634 acc=0.83594 acc_top1_avg=0.84510 acc_top5_avg=0.97021 lr=0.00010 gn=39.73351 time=62.71it/s +epoch=113 global_step=44500 loss=1.97424 loss_avg=1.78223 acc=0.82031 acc_top1_avg=0.84442 acc_top5_avg=0.97035 lr=0.00010 gn=25.04359 time=56.51it/s +epoch=113 global_step=44550 loss=1.87200 loss_avg=1.77873 acc=0.84375 acc_top1_avg=0.84458 acc_top5_avg=0.97026 lr=0.00010 gn=39.99946 time=63.69it/s +====================Eval==================== +epoch=113 global_step=44574 loss=1.03452 test_loss_avg=0.85209 acc=0.66406 test_acc_avg=0.74049 test_acc_top5_avg=0.98404 time=237.52it/s +epoch=113 global_step=44574 loss=0.42791 test_loss_avg=0.52546 acc=0.91406 test_acc_avg=0.84289 test_acc_top5_avg=0.98951 time=256.22it/s +epoch=113 global_step=44574 loss=0.17186 test_loss_avg=0.50178 acc=0.93750 test_acc_avg=0.84958 test_acc_top5_avg=0.99001 time=652.30it/s +curr_acc 0.8496 +BEST_ACC 0.8939 +curr_acc_top5 0.9900 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=1.97740 loss_avg=1.76121 acc=0.83594 acc_top1_avg=0.84525 acc_top5_avg=0.96725 lr=0.00010 gn=44.16046 time=54.13it/s +epoch=114 global_step=44650 loss=1.79865 loss_avg=1.82486 acc=0.84375 acc_top1_avg=0.83861 acc_top5_avg=0.96752 lr=0.00010 gn=39.40212 time=56.83it/s +epoch=114 global_step=44700 loss=1.77157 loss_avg=1.78758 acc=0.83594 acc_top1_avg=0.84282 acc_top5_avg=0.96943 lr=0.00010 gn=30.26393 time=52.09it/s +epoch=114 global_step=44750 loss=2.11804 loss_avg=1.79034 acc=0.79688 acc_top1_avg=0.84291 acc_top5_avg=0.97035 lr=0.00010 gn=38.19482 time=60.57it/s +epoch=114 global_step=44800 loss=1.68088 loss_avg=1.77438 acc=0.85156 acc_top1_avg=0.84472 acc_top5_avg=0.97120 lr=0.00010 gn=37.87870 time=58.42it/s +epoch=114 global_step=44850 loss=1.62515 loss_avg=1.77128 acc=0.87500 acc_top1_avg=0.84534 acc_top5_avg=0.97045 lr=0.00010 gn=34.68085 time=56.08it/s +epoch=114 global_step=44900 loss=1.78588 loss_avg=1.77269 acc=0.85938 acc_top1_avg=0.84519 acc_top5_avg=0.97014 lr=0.00010 gn=31.70085 time=60.65it/s +epoch=114 global_step=44950 loss=2.46022 loss_avg=1.77086 acc=0.78125 acc_top1_avg=0.84558 acc_top5_avg=0.97023 lr=0.00010 gn=37.30174 time=62.83it/s +====================Eval==================== +epoch=114 global_step=44965 loss=0.95703 test_loss_avg=0.68645 acc=0.76562 test_acc_avg=0.79119 test_acc_top5_avg=0.98438 time=246.91it/s +epoch=114 global_step=44965 loss=0.17993 test_loss_avg=0.50057 acc=0.93750 test_acc_avg=0.84998 test_acc_top5_avg=0.98952 time=474.90it/s +curr_acc 0.8500 +BEST_ACC 0.8939 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=1.50124 loss_avg=1.80957 acc=0.87500 acc_top1_avg=0.84107 acc_top5_avg=0.96763 lr=0.00010 gn=24.84569 time=62.28it/s +epoch=115 global_step=45050 loss=2.04355 loss_avg=1.79899 acc=0.82031 acc_top1_avg=0.84265 acc_top5_avg=0.97040 lr=0.00010 gn=33.61862 time=52.10it/s +epoch=115 global_step=45100 loss=1.21808 loss_avg=1.79133 acc=0.89844 acc_top1_avg=0.84282 acc_top5_avg=0.97135 lr=0.00010 gn=23.66730 time=53.06it/s +epoch=115 global_step=45150 loss=1.66797 loss_avg=1.79435 acc=0.85156 acc_top1_avg=0.84223 acc_top5_avg=0.97107 lr=0.00010 gn=26.32813 time=55.10it/s +epoch=115 global_step=45200 loss=1.36854 loss_avg=1.79493 acc=0.88281 acc_top1_avg=0.84219 acc_top5_avg=0.97094 lr=0.00010 gn=26.41607 time=53.70it/s +epoch=115 global_step=45250 loss=1.70442 loss_avg=1.77919 acc=0.85156 acc_top1_avg=0.84383 acc_top5_avg=0.97075 lr=0.00010 gn=29.69611 time=56.58it/s +epoch=115 global_step=45300 loss=2.19583 loss_avg=1.76887 acc=0.79688 acc_top1_avg=0.84494 acc_top5_avg=0.97055 lr=0.00010 gn=31.78844 time=49.89it/s +epoch=115 global_step=45350 loss=1.76390 loss_avg=1.76645 acc=0.83594 acc_top1_avg=0.84509 acc_top5_avg=0.97019 lr=0.00010 gn=23.83360 time=55.56it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.21687 test_loss_avg=0.67366 acc=0.94531 test_acc_avg=0.80052 test_acc_top5_avg=0.98750 time=243.71it/s +epoch=115 global_step=45356 loss=0.05988 test_loss_avg=0.54134 acc=0.98438 test_acc_avg=0.83870 test_acc_top5_avg=0.98858 time=238.22it/s +epoch=115 global_step=45356 loss=0.19713 test_loss_avg=0.48368 acc=0.93750 test_acc_avg=0.85670 test_acc_top5_avg=0.98972 time=672.06it/s +curr_acc 0.8567 +BEST_ACC 0.8939 +curr_acc_top5 0.9897 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=1.60936 loss_avg=1.78338 acc=0.85938 acc_top1_avg=0.84464 acc_top5_avg=0.96715 lr=0.00010 gn=36.57958 time=55.37it/s +epoch=116 global_step=45450 loss=2.01539 loss_avg=1.76434 acc=0.82031 acc_top1_avg=0.84533 acc_top5_avg=0.97025 lr=0.00010 gn=38.18871 time=60.44it/s +epoch=116 global_step=45500 loss=1.60442 loss_avg=1.75781 acc=0.84375 acc_top1_avg=0.84619 acc_top5_avg=0.97032 lr=0.00010 gn=25.80165 time=51.77it/s +epoch=116 global_step=45550 loss=1.10814 loss_avg=1.75616 acc=0.90625 acc_top1_avg=0.84552 acc_top5_avg=0.97000 lr=0.00010 gn=29.65328 time=52.41it/s +epoch=116 global_step=45600 loss=1.61896 loss_avg=1.76732 acc=0.86719 acc_top1_avg=0.84474 acc_top5_avg=0.97003 lr=0.00010 gn=28.47027 time=58.88it/s +epoch=116 global_step=45650 loss=1.92182 loss_avg=1.77193 acc=0.82812 acc_top1_avg=0.84455 acc_top5_avg=0.96957 lr=0.00010 gn=36.18811 time=54.16it/s +epoch=116 global_step=45700 loss=2.56137 loss_avg=1.77595 acc=0.75781 acc_top1_avg=0.84443 acc_top5_avg=0.96959 lr=0.00010 gn=32.11378 time=60.63it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.19671 test_loss_avg=0.68466 acc=0.91406 test_acc_avg=0.79058 test_acc_top5_avg=0.98698 time=239.51it/s +epoch=116 global_step=45747 loss=0.22453 test_loss_avg=0.48953 acc=0.93750 test_acc_avg=0.85423 test_acc_top5_avg=0.99041 time=524.03it/s +curr_acc 0.8542 +BEST_ACC 0.8939 +curr_acc_top5 0.9904 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=1.62422 loss_avg=1.46221 acc=0.85938 acc_top1_avg=0.87760 acc_top5_avg=0.96615 lr=0.00010 gn=23.34103 time=58.85it/s +epoch=117 global_step=45800 loss=1.50009 loss_avg=1.72143 acc=0.87500 acc_top1_avg=0.85171 acc_top5_avg=0.97022 lr=0.00010 gn=31.18081 time=54.61it/s +epoch=117 global_step=45850 loss=2.22640 loss_avg=1.73516 acc=0.79688 acc_top1_avg=0.84929 acc_top5_avg=0.97171 lr=0.00010 gn=31.05192 time=53.67it/s +epoch=117 global_step=45900 loss=2.13212 loss_avg=1.75559 acc=0.81250 acc_top1_avg=0.84732 acc_top5_avg=0.97049 lr=0.00010 gn=44.91841 time=51.59it/s +epoch=117 global_step=45950 loss=1.60675 loss_avg=1.75139 acc=0.86719 acc_top1_avg=0.84706 acc_top5_avg=0.97037 lr=0.00010 gn=29.88106 time=52.39it/s +epoch=117 global_step=46000 loss=1.78385 loss_avg=1.75350 acc=0.85156 acc_top1_avg=0.84715 acc_top5_avg=0.97039 lr=0.00010 gn=37.04441 time=63.65it/s +epoch=117 global_step=46050 loss=1.50835 loss_avg=1.75115 acc=0.85938 acc_top1_avg=0.84733 acc_top5_avg=0.97043 lr=0.00010 gn=25.23669 time=55.77it/s +epoch=117 global_step=46100 loss=1.62322 loss_avg=1.75973 acc=0.87500 acc_top1_avg=0.84623 acc_top5_avg=0.96990 lr=0.00010 gn=33.68887 time=59.01it/s +====================Eval==================== +epoch=117 global_step=46138 loss=1.31458 test_loss_avg=1.25413 acc=0.60938 test_acc_avg=0.61496 test_acc_top5_avg=0.96987 time=219.30it/s +epoch=117 global_step=46138 loss=0.28414 test_loss_avg=0.59835 acc=0.90625 test_acc_avg=0.81949 test_acc_top5_avg=0.98684 time=56.19it/s +epoch=117 global_step=46138 loss=0.24349 test_loss_avg=0.49672 acc=0.93750 test_acc_avg=0.85156 test_acc_top5_avg=0.98952 time=819.52it/s +curr_acc 0.8516 +BEST_ACC 0.8939 +curr_acc_top5 0.9895 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=1.61317 loss_avg=1.79702 acc=0.85938 acc_top1_avg=0.84049 acc_top5_avg=0.97005 lr=0.00010 gn=30.18682 time=51.95it/s +epoch=118 global_step=46200 loss=1.94521 loss_avg=1.80232 acc=0.82031 acc_top1_avg=0.84337 acc_top5_avg=0.96762 lr=0.00010 gn=32.97228 time=59.50it/s +epoch=118 global_step=46250 loss=1.40552 loss_avg=1.77722 acc=0.88281 acc_top1_avg=0.84563 acc_top5_avg=0.96777 lr=0.00010 gn=30.84337 time=52.23it/s +epoch=118 global_step=46300 loss=1.88688 loss_avg=1.78612 acc=0.82812 acc_top1_avg=0.84394 acc_top5_avg=0.96851 lr=0.00010 gn=35.70327 time=51.96it/s +epoch=118 global_step=46350 loss=1.50689 loss_avg=1.78887 acc=0.87500 acc_top1_avg=0.84401 acc_top5_avg=0.96890 lr=0.00010 gn=30.92781 time=60.02it/s +epoch=118 global_step=46400 loss=1.89989 loss_avg=1.77609 acc=0.83594 acc_top1_avg=0.84572 acc_top5_avg=0.96929 lr=0.00010 gn=27.33952 time=55.00it/s +epoch=118 global_step=46450 loss=1.68278 loss_avg=1.77216 acc=0.85156 acc_top1_avg=0.84613 acc_top5_avg=0.96908 lr=0.00010 gn=27.01401 time=54.61it/s +epoch=118 global_step=46500 loss=1.35901 loss_avg=1.76979 acc=0.88281 acc_top1_avg=0.84636 acc_top5_avg=0.96946 lr=0.00010 gn=25.34636 time=58.92it/s +====================Eval==================== +epoch=118 global_step=46529 loss=0.34706 test_loss_avg=0.76023 acc=0.88281 test_acc_avg=0.76423 test_acc_top5_avg=0.98661 time=238.30it/s +epoch=118 global_step=46529 loss=0.20593 test_loss_avg=0.49186 acc=0.94531 test_acc_avg=0.85236 test_acc_top5_avg=0.99069 time=252.79it/s +epoch=118 global_step=46529 loss=0.25042 test_loss_avg=0.48881 acc=0.93750 test_acc_avg=0.85344 test_acc_top5_avg=0.99080 time=672.81it/s +curr_acc 0.8534 +BEST_ACC 0.8939 +curr_acc_top5 0.9908 +BEST_ACC_top5 0.9943 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=1.84642 loss_avg=1.66381 acc=0.84375 acc_top1_avg=0.85714 acc_top5_avg=0.97359 lr=0.00010 gn=40.14376 time=54.34it/s +epoch=119 global_step=46600 loss=1.60940 loss_avg=1.75925 acc=0.86719 acc_top1_avg=0.84573 acc_top5_avg=0.96886 lr=0.00010 gn=34.55109 time=51.87it/s +epoch=119 global_step=46650 loss=1.83260 loss_avg=1.73878 acc=0.82812 acc_top1_avg=0.84898 acc_top5_avg=0.96985 lr=0.00010 gn=41.45619 time=55.98it/s +epoch=119 global_step=46700 loss=1.64866 loss_avg=1.72599 acc=0.85938 acc_top1_avg=0.85015 acc_top5_avg=0.97021 lr=0.00010 gn=27.50156 time=54.74it/s +epoch=119 global_step=46750 loss=1.78979 loss_avg=1.73353 acc=0.83594 acc_top1_avg=0.84905 acc_top5_avg=0.97052 lr=0.00010 gn=23.21077 time=63.61it/s +epoch=119 global_step=46800 loss=2.03306 loss_avg=1.73317 acc=0.82031 acc_top1_avg=0.84914 acc_top5_avg=0.97062 lr=0.00010 gn=29.55023 time=54.44it/s +epoch=119 global_step=46850 loss=1.62011 loss_avg=1.75085 acc=0.85938 acc_top1_avg=0.84730 acc_top5_avg=0.97050 lr=0.00010 gn=38.91562 time=57.87it/s +epoch=119 global_step=46900 loss=2.06894 loss_avg=1.75577 acc=0.80469 acc_top1_avg=0.84666 acc_top5_avg=0.97031 lr=0.00010 gn=27.59171 time=51.50it/s +====================Eval==================== +epoch=119 global_step=46920 loss=0.25382 test_loss_avg=0.71293 acc=0.93750 test_acc_avg=0.78715 test_acc_top5_avg=0.98533 time=233.60it/s +epoch=119 global_step=46920 loss=0.22030 test_loss_avg=0.52673 acc=0.93750 test_acc_avg=0.84345 test_acc_top5_avg=0.98962 time=503.52it/s +curr_acc 0.8435 +BEST_ACC 0.8939 +curr_acc_top5 0.9896 +BEST_ACC_top5 0.9943 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_6_4.log b/other_methods/sceloss/sceloss_results/out_6_4.log new file mode 100644 index 0000000..b19e6b6 --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_6_4.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.6__noise_amount__0.4.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=6.87973 loss_avg=7.26662 acc=0.31250 acc_top1_avg=0.26891 acc_top5_avg=0.71141 lr=0.01000 gn=6.41600 time=61.48it/s +epoch=0 global_step=100 loss=7.14121 loss_avg=7.15559 acc=0.27344 acc_top1_avg=0.27609 acc_top5_avg=0.74039 lr=0.01000 gn=6.07869 time=62.23it/s +epoch=0 global_step=150 loss=6.57607 loss_avg=7.07591 acc=0.32812 acc_top1_avg=0.28245 acc_top5_avg=0.75458 lr=0.01000 gn=5.59489 time=59.80it/s +epoch=0 global_step=200 loss=6.66094 loss_avg=6.96622 acc=0.31250 acc_top1_avg=0.29512 acc_top5_avg=0.76637 lr=0.01000 gn=4.86961 time=62.70it/s +epoch=0 global_step=250 loss=6.47185 loss_avg=6.90518 acc=0.34375 acc_top1_avg=0.30116 acc_top5_avg=0.77441 lr=0.01000 gn=5.43866 time=59.94it/s +epoch=0 global_step=300 loss=6.73274 loss_avg=6.83352 acc=0.32031 acc_top1_avg=0.30938 acc_top5_avg=0.78104 lr=0.01000 gn=3.88396 time=61.19it/s +epoch=0 global_step=350 loss=6.47334 loss_avg=6.77130 acc=0.32812 acc_top1_avg=0.31592 acc_top5_avg=0.78598 lr=0.01000 gn=4.27100 time=61.47it/s +====================Eval==================== +epoch=0 global_step=391 loss=6.02314 test_loss_avg=4.19116 acc=0.00000 test_acc_avg=0.20891 test_acc_top5_avg=0.78203 time=257.34it/s +epoch=0 global_step=391 loss=2.54971 test_loss_avg=3.56764 acc=0.56250 test_acc_avg=0.32625 test_acc_top5_avg=0.80607 time=31.18it/s +curr_acc 0.3262 +BEST_ACC 0.0000 +curr_acc_top5 0.8061 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.42006 loss_avg=6.37848 acc=0.34375 acc_top1_avg=0.35590 acc_top5_avg=0.81510 lr=0.01000 gn=3.60785 time=64.44it/s +epoch=1 global_step=450 loss=6.32588 loss_avg=6.32907 acc=0.35156 acc_top1_avg=0.36322 acc_top5_avg=0.82402 lr=0.01000 gn=5.36309 time=56.31it/s +epoch=1 global_step=500 loss=5.82438 loss_avg=6.27039 acc=0.39844 acc_top1_avg=0.36676 acc_top5_avg=0.83264 lr=0.01000 gn=4.73012 time=59.09it/s +epoch=1 global_step=550 loss=6.40109 loss_avg=6.24541 acc=0.34375 acc_top1_avg=0.36925 acc_top5_avg=0.83417 lr=0.01000 gn=4.85751 time=61.55it/s +epoch=1 global_step=600 loss=6.14979 loss_avg=6.20992 acc=0.39844 acc_top1_avg=0.37313 acc_top5_avg=0.83725 lr=0.01000 gn=5.14155 time=60.31it/s +epoch=1 global_step=650 loss=6.43851 loss_avg=6.19546 acc=0.35156 acc_top1_avg=0.37394 acc_top5_avg=0.83856 lr=0.01000 gn=5.29853 time=63.16it/s +epoch=1 global_step=700 loss=5.75610 loss_avg=6.16048 acc=0.39062 acc_top1_avg=0.37771 acc_top5_avg=0.84026 lr=0.01000 gn=4.20127 time=51.62it/s +epoch=1 global_step=750 loss=6.09743 loss_avg=6.13035 acc=0.39062 acc_top1_avg=0.38138 acc_top5_avg=0.84210 lr=0.01000 gn=4.48088 time=55.61it/s +====================Eval==================== +epoch=1 global_step=782 loss=5.43968 test_loss_avg=3.22921 acc=0.00000 test_acc_avg=0.27827 test_acc_top5_avg=0.94531 time=250.41it/s +epoch=1 global_step=782 loss=0.75076 test_loss_avg=3.21882 acc=0.77344 test_acc_avg=0.33737 test_acc_top5_avg=0.84562 time=246.10it/s +epoch=1 global_step=782 loss=1.37426 test_loss_avg=3.01618 acc=0.62500 test_acc_avg=0.37203 test_acc_top5_avg=0.85809 time=499.14it/s +curr_acc 0.3720 +BEST_ACC 0.3262 +curr_acc_top5 0.8581 +BEST_ACC_top5 0.8061 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=5.74102 loss_avg=5.95278 acc=0.42188 acc_top1_avg=0.40017 acc_top5_avg=0.85677 lr=0.01000 gn=4.44106 time=60.54it/s +epoch=2 global_step=850 loss=5.90342 loss_avg=5.93022 acc=0.42188 acc_top1_avg=0.40165 acc_top5_avg=0.85007 lr=0.01000 gn=4.36041 time=61.27it/s +epoch=2 global_step=900 loss=6.04215 loss_avg=5.86943 acc=0.37500 acc_top1_avg=0.40916 acc_top5_avg=0.85143 lr=0.01000 gn=5.18791 time=62.04it/s +epoch=2 global_step=950 loss=5.88720 loss_avg=5.88414 acc=0.42188 acc_top1_avg=0.40765 acc_top5_avg=0.85221 lr=0.01000 gn=5.13403 time=55.68it/s +epoch=2 global_step=1000 loss=6.06003 loss_avg=5.87485 acc=0.39062 acc_top1_avg=0.40865 acc_top5_avg=0.85318 lr=0.01000 gn=4.01665 time=61.95it/s +epoch=2 global_step=1050 loss=5.80222 loss_avg=5.84765 acc=0.39844 acc_top1_avg=0.41129 acc_top5_avg=0.85556 lr=0.01000 gn=4.91968 time=55.08it/s +epoch=2 global_step=1100 loss=5.97709 loss_avg=5.82919 acc=0.40625 acc_top1_avg=0.41335 acc_top5_avg=0.85711 lr=0.01000 gn=4.58520 time=42.28it/s +epoch=2 global_step=1150 loss=5.86568 loss_avg=5.82176 acc=0.43750 acc_top1_avg=0.41383 acc_top5_avg=0.85853 lr=0.01000 gn=4.46916 time=53.22it/s +====================Eval==================== +epoch=2 global_step=1173 loss=3.08640 test_loss_avg=3.23326 acc=0.14062 test_acc_avg=0.29781 test_acc_top5_avg=0.84189 time=242.70it/s +epoch=2 global_step=1173 loss=1.24254 test_loss_avg=2.78427 acc=0.75000 test_acc_avg=0.39092 test_acc_top5_avg=0.86640 time=815.54it/s +curr_acc 0.3909 +BEST_ACC 0.3720 +curr_acc_top5 0.8664 +BEST_ACC_top5 0.8581 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=5.63799 loss_avg=5.64761 acc=0.43750 acc_top1_avg=0.43374 acc_top5_avg=0.86719 lr=0.01000 gn=3.76061 time=58.29it/s +epoch=3 global_step=1250 loss=5.84090 loss_avg=5.68503 acc=0.39062 acc_top1_avg=0.42644 acc_top5_avg=0.86678 lr=0.01000 gn=4.74467 time=57.44it/s +epoch=3 global_step=1300 loss=5.96458 loss_avg=5.69567 acc=0.39062 acc_top1_avg=0.42661 acc_top5_avg=0.86793 lr=0.01000 gn=4.67900 time=61.45it/s +epoch=3 global_step=1350 loss=5.43347 loss_avg=5.68163 acc=0.44531 acc_top1_avg=0.42730 acc_top5_avg=0.86851 lr=0.01000 gn=4.97022 time=65.54it/s +epoch=3 global_step=1400 loss=6.27717 loss_avg=5.67663 acc=0.33594 acc_top1_avg=0.42807 acc_top5_avg=0.86963 lr=0.01000 gn=3.86847 time=59.57it/s +epoch=3 global_step=1450 loss=5.35190 loss_avg=5.67332 acc=0.48438 acc_top1_avg=0.42814 acc_top5_avg=0.86925 lr=0.01000 gn=4.51644 time=47.35it/s +epoch=3 global_step=1500 loss=5.86908 loss_avg=5.67227 acc=0.41406 acc_top1_avg=0.42847 acc_top5_avg=0.86927 lr=0.01000 gn=4.76158 time=58.64it/s +epoch=3 global_step=1550 loss=6.01308 loss_avg=5.66755 acc=0.39062 acc_top1_avg=0.42934 acc_top5_avg=0.86988 lr=0.01000 gn=3.75400 time=58.86it/s +====================Eval==================== +epoch=3 global_step=1564 loss=2.29407 test_loss_avg=3.14486 acc=0.58594 test_acc_avg=0.41947 test_acc_top5_avg=0.86899 time=227.74it/s +epoch=3 global_step=1564 loss=1.66006 test_loss_avg=4.46047 acc=0.60156 test_acc_avg=0.23140 test_acc_top5_avg=0.83780 time=230.99it/s +epoch=3 global_step=1564 loss=0.01974 test_loss_avg=3.71160 acc=1.00000 test_acc_avg=0.35552 test_acc_top5_avg=0.86808 time=822.74it/s +curr_acc 0.3555 +BEST_ACC 0.3909 +curr_acc_top5 0.8681 +BEST_ACC_top5 0.8664 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=4.98534 loss_avg=5.70233 acc=0.50000 acc_top1_avg=0.42817 acc_top5_avg=0.86719 lr=0.01000 gn=4.38066 time=61.27it/s +epoch=4 global_step=1650 loss=5.33581 loss_avg=5.55506 acc=0.46094 acc_top1_avg=0.44350 acc_top5_avg=0.87782 lr=0.01000 gn=4.47409 time=62.58it/s +epoch=4 global_step=1700 loss=5.34580 loss_avg=5.57630 acc=0.45312 acc_top1_avg=0.43876 acc_top5_avg=0.87965 lr=0.01000 gn=4.85715 time=59.10it/s +epoch=4 global_step=1750 loss=4.98405 loss_avg=5.58818 acc=0.49219 acc_top1_avg=0.43746 acc_top5_avg=0.87752 lr=0.01000 gn=5.17104 time=56.48it/s +epoch=4 global_step=1800 loss=5.59187 loss_avg=5.56632 acc=0.44531 acc_top1_avg=0.43992 acc_top5_avg=0.87745 lr=0.01000 gn=4.20895 time=56.59it/s +epoch=4 global_step=1850 loss=5.55584 loss_avg=5.56964 acc=0.44531 acc_top1_avg=0.43870 acc_top5_avg=0.87740 lr=0.01000 gn=6.22189 time=54.56it/s +epoch=4 global_step=1900 loss=5.78009 loss_avg=5.56108 acc=0.42969 acc_top1_avg=0.44017 acc_top5_avg=0.87658 lr=0.01000 gn=4.06874 time=55.64it/s +epoch=4 global_step=1950 loss=5.71462 loss_avg=5.54689 acc=0.39844 acc_top1_avg=0.44108 acc_top5_avg=0.87708 lr=0.01000 gn=4.97145 time=65.34it/s +====================Eval==================== +epoch=4 global_step=1955 loss=0.78521 test_loss_avg=2.98335 acc=0.75781 test_acc_avg=0.35294 test_acc_top5_avg=0.82606 time=244.10it/s +epoch=4 global_step=1955 loss=0.21664 test_loss_avg=2.25133 acc=0.87500 test_acc_avg=0.47973 test_acc_top5_avg=0.88212 time=496.13it/s +curr_acc 0.4797 +BEST_ACC 0.3909 +curr_acc_top5 0.8821 +BEST_ACC_top5 0.8681 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=5.35167 loss_avg=5.47911 acc=0.44531 acc_top1_avg=0.44774 acc_top5_avg=0.87934 lr=0.01000 gn=4.34870 time=60.14it/s +epoch=5 global_step=2050 loss=5.33265 loss_avg=5.46682 acc=0.44531 acc_top1_avg=0.44794 acc_top5_avg=0.87854 lr=0.01000 gn=4.15173 time=57.50it/s +epoch=5 global_step=2100 loss=5.55388 loss_avg=5.45086 acc=0.43750 acc_top1_avg=0.44892 acc_top5_avg=0.87904 lr=0.01000 gn=4.12226 time=56.05it/s +epoch=5 global_step=2150 loss=5.43771 loss_avg=5.45339 acc=0.44531 acc_top1_avg=0.45036 acc_top5_avg=0.87808 lr=0.01000 gn=4.84204 time=53.67it/s +epoch=5 global_step=2200 loss=5.99577 loss_avg=5.44776 acc=0.40625 acc_top1_avg=0.45156 acc_top5_avg=0.88010 lr=0.01000 gn=5.46378 time=59.21it/s +epoch=5 global_step=2250 loss=5.48404 loss_avg=5.45207 acc=0.45312 acc_top1_avg=0.45101 acc_top5_avg=0.87979 lr=0.01000 gn=5.12194 time=55.92it/s +epoch=5 global_step=2300 loss=5.70410 loss_avg=5.45559 acc=0.39844 acc_top1_avg=0.45034 acc_top5_avg=0.88012 lr=0.01000 gn=4.15055 time=55.57it/s +====================Eval==================== +epoch=5 global_step=2346 loss=4.42070 test_loss_avg=4.26762 acc=0.14844 test_acc_avg=0.17188 test_acc_top5_avg=0.92656 time=180.48it/s +epoch=5 global_step=2346 loss=4.63373 test_loss_avg=3.71201 acc=0.10938 test_acc_avg=0.25213 test_acc_top5_avg=0.83054 time=242.07it/s +epoch=5 global_step=2346 loss=0.24589 test_loss_avg=2.87005 acc=0.81250 test_acc_avg=0.40862 test_acc_top5_avg=0.86917 time=639.96it/s +curr_acc 0.4086 +BEST_ACC 0.4797 +curr_acc_top5 0.8692 +BEST_ACC_top5 0.8821 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=5.23031 loss_avg=5.31913 acc=0.47656 acc_top1_avg=0.47266 acc_top5_avg=0.88086 lr=0.01000 gn=5.05223 time=52.75it/s +epoch=6 global_step=2400 loss=4.35395 loss_avg=5.37353 acc=0.57031 acc_top1_avg=0.45804 acc_top5_avg=0.88455 lr=0.01000 gn=4.64895 time=57.65it/s +epoch=6 global_step=2450 loss=5.90293 loss_avg=5.39771 acc=0.39062 acc_top1_avg=0.45816 acc_top5_avg=0.88672 lr=0.01000 gn=4.08746 time=58.29it/s +epoch=6 global_step=2500 loss=5.65867 loss_avg=5.40080 acc=0.43750 acc_top1_avg=0.45723 acc_top5_avg=0.88393 lr=0.01000 gn=4.95873 time=62.70it/s +epoch=6 global_step=2550 loss=5.76612 loss_avg=5.41300 acc=0.42969 acc_top1_avg=0.45630 acc_top5_avg=0.88408 lr=0.01000 gn=5.12417 time=50.39it/s +epoch=6 global_step=2600 loss=5.60960 loss_avg=5.39937 acc=0.44531 acc_top1_avg=0.45805 acc_top5_avg=0.88250 lr=0.01000 gn=5.02664 time=53.46it/s +epoch=6 global_step=2650 loss=5.74657 loss_avg=5.39481 acc=0.40625 acc_top1_avg=0.45821 acc_top5_avg=0.88425 lr=0.01000 gn=5.18781 time=33.77it/s +epoch=6 global_step=2700 loss=5.31053 loss_avg=5.39857 acc=0.45312 acc_top1_avg=0.45754 acc_top5_avg=0.88478 lr=0.01000 gn=5.04885 time=60.20it/s +====================Eval==================== +epoch=6 global_step=2737 loss=5.66717 test_loss_avg=2.72724 acc=0.00000 test_acc_avg=0.38882 test_acc_top5_avg=0.87921 time=220.94it/s +epoch=6 global_step=2737 loss=0.61339 test_loss_avg=2.61553 acc=0.82031 test_acc_avg=0.42773 test_acc_top5_avg=0.87459 time=243.97it/s +epoch=6 global_step=2737 loss=0.36399 test_loss_avg=2.53487 acc=0.87500 test_acc_avg=0.44422 test_acc_top5_avg=0.87915 time=491.25it/s +curr_acc 0.4442 +BEST_ACC 0.4797 +curr_acc_top5 0.8792 +BEST_ACC_top5 0.8821 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=5.38272 loss_avg=5.37476 acc=0.45312 acc_top1_avg=0.45252 acc_top5_avg=0.88642 lr=0.01000 gn=4.07341 time=58.90it/s +epoch=7 global_step=2800 loss=5.04318 loss_avg=5.34472 acc=0.51562 acc_top1_avg=0.46305 acc_top5_avg=0.88740 lr=0.01000 gn=4.83706 time=61.74it/s +epoch=7 global_step=2850 loss=5.42046 loss_avg=5.32656 acc=0.44531 acc_top1_avg=0.46446 acc_top5_avg=0.88703 lr=0.01000 gn=4.82274 time=54.77it/s +epoch=7 global_step=2900 loss=5.34890 loss_avg=5.31176 acc=0.46875 acc_top1_avg=0.46592 acc_top5_avg=0.88837 lr=0.01000 gn=5.39084 time=57.10it/s +epoch=7 global_step=2950 loss=4.87976 loss_avg=5.32617 acc=0.51562 acc_top1_avg=0.46428 acc_top5_avg=0.88729 lr=0.01000 gn=5.46073 time=54.75it/s +epoch=7 global_step=3000 loss=5.49782 loss_avg=5.32791 acc=0.45312 acc_top1_avg=0.46409 acc_top5_avg=0.88697 lr=0.01000 gn=5.08208 time=55.96it/s +epoch=7 global_step=3050 loss=5.25227 loss_avg=5.34392 acc=0.48438 acc_top1_avg=0.46261 acc_top5_avg=0.88653 lr=0.01000 gn=5.44298 time=53.23it/s +epoch=7 global_step=3100 loss=4.80594 loss_avg=5.33398 acc=0.50781 acc_top1_avg=0.46361 acc_top5_avg=0.88658 lr=0.01000 gn=5.57201 time=62.88it/s +====================Eval==================== +epoch=7 global_step=3128 loss=3.39635 test_loss_avg=2.91474 acc=0.04688 test_acc_avg=0.33361 test_acc_top5_avg=0.86154 time=241.36it/s +epoch=7 global_step=3128 loss=0.37471 test_loss_avg=2.55737 acc=0.87500 test_acc_avg=0.42316 test_acc_top5_avg=0.86452 time=843.58it/s +curr_acc 0.4232 +BEST_ACC 0.4797 +curr_acc_top5 0.8645 +BEST_ACC_top5 0.8821 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=5.40854 loss_avg=5.47278 acc=0.47656 acc_top1_avg=0.45170 acc_top5_avg=0.87820 lr=0.01000 gn=5.02800 time=57.07it/s +epoch=8 global_step=3200 loss=4.81809 loss_avg=5.35579 acc=0.52344 acc_top1_avg=0.46159 acc_top5_avg=0.88845 lr=0.01000 gn=5.42429 time=58.48it/s +epoch=8 global_step=3250 loss=5.28321 loss_avg=5.36282 acc=0.46094 acc_top1_avg=0.46119 acc_top5_avg=0.88800 lr=0.01000 gn=5.25472 time=60.62it/s +epoch=8 global_step=3300 loss=5.12279 loss_avg=5.34297 acc=0.47656 acc_top1_avg=0.46357 acc_top5_avg=0.88654 lr=0.01000 gn=5.35002 time=54.19it/s +epoch=8 global_step=3350 loss=5.42706 loss_avg=5.33599 acc=0.47656 acc_top1_avg=0.46382 acc_top5_avg=0.88721 lr=0.01000 gn=5.66403 time=61.64it/s +epoch=8 global_step=3400 loss=4.39024 loss_avg=5.31857 acc=0.58594 acc_top1_avg=0.46622 acc_top5_avg=0.88778 lr=0.01000 gn=5.54560 time=58.71it/s +epoch=8 global_step=3450 loss=5.12535 loss_avg=5.32246 acc=0.46875 acc_top1_avg=0.46509 acc_top5_avg=0.88752 lr=0.01000 gn=6.33406 time=51.05it/s +epoch=8 global_step=3500 loss=5.90424 loss_avg=5.31739 acc=0.36719 acc_top1_avg=0.46533 acc_top5_avg=0.88743 lr=0.01000 gn=5.77482 time=55.79it/s +====================Eval==================== +epoch=8 global_step=3519 loss=4.23602 test_loss_avg=1.93906 acc=0.03125 test_acc_avg=0.53255 test_acc_top5_avg=0.94792 time=241.30it/s +epoch=8 global_step=3519 loss=0.13882 test_loss_avg=2.86998 acc=0.95312 test_acc_avg=0.37868 test_acc_top5_avg=0.87351 time=235.67it/s +epoch=8 global_step=3519 loss=0.67273 test_loss_avg=2.54803 acc=0.87500 test_acc_avg=0.44521 test_acc_top5_avg=0.88924 time=585.88it/s +curr_acc 0.4452 +BEST_ACC 0.4797 +curr_acc_top5 0.8892 +BEST_ACC_top5 0.8821 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=4.71791 loss_avg=5.34721 acc=0.53125 acc_top1_avg=0.46421 acc_top5_avg=0.88584 lr=0.01000 gn=5.48878 time=59.94it/s +epoch=9 global_step=3600 loss=5.50441 loss_avg=5.27047 acc=0.46094 acc_top1_avg=0.47135 acc_top5_avg=0.89169 lr=0.01000 gn=5.02398 time=57.69it/s +epoch=9 global_step=3650 loss=4.33276 loss_avg=5.23340 acc=0.57031 acc_top1_avg=0.47519 acc_top5_avg=0.89265 lr=0.01000 gn=5.33261 time=56.16it/s +epoch=9 global_step=3700 loss=4.47963 loss_avg=5.25372 acc=0.53906 acc_top1_avg=0.47333 acc_top5_avg=0.89321 lr=0.01000 gn=4.87456 time=55.73it/s +epoch=9 global_step=3750 loss=5.41521 loss_avg=5.26749 acc=0.45312 acc_top1_avg=0.47152 acc_top5_avg=0.89100 lr=0.01000 gn=6.14419 time=50.67it/s +epoch=9 global_step=3800 loss=4.93295 loss_avg=5.26728 acc=0.50000 acc_top1_avg=0.47178 acc_top5_avg=0.89099 lr=0.01000 gn=4.35683 time=60.15it/s +epoch=9 global_step=3850 loss=5.82318 loss_avg=5.25758 acc=0.40625 acc_top1_avg=0.47293 acc_top5_avg=0.89211 lr=0.01000 gn=5.80420 time=53.99it/s +epoch=9 global_step=3900 loss=5.41634 loss_avg=5.25983 acc=0.45312 acc_top1_avg=0.47302 acc_top5_avg=0.89171 lr=0.01000 gn=5.49164 time=48.84it/s +====================Eval==================== +epoch=9 global_step=3910 loss=0.83909 test_loss_avg=2.74061 acc=0.78906 test_acc_avg=0.41567 test_acc_top5_avg=0.83293 time=243.06it/s +epoch=9 global_step=3910 loss=0.32639 test_loss_avg=2.22455 acc=0.81250 test_acc_avg=0.50148 test_acc_top5_avg=0.88558 time=510.32it/s +curr_acc 0.5015 +BEST_ACC 0.4797 +curr_acc_top5 0.8856 +BEST_ACC_top5 0.8892 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=5.32464 loss_avg=5.26595 acc=0.47656 acc_top1_avg=0.47168 acc_top5_avg=0.88848 lr=0.01000 gn=5.32857 time=51.99it/s +epoch=10 global_step=4000 loss=5.43288 loss_avg=5.22479 acc=0.45312 acc_top1_avg=0.47786 acc_top5_avg=0.89106 lr=0.01000 gn=6.68026 time=55.48it/s +epoch=10 global_step=4050 loss=4.86270 loss_avg=5.22867 acc=0.50000 acc_top1_avg=0.47740 acc_top5_avg=0.89174 lr=0.01000 gn=6.45277 time=60.83it/s +epoch=10 global_step=4100 loss=5.67887 loss_avg=5.22563 acc=0.41406 acc_top1_avg=0.47743 acc_top5_avg=0.89017 lr=0.01000 gn=5.45813 time=63.77it/s +epoch=10 global_step=4150 loss=4.58956 loss_avg=5.21069 acc=0.54688 acc_top1_avg=0.47887 acc_top5_avg=0.89111 lr=0.01000 gn=6.85562 time=57.53it/s +epoch=10 global_step=4200 loss=4.99319 loss_avg=5.23506 acc=0.50000 acc_top1_avg=0.47627 acc_top5_avg=0.89100 lr=0.01000 gn=5.91314 time=58.66it/s +epoch=10 global_step=4250 loss=5.19776 loss_avg=5.23713 acc=0.46875 acc_top1_avg=0.47560 acc_top5_avg=0.89049 lr=0.01000 gn=4.81902 time=57.75it/s +epoch=10 global_step=4300 loss=5.25115 loss_avg=5.23754 acc=0.47656 acc_top1_avg=0.47534 acc_top5_avg=0.89020 lr=0.01000 gn=6.28138 time=56.42it/s +====================Eval==================== +epoch=10 global_step=4301 loss=0.28783 test_loss_avg=1.12636 acc=0.90625 test_acc_avg=0.68203 test_acc_top5_avg=0.99531 time=243.25it/s +epoch=10 global_step=4301 loss=1.99801 test_loss_avg=3.11920 acc=0.54688 test_acc_avg=0.35000 test_acc_top5_avg=0.81589 time=240.97it/s +epoch=10 global_step=4301 loss=0.69448 test_loss_avg=2.55690 acc=0.81250 test_acc_avg=0.45995 test_acc_top5_avg=0.85314 time=619.09it/s +curr_acc 0.4599 +BEST_ACC 0.5015 +curr_acc_top5 0.8531 +BEST_ACC_top5 0.8892 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=4.78247 loss_avg=5.22924 acc=0.49219 acc_top1_avg=0.47529 acc_top5_avg=0.88967 lr=0.01000 gn=6.46782 time=54.37it/s +epoch=11 global_step=4400 loss=5.97048 loss_avg=5.22757 acc=0.40625 acc_top1_avg=0.47522 acc_top5_avg=0.88920 lr=0.01000 gn=6.48640 time=60.94it/s +epoch=11 global_step=4450 loss=5.49774 loss_avg=5.22001 acc=0.43750 acc_top1_avg=0.47735 acc_top5_avg=0.88989 lr=0.01000 gn=6.19779 time=60.45it/s +epoch=11 global_step=4500 loss=5.19662 loss_avg=5.22056 acc=0.46875 acc_top1_avg=0.47762 acc_top5_avg=0.89074 lr=0.01000 gn=6.69482 time=51.78it/s +epoch=11 global_step=4550 loss=5.57590 loss_avg=5.22730 acc=0.46094 acc_top1_avg=0.47741 acc_top5_avg=0.89022 lr=0.01000 gn=6.10131 time=60.07it/s +epoch=11 global_step=4600 loss=4.96040 loss_avg=5.23571 acc=0.49219 acc_top1_avg=0.47630 acc_top5_avg=0.89044 lr=0.01000 gn=7.25953 time=54.40it/s +epoch=11 global_step=4650 loss=4.35046 loss_avg=5.23207 acc=0.60156 acc_top1_avg=0.47629 acc_top5_avg=0.89076 lr=0.01000 gn=7.22722 time=53.91it/s +====================Eval==================== +epoch=11 global_step=4692 loss=5.17565 test_loss_avg=2.43860 acc=0.00000 test_acc_avg=0.44128 test_acc_top5_avg=0.81552 time=241.51it/s +epoch=11 global_step=4692 loss=1.51941 test_loss_avg=2.14824 acc=0.62500 test_acc_avg=0.49871 test_acc_top5_avg=0.87876 time=508.77it/s +curr_acc 0.4987 +BEST_ACC 0.5015 +curr_acc_top5 0.8788 +BEST_ACC_top5 0.8892 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=4.55065 loss_avg=5.05822 acc=0.52344 acc_top1_avg=0.49219 acc_top5_avg=0.89844 lr=0.01000 gn=5.25412 time=55.96it/s +epoch=12 global_step=4750 loss=5.06492 loss_avg=5.15459 acc=0.50000 acc_top1_avg=0.48249 acc_top5_avg=0.88699 lr=0.01000 gn=6.22295 time=52.61it/s +epoch=12 global_step=4800 loss=5.39345 loss_avg=5.15151 acc=0.46875 acc_top1_avg=0.48278 acc_top5_avg=0.89055 lr=0.01000 gn=6.81486 time=50.51it/s +epoch=12 global_step=4850 loss=4.96422 loss_avg=5.14977 acc=0.50000 acc_top1_avg=0.48428 acc_top5_avg=0.89156 lr=0.01000 gn=5.86639 time=55.69it/s +epoch=12 global_step=4900 loss=5.26563 loss_avg=5.17540 acc=0.46875 acc_top1_avg=0.48043 acc_top5_avg=0.89175 lr=0.01000 gn=5.18673 time=61.47it/s +epoch=12 global_step=4950 loss=5.69574 loss_avg=5.19048 acc=0.40625 acc_top1_avg=0.47838 acc_top5_avg=0.89111 lr=0.01000 gn=5.76320 time=60.94it/s +epoch=12 global_step=5000 loss=5.54122 loss_avg=5.19733 acc=0.42969 acc_top1_avg=0.47760 acc_top5_avg=0.89123 lr=0.01000 gn=5.27711 time=51.99it/s +epoch=12 global_step=5050 loss=5.53676 loss_avg=5.21525 acc=0.41406 acc_top1_avg=0.47584 acc_top5_avg=0.89124 lr=0.01000 gn=6.06638 time=54.51it/s +====================Eval==================== +epoch=12 global_step=5083 loss=1.99098 test_loss_avg=1.96492 acc=0.52344 test_acc_avg=0.53516 test_acc_top5_avg=0.89062 time=236.27it/s +epoch=12 global_step=5083 loss=6.44327 test_loss_avg=2.99650 acc=0.00000 test_acc_avg=0.44366 test_acc_top5_avg=0.89573 time=238.57it/s +epoch=12 global_step=5083 loss=0.53781 test_loss_avg=2.51663 acc=0.81250 test_acc_avg=0.52146 test_acc_top5_avg=0.91901 time=682.22it/s +curr_acc 0.5215 +BEST_ACC 0.5015 +curr_acc_top5 0.9190 +BEST_ACC_top5 0.8892 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=5.37794 loss_avg=5.08058 acc=0.45312 acc_top1_avg=0.49219 acc_top5_avg=0.88833 lr=0.01000 gn=5.30911 time=54.92it/s +epoch=13 global_step=5150 loss=5.00458 loss_avg=5.14320 acc=0.49219 acc_top1_avg=0.48519 acc_top5_avg=0.89074 lr=0.01000 gn=7.35226 time=47.95it/s +epoch=13 global_step=5200 loss=4.56545 loss_avg=5.16806 acc=0.57031 acc_top1_avg=0.48237 acc_top5_avg=0.89223 lr=0.01000 gn=5.75187 time=61.26it/s +epoch=13 global_step=5250 loss=5.96763 loss_avg=5.15743 acc=0.39062 acc_top1_avg=0.48260 acc_top5_avg=0.89577 lr=0.01000 gn=7.49215 time=53.79it/s +epoch=13 global_step=5300 loss=5.09985 loss_avg=5.17967 acc=0.49219 acc_top1_avg=0.48045 acc_top5_avg=0.89293 lr=0.01000 gn=5.23309 time=59.74it/s +epoch=13 global_step=5350 loss=5.15935 loss_avg=5.19523 acc=0.48438 acc_top1_avg=0.47890 acc_top5_avg=0.89209 lr=0.01000 gn=5.12841 time=54.13it/s +epoch=13 global_step=5400 loss=5.14334 loss_avg=5.20324 acc=0.47656 acc_top1_avg=0.47767 acc_top5_avg=0.89228 lr=0.01000 gn=5.85581 time=61.37it/s +epoch=13 global_step=5450 loss=5.72728 loss_avg=5.20443 acc=0.42969 acc_top1_avg=0.47729 acc_top5_avg=0.89207 lr=0.01000 gn=7.42186 time=54.95it/s +====================Eval==================== +epoch=13 global_step=5474 loss=3.00205 test_loss_avg=1.57023 acc=0.19531 test_acc_avg=0.57235 test_acc_top5_avg=0.94905 time=152.99it/s +epoch=13 global_step=5474 loss=0.60734 test_loss_avg=2.11801 acc=0.86719 test_acc_avg=0.53253 test_acc_top5_avg=0.87735 time=257.46it/s +epoch=13 global_step=5474 loss=0.10871 test_loss_avg=1.98285 acc=0.93750 test_acc_avg=0.56062 test_acc_top5_avg=0.88617 time=844.09it/s +curr_acc 0.5606 +BEST_ACC 0.5215 +curr_acc_top5 0.8862 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=5.56889 loss_avg=5.12100 acc=0.44531 acc_top1_avg=0.48287 acc_top5_avg=0.88732 lr=0.01000 gn=6.04274 time=57.78it/s +epoch=14 global_step=5550 loss=4.64666 loss_avg=5.14042 acc=0.53906 acc_top1_avg=0.48314 acc_top5_avg=0.89186 lr=0.01000 gn=5.58774 time=30.09it/s +epoch=14 global_step=5600 loss=5.30139 loss_avg=5.13092 acc=0.46875 acc_top1_avg=0.48338 acc_top5_avg=0.89583 lr=0.01000 gn=5.40869 time=54.71it/s +epoch=14 global_step=5650 loss=5.45790 loss_avg=5.15672 acc=0.42188 acc_top1_avg=0.47998 acc_top5_avg=0.89662 lr=0.01000 gn=7.54348 time=55.19it/s +epoch=14 global_step=5700 loss=4.90134 loss_avg=5.16692 acc=0.54688 acc_top1_avg=0.47902 acc_top5_avg=0.89730 lr=0.01000 gn=5.31978 time=62.75it/s +epoch=14 global_step=5750 loss=5.78932 loss_avg=5.16671 acc=0.39062 acc_top1_avg=0.47979 acc_top5_avg=0.89555 lr=0.01000 gn=5.59459 time=60.00it/s +epoch=14 global_step=5800 loss=5.17415 loss_avg=5.17250 acc=0.48438 acc_top1_avg=0.47956 acc_top5_avg=0.89592 lr=0.01000 gn=6.89263 time=55.13it/s +epoch=14 global_step=5850 loss=5.38993 loss_avg=5.19858 acc=0.47656 acc_top1_avg=0.47681 acc_top5_avg=0.89547 lr=0.01000 gn=7.06995 time=54.30it/s +====================Eval==================== +epoch=14 global_step=5865 loss=1.12255 test_loss_avg=2.39913 acc=0.68750 test_acc_avg=0.48757 test_acc_top5_avg=0.86754 time=240.86it/s +epoch=14 global_step=5865 loss=0.16774 test_loss_avg=2.18254 acc=0.87500 test_acc_avg=0.53145 test_acc_top5_avg=0.89241 time=507.78it/s +curr_acc 0.5314 +BEST_ACC 0.5606 +curr_acc_top5 0.8924 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=5.09444 loss_avg=5.15156 acc=0.50781 acc_top1_avg=0.48438 acc_top5_avg=0.89754 lr=0.01000 gn=5.76134 time=55.82it/s +epoch=15 global_step=5950 loss=5.18333 loss_avg=5.09560 acc=0.46875 acc_top1_avg=0.48869 acc_top5_avg=0.90028 lr=0.01000 gn=6.01264 time=55.76it/s +epoch=15 global_step=6000 loss=4.98950 loss_avg=5.09910 acc=0.52344 acc_top1_avg=0.48738 acc_top5_avg=0.90110 lr=0.01000 gn=6.81720 time=59.81it/s +epoch=15 global_step=6050 loss=5.25022 loss_avg=5.14944 acc=0.46875 acc_top1_avg=0.48188 acc_top5_avg=0.89827 lr=0.01000 gn=6.62163 time=56.79it/s +epoch=15 global_step=6100 loss=4.79763 loss_avg=5.16034 acc=0.51562 acc_top1_avg=0.48082 acc_top5_avg=0.89757 lr=0.01000 gn=7.59505 time=50.48it/s +epoch=15 global_step=6150 loss=5.29495 loss_avg=5.17154 acc=0.45312 acc_top1_avg=0.47969 acc_top5_avg=0.89671 lr=0.01000 gn=6.62494 time=61.16it/s +epoch=15 global_step=6200 loss=5.73299 loss_avg=5.18958 acc=0.38281 acc_top1_avg=0.47761 acc_top5_avg=0.89583 lr=0.01000 gn=5.55810 time=54.19it/s +epoch=15 global_step=6250 loss=5.07491 loss_avg=5.18674 acc=0.50000 acc_top1_avg=0.47760 acc_top5_avg=0.89619 lr=0.01000 gn=6.76013 time=61.11it/s +====================Eval==================== +epoch=15 global_step=6256 loss=0.25875 test_loss_avg=1.38681 acc=0.94531 test_acc_avg=0.64583 test_acc_top5_avg=0.98750 time=239.41it/s +epoch=15 global_step=6256 loss=0.11948 test_loss_avg=2.76740 acc=0.96094 test_acc_avg=0.41695 test_acc_top5_avg=0.85084 time=234.78it/s +epoch=15 global_step=6256 loss=2.39384 test_loss_avg=2.54726 acc=0.50000 test_acc_avg=0.46539 test_acc_top5_avg=0.86936 time=780.77it/s +curr_acc 0.4654 +BEST_ACC 0.5606 +curr_acc_top5 0.8694 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=5.66109 loss_avg=5.12970 acc=0.42969 acc_top1_avg=0.48704 acc_top5_avg=0.89595 lr=0.01000 gn=6.37063 time=51.64it/s +epoch=16 global_step=6350 loss=5.14367 loss_avg=5.16691 acc=0.47656 acc_top1_avg=0.48280 acc_top5_avg=0.89461 lr=0.01000 gn=6.09022 time=60.22it/s +epoch=16 global_step=6400 loss=4.97536 loss_avg=5.17101 acc=0.50000 acc_top1_avg=0.48134 acc_top5_avg=0.89317 lr=0.01000 gn=7.47850 time=54.61it/s +epoch=16 global_step=6450 loss=5.01674 loss_avg=5.16534 acc=0.47656 acc_top1_avg=0.48148 acc_top5_avg=0.89332 lr=0.01000 gn=6.03988 time=61.09it/s +epoch=16 global_step=6500 loss=5.48928 loss_avg=5.16785 acc=0.45312 acc_top1_avg=0.48165 acc_top5_avg=0.89354 lr=0.01000 gn=5.76436 time=27.98it/s +epoch=16 global_step=6550 loss=4.39389 loss_avg=5.16391 acc=0.56250 acc_top1_avg=0.48151 acc_top5_avg=0.89355 lr=0.01000 gn=6.93937 time=36.45it/s +epoch=16 global_step=6600 loss=5.33682 loss_avg=5.16245 acc=0.46875 acc_top1_avg=0.48167 acc_top5_avg=0.89474 lr=0.01000 gn=6.82089 time=51.81it/s +====================Eval==================== +epoch=16 global_step=6647 loss=2.07513 test_loss_avg=3.05615 acc=0.52344 test_acc_avg=0.40234 test_acc_top5_avg=0.83702 time=238.20it/s +epoch=16 global_step=6647 loss=0.17460 test_loss_avg=2.49573 acc=0.93750 test_acc_avg=0.48507 test_acc_top5_avg=0.88143 time=843.92it/s +curr_acc 0.4851 +BEST_ACC 0.5606 +curr_acc_top5 0.8814 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=4.55303 loss_avg=5.09738 acc=0.56250 acc_top1_avg=0.48698 acc_top5_avg=0.89323 lr=0.01000 gn=6.09520 time=32.13it/s +epoch=17 global_step=6700 loss=4.79148 loss_avg=5.01747 acc=0.52344 acc_top1_avg=0.49602 acc_top5_avg=0.90212 lr=0.01000 gn=5.83592 time=55.64it/s +epoch=17 global_step=6750 loss=5.38090 loss_avg=5.11947 acc=0.45312 acc_top1_avg=0.48642 acc_top5_avg=0.89844 lr=0.01000 gn=5.78803 time=54.60it/s +epoch=17 global_step=6800 loss=5.28569 loss_avg=5.12157 acc=0.46094 acc_top1_avg=0.48769 acc_top5_avg=0.89706 lr=0.01000 gn=5.62378 time=44.82it/s +epoch=17 global_step=6850 loss=5.18866 loss_avg=5.13088 acc=0.48438 acc_top1_avg=0.48622 acc_top5_avg=0.89636 lr=0.01000 gn=7.53040 time=54.84it/s +epoch=17 global_step=6900 loss=4.92947 loss_avg=5.12443 acc=0.50000 acc_top1_avg=0.48749 acc_top5_avg=0.89631 lr=0.01000 gn=5.77481 time=58.49it/s +epoch=17 global_step=6950 loss=4.90160 loss_avg=5.14488 acc=0.50781 acc_top1_avg=0.48438 acc_top5_avg=0.89558 lr=0.01000 gn=5.78700 time=63.50it/s +epoch=17 global_step=7000 loss=4.31449 loss_avg=5.14323 acc=0.57031 acc_top1_avg=0.48440 acc_top5_avg=0.89547 lr=0.01000 gn=7.53606 time=55.12it/s +====================Eval==================== +epoch=17 global_step=7038 loss=1.57744 test_loss_avg=1.61867 acc=0.64062 test_acc_avg=0.62388 test_acc_top5_avg=0.92857 time=116.89it/s +epoch=17 global_step=7038 loss=1.37309 test_loss_avg=3.21177 acc=0.64844 test_acc_avg=0.37390 test_acc_top5_avg=0.87706 time=242.96it/s +epoch=17 global_step=7038 loss=0.32849 test_loss_avg=2.50062 acc=0.93750 test_acc_avg=0.49842 test_acc_top5_avg=0.90852 time=801.20it/s +curr_acc 0.4984 +BEST_ACC 0.5606 +curr_acc_top5 0.9085 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=5.23993 loss_avg=5.02923 acc=0.46875 acc_top1_avg=0.49414 acc_top5_avg=0.89453 lr=0.01000 gn=6.38863 time=54.06it/s +epoch=18 global_step=7100 loss=5.06801 loss_avg=5.11932 acc=0.50000 acc_top1_avg=0.48677 acc_top5_avg=0.89756 lr=0.01000 gn=6.68342 time=55.43it/s +epoch=18 global_step=7150 loss=5.11861 loss_avg=5.13184 acc=0.46875 acc_top1_avg=0.48570 acc_top5_avg=0.89676 lr=0.01000 gn=6.90767 time=63.04it/s +epoch=18 global_step=7200 loss=5.47166 loss_avg=5.16011 acc=0.45312 acc_top1_avg=0.48206 acc_top5_avg=0.89612 lr=0.01000 gn=6.47801 time=63.18it/s +epoch=18 global_step=7250 loss=4.66709 loss_avg=5.14755 acc=0.53125 acc_top1_avg=0.48253 acc_top5_avg=0.89545 lr=0.01000 gn=6.43081 time=51.92it/s +epoch=18 global_step=7300 loss=5.75294 loss_avg=5.14742 acc=0.41406 acc_top1_avg=0.48193 acc_top5_avg=0.89540 lr=0.01000 gn=6.05678 time=52.09it/s +epoch=18 global_step=7350 loss=4.95905 loss_avg=5.14906 acc=0.50000 acc_top1_avg=0.48205 acc_top5_avg=0.89583 lr=0.01000 gn=7.21800 time=60.92it/s +epoch=18 global_step=7400 loss=5.88282 loss_avg=5.15925 acc=0.38281 acc_top1_avg=0.48116 acc_top5_avg=0.89637 lr=0.01000 gn=6.06638 time=61.57it/s +====================Eval==================== +epoch=18 global_step=7429 loss=5.82987 test_loss_avg=2.21143 acc=0.00000 test_acc_avg=0.46456 test_acc_top5_avg=0.88923 time=212.23it/s +epoch=18 global_step=7429 loss=0.84214 test_loss_avg=2.37758 acc=0.76562 test_acc_avg=0.46214 test_acc_top5_avg=0.88311 time=256.96it/s +epoch=18 global_step=7429 loss=0.47366 test_loss_avg=2.35348 acc=0.81250 test_acc_avg=0.46657 test_acc_top5_avg=0.88459 time=506.93it/s +curr_acc 0.4666 +BEST_ACC 0.5606 +curr_acc_top5 0.8846 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=5.10538 loss_avg=5.13649 acc=0.49219 acc_top1_avg=0.48698 acc_top5_avg=0.88876 lr=0.01000 gn=5.51248 time=54.54it/s +epoch=19 global_step=7500 loss=5.26559 loss_avg=5.11073 acc=0.47656 acc_top1_avg=0.48900 acc_top5_avg=0.89162 lr=0.01000 gn=6.60010 time=56.57it/s +epoch=19 global_step=7550 loss=5.24146 loss_avg=5.08854 acc=0.49219 acc_top1_avg=0.49264 acc_top5_avg=0.89443 lr=0.01000 gn=7.45831 time=52.22it/s +epoch=19 global_step=7600 loss=5.11279 loss_avg=5.12264 acc=0.46875 acc_top1_avg=0.48771 acc_top5_avg=0.89515 lr=0.01000 gn=6.36364 time=57.23it/s +epoch=19 global_step=7650 loss=5.51034 loss_avg=5.12380 acc=0.44531 acc_top1_avg=0.48681 acc_top5_avg=0.89610 lr=0.01000 gn=7.10449 time=51.92it/s +epoch=19 global_step=7700 loss=5.45889 loss_avg=5.12402 acc=0.44531 acc_top1_avg=0.48683 acc_top5_avg=0.89746 lr=0.01000 gn=6.27729 time=63.07it/s +epoch=19 global_step=7750 loss=5.22031 loss_avg=5.14266 acc=0.47656 acc_top1_avg=0.48457 acc_top5_avg=0.89605 lr=0.01000 gn=5.87607 time=59.18it/s +epoch=19 global_step=7800 loss=4.75709 loss_avg=5.14281 acc=0.53125 acc_top1_avg=0.48446 acc_top5_avg=0.89631 lr=0.01000 gn=5.54106 time=54.48it/s +====================Eval==================== +epoch=19 global_step=7820 loss=5.36330 test_loss_avg=2.14763 acc=0.00000 test_acc_avg=0.54576 test_acc_top5_avg=0.85172 time=236.25it/s +epoch=19 global_step=7820 loss=0.03267 test_loss_avg=1.96933 acc=1.00000 test_acc_avg=0.58386 test_acc_top5_avg=0.88083 time=840.88it/s +curr_acc 0.5839 +BEST_ACC 0.5606 +curr_acc_top5 0.8808 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=5.19641 loss_avg=5.03420 acc=0.49219 acc_top1_avg=0.49974 acc_top5_avg=0.90339 lr=0.01000 gn=7.45958 time=54.46it/s +epoch=20 global_step=7900 loss=5.57209 loss_avg=5.06342 acc=0.44531 acc_top1_avg=0.49365 acc_top5_avg=0.90088 lr=0.01000 gn=6.04847 time=58.17it/s +epoch=20 global_step=7950 loss=5.14892 loss_avg=5.06892 acc=0.47656 acc_top1_avg=0.49237 acc_top5_avg=0.89615 lr=0.01000 gn=6.05349 time=59.85it/s +epoch=20 global_step=8000 loss=5.29852 loss_avg=5.09416 acc=0.46094 acc_top1_avg=0.48945 acc_top5_avg=0.89657 lr=0.01000 gn=7.05898 time=60.59it/s +epoch=20 global_step=8050 loss=5.00085 loss_avg=5.11347 acc=0.53125 acc_top1_avg=0.48787 acc_top5_avg=0.89677 lr=0.01000 gn=6.93285 time=32.01it/s +epoch=20 global_step=8100 loss=5.29940 loss_avg=5.11946 acc=0.46094 acc_top1_avg=0.48700 acc_top5_avg=0.89618 lr=0.01000 gn=6.90297 time=51.17it/s +epoch=20 global_step=8150 loss=5.28758 loss_avg=5.12656 acc=0.47656 acc_top1_avg=0.48601 acc_top5_avg=0.89598 lr=0.01000 gn=6.94896 time=61.82it/s +epoch=20 global_step=8200 loss=5.21005 loss_avg=5.14117 acc=0.46875 acc_top1_avg=0.48462 acc_top5_avg=0.89523 lr=0.01000 gn=6.67435 time=56.68it/s +====================Eval==================== +epoch=20 global_step=8211 loss=2.48375 test_loss_avg=1.92122 acc=0.42969 test_acc_avg=0.52305 test_acc_top5_avg=0.91133 time=239.01it/s +epoch=20 global_step=8211 loss=0.69899 test_loss_avg=2.25482 acc=0.80469 test_acc_avg=0.51908 test_acc_top5_avg=0.89353 time=243.42it/s +epoch=20 global_step=8211 loss=0.25003 test_loss_avg=2.03097 acc=0.87500 test_acc_avg=0.56240 test_acc_top5_avg=0.90516 time=855.11it/s +curr_acc 0.5624 +BEST_ACC 0.5839 +curr_acc_top5 0.9052 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=6.01417 loss_avg=5.09376 acc=0.39062 acc_top1_avg=0.49079 acc_top5_avg=0.89143 lr=0.01000 gn=6.70571 time=56.97it/s +epoch=21 global_step=8300 loss=4.84218 loss_avg=5.11641 acc=0.52344 acc_top1_avg=0.48771 acc_top5_avg=0.89326 lr=0.01000 gn=6.06394 time=50.67it/s +epoch=21 global_step=8350 loss=5.00377 loss_avg=5.11624 acc=0.49219 acc_top1_avg=0.48837 acc_top5_avg=0.89237 lr=0.01000 gn=6.94958 time=57.04it/s +epoch=21 global_step=8400 loss=5.11217 loss_avg=5.12850 acc=0.50000 acc_top1_avg=0.48665 acc_top5_avg=0.89306 lr=0.01000 gn=7.37671 time=59.49it/s +epoch=21 global_step=8450 loss=4.80137 loss_avg=5.11489 acc=0.50000 acc_top1_avg=0.48791 acc_top5_avg=0.89295 lr=0.01000 gn=6.34566 time=59.44it/s +epoch=21 global_step=8500 loss=4.96450 loss_avg=5.12909 acc=0.51562 acc_top1_avg=0.48570 acc_top5_avg=0.89292 lr=0.01000 gn=6.65432 time=59.45it/s +epoch=21 global_step=8550 loss=4.93511 loss_avg=5.13200 acc=0.49219 acc_top1_avg=0.48564 acc_top5_avg=0.89334 lr=0.01000 gn=6.10193 time=57.42it/s +epoch=21 global_step=8600 loss=5.19932 loss_avg=5.13652 acc=0.48438 acc_top1_avg=0.48538 acc_top5_avg=0.89318 lr=0.01000 gn=6.64471 time=60.66it/s +====================Eval==================== +epoch=21 global_step=8602 loss=2.26043 test_loss_avg=2.19427 acc=0.35938 test_acc_avg=0.50953 test_acc_top5_avg=0.83746 time=239.28it/s +epoch=21 global_step=8602 loss=0.83323 test_loss_avg=2.16570 acc=0.68750 test_acc_avg=0.52581 test_acc_top5_avg=0.87144 time=473.08it/s +curr_acc 0.5258 +BEST_ACC 0.5839 +curr_acc_top5 0.8714 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=4.30257 loss_avg=5.05538 acc=0.57812 acc_top1_avg=0.49349 acc_top5_avg=0.89665 lr=0.01000 gn=6.76326 time=35.27it/s +epoch=22 global_step=8700 loss=5.83393 loss_avg=5.04304 acc=0.39062 acc_top1_avg=0.49609 acc_top5_avg=0.90067 lr=0.01000 gn=6.25971 time=23.51it/s +epoch=22 global_step=8750 loss=5.03081 loss_avg=5.06819 acc=0.49219 acc_top1_avg=0.49213 acc_top5_avg=0.89923 lr=0.01000 gn=6.84542 time=55.62it/s +epoch=22 global_step=8800 loss=5.05401 loss_avg=5.09499 acc=0.49219 acc_top1_avg=0.48907 acc_top5_avg=0.89919 lr=0.01000 gn=5.68359 time=51.63it/s +epoch=22 global_step=8850 loss=4.76832 loss_avg=5.09048 acc=0.50781 acc_top1_avg=0.48976 acc_top5_avg=0.89885 lr=0.01000 gn=6.59359 time=36.70it/s +epoch=22 global_step=8900 loss=5.63822 loss_avg=5.08763 acc=0.41406 acc_top1_avg=0.48962 acc_top5_avg=0.89920 lr=0.01000 gn=6.43837 time=51.66it/s +epoch=22 global_step=8950 loss=5.22314 loss_avg=5.08493 acc=0.47656 acc_top1_avg=0.49014 acc_top5_avg=0.89886 lr=0.01000 gn=5.62373 time=50.36it/s +====================Eval==================== +epoch=22 global_step=8993 loss=0.39223 test_loss_avg=1.40320 acc=0.89062 test_acc_avg=0.64583 test_acc_top5_avg=0.95117 time=242.77it/s +epoch=22 global_step=8993 loss=0.63435 test_loss_avg=2.72915 acc=0.85156 test_acc_avg=0.42805 test_acc_top5_avg=0.85862 time=254.40it/s +epoch=22 global_step=8993 loss=0.02352 test_loss_avg=2.21059 acc=1.00000 test_acc_avg=0.53362 test_acc_top5_avg=0.88766 time=684.11it/s +curr_acc 0.5336 +BEST_ACC 0.5839 +curr_acc_top5 0.8877 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=4.90740 loss_avg=4.97186 acc=0.50781 acc_top1_avg=0.50446 acc_top5_avg=0.90848 lr=0.01000 gn=8.02076 time=51.40it/s +epoch=23 global_step=9050 loss=5.12509 loss_avg=5.14969 acc=0.48438 acc_top1_avg=0.48396 acc_top5_avg=0.89350 lr=0.01000 gn=7.13675 time=40.42it/s +epoch=23 global_step=9100 loss=5.23479 loss_avg=5.13894 acc=0.50781 acc_top1_avg=0.48481 acc_top5_avg=0.89690 lr=0.01000 gn=7.42876 time=59.52it/s +epoch=23 global_step=9150 loss=5.70789 loss_avg=5.14683 acc=0.43750 acc_top1_avg=0.48363 acc_top5_avg=0.89709 lr=0.01000 gn=7.56144 time=53.63it/s +epoch=23 global_step=9200 loss=5.42590 loss_avg=5.12979 acc=0.42969 acc_top1_avg=0.48554 acc_top5_avg=0.89659 lr=0.01000 gn=7.11395 time=53.76it/s +epoch=23 global_step=9250 loss=5.22562 loss_avg=5.13011 acc=0.46875 acc_top1_avg=0.48520 acc_top5_avg=0.89573 lr=0.01000 gn=5.29183 time=56.93it/s +epoch=23 global_step=9300 loss=5.75179 loss_avg=5.12585 acc=0.43750 acc_top1_avg=0.48544 acc_top5_avg=0.89592 lr=0.01000 gn=5.48842 time=58.05it/s +epoch=23 global_step=9350 loss=4.39441 loss_avg=5.12118 acc=0.57031 acc_top1_avg=0.48615 acc_top5_avg=0.89625 lr=0.01000 gn=6.37713 time=59.32it/s +====================Eval==================== +epoch=23 global_step=9384 loss=0.97866 test_loss_avg=2.95305 acc=0.77344 test_acc_avg=0.36245 test_acc_top5_avg=0.81487 time=244.28it/s +epoch=23 global_step=9384 loss=0.23184 test_loss_avg=2.20477 acc=0.87500 test_acc_avg=0.52146 test_acc_top5_avg=0.88983 time=667.46it/s +curr_acc 0.5215 +BEST_ACC 0.5839 +curr_acc_top5 0.8898 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=4.39152 loss_avg=5.08550 acc=0.57031 acc_top1_avg=0.48828 acc_top5_avg=0.89014 lr=0.01000 gn=6.15498 time=63.09it/s +epoch=24 global_step=9450 loss=5.51197 loss_avg=5.09089 acc=0.40625 acc_top1_avg=0.48828 acc_top5_avg=0.89370 lr=0.01000 gn=5.90395 time=54.56it/s +epoch=24 global_step=9500 loss=5.46982 loss_avg=5.07998 acc=0.46875 acc_top1_avg=0.49037 acc_top5_avg=0.89702 lr=0.01000 gn=6.90262 time=58.58it/s +epoch=24 global_step=9550 loss=5.27923 loss_avg=5.09581 acc=0.47656 acc_top1_avg=0.48899 acc_top5_avg=0.89726 lr=0.01000 gn=9.27512 time=63.24it/s +epoch=24 global_step=9600 loss=4.96090 loss_avg=5.09548 acc=0.49219 acc_top1_avg=0.48872 acc_top5_avg=0.89663 lr=0.01000 gn=8.13965 time=61.15it/s +epoch=24 global_step=9650 loss=4.63950 loss_avg=5.08639 acc=0.54688 acc_top1_avg=0.49025 acc_top5_avg=0.89762 lr=0.01000 gn=6.10589 time=54.88it/s +epoch=24 global_step=9700 loss=5.61302 loss_avg=5.09329 acc=0.41406 acc_top1_avg=0.48994 acc_top5_avg=0.89765 lr=0.01000 gn=7.04365 time=55.02it/s +epoch=24 global_step=9750 loss=5.14687 loss_avg=5.10722 acc=0.48438 acc_top1_avg=0.48847 acc_top5_avg=0.89724 lr=0.01000 gn=6.91369 time=54.75it/s +====================Eval==================== +epoch=24 global_step=9775 loss=3.03545 test_loss_avg=3.07915 acc=0.33594 test_acc_avg=0.31055 test_acc_top5_avg=0.74609 time=63.05it/s +epoch=24 global_step=9775 loss=4.96287 test_loss_avg=2.77303 acc=0.00000 test_acc_avg=0.43099 test_acc_top5_avg=0.85344 time=238.12it/s +epoch=24 global_step=9775 loss=0.05293 test_loss_avg=2.18538 acc=0.93750 test_acc_avg=0.54193 test_acc_top5_avg=0.88795 time=507.54it/s +curr_acc 0.5419 +BEST_ACC 0.5839 +curr_acc_top5 0.8880 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=5.11923 loss_avg=5.07315 acc=0.48438 acc_top1_avg=0.48812 acc_top5_avg=0.89125 lr=0.01000 gn=6.22849 time=56.93it/s +epoch=25 global_step=9850 loss=4.41102 loss_avg=5.08895 acc=0.57031 acc_top1_avg=0.48615 acc_top5_avg=0.89490 lr=0.01000 gn=7.72517 time=54.60it/s +epoch=25 global_step=9900 loss=5.03307 loss_avg=5.13190 acc=0.48438 acc_top1_avg=0.48094 acc_top5_avg=0.89744 lr=0.01000 gn=5.64413 time=53.77it/s +epoch=25 global_step=9950 loss=4.68736 loss_avg=5.12936 acc=0.51562 acc_top1_avg=0.48161 acc_top5_avg=0.89580 lr=0.01000 gn=8.29923 time=50.94it/s +epoch=25 global_step=10000 loss=5.24030 loss_avg=5.12322 acc=0.46875 acc_top1_avg=0.48243 acc_top5_avg=0.89660 lr=0.01000 gn=7.35389 time=43.10it/s +epoch=25 global_step=10050 loss=5.59568 loss_avg=5.13506 acc=0.42969 acc_top1_avg=0.48182 acc_top5_avg=0.89605 lr=0.01000 gn=6.68163 time=55.29it/s +epoch=25 global_step=10100 loss=5.13103 loss_avg=5.13880 acc=0.47656 acc_top1_avg=0.48190 acc_top5_avg=0.89608 lr=0.01000 gn=6.86966 time=58.68it/s +epoch=25 global_step=10150 loss=5.96907 loss_avg=5.12739 acc=0.39062 acc_top1_avg=0.48340 acc_top5_avg=0.89525 lr=0.01000 gn=6.60341 time=50.76it/s +====================Eval==================== +epoch=25 global_step=10166 loss=6.27137 test_loss_avg=2.36673 acc=0.00000 test_acc_avg=0.53375 test_acc_top5_avg=0.85156 time=244.67it/s +epoch=25 global_step=10166 loss=0.15980 test_loss_avg=2.89664 acc=0.95312 test_acc_avg=0.48104 test_acc_top5_avg=0.81667 time=235.19it/s +epoch=25 global_step=10166 loss=0.10598 test_loss_avg=2.75871 acc=0.93750 test_acc_avg=0.50465 test_acc_top5_avg=0.82575 time=505.28it/s +curr_acc 0.5046 +BEST_ACC 0.5839 +curr_acc_top5 0.8258 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=5.07270 loss_avg=5.05608 acc=0.46094 acc_top1_avg=0.49701 acc_top5_avg=0.90349 lr=0.01000 gn=6.94249 time=56.35it/s +epoch=26 global_step=10250 loss=5.50324 loss_avg=5.03280 acc=0.43750 acc_top1_avg=0.49702 acc_top5_avg=0.89927 lr=0.01000 gn=6.72059 time=62.92it/s +epoch=26 global_step=10300 loss=5.09546 loss_avg=5.01030 acc=0.49219 acc_top1_avg=0.50023 acc_top5_avg=0.90071 lr=0.01000 gn=7.07813 time=63.37it/s +epoch=26 global_step=10350 loss=5.30252 loss_avg=5.03693 acc=0.47656 acc_top1_avg=0.49783 acc_top5_avg=0.89971 lr=0.01000 gn=6.37269 time=57.74it/s +epoch=26 global_step=10400 loss=4.93346 loss_avg=5.08091 acc=0.53125 acc_top1_avg=0.49296 acc_top5_avg=0.89784 lr=0.01000 gn=7.16296 time=54.00it/s +epoch=26 global_step=10450 loss=4.93462 loss_avg=5.08430 acc=0.49219 acc_top1_avg=0.49255 acc_top5_avg=0.89756 lr=0.01000 gn=8.56107 time=61.61it/s +epoch=26 global_step=10500 loss=5.19557 loss_avg=5.08241 acc=0.49219 acc_top1_avg=0.49240 acc_top5_avg=0.89764 lr=0.01000 gn=6.05328 time=56.59it/s +epoch=26 global_step=10550 loss=4.89571 loss_avg=5.09255 acc=0.51562 acc_top1_avg=0.49131 acc_top5_avg=0.89705 lr=0.01000 gn=5.00693 time=63.93it/s +====================Eval==================== +epoch=26 global_step=10557 loss=1.21231 test_loss_avg=2.04178 acc=0.63281 test_acc_avg=0.53974 test_acc_top5_avg=0.82812 time=237.54it/s +epoch=26 global_step=10557 loss=0.01735 test_loss_avg=1.88358 acc=1.00000 test_acc_avg=0.58782 test_acc_top5_avg=0.88370 time=638.69it/s +curr_acc 0.5878 +BEST_ACC 0.5839 +curr_acc_top5 0.8837 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=4.58441 loss_avg=5.15439 acc=0.56250 acc_top1_avg=0.48347 acc_top5_avg=0.89353 lr=0.01000 gn=6.44094 time=49.32it/s +epoch=27 global_step=10650 loss=5.62742 loss_avg=5.07538 acc=0.42188 acc_top1_avg=0.49252 acc_top5_avg=0.89751 lr=0.01000 gn=7.01572 time=59.55it/s +epoch=27 global_step=10700 loss=5.87799 loss_avg=5.09027 acc=0.39844 acc_top1_avg=0.49022 acc_top5_avg=0.89592 lr=0.01000 gn=6.23780 time=43.96it/s +epoch=27 global_step=10750 loss=5.36157 loss_avg=5.08534 acc=0.46094 acc_top1_avg=0.49008 acc_top5_avg=0.89791 lr=0.01000 gn=8.38902 time=60.79it/s +epoch=27 global_step=10800 loss=5.20210 loss_avg=5.10378 acc=0.47656 acc_top1_avg=0.48746 acc_top5_avg=0.89718 lr=0.01000 gn=7.16682 time=62.97it/s +epoch=27 global_step=10850 loss=5.28605 loss_avg=5.10624 acc=0.42969 acc_top1_avg=0.48752 acc_top5_avg=0.89742 lr=0.01000 gn=7.17710 time=56.44it/s +epoch=27 global_step=10900 loss=5.22541 loss_avg=5.10258 acc=0.46875 acc_top1_avg=0.48829 acc_top5_avg=0.89689 lr=0.01000 gn=7.64595 time=58.41it/s +====================Eval==================== +epoch=27 global_step=10948 loss=3.01814 test_loss_avg=1.74788 acc=0.31250 test_acc_avg=0.58778 test_acc_top5_avg=0.96094 time=242.38it/s +epoch=27 global_step=10948 loss=0.49912 test_loss_avg=2.71413 acc=0.88281 test_acc_avg=0.42596 test_acc_top5_avg=0.88514 time=232.00it/s +epoch=27 global_step=10948 loss=0.08771 test_loss_avg=2.32955 acc=0.93750 test_acc_avg=0.50435 test_acc_top5_avg=0.90259 time=500.87it/s +curr_acc 0.5044 +BEST_ACC 0.5878 +curr_acc_top5 0.9026 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=5.83285 loss_avg=5.54305 acc=0.42188 acc_top1_avg=0.44922 acc_top5_avg=0.91016 lr=0.01000 gn=6.74368 time=50.15it/s +epoch=28 global_step=11000 loss=4.91209 loss_avg=5.11542 acc=0.52344 acc_top1_avg=0.48813 acc_top5_avg=0.89468 lr=0.01000 gn=6.63533 time=54.63it/s +epoch=28 global_step=11050 loss=4.89918 loss_avg=5.09658 acc=0.49219 acc_top1_avg=0.49081 acc_top5_avg=0.89668 lr=0.01000 gn=6.47208 time=55.39it/s +epoch=28 global_step=11100 loss=5.37098 loss_avg=5.10821 acc=0.45312 acc_top1_avg=0.48890 acc_top5_avg=0.89690 lr=0.01000 gn=8.24043 time=50.62it/s +epoch=28 global_step=11150 loss=5.40067 loss_avg=5.06591 acc=0.46875 acc_top1_avg=0.49362 acc_top5_avg=0.89697 lr=0.01000 gn=8.78546 time=54.58it/s +epoch=28 global_step=11200 loss=4.98788 loss_avg=5.07157 acc=0.48438 acc_top1_avg=0.49278 acc_top5_avg=0.89723 lr=0.01000 gn=6.52393 time=59.84it/s +epoch=28 global_step=11250 loss=4.83491 loss_avg=5.08325 acc=0.51562 acc_top1_avg=0.49108 acc_top5_avg=0.89725 lr=0.01000 gn=6.35298 time=59.34it/s +epoch=28 global_step=11300 loss=5.30372 loss_avg=5.09633 acc=0.48438 acc_top1_avg=0.48957 acc_top5_avg=0.89691 lr=0.01000 gn=8.87939 time=57.00it/s +====================Eval==================== +epoch=28 global_step=11339 loss=0.39355 test_loss_avg=2.35900 acc=0.90625 test_acc_avg=0.50123 test_acc_top5_avg=0.83676 time=240.89it/s +epoch=28 global_step=11339 loss=0.37998 test_loss_avg=2.12806 acc=0.87500 test_acc_avg=0.54757 test_acc_top5_avg=0.87816 time=851.29it/s +curr_acc 0.5476 +BEST_ACC 0.5878 +curr_acc_top5 0.8782 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=5.17005 loss_avg=5.32641 acc=0.50781 acc_top1_avg=0.46165 acc_top5_avg=0.90270 lr=0.01000 gn=5.81374 time=57.44it/s +epoch=29 global_step=11400 loss=5.29592 loss_avg=5.08706 acc=0.47656 acc_top1_avg=0.48835 acc_top5_avg=0.89677 lr=0.01000 gn=7.14173 time=61.39it/s +epoch=29 global_step=11450 loss=5.55740 loss_avg=5.11286 acc=0.43750 acc_top1_avg=0.48867 acc_top5_avg=0.89541 lr=0.01000 gn=6.01493 time=57.48it/s +epoch=29 global_step=11500 loss=5.70267 loss_avg=5.11731 acc=0.42969 acc_top1_avg=0.48695 acc_top5_avg=0.89528 lr=0.01000 gn=7.46681 time=62.68it/s +epoch=29 global_step=11550 loss=4.23101 loss_avg=5.10595 acc=0.60156 acc_top1_avg=0.48867 acc_top5_avg=0.89636 lr=0.01000 gn=5.84501 time=52.59it/s +epoch=29 global_step=11600 loss=5.24217 loss_avg=5.10630 acc=0.46875 acc_top1_avg=0.48818 acc_top5_avg=0.89691 lr=0.01000 gn=6.37047 time=49.13it/s +epoch=29 global_step=11650 loss=4.81313 loss_avg=5.10642 acc=0.53125 acc_top1_avg=0.48839 acc_top5_avg=0.89713 lr=0.01000 gn=7.43543 time=55.37it/s +epoch=29 global_step=11700 loss=5.28281 loss_avg=5.10485 acc=0.45312 acc_top1_avg=0.48857 acc_top5_avg=0.89762 lr=0.01000 gn=6.13412 time=53.68it/s +====================Eval==================== +epoch=29 global_step=11730 loss=0.92001 test_loss_avg=2.29345 acc=0.75781 test_acc_avg=0.46354 test_acc_top5_avg=0.85851 time=241.98it/s +epoch=29 global_step=11730 loss=0.59153 test_loss_avg=2.81648 acc=0.82812 test_acc_avg=0.45670 test_acc_top5_avg=0.85501 time=245.74it/s +epoch=29 global_step=11730 loss=0.19297 test_loss_avg=2.21265 acc=0.93750 test_acc_avg=0.56141 test_acc_top5_avg=0.88964 time=455.65it/s +curr_acc 0.5614 +BEST_ACC 0.5878 +curr_acc_top5 0.8896 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=5.03869 loss_avg=5.06737 acc=0.49219 acc_top1_avg=0.49414 acc_top5_avg=0.90117 lr=0.01000 gn=5.54410 time=54.77it/s +epoch=30 global_step=11800 loss=5.85853 loss_avg=5.08550 acc=0.39844 acc_top1_avg=0.49342 acc_top5_avg=0.90033 lr=0.01000 gn=5.26239 time=59.09it/s +epoch=30 global_step=11850 loss=5.56119 loss_avg=5.09015 acc=0.44531 acc_top1_avg=0.49212 acc_top5_avg=0.89668 lr=0.01000 gn=7.67713 time=58.00it/s +epoch=30 global_step=11900 loss=4.75368 loss_avg=5.09881 acc=0.54688 acc_top1_avg=0.49145 acc_top5_avg=0.89577 lr=0.01000 gn=8.43809 time=60.78it/s +epoch=30 global_step=11950 loss=5.82525 loss_avg=5.09128 acc=0.39062 acc_top1_avg=0.49208 acc_top5_avg=0.89790 lr=0.01000 gn=7.21366 time=56.63it/s +epoch=30 global_step=12000 loss=5.42154 loss_avg=5.08718 acc=0.45312 acc_top1_avg=0.49265 acc_top5_avg=0.89734 lr=0.01000 gn=7.26429 time=55.03it/s +epoch=30 global_step=12050 loss=4.63726 loss_avg=5.07810 acc=0.55469 acc_top1_avg=0.49336 acc_top5_avg=0.89629 lr=0.01000 gn=6.37286 time=52.35it/s +epoch=30 global_step=12100 loss=5.22075 loss_avg=5.09218 acc=0.48438 acc_top1_avg=0.49162 acc_top5_avg=0.89603 lr=0.01000 gn=5.88062 time=59.65it/s +====================Eval==================== +epoch=30 global_step=12121 loss=6.01524 test_loss_avg=2.87726 acc=0.00000 test_acc_avg=0.44062 test_acc_top5_avg=0.82578 time=226.06it/s +epoch=30 global_step=12121 loss=0.00719 test_loss_avg=2.11405 acc=1.00000 test_acc_avg=0.56863 test_acc_top5_avg=0.88687 time=463.41it/s +curr_acc 0.5686 +BEST_ACC 0.5878 +curr_acc_top5 0.8869 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=4.86922 loss_avg=5.05838 acc=0.49219 acc_top1_avg=0.48976 acc_top5_avg=0.89574 lr=0.01000 gn=7.06957 time=54.35it/s +epoch=31 global_step=12200 loss=4.77679 loss_avg=5.06963 acc=0.52344 acc_top1_avg=0.49031 acc_top5_avg=0.89864 lr=0.01000 gn=6.38286 time=41.44it/s +epoch=31 global_step=12250 loss=5.18246 loss_avg=5.10628 acc=0.47656 acc_top1_avg=0.48758 acc_top5_avg=0.89947 lr=0.01000 gn=6.98220 time=55.90it/s +epoch=31 global_step=12300 loss=5.10619 loss_avg=5.06943 acc=0.49219 acc_top1_avg=0.49179 acc_top5_avg=0.89997 lr=0.01000 gn=7.13282 time=54.92it/s +epoch=31 global_step=12350 loss=4.75943 loss_avg=5.07699 acc=0.51562 acc_top1_avg=0.49191 acc_top5_avg=0.89909 lr=0.01000 gn=7.71723 time=55.81it/s +epoch=31 global_step=12400 loss=4.34820 loss_avg=5.06554 acc=0.57812 acc_top1_avg=0.49322 acc_top5_avg=0.89942 lr=0.01000 gn=6.64779 time=59.88it/s +epoch=31 global_step=12450 loss=4.81883 loss_avg=5.08915 acc=0.49219 acc_top1_avg=0.49050 acc_top5_avg=0.89794 lr=0.01000 gn=5.79689 time=63.21it/s +epoch=31 global_step=12500 loss=5.42420 loss_avg=5.09813 acc=0.47656 acc_top1_avg=0.48938 acc_top5_avg=0.89664 lr=0.01000 gn=6.28578 time=58.81it/s +====================Eval==================== +epoch=31 global_step=12512 loss=3.04685 test_loss_avg=3.04685 acc=0.31250 test_acc_avg=0.31250 test_acc_top5_avg=0.91406 time=212.59it/s +epoch=31 global_step=12512 loss=5.02299 test_loss_avg=2.89919 acc=0.00000 test_acc_avg=0.37669 test_acc_top5_avg=0.81572 time=236.33it/s +epoch=31 global_step=12512 loss=0.00017 test_loss_avg=2.34350 acc=1.00000 test_acc_avg=0.48932 test_acc_top5_avg=0.85928 time=464.95it/s +curr_acc 0.4893 +BEST_ACC 0.5878 +curr_acc_top5 0.8593 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=5.35623 loss_avg=5.11538 acc=0.44531 acc_top1_avg=0.49075 acc_top5_avg=0.89515 lr=0.01000 gn=7.58401 time=60.64it/s +epoch=32 global_step=12600 loss=4.97252 loss_avg=5.06213 acc=0.51562 acc_top1_avg=0.49281 acc_top5_avg=0.89640 lr=0.01000 gn=7.46202 time=63.19it/s +epoch=32 global_step=12650 loss=5.10298 loss_avg=5.05268 acc=0.49219 acc_top1_avg=0.49400 acc_top5_avg=0.89714 lr=0.01000 gn=6.47954 time=54.87it/s +epoch=32 global_step=12700 loss=4.40050 loss_avg=5.06687 acc=0.56250 acc_top1_avg=0.49194 acc_top5_avg=0.89773 lr=0.01000 gn=8.13202 time=53.87it/s +epoch=32 global_step=12750 loss=5.01672 loss_avg=5.06729 acc=0.49219 acc_top1_avg=0.49209 acc_top5_avg=0.89867 lr=0.01000 gn=7.50807 time=55.98it/s +epoch=32 global_step=12800 loss=4.91476 loss_avg=5.08247 acc=0.52344 acc_top1_avg=0.49061 acc_top5_avg=0.89792 lr=0.01000 gn=7.24167 time=57.64it/s +epoch=32 global_step=12850 loss=5.33994 loss_avg=5.09366 acc=0.47656 acc_top1_avg=0.48937 acc_top5_avg=0.89705 lr=0.01000 gn=7.38404 time=53.52it/s +epoch=32 global_step=12900 loss=5.22892 loss_avg=5.10166 acc=0.47656 acc_top1_avg=0.48893 acc_top5_avg=0.89719 lr=0.01000 gn=6.23608 time=62.92it/s +====================Eval==================== +epoch=32 global_step=12903 loss=2.85527 test_loss_avg=1.40633 acc=0.24219 test_acc_avg=0.63459 test_acc_top5_avg=0.92401 time=221.62it/s +epoch=32 global_step=12903 loss=0.74319 test_loss_avg=2.04513 acc=0.76562 test_acc_avg=0.53906 test_acc_top5_avg=0.86447 time=240.47it/s +epoch=32 global_step=12903 loss=0.61052 test_loss_avg=1.91889 acc=0.87500 test_acc_avg=0.56596 test_acc_top5_avg=0.87599 time=850.25it/s +curr_acc 0.5660 +BEST_ACC 0.5878 +curr_acc_top5 0.8760 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=4.94538 loss_avg=5.13159 acc=0.48438 acc_top1_avg=0.48753 acc_top5_avg=0.89644 lr=0.01000 gn=7.63264 time=51.97it/s +epoch=33 global_step=13000 loss=4.73129 loss_avg=5.09894 acc=0.53906 acc_top1_avg=0.49138 acc_top5_avg=0.89634 lr=0.01000 gn=7.87217 time=59.92it/s +epoch=33 global_step=13050 loss=4.66374 loss_avg=5.06464 acc=0.53125 acc_top1_avg=0.49421 acc_top5_avg=0.89658 lr=0.01000 gn=7.21089 time=52.63it/s +epoch=33 global_step=13100 loss=5.85195 loss_avg=5.07377 acc=0.39062 acc_top1_avg=0.49258 acc_top5_avg=0.89887 lr=0.01000 gn=7.63468 time=52.75it/s +epoch=33 global_step=13150 loss=4.93932 loss_avg=5.05972 acc=0.51562 acc_top1_avg=0.49409 acc_top5_avg=0.89910 lr=0.01000 gn=6.63436 time=53.48it/s +epoch=33 global_step=13200 loss=4.41734 loss_avg=5.06784 acc=0.57031 acc_top1_avg=0.49306 acc_top5_avg=0.89744 lr=0.01000 gn=5.82461 time=61.28it/s +epoch=33 global_step=13250 loss=5.02897 loss_avg=5.08508 acc=0.49219 acc_top1_avg=0.49099 acc_top5_avg=0.89688 lr=0.01000 gn=6.94374 time=52.53it/s +====================Eval==================== +epoch=33 global_step=13294 loss=2.65040 test_loss_avg=2.56089 acc=0.08594 test_acc_avg=0.45658 test_acc_top5_avg=0.86283 time=234.10it/s +epoch=33 global_step=13294 loss=0.00555 test_loss_avg=2.19993 acc=1.00000 test_acc_avg=0.51889 test_acc_top5_avg=0.87371 time=459.45it/s +curr_acc 0.5189 +BEST_ACC 0.5878 +curr_acc_top5 0.8737 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=4.83390 loss_avg=5.04212 acc=0.51562 acc_top1_avg=0.49089 acc_top5_avg=0.90495 lr=0.01000 gn=8.25428 time=56.68it/s +epoch=34 global_step=13350 loss=5.53250 loss_avg=5.02095 acc=0.43750 acc_top1_avg=0.49819 acc_top5_avg=0.90220 lr=0.01000 gn=5.90803 time=58.38it/s +epoch=34 global_step=13400 loss=5.12956 loss_avg=5.08789 acc=0.50000 acc_top1_avg=0.48998 acc_top5_avg=0.89844 lr=0.01000 gn=6.40154 time=53.42it/s +epoch=34 global_step=13450 loss=5.71469 loss_avg=5.08205 acc=0.43750 acc_top1_avg=0.49104 acc_top5_avg=0.89648 lr=0.01000 gn=6.56313 time=55.52it/s +epoch=34 global_step=13500 loss=5.00303 loss_avg=5.10294 acc=0.50000 acc_top1_avg=0.48855 acc_top5_avg=0.89639 lr=0.01000 gn=7.30420 time=57.55it/s +epoch=34 global_step=13550 loss=5.33415 loss_avg=5.10937 acc=0.44531 acc_top1_avg=0.48761 acc_top5_avg=0.89673 lr=0.01000 gn=6.60915 time=53.55it/s +epoch=34 global_step=13600 loss=5.15068 loss_avg=5.12366 acc=0.46875 acc_top1_avg=0.48609 acc_top5_avg=0.89711 lr=0.01000 gn=6.09203 time=56.63it/s +epoch=34 global_step=13650 loss=4.83201 loss_avg=5.11713 acc=0.50000 acc_top1_avg=0.48685 acc_top5_avg=0.89765 lr=0.01000 gn=6.54635 time=56.97it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.12307 test_loss_avg=1.16617 acc=0.96094 test_acc_avg=0.69810 test_acc_top5_avg=0.94643 time=87.20it/s +epoch=34 global_step=13685 loss=0.48540 test_loss_avg=2.19150 acc=0.89844 test_acc_avg=0.51611 test_acc_top5_avg=0.85938 time=242.25it/s +epoch=34 global_step=13685 loss=0.48224 test_loss_avg=1.87531 acc=0.81250 test_acc_avg=0.58060 test_acc_top5_avg=0.88509 time=504.55it/s +curr_acc 0.5806 +BEST_ACC 0.5878 +curr_acc_top5 0.8851 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=5.22708 loss_avg=5.11312 acc=0.46094 acc_top1_avg=0.48125 acc_top5_avg=0.89531 lr=0.01000 gn=6.10244 time=60.41it/s +epoch=35 global_step=13750 loss=4.52558 loss_avg=5.05458 acc=0.53125 acc_top1_avg=0.49099 acc_top5_avg=0.89579 lr=0.01000 gn=6.11898 time=50.93it/s +epoch=35 global_step=13800 loss=5.34482 loss_avg=5.06043 acc=0.44531 acc_top1_avg=0.49171 acc_top5_avg=0.89477 lr=0.01000 gn=8.06320 time=63.18it/s +epoch=35 global_step=13850 loss=4.81641 loss_avg=5.02944 acc=0.51562 acc_top1_avg=0.49560 acc_top5_avg=0.89820 lr=0.01000 gn=6.67477 time=57.18it/s +epoch=35 global_step=13900 loss=4.53561 loss_avg=5.02989 acc=0.56250 acc_top1_avg=0.49629 acc_top5_avg=0.89927 lr=0.01000 gn=6.73421 time=60.06it/s +epoch=35 global_step=13950 loss=5.41512 loss_avg=5.05677 acc=0.48438 acc_top1_avg=0.49410 acc_top5_avg=0.89841 lr=0.01000 gn=7.66732 time=61.28it/s +epoch=35 global_step=14000 loss=5.20077 loss_avg=5.07046 acc=0.47656 acc_top1_avg=0.49263 acc_top5_avg=0.89767 lr=0.01000 gn=7.00301 time=63.40it/s +epoch=35 global_step=14050 loss=5.16379 loss_avg=5.09338 acc=0.48438 acc_top1_avg=0.49011 acc_top5_avg=0.89698 lr=0.01000 gn=6.58212 time=54.32it/s +====================Eval==================== +epoch=35 global_step=14076 loss=0.45286 test_loss_avg=2.21016 acc=0.89844 test_acc_avg=0.49688 test_acc_top5_avg=0.81451 time=222.13it/s +epoch=35 global_step=14076 loss=0.47698 test_loss_avg=1.97544 acc=0.81250 test_acc_avg=0.53649 test_acc_top5_avg=0.88548 time=650.38it/s +curr_acc 0.5365 +BEST_ACC 0.5878 +curr_acc_top5 0.8855 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=5.20991 loss_avg=5.17008 acc=0.47656 acc_top1_avg=0.48275 acc_top5_avg=0.90462 lr=0.01000 gn=5.46026 time=55.68it/s +epoch=36 global_step=14150 loss=5.42262 loss_avg=5.08205 acc=0.46094 acc_top1_avg=0.49166 acc_top5_avg=0.90308 lr=0.01000 gn=8.20819 time=53.94it/s +epoch=36 global_step=14200 loss=4.11360 loss_avg=5.07226 acc=0.60156 acc_top1_avg=0.49263 acc_top5_avg=0.90045 lr=0.01000 gn=6.63007 time=58.88it/s +epoch=36 global_step=14250 loss=4.38312 loss_avg=5.08308 acc=0.57031 acc_top1_avg=0.49066 acc_top5_avg=0.89974 lr=0.01000 gn=6.91951 time=54.70it/s +epoch=36 global_step=14300 loss=5.90030 loss_avg=5.05975 acc=0.40625 acc_top1_avg=0.49299 acc_top5_avg=0.89962 lr=0.01000 gn=6.41876 time=53.23it/s +epoch=36 global_step=14350 loss=5.53075 loss_avg=5.06462 acc=0.42969 acc_top1_avg=0.49287 acc_top5_avg=0.89847 lr=0.01000 gn=6.08379 time=49.86it/s +epoch=36 global_step=14400 loss=4.89380 loss_avg=5.07359 acc=0.53125 acc_top1_avg=0.49212 acc_top5_avg=0.89752 lr=0.01000 gn=7.07537 time=62.59it/s +epoch=36 global_step=14450 loss=4.70462 loss_avg=5.07337 acc=0.50781 acc_top1_avg=0.49156 acc_top5_avg=0.89825 lr=0.01000 gn=6.55563 time=57.09it/s +====================Eval==================== +epoch=36 global_step=14467 loss=3.21851 test_loss_avg=3.30754 acc=0.25000 test_acc_avg=0.24219 test_acc_top5_avg=0.92708 time=242.85it/s +epoch=36 global_step=14467 loss=0.75823 test_loss_avg=3.14422 acc=0.78906 test_acc_avg=0.34710 test_acc_top5_avg=0.86398 time=239.98it/s +epoch=36 global_step=14467 loss=1.01183 test_loss_avg=2.46883 acc=0.68750 test_acc_avg=0.47696 test_acc_top5_avg=0.89428 time=839.36it/s +curr_acc 0.4770 +BEST_ACC 0.5878 +curr_acc_top5 0.8943 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=4.86681 loss_avg=5.11317 acc=0.48438 acc_top1_avg=0.48864 acc_top5_avg=0.90578 lr=0.01000 gn=6.58981 time=63.21it/s +epoch=37 global_step=14550 loss=5.12887 loss_avg=5.07019 acc=0.46875 acc_top1_avg=0.49200 acc_top5_avg=0.90324 lr=0.01000 gn=6.77351 time=63.26it/s +epoch=37 global_step=14600 loss=5.09172 loss_avg=5.07767 acc=0.47656 acc_top1_avg=0.49154 acc_top5_avg=0.90055 lr=0.01000 gn=5.43360 time=53.40it/s +epoch=37 global_step=14650 loss=4.99815 loss_avg=5.10760 acc=0.48438 acc_top1_avg=0.48783 acc_top5_avg=0.89630 lr=0.01000 gn=6.83610 time=60.04it/s +epoch=37 global_step=14700 loss=4.83463 loss_avg=5.09533 acc=0.49219 acc_top1_avg=0.48877 acc_top5_avg=0.89639 lr=0.01000 gn=6.67484 time=52.81it/s +epoch=37 global_step=14750 loss=5.38491 loss_avg=5.11435 acc=0.46094 acc_top1_avg=0.48669 acc_top5_avg=0.89606 lr=0.01000 gn=6.89506 time=52.56it/s +epoch=37 global_step=14800 loss=4.86086 loss_avg=5.11537 acc=0.51562 acc_top1_avg=0.48719 acc_top5_avg=0.89574 lr=0.01000 gn=6.06843 time=59.19it/s +epoch=37 global_step=14850 loss=4.99187 loss_avg=5.10677 acc=0.49219 acc_top1_avg=0.48784 acc_top5_avg=0.89630 lr=0.01000 gn=6.09995 time=54.66it/s +====================Eval==================== +epoch=37 global_step=14858 loss=6.00890 test_loss_avg=2.24907 acc=0.00000 test_acc_avg=0.53096 test_acc_top5_avg=0.85822 time=241.96it/s +epoch=37 global_step=14858 loss=0.16880 test_loss_avg=2.13583 acc=0.96094 test_acc_avg=0.54495 test_acc_top5_avg=0.87317 time=250.27it/s +epoch=37 global_step=14858 loss=0.04150 test_loss_avg=2.08464 acc=1.00000 test_acc_avg=0.55597 test_acc_top5_avg=0.87638 time=442.06it/s +curr_acc 0.5560 +BEST_ACC 0.5878 +curr_acc_top5 0.8764 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=4.71476 loss_avg=5.14345 acc=0.52344 acc_top1_avg=0.47935 acc_top5_avg=0.89230 lr=0.01000 gn=6.99968 time=53.26it/s +epoch=38 global_step=14950 loss=5.22730 loss_avg=5.06003 acc=0.44531 acc_top1_avg=0.49100 acc_top5_avg=0.89682 lr=0.01000 gn=6.01101 time=61.43it/s +epoch=38 global_step=15000 loss=4.40444 loss_avg=5.07152 acc=0.57812 acc_top1_avg=0.49098 acc_top5_avg=0.89602 lr=0.01000 gn=8.08649 time=52.38it/s +epoch=38 global_step=15050 loss=5.40295 loss_avg=5.05358 acc=0.43750 acc_top1_avg=0.49349 acc_top5_avg=0.89640 lr=0.01000 gn=7.11143 time=56.94it/s +epoch=38 global_step=15100 loss=4.85203 loss_avg=5.07867 acc=0.51562 acc_top1_avg=0.49073 acc_top5_avg=0.89608 lr=0.01000 gn=6.50006 time=47.59it/s +epoch=38 global_step=15150 loss=5.31956 loss_avg=5.08535 acc=0.49219 acc_top1_avg=0.49066 acc_top5_avg=0.89582 lr=0.01000 gn=6.59236 time=59.27it/s +epoch=38 global_step=15200 loss=4.70077 loss_avg=5.08492 acc=0.54688 acc_top1_avg=0.49025 acc_top5_avg=0.89560 lr=0.01000 gn=7.28247 time=49.82it/s +====================Eval==================== +epoch=38 global_step=15249 loss=5.27995 test_loss_avg=2.43260 acc=0.00000 test_acc_avg=0.44678 test_acc_top5_avg=0.85335 time=153.01it/s +epoch=38 global_step=15249 loss=0.20550 test_loss_avg=2.09578 acc=0.93750 test_acc_avg=0.52967 test_acc_top5_avg=0.88746 time=450.95it/s +curr_acc 0.5297 +BEST_ACC 0.5878 +curr_acc_top5 0.8875 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=4.76535 loss_avg=4.76535 acc=0.52344 acc_top1_avg=0.52344 acc_top5_avg=0.92969 lr=0.01000 gn=5.89571 time=52.81it/s +epoch=39 global_step=15300 loss=5.60546 loss_avg=4.97439 acc=0.43750 acc_top1_avg=0.50306 acc_top5_avg=0.90119 lr=0.01000 gn=6.66454 time=51.07it/s +epoch=39 global_step=15350 loss=5.33502 loss_avg=5.06784 acc=0.43750 acc_top1_avg=0.49188 acc_top5_avg=0.90060 lr=0.01000 gn=5.73445 time=44.62it/s +epoch=39 global_step=15400 loss=5.21716 loss_avg=5.07094 acc=0.46875 acc_top1_avg=0.49151 acc_top5_avg=0.89808 lr=0.01000 gn=6.02321 time=58.82it/s +epoch=39 global_step=15450 loss=5.37829 loss_avg=5.09531 acc=0.46094 acc_top1_avg=0.48943 acc_top5_avg=0.89817 lr=0.01000 gn=7.00752 time=58.48it/s +epoch=39 global_step=15500 loss=4.91616 loss_avg=5.12152 acc=0.50000 acc_top1_avg=0.48637 acc_top5_avg=0.89679 lr=0.01000 gn=5.99571 time=57.77it/s +epoch=39 global_step=15550 loss=5.01332 loss_avg=5.12486 acc=0.49219 acc_top1_avg=0.48630 acc_top5_avg=0.89659 lr=0.01000 gn=5.87626 time=54.35it/s +epoch=39 global_step=15600 loss=5.53043 loss_avg=5.10528 acc=0.44531 acc_top1_avg=0.48831 acc_top5_avg=0.89724 lr=0.01000 gn=7.06244 time=56.75it/s +====================Eval==================== +epoch=39 global_step=15640 loss=2.36172 test_loss_avg=2.08110 acc=0.32812 test_acc_avg=0.52878 test_acc_top5_avg=0.94120 time=245.42it/s +epoch=39 global_step=15640 loss=0.10881 test_loss_avg=2.64360 acc=0.96875 test_acc_avg=0.46264 test_acc_top5_avg=0.89142 time=237.80it/s +epoch=39 global_step=15640 loss=0.00369 test_loss_avg=2.32680 acc=1.00000 test_acc_avg=0.52631 test_acc_top5_avg=0.90516 time=842.57it/s +curr_acc 0.5263 +BEST_ACC 0.5878 +curr_acc_top5 0.9052 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=5.21286 loss_avg=4.97202 acc=0.46875 acc_top1_avg=0.50703 acc_top5_avg=0.89766 lr=0.00100 gn=5.31773 time=57.31it/s +epoch=40 global_step=15700 loss=5.28490 loss_avg=4.83153 acc=0.46094 acc_top1_avg=0.51927 acc_top5_avg=0.90781 lr=0.00100 gn=7.64831 time=55.30it/s +epoch=40 global_step=15750 loss=5.19645 loss_avg=4.80437 acc=0.50000 acc_top1_avg=0.52187 acc_top5_avg=0.91065 lr=0.00100 gn=6.40126 time=53.97it/s +epoch=40 global_step=15800 loss=4.13563 loss_avg=4.78653 acc=0.57031 acc_top1_avg=0.52310 acc_top5_avg=0.91128 lr=0.00100 gn=6.99965 time=56.13it/s +epoch=40 global_step=15850 loss=4.23331 loss_avg=4.79095 acc=0.60938 acc_top1_avg=0.52206 acc_top5_avg=0.91090 lr=0.00100 gn=5.58737 time=59.84it/s +epoch=40 global_step=15900 loss=4.37400 loss_avg=4.78606 acc=0.53906 acc_top1_avg=0.52218 acc_top5_avg=0.91148 lr=0.00100 gn=5.52904 time=54.67it/s +epoch=40 global_step=15950 loss=5.12857 loss_avg=4.77816 acc=0.50000 acc_top1_avg=0.52356 acc_top5_avg=0.91121 lr=0.00100 gn=5.19262 time=54.06it/s +epoch=40 global_step=16000 loss=4.97531 loss_avg=4.77697 acc=0.47656 acc_top1_avg=0.52324 acc_top5_avg=0.91131 lr=0.00100 gn=6.10670 time=55.41it/s +====================Eval==================== +epoch=40 global_step=16031 loss=1.55471 test_loss_avg=1.79497 acc=0.49219 test_acc_avg=0.59316 test_acc_top5_avg=0.87676 time=236.41it/s +epoch=40 global_step=16031 loss=0.10642 test_loss_avg=1.63516 acc=0.93750 test_acc_avg=0.62282 test_acc_top5_avg=0.92098 time=505.70it/s +curr_acc 0.6228 +BEST_ACC 0.5878 +curr_acc_top5 0.9210 +BEST_ACC_top5 0.9190 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=4.20564 loss_avg=4.56425 acc=0.58594 acc_top1_avg=0.54975 acc_top5_avg=0.91982 lr=0.00100 gn=7.49408 time=53.98it/s +epoch=41 global_step=16100 loss=5.25638 loss_avg=4.65448 acc=0.47656 acc_top1_avg=0.53782 acc_top5_avg=0.92040 lr=0.00100 gn=7.11253 time=63.50it/s +epoch=41 global_step=16150 loss=4.78837 loss_avg=4.65929 acc=0.52344 acc_top1_avg=0.53676 acc_top5_avg=0.91570 lr=0.00100 gn=6.26638 time=60.13it/s +epoch=41 global_step=16200 loss=4.56677 loss_avg=4.64864 acc=0.56250 acc_top1_avg=0.53694 acc_top5_avg=0.91568 lr=0.00100 gn=6.76293 time=59.07it/s +epoch=41 global_step=16250 loss=4.58811 loss_avg=4.65322 acc=0.54688 acc_top1_avg=0.53578 acc_top5_avg=0.91610 lr=0.00100 gn=7.84292 time=52.56it/s +epoch=41 global_step=16300 loss=5.35741 loss_avg=4.65295 acc=0.47656 acc_top1_avg=0.53569 acc_top5_avg=0.91540 lr=0.00100 gn=8.04022 time=54.87it/s +epoch=41 global_step=16350 loss=3.86453 loss_avg=4.65829 acc=0.61719 acc_top1_avg=0.53458 acc_top5_avg=0.91565 lr=0.00100 gn=7.23956 time=59.03it/s +epoch=41 global_step=16400 loss=4.73819 loss_avg=4.65713 acc=0.49219 acc_top1_avg=0.53453 acc_top5_avg=0.91508 lr=0.00100 gn=8.18800 time=55.60it/s +====================Eval==================== +epoch=41 global_step=16422 loss=0.19884 test_loss_avg=0.75229 acc=0.93750 test_acc_avg=0.78125 test_acc_top5_avg=0.98509 time=240.22it/s +epoch=41 global_step=16422 loss=0.38320 test_loss_avg=2.04952 acc=0.89844 test_acc_avg=0.53573 test_acc_top5_avg=0.90023 time=227.78it/s +epoch=41 global_step=16422 loss=0.04243 test_loss_avg=1.63515 acc=1.00000 test_acc_avg=0.62688 test_acc_top5_avg=0.92188 time=721.17it/s +curr_acc 0.6269 +BEST_ACC 0.6228 +curr_acc_top5 0.9219 +BEST_ACC_top5 0.9210 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=4.45020 loss_avg=4.53560 acc=0.55469 acc_top1_avg=0.55329 acc_top5_avg=0.91267 lr=0.00100 gn=6.90879 time=52.98it/s +epoch=42 global_step=16500 loss=3.99070 loss_avg=4.53681 acc=0.62500 acc_top1_avg=0.54948 acc_top5_avg=0.91446 lr=0.00100 gn=7.75060 time=54.55it/s +epoch=42 global_step=16550 loss=4.82293 loss_avg=4.57491 acc=0.51562 acc_top1_avg=0.54480 acc_top5_avg=0.91333 lr=0.00100 gn=6.56267 time=59.83it/s +epoch=42 global_step=16600 loss=4.09370 loss_avg=4.57716 acc=0.60156 acc_top1_avg=0.54490 acc_top5_avg=0.91534 lr=0.00100 gn=6.68680 time=50.55it/s +epoch=42 global_step=16650 loss=4.30788 loss_avg=4.58092 acc=0.58594 acc_top1_avg=0.54379 acc_top5_avg=0.91506 lr=0.00100 gn=7.59508 time=53.57it/s +epoch=42 global_step=16700 loss=4.96364 loss_avg=4.59955 acc=0.50781 acc_top1_avg=0.54128 acc_top5_avg=0.91544 lr=0.00100 gn=8.44195 time=53.18it/s +epoch=42 global_step=16750 loss=3.90940 loss_avg=4.58969 acc=0.62500 acc_top1_avg=0.54275 acc_top5_avg=0.91535 lr=0.00100 gn=7.89511 time=53.60it/s +epoch=42 global_step=16800 loss=4.66511 loss_avg=4.59067 acc=0.51562 acc_top1_avg=0.54256 acc_top5_avg=0.91574 lr=0.00100 gn=7.91017 time=55.60it/s +====================Eval==================== +epoch=42 global_step=16813 loss=1.80007 test_loss_avg=2.06394 acc=0.63281 test_acc_avg=0.54907 test_acc_top5_avg=0.86865 time=126.03it/s +epoch=42 global_step=16813 loss=0.09866 test_loss_avg=1.65691 acc=0.93750 test_acc_avg=0.62787 test_acc_top5_avg=0.92959 time=757.78it/s +curr_acc 0.6279 +BEST_ACC 0.6269 +curr_acc_top5 0.9296 +BEST_ACC_top5 0.9219 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=4.18339 loss_avg=4.52923 acc=0.59375 acc_top1_avg=0.54962 acc_top5_avg=0.92378 lr=0.00100 gn=8.64073 time=53.41it/s +epoch=43 global_step=16900 loss=4.21157 loss_avg=4.55275 acc=0.57812 acc_top1_avg=0.54580 acc_top5_avg=0.92008 lr=0.00100 gn=8.14153 time=53.96it/s +epoch=43 global_step=16950 loss=4.89889 loss_avg=4.55767 acc=0.53125 acc_top1_avg=0.54682 acc_top5_avg=0.92028 lr=0.00100 gn=6.59201 time=62.28it/s +epoch=43 global_step=17000 loss=4.23002 loss_avg=4.55502 acc=0.58594 acc_top1_avg=0.54679 acc_top5_avg=0.91795 lr=0.00100 gn=8.18170 time=54.92it/s +epoch=43 global_step=17050 loss=4.08039 loss_avg=4.57655 acc=0.59375 acc_top1_avg=0.54381 acc_top5_avg=0.91769 lr=0.00100 gn=8.80068 time=58.01it/s +epoch=43 global_step=17100 loss=4.29819 loss_avg=4.57075 acc=0.57031 acc_top1_avg=0.54418 acc_top5_avg=0.91774 lr=0.00100 gn=8.66232 time=61.03it/s +epoch=43 global_step=17150 loss=4.36599 loss_avg=4.56190 acc=0.56250 acc_top1_avg=0.54546 acc_top5_avg=0.91719 lr=0.00100 gn=7.58525 time=57.81it/s +epoch=43 global_step=17200 loss=4.35874 loss_avg=4.55529 acc=0.56250 acc_top1_avg=0.54589 acc_top5_avg=0.91838 lr=0.00100 gn=9.12073 time=54.60it/s +====================Eval==================== +epoch=43 global_step=17204 loss=0.83433 test_loss_avg=0.94174 acc=0.72656 test_acc_avg=0.72656 test_acc_top5_avg=0.97396 time=240.97it/s +epoch=43 global_step=17204 loss=5.41076 test_loss_avg=2.30459 acc=0.00000 test_acc_avg=0.49145 test_acc_top5_avg=0.90448 time=216.86it/s +epoch=43 global_step=17204 loss=0.08207 test_loss_avg=1.76314 acc=0.93750 test_acc_avg=0.60641 test_acc_top5_avg=0.93147 time=509.95it/s +curr_acc 0.6064 +BEST_ACC 0.6279 +curr_acc_top5 0.9315 +BEST_ACC_top5 0.9296 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=4.71030 loss_avg=4.54609 acc=0.50781 acc_top1_avg=0.54738 acc_top5_avg=0.91559 lr=0.00100 gn=7.31901 time=51.22it/s +epoch=44 global_step=17300 loss=4.62390 loss_avg=4.52759 acc=0.54688 acc_top1_avg=0.55054 acc_top5_avg=0.91813 lr=0.00100 gn=8.94832 time=54.14it/s +epoch=44 global_step=17350 loss=4.48801 loss_avg=4.52468 acc=0.53906 acc_top1_avg=0.55073 acc_top5_avg=0.91754 lr=0.00100 gn=7.55342 time=58.05it/s +epoch=44 global_step=17400 loss=4.24118 loss_avg=4.53059 acc=0.58594 acc_top1_avg=0.54963 acc_top5_avg=0.91785 lr=0.00100 gn=8.57412 time=56.48it/s +epoch=44 global_step=17450 loss=4.47415 loss_avg=4.54289 acc=0.57031 acc_top1_avg=0.54830 acc_top5_avg=0.91860 lr=0.00100 gn=8.42508 time=54.15it/s +epoch=44 global_step=17500 loss=4.82745 loss_avg=4.51728 acc=0.50781 acc_top1_avg=0.55086 acc_top5_avg=0.91926 lr=0.00100 gn=8.53296 time=54.52it/s +epoch=44 global_step=17550 loss=4.26518 loss_avg=4.50150 acc=0.59375 acc_top1_avg=0.55308 acc_top5_avg=0.91944 lr=0.00100 gn=9.47833 time=61.93it/s +====================Eval==================== +epoch=44 global_step=17595 loss=3.76747 test_loss_avg=0.94133 acc=0.23438 test_acc_avg=0.73991 test_acc_top5_avg=0.96484 time=235.50it/s +epoch=44 global_step=17595 loss=0.28962 test_loss_avg=1.72652 acc=0.91406 test_acc_avg=0.61983 test_acc_top5_avg=0.92430 time=243.11it/s +epoch=44 global_step=17595 loss=0.14226 test_loss_avg=1.62767 acc=0.87500 test_acc_avg=0.63983 test_acc_top5_avg=0.92909 time=503.94it/s +curr_acc 0.6398 +BEST_ACC 0.6279 +curr_acc_top5 0.9291 +BEST_ACC_top5 0.9315 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=4.08055 loss_avg=4.35698 acc=0.60156 acc_top1_avg=0.56563 acc_top5_avg=0.92969 lr=0.00100 gn=7.22254 time=51.51it/s +epoch=45 global_step=17650 loss=4.92490 loss_avg=4.49053 acc=0.49219 acc_top1_avg=0.55156 acc_top5_avg=0.92330 lr=0.00100 gn=8.65777 time=56.50it/s +epoch=45 global_step=17700 loss=4.63381 loss_avg=4.47670 acc=0.53125 acc_top1_avg=0.55327 acc_top5_avg=0.92039 lr=0.00100 gn=6.94339 time=53.05it/s +epoch=45 global_step=17750 loss=4.09317 loss_avg=4.46513 acc=0.61719 acc_top1_avg=0.55605 acc_top5_avg=0.92097 lr=0.00100 gn=9.24064 time=56.46it/s +epoch=45 global_step=17800 loss=4.07231 loss_avg=4.47155 acc=0.60156 acc_top1_avg=0.55579 acc_top5_avg=0.92085 lr=0.00100 gn=7.82695 time=57.95it/s +epoch=45 global_step=17850 loss=4.35586 loss_avg=4.48061 acc=0.58594 acc_top1_avg=0.55545 acc_top5_avg=0.92059 lr=0.00100 gn=9.57110 time=57.94it/s +epoch=45 global_step=17900 loss=4.14171 loss_avg=4.47450 acc=0.60156 acc_top1_avg=0.55648 acc_top5_avg=0.92034 lr=0.00100 gn=9.72893 time=54.67it/s +epoch=45 global_step=17950 loss=4.18122 loss_avg=4.46904 acc=0.60156 acc_top1_avg=0.55673 acc_top5_avg=0.92091 lr=0.00100 gn=10.94280 time=56.43it/s +====================Eval==================== +epoch=45 global_step=17986 loss=1.60428 test_loss_avg=1.75840 acc=0.52344 test_acc_avg=0.59635 test_acc_top5_avg=0.90660 time=249.16it/s +epoch=45 global_step=17986 loss=0.04912 test_loss_avg=1.65712 acc=1.00000 test_acc_avg=0.63736 test_acc_top5_avg=0.93562 time=537.04it/s +curr_acc 0.6374 +BEST_ACC 0.6398 +curr_acc_top5 0.9356 +BEST_ACC_top5 0.9315 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=5.06777 loss_avg=4.43038 acc=0.48438 acc_top1_avg=0.55246 acc_top5_avg=0.92132 lr=0.00100 gn=10.75988 time=51.25it/s +epoch=46 global_step=18050 loss=5.05343 loss_avg=4.45907 acc=0.52344 acc_top1_avg=0.55518 acc_top5_avg=0.92114 lr=0.00100 gn=10.50879 time=50.46it/s +epoch=46 global_step=18100 loss=5.58646 loss_avg=4.44884 acc=0.41406 acc_top1_avg=0.55695 acc_top5_avg=0.91941 lr=0.00100 gn=9.59679 time=55.80it/s +epoch=46 global_step=18150 loss=4.67190 loss_avg=4.42236 acc=0.53906 acc_top1_avg=0.56007 acc_top5_avg=0.92145 lr=0.00100 gn=11.13352 time=57.29it/s +epoch=46 global_step=18200 loss=4.84455 loss_avg=4.40319 acc=0.50781 acc_top1_avg=0.56232 acc_top5_avg=0.92180 lr=0.00100 gn=9.73396 time=56.47it/s +epoch=46 global_step=18250 loss=4.28099 loss_avg=4.41434 acc=0.57812 acc_top1_avg=0.56117 acc_top5_avg=0.92114 lr=0.00100 gn=10.79585 time=51.07it/s +epoch=46 global_step=18300 loss=4.42457 loss_avg=4.42622 acc=0.56250 acc_top1_avg=0.56006 acc_top5_avg=0.92133 lr=0.00100 gn=10.87706 time=61.05it/s +epoch=46 global_step=18350 loss=4.93731 loss_avg=4.44373 acc=0.46875 acc_top1_avg=0.55771 acc_top5_avg=0.92065 lr=0.00100 gn=8.96999 time=55.13it/s +====================Eval==================== +epoch=46 global_step=18377 loss=1.04277 test_loss_avg=0.65242 acc=0.68750 test_acc_avg=0.81348 test_acc_top5_avg=0.98779 time=102.58it/s +epoch=46 global_step=18377 loss=0.09821 test_loss_avg=1.97230 acc=0.96875 test_acc_avg=0.56001 test_acc_top5_avg=0.91181 time=234.92it/s +epoch=46 global_step=18377 loss=0.11176 test_loss_avg=1.67975 acc=0.93750 test_acc_avg=0.62233 test_acc_top5_avg=0.92583 time=502.07it/s +curr_acc 0.6223 +BEST_ACC 0.6398 +curr_acc_top5 0.9258 +BEST_ACC_top5 0.9356 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=4.44730 loss_avg=4.41244 acc=0.55469 acc_top1_avg=0.56556 acc_top5_avg=0.91848 lr=0.00100 gn=7.85083 time=54.44it/s +epoch=47 global_step=18450 loss=4.29139 loss_avg=4.36562 acc=0.60156 acc_top1_avg=0.57042 acc_top5_avg=0.91695 lr=0.00100 gn=11.36391 time=47.47it/s +epoch=47 global_step=18500 loss=3.90087 loss_avg=4.35606 acc=0.63281 acc_top1_avg=0.57139 acc_top5_avg=0.91864 lr=0.00100 gn=9.53447 time=58.55it/s +epoch=47 global_step=18550 loss=3.51731 loss_avg=4.35539 acc=0.67188 acc_top1_avg=0.57085 acc_top5_avg=0.91957 lr=0.00100 gn=10.94397 time=58.26it/s +epoch=47 global_step=18600 loss=4.63513 loss_avg=4.37506 acc=0.55469 acc_top1_avg=0.56825 acc_top5_avg=0.92051 lr=0.00100 gn=11.69829 time=51.78it/s +epoch=47 global_step=18650 loss=4.35327 loss_avg=4.39620 acc=0.58594 acc_top1_avg=0.56571 acc_top5_avg=0.92016 lr=0.00100 gn=11.48473 time=63.53it/s +epoch=47 global_step=18700 loss=3.71753 loss_avg=4.40859 acc=0.63281 acc_top1_avg=0.56393 acc_top5_avg=0.92038 lr=0.00100 gn=10.88969 time=62.46it/s +epoch=47 global_step=18750 loss=4.37448 loss_avg=4.40869 acc=0.53125 acc_top1_avg=0.56388 acc_top5_avg=0.92014 lr=0.00100 gn=11.29631 time=62.57it/s +====================Eval==================== +epoch=47 global_step=18768 loss=0.44943 test_loss_avg=1.75828 acc=0.89062 test_acc_avg=0.61212 test_acc_top5_avg=0.88894 time=243.80it/s +epoch=47 global_step=18768 loss=0.12441 test_loss_avg=1.64508 acc=0.87500 test_acc_avg=0.62678 test_acc_top5_avg=0.92979 time=417.09it/s +curr_acc 0.6268 +BEST_ACC 0.6398 +curr_acc_top5 0.9298 +BEST_ACC_top5 0.9356 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=4.50888 loss_avg=4.36887 acc=0.55469 acc_top1_avg=0.56641 acc_top5_avg=0.92090 lr=0.00100 gn=12.77834 time=60.64it/s +epoch=48 global_step=18850 loss=4.14040 loss_avg=4.36290 acc=0.57812 acc_top1_avg=0.56717 acc_top5_avg=0.92045 lr=0.00100 gn=10.87639 time=51.73it/s +epoch=48 global_step=18900 loss=4.23679 loss_avg=4.34300 acc=0.57031 acc_top1_avg=0.57079 acc_top5_avg=0.92146 lr=0.00100 gn=11.70863 time=51.75it/s +epoch=48 global_step=18950 loss=3.92510 loss_avg=4.34526 acc=0.60938 acc_top1_avg=0.57057 acc_top5_avg=0.92106 lr=0.00100 gn=10.13471 time=53.78it/s +epoch=48 global_step=19000 loss=4.33839 loss_avg=4.35071 acc=0.60156 acc_top1_avg=0.57014 acc_top5_avg=0.92134 lr=0.00100 gn=12.41592 time=52.92it/s +epoch=48 global_step=19050 loss=3.21378 loss_avg=4.35620 acc=0.71875 acc_top1_avg=0.56990 acc_top5_avg=0.91980 lr=0.00100 gn=12.13624 time=54.32it/s +epoch=48 global_step=19100 loss=4.35542 loss_avg=4.35456 acc=0.57031 acc_top1_avg=0.56991 acc_top5_avg=0.91955 lr=0.00100 gn=10.70616 time=51.47it/s +epoch=48 global_step=19150 loss=4.55675 loss_avg=4.36580 acc=0.54688 acc_top1_avg=0.56849 acc_top5_avg=0.91985 lr=0.00100 gn=9.34451 time=61.07it/s +====================Eval==================== +epoch=48 global_step=19159 loss=0.78755 test_loss_avg=1.10348 acc=0.77344 test_acc_avg=0.66113 test_acc_top5_avg=0.97363 time=243.70it/s +epoch=48 global_step=19159 loss=0.45795 test_loss_avg=2.25387 acc=0.87500 test_acc_avg=0.50013 test_acc_top5_avg=0.90517 time=237.23it/s +epoch=48 global_step=19159 loss=0.02792 test_loss_avg=1.71472 acc=1.00000 test_acc_avg=0.61561 test_acc_top5_avg=0.92949 time=651.59it/s +curr_acc 0.6156 +BEST_ACC 0.6398 +curr_acc_top5 0.9295 +BEST_ACC_top5 0.9356 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=4.12060 loss_avg=4.34426 acc=0.58594 acc_top1_avg=0.57184 acc_top5_avg=0.91940 lr=0.00100 gn=14.77897 time=53.18it/s +epoch=49 global_step=19250 loss=5.01126 loss_avg=4.35960 acc=0.48438 acc_top1_avg=0.56920 acc_top5_avg=0.92016 lr=0.00100 gn=12.30781 time=58.69it/s +epoch=49 global_step=19300 loss=3.79066 loss_avg=4.33741 acc=0.63281 acc_top1_avg=0.57186 acc_top5_avg=0.91966 lr=0.00100 gn=11.95633 time=52.66it/s +epoch=49 global_step=19350 loss=4.79244 loss_avg=4.34843 acc=0.51562 acc_top1_avg=0.57105 acc_top5_avg=0.91909 lr=0.00100 gn=14.14619 time=61.81it/s +epoch=49 global_step=19400 loss=4.19234 loss_avg=4.34560 acc=0.59375 acc_top1_avg=0.57154 acc_top5_avg=0.91909 lr=0.00100 gn=13.02276 time=52.87it/s +epoch=49 global_step=19450 loss=4.41639 loss_avg=4.34647 acc=0.57812 acc_top1_avg=0.57112 acc_top5_avg=0.91924 lr=0.00100 gn=12.62447 time=58.65it/s +epoch=49 global_step=19500 loss=3.95256 loss_avg=4.34341 acc=0.63281 acc_top1_avg=0.57093 acc_top5_avg=0.92066 lr=0.00100 gn=11.48940 time=53.89it/s +epoch=49 global_step=19550 loss=5.31044 loss_avg=4.34753 acc=0.45000 acc_top1_avg=0.57082 acc_top5_avg=0.92052 lr=0.00100 gn=15.87599 time=79.42it/s +====================Eval==================== +epoch=49 global_step=19550 loss=5.39714 test_loss_avg=1.84408 acc=0.00000 test_acc_avg=0.57893 test_acc_top5_avg=0.89520 time=242.98it/s +epoch=49 global_step=19550 loss=0.05826 test_loss_avg=1.60599 acc=1.00000 test_acc_avg=0.64250 test_acc_top5_avg=0.93354 time=688.27it/s +epoch=49 global_step=19550 loss=0.05826 test_loss_avg=1.60599 acc=1.00000 test_acc_avg=0.64250 test_acc_top5_avg=0.93354 time=688.27it/s +curr_acc 0.6425 +BEST_ACC 0.6398 +curr_acc_top5 0.9335 +BEST_ACC_top5 0.9356 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=4.15661 loss_avg=4.22879 acc=0.56250 acc_top1_avg=0.58500 acc_top5_avg=0.92281 lr=0.00100 gn=12.88037 time=53.05it/s +epoch=50 global_step=19650 loss=4.61727 loss_avg=4.29013 acc=0.56250 acc_top1_avg=0.57898 acc_top5_avg=0.91859 lr=0.00100 gn=13.90286 time=57.78it/s +epoch=50 global_step=19700 loss=4.86036 loss_avg=4.32285 acc=0.51562 acc_top1_avg=0.57484 acc_top5_avg=0.91901 lr=0.00100 gn=13.09550 time=53.12it/s +epoch=50 global_step=19750 loss=4.61936 loss_avg=4.30648 acc=0.56250 acc_top1_avg=0.57684 acc_top5_avg=0.91945 lr=0.00100 gn=14.95738 time=62.29it/s +epoch=50 global_step=19800 loss=3.96809 loss_avg=4.32203 acc=0.63281 acc_top1_avg=0.57484 acc_top5_avg=0.92012 lr=0.00100 gn=14.75005 time=60.42it/s +epoch=50 global_step=19850 loss=4.05693 loss_avg=4.31811 acc=0.58594 acc_top1_avg=0.57529 acc_top5_avg=0.92021 lr=0.00100 gn=12.50133 time=59.56it/s +epoch=50 global_step=19900 loss=4.03393 loss_avg=4.32272 acc=0.57812 acc_top1_avg=0.57415 acc_top5_avg=0.92045 lr=0.00100 gn=13.81758 time=55.68it/s +====================Eval==================== +epoch=50 global_step=19941 loss=5.28965 test_loss_avg=1.97713 acc=0.00000 test_acc_avg=0.55922 test_acc_top5_avg=0.91406 time=237.66it/s +epoch=50 global_step=19941 loss=0.20073 test_loss_avg=1.66341 acc=0.93750 test_acc_avg=0.63341 test_acc_top5_avg=0.93750 time=512.69it/s +curr_acc 0.6334 +BEST_ACC 0.6425 +curr_acc_top5 0.9375 +BEST_ACC_top5 0.9356 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=4.54575 loss_avg=4.29624 acc=0.54688 acc_top1_avg=0.57552 acc_top5_avg=0.92622 lr=0.00100 gn=11.50795 time=55.17it/s +epoch=51 global_step=20000 loss=3.85464 loss_avg=4.23788 acc=0.63281 acc_top1_avg=0.58382 acc_top5_avg=0.92161 lr=0.00100 gn=12.21329 time=54.82it/s +epoch=51 global_step=20050 loss=3.72496 loss_avg=4.25123 acc=0.64062 acc_top1_avg=0.58235 acc_top5_avg=0.92395 lr=0.00100 gn=13.86969 time=56.06it/s +epoch=51 global_step=20100 loss=4.78523 loss_avg=4.25338 acc=0.51562 acc_top1_avg=0.58250 acc_top5_avg=0.92335 lr=0.00100 gn=13.10784 time=61.74it/s +epoch=51 global_step=20150 loss=4.29899 loss_avg=4.23838 acc=0.54688 acc_top1_avg=0.58429 acc_top5_avg=0.92311 lr=0.00100 gn=14.48613 time=54.32it/s +epoch=51 global_step=20200 loss=4.25705 loss_avg=4.25539 acc=0.58594 acc_top1_avg=0.58265 acc_top5_avg=0.92302 lr=0.00100 gn=16.84229 time=55.60it/s +epoch=51 global_step=20250 loss=4.66274 loss_avg=4.26280 acc=0.52344 acc_top1_avg=0.58141 acc_top5_avg=0.92294 lr=0.00100 gn=12.58212 time=54.23it/s +epoch=51 global_step=20300 loss=4.21396 loss_avg=4.26484 acc=0.58594 acc_top1_avg=0.58122 acc_top5_avg=0.92307 lr=0.00100 gn=14.00984 time=63.19it/s +====================Eval==================== +epoch=51 global_step=20332 loss=1.67354 test_loss_avg=0.95633 acc=0.47656 test_acc_avg=0.71763 test_acc_top5_avg=0.97135 time=238.26it/s +epoch=51 global_step=20332 loss=0.15546 test_loss_avg=1.92658 acc=0.94531 test_acc_avg=0.57504 test_acc_top5_avg=0.92716 time=252.50it/s +epoch=51 global_step=20332 loss=0.09816 test_loss_avg=1.74989 acc=0.93750 test_acc_avg=0.61224 test_acc_top5_avg=0.93414 time=538.15it/s +curr_acc 0.6122 +BEST_ACC 0.6425 +curr_acc_top5 0.9341 +BEST_ACC_top5 0.9375 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=4.17051 loss_avg=4.12534 acc=0.57812 acc_top1_avg=0.59852 acc_top5_avg=0.92665 lr=0.00100 gn=13.84459 time=53.64it/s +epoch=52 global_step=20400 loss=4.00876 loss_avg=4.23677 acc=0.59375 acc_top1_avg=0.58433 acc_top5_avg=0.92567 lr=0.00100 gn=12.28285 time=32.98it/s +epoch=52 global_step=20450 loss=4.22105 loss_avg=4.25526 acc=0.57031 acc_top1_avg=0.58104 acc_top5_avg=0.92346 lr=0.00100 gn=16.88784 time=52.96it/s +epoch=52 global_step=20500 loss=4.56336 loss_avg=4.28071 acc=0.55469 acc_top1_avg=0.57873 acc_top5_avg=0.92220 lr=0.00100 gn=11.86548 time=53.85it/s +epoch=52 global_step=20550 loss=4.17786 loss_avg=4.28389 acc=0.58594 acc_top1_avg=0.57827 acc_top5_avg=0.92327 lr=0.00100 gn=15.48962 time=52.53it/s +epoch=52 global_step=20600 loss=4.18246 loss_avg=4.27035 acc=0.57812 acc_top1_avg=0.57979 acc_top5_avg=0.92292 lr=0.00100 gn=14.92479 time=54.17it/s +epoch=52 global_step=20650 loss=5.10970 loss_avg=4.27105 acc=0.47656 acc_top1_avg=0.57962 acc_top5_avg=0.92210 lr=0.00100 gn=16.00216 time=55.55it/s +epoch=52 global_step=20700 loss=4.54227 loss_avg=4.26758 acc=0.53906 acc_top1_avg=0.57944 acc_top5_avg=0.92236 lr=0.00100 gn=13.09213 time=45.30it/s +====================Eval==================== +epoch=52 global_step=20723 loss=1.76923 test_loss_avg=1.76213 acc=0.48438 test_acc_avg=0.60119 test_acc_top5_avg=0.90216 time=237.76it/s +epoch=52 global_step=20723 loss=0.34412 test_loss_avg=1.69990 acc=0.93750 test_acc_avg=0.62352 test_acc_top5_avg=0.93354 time=848.53it/s +curr_acc 0.6235 +BEST_ACC 0.6425 +curr_acc_top5 0.9335 +BEST_ACC_top5 0.9375 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=4.34869 loss_avg=4.16422 acc=0.56250 acc_top1_avg=0.58825 acc_top5_avg=0.91696 lr=0.00100 gn=13.31357 time=54.96it/s +epoch=53 global_step=20800 loss=4.21065 loss_avg=4.22038 acc=0.60156 acc_top1_avg=0.58594 acc_top5_avg=0.91772 lr=0.00100 gn=15.30866 time=58.58it/s +epoch=53 global_step=20850 loss=4.08113 loss_avg=4.22218 acc=0.60938 acc_top1_avg=0.58612 acc_top5_avg=0.91745 lr=0.00100 gn=14.65927 time=57.69it/s +epoch=53 global_step=20900 loss=4.78328 loss_avg=4.20290 acc=0.52344 acc_top1_avg=0.58770 acc_top5_avg=0.91892 lr=0.00100 gn=14.54045 time=55.96it/s +epoch=53 global_step=20950 loss=4.46845 loss_avg=4.20014 acc=0.55469 acc_top1_avg=0.58814 acc_top5_avg=0.92074 lr=0.00100 gn=15.28370 time=59.21it/s +epoch=53 global_step=21000 loss=3.66822 loss_avg=4.21999 acc=0.65625 acc_top1_avg=0.58591 acc_top5_avg=0.92134 lr=0.00100 gn=15.94289 time=55.04it/s +epoch=53 global_step=21050 loss=3.64792 loss_avg=4.21960 acc=0.66406 acc_top1_avg=0.58577 acc_top5_avg=0.92116 lr=0.00100 gn=14.85266 time=57.21it/s +epoch=53 global_step=21100 loss=3.77810 loss_avg=4.23543 acc=0.62500 acc_top1_avg=0.58426 acc_top5_avg=0.92188 lr=0.00100 gn=15.11053 time=61.38it/s +====================Eval==================== +epoch=53 global_step=21114 loss=0.28161 test_loss_avg=0.87587 acc=0.90625 test_acc_avg=0.73077 test_acc_top5_avg=0.98498 time=238.02it/s +epoch=53 global_step=21114 loss=0.43731 test_loss_avg=2.16681 acc=0.89062 test_acc_avg=0.51885 test_acc_top5_avg=0.92882 time=237.42it/s +epoch=53 global_step=21114 loss=0.16741 test_loss_avg=1.76987 acc=0.87500 test_acc_avg=0.60305 test_acc_top5_avg=0.94264 time=751.80it/s +curr_acc 0.6030 +BEST_ACC 0.6425 +curr_acc_top5 0.9426 +BEST_ACC_top5 0.9375 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=4.17771 loss_avg=4.22724 acc=0.60156 acc_top1_avg=0.58811 acc_top5_avg=0.91840 lr=0.00100 gn=17.05050 time=59.29it/s +epoch=54 global_step=21200 loss=4.12705 loss_avg=4.14523 acc=0.58594 acc_top1_avg=0.59493 acc_top5_avg=0.92442 lr=0.00100 gn=15.35374 time=52.44it/s +epoch=54 global_step=21250 loss=3.99550 loss_avg=4.16221 acc=0.59375 acc_top1_avg=0.59272 acc_top5_avg=0.92366 lr=0.00100 gn=16.36692 time=59.66it/s +epoch=54 global_step=21300 loss=4.06878 loss_avg=4.17624 acc=0.59375 acc_top1_avg=0.59005 acc_top5_avg=0.92204 lr=0.00100 gn=13.91394 time=60.43it/s +epoch=54 global_step=21350 loss=3.77409 loss_avg=4.17180 acc=0.63281 acc_top1_avg=0.59113 acc_top5_avg=0.92178 lr=0.00100 gn=14.96598 time=55.76it/s +epoch=54 global_step=21400 loss=3.67712 loss_avg=4.19155 acc=0.63281 acc_top1_avg=0.58927 acc_top5_avg=0.92209 lr=0.00100 gn=15.36747 time=57.98it/s +epoch=54 global_step=21450 loss=4.34005 loss_avg=4.19584 acc=0.56250 acc_top1_avg=0.58833 acc_top5_avg=0.92253 lr=0.00100 gn=15.07216 time=50.39it/s +epoch=54 global_step=21500 loss=3.69661 loss_avg=4.19481 acc=0.62500 acc_top1_avg=0.58829 acc_top5_avg=0.92232 lr=0.00100 gn=15.53724 time=55.40it/s +====================Eval==================== +epoch=54 global_step=21505 loss=0.38687 test_loss_avg=1.96766 acc=0.89062 test_acc_avg=0.55492 test_acc_top5_avg=0.88971 time=223.96it/s +epoch=54 global_step=21505 loss=0.07944 test_loss_avg=1.66557 acc=0.93750 test_acc_avg=0.61966 test_acc_top5_avg=0.93503 time=641.53it/s +curr_acc 0.6197 +BEST_ACC 0.6425 +curr_acc_top5 0.9350 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=4.46632 loss_avg=4.15849 acc=0.53125 acc_top1_avg=0.58976 acc_top5_avg=0.91892 lr=0.00100 gn=12.50982 time=56.98it/s +epoch=55 global_step=21600 loss=4.30490 loss_avg=4.17560 acc=0.57812 acc_top1_avg=0.58980 acc_top5_avg=0.92188 lr=0.00100 gn=16.85938 time=60.20it/s +epoch=55 global_step=21650 loss=4.38377 loss_avg=4.17094 acc=0.57812 acc_top1_avg=0.58987 acc_top5_avg=0.92214 lr=0.00100 gn=17.77858 time=58.60it/s +epoch=55 global_step=21700 loss=3.67211 loss_avg=4.16458 acc=0.64844 acc_top1_avg=0.59167 acc_top5_avg=0.92192 lr=0.00100 gn=14.76931 time=54.37it/s +epoch=55 global_step=21750 loss=4.55190 loss_avg=4.16398 acc=0.53125 acc_top1_avg=0.59263 acc_top5_avg=0.92264 lr=0.00100 gn=14.96526 time=52.11it/s +epoch=55 global_step=21800 loss=3.44105 loss_avg=4.17410 acc=0.67188 acc_top1_avg=0.59126 acc_top5_avg=0.92256 lr=0.00100 gn=15.48355 time=55.07it/s +epoch=55 global_step=21850 loss=4.58413 loss_avg=4.18956 acc=0.52344 acc_top1_avg=0.58886 acc_top5_avg=0.92305 lr=0.00100 gn=17.30287 time=54.31it/s +====================Eval==================== +epoch=55 global_step=21896 loss=1.17776 test_loss_avg=1.26814 acc=0.57812 test_acc_avg=0.63125 test_acc_top5_avg=0.97031 time=250.38it/s +epoch=55 global_step=21896 loss=4.13956 test_loss_avg=2.32691 acc=0.22656 test_acc_avg=0.49744 test_acc_top5_avg=0.91264 time=227.46it/s +epoch=55 global_step=21896 loss=0.06042 test_loss_avg=1.70742 acc=0.93750 test_acc_avg=0.62391 test_acc_top5_avg=0.93661 time=800.90it/s +curr_acc 0.6239 +BEST_ACC 0.6425 +curr_acc_top5 0.9366 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=3.38707 loss_avg=3.76115 acc=0.68750 acc_top1_avg=0.64258 acc_top5_avg=0.93555 lr=0.00100 gn=18.58423 time=53.66it/s +epoch=56 global_step=21950 loss=3.72153 loss_avg=4.05670 acc=0.62500 acc_top1_avg=0.60663 acc_top5_avg=0.92679 lr=0.00100 gn=17.43663 time=56.78it/s +epoch=56 global_step=22000 loss=4.45402 loss_avg=4.10208 acc=0.54688 acc_top1_avg=0.60141 acc_top5_avg=0.92435 lr=0.00100 gn=14.94115 time=52.27it/s +epoch=56 global_step=22050 loss=4.32121 loss_avg=4.13999 acc=0.56250 acc_top1_avg=0.59624 acc_top5_avg=0.92289 lr=0.00100 gn=19.17610 time=57.07it/s +epoch=56 global_step=22100 loss=4.45578 loss_avg=4.15544 acc=0.55469 acc_top1_avg=0.59432 acc_top5_avg=0.92260 lr=0.00100 gn=15.83923 time=60.88it/s +epoch=56 global_step=22150 loss=4.34379 loss_avg=4.16272 acc=0.57812 acc_top1_avg=0.59335 acc_top5_avg=0.92283 lr=0.00100 gn=15.57114 time=53.70it/s +epoch=56 global_step=22200 loss=3.57991 loss_avg=4.16035 acc=0.65625 acc_top1_avg=0.59375 acc_top5_avg=0.92231 lr=0.00100 gn=18.03945 time=55.82it/s +epoch=56 global_step=22250 loss=3.59232 loss_avg=4.15097 acc=0.64844 acc_top1_avg=0.59474 acc_top5_avg=0.92316 lr=0.00100 gn=17.69319 time=54.66it/s +====================Eval==================== +epoch=56 global_step=22287 loss=5.58709 test_loss_avg=1.46171 acc=0.00000 test_acc_avg=0.63522 test_acc_top5_avg=0.93239 time=236.78it/s +epoch=56 global_step=22287 loss=0.16088 test_loss_avg=1.86163 acc=0.94531 test_acc_avg=0.59097 test_acc_top5_avg=0.92444 time=259.36it/s +epoch=56 global_step=22287 loss=0.06185 test_loss_avg=1.79477 acc=0.93750 test_acc_avg=0.60473 test_acc_top5_avg=0.92731 time=845.46it/s +curr_acc 0.6047 +BEST_ACC 0.6425 +curr_acc_top5 0.9273 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=3.83726 loss_avg=4.01280 acc=0.63281 acc_top1_avg=0.60637 acc_top5_avg=0.91707 lr=0.00100 gn=18.79101 time=54.25it/s +epoch=57 global_step=22350 loss=3.89168 loss_avg=4.15043 acc=0.59375 acc_top1_avg=0.59363 acc_top5_avg=0.91605 lr=0.00100 gn=18.32986 time=54.37it/s +epoch=57 global_step=22400 loss=4.01615 loss_avg=4.14676 acc=0.60938 acc_top1_avg=0.59444 acc_top5_avg=0.91759 lr=0.00100 gn=19.31374 time=51.91it/s +epoch=57 global_step=22450 loss=4.01878 loss_avg=4.12742 acc=0.60938 acc_top1_avg=0.59610 acc_top5_avg=0.91818 lr=0.00100 gn=19.27925 time=57.08it/s +epoch=57 global_step=22500 loss=4.60096 loss_avg=4.11832 acc=0.52344 acc_top1_avg=0.59797 acc_top5_avg=0.91945 lr=0.00100 gn=21.31806 time=55.49it/s +epoch=57 global_step=22550 loss=4.89805 loss_avg=4.11434 acc=0.51562 acc_top1_avg=0.59821 acc_top5_avg=0.92104 lr=0.00100 gn=15.60331 time=55.26it/s +epoch=57 global_step=22600 loss=4.16189 loss_avg=4.10254 acc=0.57031 acc_top1_avg=0.59947 acc_top5_avg=0.92168 lr=0.00100 gn=18.64909 time=63.46it/s +epoch=57 global_step=22650 loss=4.33312 loss_avg=4.11093 acc=0.54688 acc_top1_avg=0.59848 acc_top5_avg=0.92134 lr=0.00100 gn=20.49096 time=55.51it/s +====================Eval==================== +epoch=57 global_step=22678 loss=2.08670 test_loss_avg=1.86026 acc=0.42188 test_acc_avg=0.57015 test_acc_top5_avg=0.89811 time=242.49it/s +epoch=57 global_step=22678 loss=0.09275 test_loss_avg=1.72769 acc=0.93750 test_acc_avg=0.61709 test_acc_top5_avg=0.92741 time=501.47it/s +curr_acc 0.6171 +BEST_ACC 0.6425 +curr_acc_top5 0.9274 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=3.44340 loss_avg=4.06611 acc=0.67188 acc_top1_avg=0.59943 acc_top5_avg=0.92756 lr=0.00100 gn=22.44747 time=48.13it/s +epoch=58 global_step=22750 loss=3.95985 loss_avg=4.05023 acc=0.61719 acc_top1_avg=0.60297 acc_top5_avg=0.92307 lr=0.00100 gn=19.21046 time=56.75it/s +epoch=58 global_step=22800 loss=3.73349 loss_avg=4.07198 acc=0.66406 acc_top1_avg=0.60207 acc_top5_avg=0.92424 lr=0.00100 gn=20.94544 time=59.03it/s +epoch=58 global_step=22850 loss=4.52216 loss_avg=4.08212 acc=0.52344 acc_top1_avg=0.60184 acc_top5_avg=0.92374 lr=0.00100 gn=22.53384 time=51.56it/s +epoch=58 global_step=22900 loss=3.92524 loss_avg=4.08887 acc=0.59375 acc_top1_avg=0.60079 acc_top5_avg=0.92335 lr=0.00100 gn=18.34367 time=46.89it/s +epoch=58 global_step=22950 loss=3.74215 loss_avg=4.10189 acc=0.62500 acc_top1_avg=0.59915 acc_top5_avg=0.92345 lr=0.00100 gn=20.56232 time=59.52it/s +epoch=58 global_step=23000 loss=3.84071 loss_avg=4.09456 acc=0.63281 acc_top1_avg=0.60008 acc_top5_avg=0.92382 lr=0.00100 gn=23.92850 time=53.84it/s +epoch=58 global_step=23050 loss=3.55659 loss_avg=4.10541 acc=0.67188 acc_top1_avg=0.59850 acc_top5_avg=0.92343 lr=0.00100 gn=21.87144 time=60.69it/s +====================Eval==================== +epoch=58 global_step=23069 loss=1.99114 test_loss_avg=0.83632 acc=0.48438 test_acc_avg=0.76910 test_acc_top5_avg=0.97309 time=238.95it/s +epoch=58 global_step=23069 loss=0.31520 test_loss_avg=2.02701 acc=0.91406 test_acc_avg=0.55653 test_acc_top5_avg=0.92980 time=220.25it/s +epoch=58 global_step=23069 loss=0.16164 test_loss_avg=1.77213 acc=0.87500 test_acc_avg=0.61027 test_acc_top5_avg=0.93908 time=854.24it/s +curr_acc 0.6103 +BEST_ACC 0.6425 +curr_acc_top5 0.9391 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=4.77908 loss_avg=4.10980 acc=0.54688 acc_top1_avg=0.60005 acc_top5_avg=0.91457 lr=0.00100 gn=21.53600 time=55.12it/s +epoch=59 global_step=23150 loss=4.15606 loss_avg=4.09031 acc=0.59375 acc_top1_avg=0.60204 acc_top5_avg=0.91512 lr=0.00100 gn=21.09083 time=55.38it/s +epoch=59 global_step=23200 loss=4.38666 loss_avg=4.09810 acc=0.56250 acc_top1_avg=0.59959 acc_top5_avg=0.91502 lr=0.00100 gn=18.59173 time=54.55it/s +epoch=59 global_step=23250 loss=4.39120 loss_avg=4.10439 acc=0.54688 acc_top1_avg=0.59893 acc_top5_avg=0.91790 lr=0.00100 gn=20.19014 time=53.14it/s +epoch=59 global_step=23300 loss=3.97549 loss_avg=4.10696 acc=0.62500 acc_top1_avg=0.59859 acc_top5_avg=0.91991 lr=0.00100 gn=21.62448 time=55.57it/s +epoch=59 global_step=23350 loss=4.64644 loss_avg=4.09020 acc=0.54688 acc_top1_avg=0.60095 acc_top5_avg=0.92079 lr=0.00100 gn=21.65409 time=62.64it/s +epoch=59 global_step=23400 loss=4.02072 loss_avg=4.08113 acc=0.62500 acc_top1_avg=0.60211 acc_top5_avg=0.92098 lr=0.00100 gn=20.82879 time=52.44it/s +epoch=59 global_step=23450 loss=4.77898 loss_avg=4.07965 acc=0.51562 acc_top1_avg=0.60228 acc_top5_avg=0.92112 lr=0.00100 gn=22.68281 time=57.12it/s +====================Eval==================== +epoch=59 global_step=23460 loss=0.38373 test_loss_avg=1.98298 acc=0.90625 test_acc_avg=0.56050 test_acc_top5_avg=0.90405 time=222.44it/s +epoch=59 global_step=23460 loss=0.22136 test_loss_avg=1.86798 acc=0.87500 test_acc_avg=0.59316 test_acc_top5_avg=0.93898 time=841.72it/s +curr_acc 0.5932 +BEST_ACC 0.6425 +curr_acc_top5 0.9390 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=3.96455 loss_avg=4.05400 acc=0.62500 acc_top1_avg=0.60684 acc_top5_avg=0.92207 lr=0.00100 gn=19.98022 time=57.74it/s +epoch=60 global_step=23550 loss=3.81006 loss_avg=4.06831 acc=0.60156 acc_top1_avg=0.60217 acc_top5_avg=0.92031 lr=0.00100 gn=22.44784 time=57.49it/s +epoch=60 global_step=23600 loss=4.14816 loss_avg=4.02422 acc=0.59375 acc_top1_avg=0.60731 acc_top5_avg=0.92254 lr=0.00100 gn=16.87427 time=54.68it/s +epoch=60 global_step=23650 loss=4.08476 loss_avg=4.02116 acc=0.58594 acc_top1_avg=0.60744 acc_top5_avg=0.92216 lr=0.00100 gn=17.92070 time=53.65it/s +epoch=60 global_step=23700 loss=3.36817 loss_avg=4.02598 acc=0.66406 acc_top1_avg=0.60765 acc_top5_avg=0.92168 lr=0.00100 gn=19.81490 time=60.17it/s +epoch=60 global_step=23750 loss=3.83838 loss_avg=4.03488 acc=0.64062 acc_top1_avg=0.60687 acc_top5_avg=0.92161 lr=0.00100 gn=22.13180 time=52.52it/s +epoch=60 global_step=23800 loss=3.85448 loss_avg=4.04873 acc=0.64844 acc_top1_avg=0.60556 acc_top5_avg=0.92080 lr=0.00100 gn=19.48882 time=53.57it/s +epoch=60 global_step=23850 loss=4.34203 loss_avg=4.05646 acc=0.57812 acc_top1_avg=0.60489 acc_top5_avg=0.92067 lr=0.00100 gn=22.72168 time=56.69it/s +====================Eval==================== +epoch=60 global_step=23851 loss=0.24532 test_loss_avg=1.15206 acc=0.92188 test_acc_avg=0.65859 test_acc_top5_avg=0.96719 time=242.42it/s +epoch=60 global_step=23851 loss=0.52039 test_loss_avg=2.22575 acc=0.82031 test_acc_avg=0.50195 test_acc_top5_avg=0.90977 time=242.49it/s +epoch=60 global_step=23851 loss=0.14409 test_loss_avg=1.74943 acc=0.87500 test_acc_avg=0.60265 test_acc_top5_avg=0.93028 time=829.24it/s +curr_acc 0.6027 +BEST_ACC 0.6425 +curr_acc_top5 0.9303 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=4.27238 loss_avg=4.02813 acc=0.60938 acc_top1_avg=0.60969 acc_top5_avg=0.92124 lr=0.00100 gn=23.91512 time=63.38it/s +epoch=61 global_step=23950 loss=4.25957 loss_avg=4.01319 acc=0.59375 acc_top1_avg=0.61182 acc_top5_avg=0.92132 lr=0.00100 gn=24.32038 time=52.51it/s +epoch=61 global_step=24000 loss=4.23854 loss_avg=3.99974 acc=0.60156 acc_top1_avg=0.61352 acc_top5_avg=0.92172 lr=0.00100 gn=21.06033 time=50.58it/s +epoch=61 global_step=24050 loss=3.59220 loss_avg=4.00891 acc=0.66406 acc_top1_avg=0.61220 acc_top5_avg=0.92262 lr=0.00100 gn=23.46288 time=55.73it/s +epoch=61 global_step=24100 loss=4.71442 loss_avg=4.01788 acc=0.54688 acc_top1_avg=0.61145 acc_top5_avg=0.92275 lr=0.00100 gn=21.05235 time=55.17it/s +epoch=61 global_step=24150 loss=4.58233 loss_avg=4.03021 acc=0.50781 acc_top1_avg=0.60938 acc_top5_avg=0.92180 lr=0.00100 gn=21.69665 time=55.59it/s +epoch=61 global_step=24200 loss=4.28996 loss_avg=4.03304 acc=0.58594 acc_top1_avg=0.60920 acc_top5_avg=0.92179 lr=0.00100 gn=21.12389 time=53.88it/s +====================Eval==================== +epoch=61 global_step=24242 loss=5.44932 test_loss_avg=2.20075 acc=0.00000 test_acc_avg=0.49672 test_acc_top5_avg=0.88533 time=230.36it/s +epoch=61 global_step=24242 loss=0.15043 test_loss_avg=1.73886 acc=0.93750 test_acc_avg=0.60433 test_acc_top5_avg=0.92949 time=854.06it/s +curr_acc 0.6043 +BEST_ACC 0.6425 +curr_acc_top5 0.9295 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=4.12622 loss_avg=4.24705 acc=0.58594 acc_top1_avg=0.57910 acc_top5_avg=0.90137 lr=0.00100 gn=19.56984 time=48.73it/s +epoch=62 global_step=24300 loss=3.60711 loss_avg=4.05167 acc=0.64844 acc_top1_avg=0.60709 acc_top5_avg=0.91595 lr=0.00100 gn=22.57238 time=55.30it/s +epoch=62 global_step=24350 loss=4.06560 loss_avg=4.03675 acc=0.60156 acc_top1_avg=0.60786 acc_top5_avg=0.92057 lr=0.00100 gn=23.30485 time=52.11it/s +epoch=62 global_step=24400 loss=4.37589 loss_avg=4.00919 acc=0.57812 acc_top1_avg=0.61170 acc_top5_avg=0.92039 lr=0.00100 gn=23.33412 time=56.43it/s +epoch=62 global_step=24450 loss=4.86055 loss_avg=4.00946 acc=0.50000 acc_top1_avg=0.61137 acc_top5_avg=0.92214 lr=0.00100 gn=22.30805 time=55.01it/s +epoch=62 global_step=24500 loss=3.65763 loss_avg=4.01238 acc=0.64062 acc_top1_avg=0.61089 acc_top5_avg=0.92018 lr=0.00100 gn=25.06022 time=62.18it/s +epoch=62 global_step=24550 loss=4.43262 loss_avg=4.01513 acc=0.57031 acc_top1_avg=0.61019 acc_top5_avg=0.91962 lr=0.00100 gn=24.64861 time=54.30it/s +epoch=62 global_step=24600 loss=4.03104 loss_avg=4.01467 acc=0.60156 acc_top1_avg=0.61007 acc_top5_avg=0.91969 lr=0.00100 gn=23.30301 time=59.80it/s +====================Eval==================== +epoch=62 global_step=24633 loss=1.72942 test_loss_avg=1.62614 acc=0.52344 test_acc_avg=0.53516 test_acc_top5_avg=0.95312 time=240.61it/s +epoch=62 global_step=24633 loss=5.54599 test_loss_avg=2.18393 acc=0.00000 test_acc_avg=0.51547 test_acc_top5_avg=0.90249 time=244.17it/s +epoch=62 global_step=24633 loss=0.24277 test_loss_avg=1.71196 acc=0.87500 test_acc_avg=0.62203 test_acc_top5_avg=0.93147 time=811.75it/s +curr_acc 0.6220 +BEST_ACC 0.6425 +curr_acc_top5 0.9315 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=3.57818 loss_avg=3.91467 acc=0.65625 acc_top1_avg=0.61994 acc_top5_avg=0.92601 lr=0.00100 gn=20.88750 time=51.13it/s +epoch=63 global_step=24700 loss=4.24413 loss_avg=3.97742 acc=0.57812 acc_top1_avg=0.61521 acc_top5_avg=0.92269 lr=0.00100 gn=24.85784 time=52.33it/s +epoch=63 global_step=24750 loss=4.09297 loss_avg=3.95411 acc=0.57812 acc_top1_avg=0.61792 acc_top5_avg=0.92261 lr=0.00100 gn=25.25329 time=52.65it/s +epoch=63 global_step=24800 loss=4.13299 loss_avg=3.94834 acc=0.60156 acc_top1_avg=0.61850 acc_top5_avg=0.92169 lr=0.00100 gn=25.34225 time=57.67it/s +epoch=63 global_step=24850 loss=3.54061 loss_avg=3.96645 acc=0.68750 acc_top1_avg=0.61643 acc_top5_avg=0.92151 lr=0.00100 gn=21.71340 time=59.34it/s +epoch=63 global_step=24900 loss=4.11237 loss_avg=3.95547 acc=0.58594 acc_top1_avg=0.61728 acc_top5_avg=0.92173 lr=0.00100 gn=17.04854 time=58.46it/s +epoch=63 global_step=24950 loss=4.14941 loss_avg=3.97025 acc=0.62500 acc_top1_avg=0.61563 acc_top5_avg=0.92126 lr=0.00100 gn=28.08132 time=63.44it/s +epoch=63 global_step=25000 loss=3.98457 loss_avg=3.97975 acc=0.62500 acc_top1_avg=0.61517 acc_top5_avg=0.92087 lr=0.00100 gn=20.44537 time=56.60it/s +====================Eval==================== +epoch=63 global_step=25024 loss=1.66341 test_loss_avg=1.06159 acc=0.53906 test_acc_avg=0.69090 test_acc_top5_avg=0.96977 time=242.64it/s +epoch=63 global_step=25024 loss=0.29825 test_loss_avg=1.89665 acc=0.90625 test_acc_avg=0.56464 test_acc_top5_avg=0.91642 time=253.26it/s +epoch=63 global_step=25024 loss=0.25809 test_loss_avg=1.76829 acc=0.87500 test_acc_avg=0.59187 test_acc_top5_avg=0.92267 time=844.77it/s +curr_acc 0.5919 +BEST_ACC 0.6425 +curr_acc_top5 0.9227 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=3.79978 loss_avg=3.99920 acc=0.65625 acc_top1_avg=0.61328 acc_top5_avg=0.91376 lr=0.00100 gn=33.29833 time=50.64it/s +epoch=64 global_step=25100 loss=3.21739 loss_avg=3.94439 acc=0.67969 acc_top1_avg=0.61883 acc_top5_avg=0.92044 lr=0.00100 gn=23.79202 time=56.69it/s +epoch=64 global_step=25150 loss=3.64582 loss_avg=3.95067 acc=0.64062 acc_top1_avg=0.61756 acc_top5_avg=0.92063 lr=0.00100 gn=24.46660 time=61.06it/s +epoch=64 global_step=25200 loss=3.68270 loss_avg=3.96292 acc=0.64062 acc_top1_avg=0.61599 acc_top5_avg=0.92041 lr=0.00100 gn=24.14977 time=58.58it/s +epoch=64 global_step=25250 loss=4.40933 loss_avg=3.97757 acc=0.57031 acc_top1_avg=0.61459 acc_top5_avg=0.92025 lr=0.00100 gn=25.25537 time=62.16it/s +epoch=64 global_step=25300 loss=3.70732 loss_avg=3.97769 acc=0.62500 acc_top1_avg=0.61487 acc_top5_avg=0.92023 lr=0.00100 gn=28.39444 time=58.22it/s +epoch=64 global_step=25350 loss=4.14570 loss_avg=3.96840 acc=0.60938 acc_top1_avg=0.61611 acc_top5_avg=0.92113 lr=0.00100 gn=25.91796 time=60.72it/s +epoch=64 global_step=25400 loss=3.68091 loss_avg=3.97707 acc=0.64844 acc_top1_avg=0.61546 acc_top5_avg=0.92131 lr=0.00100 gn=23.26181 time=59.48it/s +====================Eval==================== +epoch=64 global_step=25415 loss=1.41753 test_loss_avg=1.78874 acc=0.57031 test_acc_avg=0.57706 test_acc_top5_avg=0.89648 time=239.84it/s +epoch=64 global_step=25415 loss=0.15987 test_loss_avg=1.71934 acc=0.93750 test_acc_avg=0.61284 test_acc_top5_avg=0.92820 time=495.84it/s +curr_acc 0.6128 +BEST_ACC 0.6425 +curr_acc_top5 0.9282 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=3.92512 loss_avg=3.95800 acc=0.63281 acc_top1_avg=0.61540 acc_top5_avg=0.91719 lr=0.00100 gn=26.65748 time=56.24it/s +epoch=65 global_step=25500 loss=4.59326 loss_avg=3.92725 acc=0.55469 acc_top1_avg=0.61884 acc_top5_avg=0.92215 lr=0.00100 gn=23.62458 time=55.16it/s +epoch=65 global_step=25550 loss=3.65510 loss_avg=3.87257 acc=0.64844 acc_top1_avg=0.62529 acc_top5_avg=0.92269 lr=0.00100 gn=21.26263 time=51.63it/s +epoch=65 global_step=25600 loss=3.71603 loss_avg=3.90083 acc=0.64844 acc_top1_avg=0.62179 acc_top5_avg=0.92175 lr=0.00100 gn=26.80381 time=54.59it/s +epoch=65 global_step=25650 loss=3.52296 loss_avg=3.91768 acc=0.66406 acc_top1_avg=0.62064 acc_top5_avg=0.92101 lr=0.00100 gn=24.31812 time=57.33it/s +epoch=65 global_step=25700 loss=4.52839 loss_avg=3.92729 acc=0.53906 acc_top1_avg=0.61996 acc_top5_avg=0.92108 lr=0.00100 gn=23.80996 time=63.09it/s +epoch=65 global_step=25750 loss=3.89576 loss_avg=3.93599 acc=0.62500 acc_top1_avg=0.61931 acc_top5_avg=0.92201 lr=0.00100 gn=27.39553 time=54.27it/s +epoch=65 global_step=25800 loss=3.82199 loss_avg=3.93987 acc=0.64062 acc_top1_avg=0.61881 acc_top5_avg=0.92161 lr=0.00100 gn=24.24449 time=56.79it/s +====================Eval==================== +epoch=65 global_step=25806 loss=0.54821 test_loss_avg=0.77110 acc=0.89844 test_acc_avg=0.78177 test_acc_top5_avg=0.98333 time=241.44it/s +epoch=65 global_step=25806 loss=0.14876 test_loss_avg=2.11130 acc=0.95312 test_acc_avg=0.52560 test_acc_top5_avg=0.91779 time=241.07it/s +epoch=65 global_step=25806 loss=0.02312 test_loss_avg=1.76975 acc=1.00000 test_acc_avg=0.60097 test_acc_top5_avg=0.93206 time=820.48it/s +curr_acc 0.6010 +BEST_ACC 0.6425 +curr_acc_top5 0.9321 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=3.88440 loss_avg=3.90187 acc=0.61719 acc_top1_avg=0.62180 acc_top5_avg=0.92259 lr=0.00100 gn=26.74891 time=63.39it/s +epoch=66 global_step=25900 loss=3.56088 loss_avg=3.88674 acc=0.63281 acc_top1_avg=0.62375 acc_top5_avg=0.92354 lr=0.00100 gn=25.84534 time=56.23it/s +epoch=66 global_step=25950 loss=3.59644 loss_avg=3.89613 acc=0.65625 acc_top1_avg=0.62348 acc_top5_avg=0.92263 lr=0.00100 gn=24.21997 time=53.36it/s +epoch=66 global_step=26000 loss=4.28527 loss_avg=3.90201 acc=0.57812 acc_top1_avg=0.62295 acc_top5_avg=0.92328 lr=0.00100 gn=27.37778 time=60.41it/s +epoch=66 global_step=26050 loss=3.74119 loss_avg=3.89840 acc=0.64844 acc_top1_avg=0.62391 acc_top5_avg=0.92220 lr=0.00100 gn=27.50099 time=59.83it/s +epoch=66 global_step=26100 loss=3.96122 loss_avg=3.91698 acc=0.60938 acc_top1_avg=0.62205 acc_top5_avg=0.92214 lr=0.00100 gn=30.20137 time=62.82it/s +epoch=66 global_step=26150 loss=3.48927 loss_avg=3.91367 acc=0.68750 acc_top1_avg=0.62259 acc_top5_avg=0.92231 lr=0.00100 gn=27.84080 time=50.19it/s +====================Eval==================== +epoch=66 global_step=26197 loss=0.49931 test_loss_avg=2.03528 acc=0.88281 test_acc_avg=0.53385 test_acc_top5_avg=0.87283 time=44.62it/s +epoch=66 global_step=26197 loss=0.26805 test_loss_avg=1.80645 acc=0.87500 test_acc_avg=0.60661 test_acc_top5_avg=0.92791 time=501.77it/s +curr_acc 0.6066 +BEST_ACC 0.6425 +curr_acc_top5 0.9279 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=3.52788 loss_avg=3.76245 acc=0.66406 acc_top1_avg=0.63802 acc_top5_avg=0.92969 lr=0.00100 gn=23.29071 time=48.54it/s +epoch=67 global_step=26250 loss=4.16208 loss_avg=3.85899 acc=0.59375 acc_top1_avg=0.62839 acc_top5_avg=0.92143 lr=0.00100 gn=24.07092 time=59.69it/s +epoch=67 global_step=26300 loss=3.94612 loss_avg=3.85819 acc=0.60938 acc_top1_avg=0.62856 acc_top5_avg=0.92339 lr=0.00100 gn=34.27399 time=53.50it/s +epoch=67 global_step=26350 loss=3.74110 loss_avg=3.85048 acc=0.65625 acc_top1_avg=0.62908 acc_top5_avg=0.92264 lr=0.00100 gn=26.86581 time=57.51it/s +epoch=67 global_step=26400 loss=4.12708 loss_avg=3.86998 acc=0.60156 acc_top1_avg=0.62739 acc_top5_avg=0.92222 lr=0.00100 gn=24.67000 time=58.32it/s +epoch=67 global_step=26450 loss=4.11348 loss_avg=3.87244 acc=0.60156 acc_top1_avg=0.62735 acc_top5_avg=0.92200 lr=0.00100 gn=26.85348 time=52.05it/s +epoch=67 global_step=26500 loss=4.16067 loss_avg=3.88568 acc=0.59375 acc_top1_avg=0.62611 acc_top5_avg=0.92136 lr=0.00100 gn=34.58439 time=61.07it/s +epoch=67 global_step=26550 loss=3.87351 loss_avg=3.90079 acc=0.62500 acc_top1_avg=0.62429 acc_top5_avg=0.92097 lr=0.00100 gn=25.61045 time=56.02it/s +====================Eval==================== +epoch=67 global_step=26588 loss=1.32412 test_loss_avg=1.40331 acc=0.57812 test_acc_avg=0.56362 test_acc_top5_avg=0.96875 time=216.08it/s +epoch=67 global_step=26588 loss=0.90303 test_loss_avg=2.30160 acc=0.75781 test_acc_avg=0.47423 test_acc_top5_avg=0.89158 time=220.47it/s +epoch=67 global_step=26588 loss=0.08646 test_loss_avg=1.74959 acc=0.93750 test_acc_avg=0.59395 test_acc_top5_avg=0.91960 time=550.00it/s +curr_acc 0.5939 +BEST_ACC 0.6425 +curr_acc_top5 0.9196 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=3.98331 loss_avg=3.83941 acc=0.62500 acc_top1_avg=0.63477 acc_top5_avg=0.93555 lr=0.00100 gn=29.13306 time=52.00it/s +epoch=68 global_step=26650 loss=3.01640 loss_avg=3.87324 acc=0.72656 acc_top1_avg=0.62916 acc_top5_avg=0.92288 lr=0.00100 gn=24.57465 time=55.15it/s +epoch=68 global_step=26700 loss=3.56569 loss_avg=3.86422 acc=0.65625 acc_top1_avg=0.62953 acc_top5_avg=0.92181 lr=0.00100 gn=24.24659 time=63.46it/s +epoch=68 global_step=26750 loss=3.96774 loss_avg=3.86007 acc=0.60938 acc_top1_avg=0.63011 acc_top5_avg=0.92009 lr=0.00100 gn=28.96371 time=55.60it/s +epoch=68 global_step=26800 loss=4.28567 loss_avg=3.87904 acc=0.57812 acc_top1_avg=0.62824 acc_top5_avg=0.91970 lr=0.00100 gn=24.35863 time=55.46it/s +epoch=68 global_step=26850 loss=3.40068 loss_avg=3.87423 acc=0.66406 acc_top1_avg=0.62879 acc_top5_avg=0.92068 lr=0.00100 gn=33.49151 time=55.63it/s +epoch=68 global_step=26900 loss=4.00863 loss_avg=3.88576 acc=0.64062 acc_top1_avg=0.62798 acc_top5_avg=0.91990 lr=0.00100 gn=33.72494 time=61.06it/s +epoch=68 global_step=26950 loss=4.24829 loss_avg=3.88671 acc=0.58594 acc_top1_avg=0.62787 acc_top5_avg=0.92030 lr=0.00100 gn=28.51307 time=43.71it/s +====================Eval==================== +epoch=68 global_step=26979 loss=5.28810 test_loss_avg=2.15545 acc=0.00000 test_acc_avg=0.48131 test_acc_top5_avg=0.89704 time=238.83it/s +epoch=68 global_step=26979 loss=0.19840 test_loss_avg=1.85658 acc=0.94531 test_acc_avg=0.58804 test_acc_top5_avg=0.93049 time=251.31it/s +epoch=68 global_step=26979 loss=0.08581 test_loss_avg=1.83417 acc=0.93750 test_acc_avg=0.59246 test_acc_top5_avg=0.93137 time=819.52it/s +curr_acc 0.5925 +BEST_ACC 0.6425 +curr_acc_top5 0.9314 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=4.20987 loss_avg=3.84886 acc=0.57812 acc_top1_avg=0.63132 acc_top5_avg=0.91146 lr=0.00100 gn=22.94313 time=57.28it/s +epoch=69 global_step=27050 loss=3.55479 loss_avg=3.82958 acc=0.67969 acc_top1_avg=0.63380 acc_top5_avg=0.92000 lr=0.00100 gn=26.55954 time=55.22it/s +epoch=69 global_step=27100 loss=3.92602 loss_avg=3.81681 acc=0.60156 acc_top1_avg=0.63527 acc_top5_avg=0.91910 lr=0.00100 gn=33.16019 time=53.31it/s +epoch=69 global_step=27150 loss=3.95938 loss_avg=3.81820 acc=0.59375 acc_top1_avg=0.63405 acc_top5_avg=0.92000 lr=0.00100 gn=30.28152 time=52.36it/s +epoch=69 global_step=27200 loss=3.37378 loss_avg=3.84939 acc=0.70312 acc_top1_avg=0.63062 acc_top5_avg=0.91951 lr=0.00100 gn=33.01164 time=58.56it/s +epoch=69 global_step=27250 loss=4.22238 loss_avg=3.85369 acc=0.60156 acc_top1_avg=0.63042 acc_top5_avg=0.91968 lr=0.00100 gn=30.53456 time=53.21it/s +epoch=69 global_step=27300 loss=4.28477 loss_avg=3.86571 acc=0.56250 acc_top1_avg=0.62911 acc_top5_avg=0.91930 lr=0.00100 gn=24.98391 time=58.16it/s +epoch=69 global_step=27350 loss=4.37565 loss_avg=3.86667 acc=0.57812 acc_top1_avg=0.62936 acc_top5_avg=0.91886 lr=0.00100 gn=31.08676 time=55.42it/s +====================Eval==================== +epoch=69 global_step=27370 loss=5.45993 test_loss_avg=2.12356 acc=0.00000 test_acc_avg=0.50909 test_acc_top5_avg=0.90737 time=231.54it/s +epoch=69 global_step=27370 loss=0.06662 test_loss_avg=1.79052 acc=1.00000 test_acc_avg=0.59355 test_acc_top5_avg=0.92890 time=636.95it/s +curr_acc 0.5936 +BEST_ACC 0.6425 +curr_acc_top5 0.9289 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=3.33574 loss_avg=3.79117 acc=0.67969 acc_top1_avg=0.63516 acc_top5_avg=0.91875 lr=0.00100 gn=27.69509 time=55.04it/s +epoch=70 global_step=27450 loss=4.25365 loss_avg=3.78610 acc=0.60938 acc_top1_avg=0.63750 acc_top5_avg=0.92119 lr=0.00100 gn=29.39062 time=59.90it/s +epoch=70 global_step=27500 loss=3.73120 loss_avg=3.80237 acc=0.64062 acc_top1_avg=0.63708 acc_top5_avg=0.91983 lr=0.00100 gn=23.92354 time=54.94it/s +epoch=70 global_step=27550 loss=3.61646 loss_avg=3.78695 acc=0.64844 acc_top1_avg=0.63785 acc_top5_avg=0.91975 lr=0.00100 gn=26.89762 time=52.46it/s +epoch=70 global_step=27600 loss=3.79619 loss_avg=3.79806 acc=0.64062 acc_top1_avg=0.63706 acc_top5_avg=0.91943 lr=0.00100 gn=29.08220 time=51.26it/s +epoch=70 global_step=27650 loss=4.25728 loss_avg=3.82457 acc=0.58594 acc_top1_avg=0.63410 acc_top5_avg=0.91992 lr=0.00100 gn=34.50916 time=63.18it/s +epoch=70 global_step=27700 loss=4.17518 loss_avg=3.82812 acc=0.59375 acc_top1_avg=0.63393 acc_top5_avg=0.91998 lr=0.00100 gn=32.82275 time=54.36it/s +epoch=70 global_step=27750 loss=3.73807 loss_avg=3.84092 acc=0.64844 acc_top1_avg=0.63267 acc_top5_avg=0.91926 lr=0.00100 gn=32.14289 time=58.63it/s +====================Eval==================== +epoch=70 global_step=27761 loss=1.50835 test_loss_avg=1.14615 acc=0.56250 test_acc_avg=0.66914 test_acc_top5_avg=0.96758 time=242.92it/s +epoch=70 global_step=27761 loss=0.12715 test_loss_avg=1.89788 acc=0.95312 test_acc_avg=0.56406 test_acc_top5_avg=0.91328 time=236.71it/s +epoch=70 global_step=27761 loss=0.04743 test_loss_avg=1.70133 acc=1.00000 test_acc_avg=0.60809 test_acc_top5_avg=0.92296 time=806.13it/s +curr_acc 0.6081 +BEST_ACC 0.6425 +curr_acc_top5 0.9230 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=4.28410 loss_avg=3.85328 acc=0.58594 acc_top1_avg=0.62760 acc_top5_avg=0.91627 lr=0.00100 gn=32.81502 time=62.85it/s +epoch=71 global_step=27850 loss=3.73017 loss_avg=3.85307 acc=0.64062 acc_top1_avg=0.63027 acc_top5_avg=0.91968 lr=0.00100 gn=31.23283 time=56.25it/s +epoch=71 global_step=27900 loss=3.50271 loss_avg=3.82047 acc=0.67188 acc_top1_avg=0.63478 acc_top5_avg=0.92064 lr=0.00100 gn=26.89874 time=59.32it/s +epoch=71 global_step=27950 loss=3.87883 loss_avg=3.80198 acc=0.63281 acc_top1_avg=0.63670 acc_top5_avg=0.92076 lr=0.00100 gn=27.68109 time=54.72it/s +epoch=71 global_step=28000 loss=4.23509 loss_avg=3.81271 acc=0.60938 acc_top1_avg=0.63585 acc_top5_avg=0.92119 lr=0.00100 gn=38.76059 time=55.47it/s +epoch=71 global_step=28050 loss=4.13715 loss_avg=3.82385 acc=0.59375 acc_top1_avg=0.63468 acc_top5_avg=0.92158 lr=0.00100 gn=28.21391 time=61.12it/s +epoch=71 global_step=28100 loss=4.00260 loss_avg=3.83975 acc=0.63281 acc_top1_avg=0.63327 acc_top5_avg=0.92081 lr=0.00100 gn=37.47520 time=52.84it/s +epoch=71 global_step=28150 loss=3.44814 loss_avg=3.83945 acc=0.68750 acc_top1_avg=0.63305 acc_top5_avg=0.92041 lr=0.00100 gn=31.66606 time=61.08it/s +====================Eval==================== +epoch=71 global_step=28152 loss=2.09097 test_loss_avg=1.80168 acc=0.41406 test_acc_avg=0.57222 test_acc_top5_avg=0.87824 time=227.77it/s +epoch=71 global_step=28152 loss=0.02434 test_loss_avg=1.71011 acc=1.00000 test_acc_avg=0.60562 test_acc_top5_avg=0.91505 time=554.22it/s +curr_acc 0.6056 +BEST_ACC 0.6425 +curr_acc_top5 0.9151 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=3.53682 loss_avg=3.74280 acc=0.67188 acc_top1_avg=0.64274 acc_top5_avg=0.91715 lr=0.00100 gn=25.18407 time=62.84it/s +epoch=72 global_step=28250 loss=3.57851 loss_avg=3.80851 acc=0.64062 acc_top1_avg=0.63481 acc_top5_avg=0.91757 lr=0.00100 gn=32.34918 time=47.99it/s +epoch=72 global_step=28300 loss=3.13164 loss_avg=3.79704 acc=0.69531 acc_top1_avg=0.63582 acc_top5_avg=0.92019 lr=0.00100 gn=26.74298 time=52.82it/s +epoch=72 global_step=28350 loss=4.26399 loss_avg=3.80183 acc=0.59375 acc_top1_avg=0.63538 acc_top5_avg=0.92010 lr=0.00100 gn=36.04864 time=56.13it/s +epoch=72 global_step=28400 loss=3.97856 loss_avg=3.81577 acc=0.61719 acc_top1_avg=0.63432 acc_top5_avg=0.92027 lr=0.00100 gn=30.75603 time=54.85it/s +epoch=72 global_step=28450 loss=4.02880 loss_avg=3.82515 acc=0.60938 acc_top1_avg=0.63355 acc_top5_avg=0.91944 lr=0.00100 gn=35.46079 time=59.57it/s +epoch=72 global_step=28500 loss=3.54386 loss_avg=3.81931 acc=0.69531 acc_top1_avg=0.63447 acc_top5_avg=0.91905 lr=0.00100 gn=33.97118 time=55.05it/s +====================Eval==================== +epoch=72 global_step=28543 loss=0.18642 test_loss_avg=1.05652 acc=0.94531 test_acc_avg=0.70508 test_acc_top5_avg=0.97135 time=238.99it/s +epoch=72 global_step=28543 loss=0.81253 test_loss_avg=2.22536 acc=0.72656 test_acc_avg=0.50794 test_acc_top5_avg=0.91167 time=241.07it/s +epoch=72 global_step=28543 loss=0.02328 test_loss_avg=1.78514 acc=1.00000 test_acc_avg=0.60285 test_acc_top5_avg=0.93018 time=812.69it/s +curr_acc 0.6028 +BEST_ACC 0.6425 +curr_acc_top5 0.9302 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=2.99409 loss_avg=3.68209 acc=0.73438 acc_top1_avg=0.65737 acc_top5_avg=0.91183 lr=0.00100 gn=26.76324 time=59.01it/s +epoch=73 global_step=28600 loss=3.59804 loss_avg=3.80452 acc=0.63281 acc_top1_avg=0.63747 acc_top5_avg=0.91530 lr=0.00100 gn=28.02050 time=55.48it/s +epoch=73 global_step=28650 loss=3.94909 loss_avg=3.78672 acc=0.64062 acc_top1_avg=0.64019 acc_top5_avg=0.91859 lr=0.00100 gn=31.12064 time=49.26it/s +epoch=73 global_step=28700 loss=3.97054 loss_avg=3.77976 acc=0.60938 acc_top1_avg=0.64077 acc_top5_avg=0.91824 lr=0.00100 gn=31.80676 time=51.30it/s +epoch=73 global_step=28750 loss=3.87369 loss_avg=3.78159 acc=0.63281 acc_top1_avg=0.64055 acc_top5_avg=0.91840 lr=0.00100 gn=31.10392 time=58.58it/s +epoch=73 global_step=28800 loss=3.97105 loss_avg=3.79397 acc=0.60938 acc_top1_avg=0.63996 acc_top5_avg=0.91829 lr=0.00100 gn=32.10359 time=59.09it/s +epoch=73 global_step=28850 loss=3.50965 loss_avg=3.79458 acc=0.66406 acc_top1_avg=0.64042 acc_top5_avg=0.91846 lr=0.00100 gn=34.68369 time=55.89it/s +epoch=73 global_step=28900 loss=3.60821 loss_avg=3.80077 acc=0.67188 acc_top1_avg=0.63927 acc_top5_avg=0.91875 lr=0.00100 gn=33.11522 time=52.14it/s +====================Eval==================== +epoch=73 global_step=28934 loss=0.35827 test_loss_avg=2.10810 acc=0.89062 test_acc_avg=0.51491 test_acc_top5_avg=0.87784 time=247.83it/s +epoch=73 global_step=28934 loss=0.21859 test_loss_avg=1.80749 acc=0.87500 test_acc_avg=0.59454 test_acc_top5_avg=0.93058 time=674.54it/s +curr_acc 0.5945 +BEST_ACC 0.6425 +curr_acc_top5 0.9306 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=4.11768 loss_avg=3.73287 acc=0.59375 acc_top1_avg=0.64746 acc_top5_avg=0.92480 lr=0.00100 gn=30.20235 time=51.14it/s +epoch=74 global_step=29000 loss=3.84693 loss_avg=3.77137 acc=0.65625 acc_top1_avg=0.64228 acc_top5_avg=0.91797 lr=0.00100 gn=36.76108 time=62.35it/s +epoch=74 global_step=29050 loss=3.39849 loss_avg=3.74165 acc=0.67188 acc_top1_avg=0.64601 acc_top5_avg=0.91898 lr=0.00100 gn=29.30416 time=44.86it/s +epoch=74 global_step=29100 loss=3.64377 loss_avg=3.74031 acc=0.67188 acc_top1_avg=0.64604 acc_top5_avg=0.91806 lr=0.00100 gn=31.45210 time=49.55it/s +epoch=74 global_step=29150 loss=4.10443 loss_avg=3.78147 acc=0.60156 acc_top1_avg=0.64088 acc_top5_avg=0.91641 lr=0.00100 gn=29.20872 time=56.49it/s +epoch=74 global_step=29200 loss=3.69479 loss_avg=3.77651 acc=0.63281 acc_top1_avg=0.64112 acc_top5_avg=0.91671 lr=0.00100 gn=34.85465 time=61.29it/s +epoch=74 global_step=29250 loss=3.98411 loss_avg=3.78390 acc=0.61719 acc_top1_avg=0.64001 acc_top5_avg=0.91708 lr=0.00100 gn=31.36569 time=52.48it/s +epoch=74 global_step=29300 loss=3.78279 loss_avg=3.79103 acc=0.64844 acc_top1_avg=0.63926 acc_top5_avg=0.91733 lr=0.00100 gn=30.41176 time=62.04it/s +====================Eval==================== +epoch=74 global_step=29325 loss=1.58560 test_loss_avg=1.68821 acc=0.53906 test_acc_avg=0.50000 test_acc_top5_avg=0.96680 time=244.05it/s +epoch=74 global_step=29325 loss=5.16031 test_loss_avg=2.42878 acc=0.00000 test_acc_avg=0.45920 test_acc_top5_avg=0.88542 time=241.62it/s +epoch=74 global_step=29325 loss=0.17956 test_loss_avg=1.82786 acc=0.87500 test_acc_avg=0.58307 test_acc_top5_avg=0.91782 time=638.21it/s +curr_acc 0.5831 +BEST_ACC 0.6425 +curr_acc_top5 0.9178 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=3.41488 loss_avg=3.75531 acc=0.67188 acc_top1_avg=0.64219 acc_top5_avg=0.92000 lr=0.00100 gn=28.61795 time=53.06it/s +epoch=75 global_step=29400 loss=2.92761 loss_avg=3.76972 acc=0.71094 acc_top1_avg=0.64312 acc_top5_avg=0.91802 lr=0.00100 gn=34.95644 time=57.82it/s +epoch=75 global_step=29450 loss=3.39903 loss_avg=3.74195 acc=0.68750 acc_top1_avg=0.64644 acc_top5_avg=0.91956 lr=0.00100 gn=39.16194 time=60.18it/s +epoch=75 global_step=29500 loss=4.23823 loss_avg=3.73938 acc=0.57812 acc_top1_avg=0.64643 acc_top5_avg=0.91893 lr=0.00100 gn=33.21478 time=59.98it/s +epoch=75 global_step=29550 loss=3.54979 loss_avg=3.74291 acc=0.68750 acc_top1_avg=0.64611 acc_top5_avg=0.91854 lr=0.00100 gn=37.96476 time=57.66it/s +epoch=75 global_step=29600 loss=3.39911 loss_avg=3.74400 acc=0.68750 acc_top1_avg=0.64639 acc_top5_avg=0.91901 lr=0.00100 gn=38.18630 time=56.52it/s +epoch=75 global_step=29650 loss=3.99168 loss_avg=3.75393 acc=0.61719 acc_top1_avg=0.64483 acc_top5_avg=0.91901 lr=0.00100 gn=36.50350 time=52.00it/s +epoch=75 global_step=29700 loss=3.39309 loss_avg=3.75949 acc=0.69531 acc_top1_avg=0.64373 acc_top5_avg=0.91904 lr=0.00100 gn=38.16209 time=56.30it/s +====================Eval==================== +epoch=75 global_step=29716 loss=5.13286 test_loss_avg=1.73476 acc=0.00000 test_acc_avg=0.55031 test_acc_top5_avg=0.92656 time=235.86it/s +epoch=75 global_step=29716 loss=0.13108 test_loss_avg=1.95048 acc=0.96875 test_acc_avg=0.55917 test_acc_top5_avg=0.91781 time=256.99it/s +epoch=75 global_step=29716 loss=0.14445 test_loss_avg=1.85965 acc=0.87500 test_acc_avg=0.57803 test_acc_top5_avg=0.92197 time=838.02it/s +curr_acc 0.5780 +BEST_ACC 0.6425 +curr_acc_top5 0.9220 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=3.62356 loss_avg=3.69406 acc=0.67188 acc_top1_avg=0.64384 acc_top5_avg=0.91797 lr=0.00100 gn=34.11885 time=60.76it/s +epoch=76 global_step=29800 loss=3.18483 loss_avg=3.69743 acc=0.71094 acc_top1_avg=0.64676 acc_top5_avg=0.91955 lr=0.00100 gn=34.56723 time=53.34it/s +epoch=76 global_step=29850 loss=3.66173 loss_avg=3.73759 acc=0.64844 acc_top1_avg=0.64389 acc_top5_avg=0.91943 lr=0.00100 gn=33.96563 time=54.84it/s +epoch=76 global_step=29900 loss=4.37720 loss_avg=3.74033 acc=0.55469 acc_top1_avg=0.64351 acc_top5_avg=0.92030 lr=0.00100 gn=37.89879 time=55.33it/s +epoch=76 global_step=29950 loss=3.77191 loss_avg=3.74203 acc=0.63281 acc_top1_avg=0.64390 acc_top5_avg=0.92174 lr=0.00100 gn=32.17945 time=54.16it/s +epoch=76 global_step=30000 loss=3.53327 loss_avg=3.74622 acc=0.67188 acc_top1_avg=0.64346 acc_top5_avg=0.92130 lr=0.00100 gn=34.68326 time=54.73it/s +epoch=76 global_step=30050 loss=3.75321 loss_avg=3.74963 acc=0.62500 acc_top1_avg=0.64348 acc_top5_avg=0.92099 lr=0.00100 gn=31.64351 time=53.98it/s +epoch=76 global_step=30100 loss=4.29717 loss_avg=3.74967 acc=0.57812 acc_top1_avg=0.64360 acc_top5_avg=0.92069 lr=0.00100 gn=32.84160 time=55.69it/s +====================Eval==================== +epoch=76 global_step=30107 loss=1.90886 test_loss_avg=1.93465 acc=0.46094 test_acc_avg=0.53176 test_acc_top5_avg=0.87942 time=242.42it/s +epoch=76 global_step=30107 loss=0.09575 test_loss_avg=1.75209 acc=0.93750 test_acc_avg=0.58831 test_acc_top5_avg=0.91119 time=816.65it/s +curr_acc 0.5883 +BEST_ACC 0.6425 +curr_acc_top5 0.9112 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=3.78380 loss_avg=3.74931 acc=0.64844 acc_top1_avg=0.64099 acc_top5_avg=0.91352 lr=0.00100 gn=32.84468 time=61.37it/s +epoch=77 global_step=30200 loss=3.30498 loss_avg=3.69634 acc=0.70312 acc_top1_avg=0.64869 acc_top5_avg=0.91877 lr=0.00100 gn=33.27132 time=60.00it/s +epoch=77 global_step=30250 loss=3.39475 loss_avg=3.71109 acc=0.68750 acc_top1_avg=0.64680 acc_top5_avg=0.91920 lr=0.00100 gn=33.71032 time=54.94it/s +epoch=77 global_step=30300 loss=3.15582 loss_avg=3.71493 acc=0.71094 acc_top1_avg=0.64670 acc_top5_avg=0.91884 lr=0.00100 gn=29.57540 time=55.00it/s +epoch=77 global_step=30350 loss=4.22336 loss_avg=3.72265 acc=0.61719 acc_top1_avg=0.64609 acc_top5_avg=0.91869 lr=0.00100 gn=35.35604 time=59.13it/s +epoch=77 global_step=30400 loss=3.40596 loss_avg=3.74571 acc=0.67969 acc_top1_avg=0.64385 acc_top5_avg=0.91806 lr=0.00100 gn=32.25918 time=55.09it/s +epoch=77 global_step=30450 loss=3.52744 loss_avg=3.75125 acc=0.67188 acc_top1_avg=0.64404 acc_top5_avg=0.91828 lr=0.00100 gn=33.10871 time=59.09it/s +====================Eval==================== +epoch=77 global_step=30498 loss=2.20558 test_loss_avg=0.93042 acc=0.41406 test_acc_avg=0.73759 test_acc_top5_avg=0.96875 time=230.11it/s +epoch=77 global_step=30498 loss=0.31761 test_loss_avg=2.00364 acc=0.89844 test_acc_avg=0.56227 test_acc_top5_avg=0.90730 time=236.94it/s +epoch=77 global_step=30498 loss=0.18851 test_loss_avg=1.73132 acc=0.87500 test_acc_avg=0.61828 test_acc_top5_avg=0.92098 time=648.07it/s +curr_acc 0.6183 +BEST_ACC 0.6425 +curr_acc_top5 0.9210 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=3.36912 loss_avg=3.40463 acc=0.69531 acc_top1_avg=0.68359 acc_top5_avg=0.90625 lr=0.00100 gn=34.84430 time=39.87it/s +epoch=78 global_step=30550 loss=3.64379 loss_avg=3.73275 acc=0.64844 acc_top1_avg=0.64889 acc_top5_avg=0.92188 lr=0.00100 gn=38.98209 time=60.26it/s +epoch=78 global_step=30600 loss=3.42058 loss_avg=3.71413 acc=0.68750 acc_top1_avg=0.64920 acc_top5_avg=0.92172 lr=0.00100 gn=38.59193 time=50.82it/s +epoch=78 global_step=30650 loss=3.53471 loss_avg=3.71932 acc=0.67969 acc_top1_avg=0.64833 acc_top5_avg=0.92069 lr=0.00100 gn=35.79234 time=48.27it/s +epoch=78 global_step=30700 loss=3.62766 loss_avg=3.71629 acc=0.63281 acc_top1_avg=0.64817 acc_top5_avg=0.92091 lr=0.00100 gn=36.67663 time=52.48it/s +epoch=78 global_step=30750 loss=3.71068 loss_avg=3.71757 acc=0.64062 acc_top1_avg=0.64869 acc_top5_avg=0.92032 lr=0.00100 gn=35.41283 time=54.63it/s +epoch=78 global_step=30800 loss=3.83425 loss_avg=3.71994 acc=0.64844 acc_top1_avg=0.64893 acc_top5_avg=0.92009 lr=0.00100 gn=37.56315 time=62.64it/s +epoch=78 global_step=30850 loss=3.29234 loss_avg=3.71680 acc=0.70312 acc_top1_avg=0.64944 acc_top5_avg=0.91939 lr=0.00100 gn=38.40673 time=59.26it/s +====================Eval==================== +epoch=78 global_step=30889 loss=0.61063 test_loss_avg=1.94171 acc=0.80469 test_acc_avg=0.54852 test_acc_top5_avg=0.86123 time=204.58it/s +epoch=78 global_step=30889 loss=0.29874 test_loss_avg=1.80158 acc=0.87500 test_acc_avg=0.59316 test_acc_top5_avg=0.91495 time=680.45it/s +curr_acc 0.5932 +BEST_ACC 0.6425 +curr_acc_top5 0.9150 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=3.99280 loss_avg=3.72509 acc=0.60156 acc_top1_avg=0.64844 acc_top5_avg=0.92045 lr=0.00100 gn=41.69592 time=53.83it/s +epoch=79 global_step=30950 loss=3.84214 loss_avg=3.65477 acc=0.61719 acc_top1_avg=0.65394 acc_top5_avg=0.92367 lr=0.00100 gn=31.99114 time=61.38it/s +epoch=79 global_step=31000 loss=3.75771 loss_avg=3.70651 acc=0.64062 acc_top1_avg=0.64956 acc_top5_avg=0.92005 lr=0.00100 gn=33.77970 time=53.87it/s +epoch=79 global_step=31050 loss=3.76622 loss_avg=3.71029 acc=0.64062 acc_top1_avg=0.64858 acc_top5_avg=0.91896 lr=0.00100 gn=34.16918 time=53.05it/s +epoch=79 global_step=31100 loss=3.63419 loss_avg=3.70706 acc=0.64844 acc_top1_avg=0.64903 acc_top5_avg=0.91795 lr=0.00100 gn=35.38958 time=52.48it/s +epoch=79 global_step=31150 loss=3.36377 loss_avg=3.70356 acc=0.68750 acc_top1_avg=0.64946 acc_top5_avg=0.91822 lr=0.00100 gn=30.89757 time=51.26it/s +epoch=79 global_step=31200 loss=4.15066 loss_avg=3.71584 acc=0.61719 acc_top1_avg=0.64841 acc_top5_avg=0.91881 lr=0.00100 gn=42.17402 time=47.30it/s +epoch=79 global_step=31250 loss=3.85658 loss_avg=3.72928 acc=0.64062 acc_top1_avg=0.64690 acc_top5_avg=0.91898 lr=0.00100 gn=31.94675 time=58.79it/s +====================Eval==================== +epoch=79 global_step=31280 loss=0.38896 test_loss_avg=1.18485 acc=0.91406 test_acc_avg=0.65278 test_acc_top5_avg=0.96962 time=43.00it/s +epoch=79 global_step=31280 loss=0.88055 test_loss_avg=2.39335 acc=0.76562 test_acc_avg=0.47352 test_acc_top5_avg=0.91155 time=237.07it/s +epoch=79 global_step=31280 loss=0.04278 test_loss_avg=1.86623 acc=1.00000 test_acc_avg=0.58366 test_acc_top5_avg=0.93236 time=505.89it/s +curr_acc 0.5837 +BEST_ACC 0.6425 +curr_acc_top5 0.9324 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=3.24789 loss_avg=3.63253 acc=0.71094 acc_top1_avg=0.65469 acc_top5_avg=0.93242 lr=0.00010 gn=36.32786 time=52.19it/s +epoch=80 global_step=31350 loss=3.88561 loss_avg=3.60701 acc=0.63281 acc_top1_avg=0.65982 acc_top5_avg=0.92567 lr=0.00010 gn=29.88466 time=57.27it/s +epoch=80 global_step=31400 loss=3.54771 loss_avg=3.59470 acc=0.67188 acc_top1_avg=0.66094 acc_top5_avg=0.92181 lr=0.00010 gn=34.90164 time=51.39it/s +epoch=80 global_step=31450 loss=3.68431 loss_avg=3.58846 acc=0.66406 acc_top1_avg=0.66186 acc_top5_avg=0.92105 lr=0.00010 gn=37.75694 time=54.53it/s +epoch=80 global_step=31500 loss=3.45623 loss_avg=3.59045 acc=0.66406 acc_top1_avg=0.66193 acc_top5_avg=0.92099 lr=0.00010 gn=33.39305 time=57.69it/s +epoch=80 global_step=31550 loss=3.05078 loss_avg=3.57908 acc=0.71094 acc_top1_avg=0.66305 acc_top5_avg=0.92078 lr=0.00010 gn=31.86103 time=53.20it/s +epoch=80 global_step=31600 loss=3.46072 loss_avg=3.57728 acc=0.68750 acc_top1_avg=0.66306 acc_top5_avg=0.92046 lr=0.00010 gn=35.28960 time=56.31it/s +epoch=80 global_step=31650 loss=3.62022 loss_avg=3.57245 acc=0.65625 acc_top1_avg=0.66368 acc_top5_avg=0.92050 lr=0.00010 gn=32.09846 time=54.13it/s +====================Eval==================== +epoch=80 global_step=31671 loss=5.09939 test_loss_avg=2.12722 acc=0.00000 test_acc_avg=0.49036 test_acc_top5_avg=0.86484 time=233.37it/s +epoch=80 global_step=31671 loss=0.07278 test_loss_avg=1.76777 acc=1.00000 test_acc_avg=0.59088 test_acc_top5_avg=0.91881 time=843.25it/s +curr_acc 0.5909 +BEST_ACC 0.6425 +curr_acc_top5 0.9188 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=4.16217 loss_avg=3.54449 acc=0.59375 acc_top1_avg=0.66810 acc_top5_avg=0.92241 lr=0.00010 gn=36.70411 time=57.65it/s +epoch=81 global_step=31750 loss=3.33009 loss_avg=3.51138 acc=0.67188 acc_top1_avg=0.66960 acc_top5_avg=0.91911 lr=0.00010 gn=32.62096 time=54.43it/s +epoch=81 global_step=31800 loss=3.50682 loss_avg=3.49985 acc=0.68750 acc_top1_avg=0.67085 acc_top5_avg=0.92000 lr=0.00010 gn=39.73031 time=52.95it/s +epoch=81 global_step=31850 loss=4.00101 loss_avg=3.51138 acc=0.60938 acc_top1_avg=0.66969 acc_top5_avg=0.91817 lr=0.00010 gn=36.35868 time=56.80it/s +epoch=81 global_step=31900 loss=3.36848 loss_avg=3.49964 acc=0.67969 acc_top1_avg=0.67143 acc_top5_avg=0.91846 lr=0.00010 gn=34.93155 time=51.79it/s +epoch=81 global_step=31950 loss=3.60670 loss_avg=3.49271 acc=0.67188 acc_top1_avg=0.67218 acc_top5_avg=0.91860 lr=0.00010 gn=37.59511 time=55.19it/s +epoch=81 global_step=32000 loss=3.98288 loss_avg=3.48681 acc=0.63281 acc_top1_avg=0.67304 acc_top5_avg=0.91886 lr=0.00010 gn=42.01817 time=54.20it/s +epoch=81 global_step=32050 loss=3.49618 loss_avg=3.48064 acc=0.68750 acc_top1_avg=0.67387 acc_top5_avg=0.91899 lr=0.00010 gn=38.69527 time=60.67it/s +====================Eval==================== +epoch=81 global_step=32062 loss=1.56990 test_loss_avg=1.56990 acc=0.52344 test_acc_avg=0.52344 test_acc_top5_avg=0.96875 time=213.01it/s +epoch=81 global_step=32062 loss=5.33628 test_loss_avg=2.19779 acc=0.00000 test_acc_avg=0.49663 test_acc_top5_avg=0.89553 time=221.45it/s +epoch=81 global_step=32062 loss=0.05935 test_loss_avg=1.77427 acc=1.00000 test_acc_avg=0.59454 test_acc_top5_avg=0.92237 time=447.34it/s +curr_acc 0.5945 +BEST_ACC 0.6425 +curr_acc_top5 0.9224 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=3.65890 loss_avg=3.40799 acc=0.65625 acc_top1_avg=0.68133 acc_top5_avg=0.91900 lr=0.00010 gn=41.75348 time=51.94it/s +epoch=82 global_step=32150 loss=3.59465 loss_avg=3.44058 acc=0.66406 acc_top1_avg=0.67827 acc_top5_avg=0.92125 lr=0.00010 gn=31.64662 time=63.34it/s +epoch=82 global_step=32200 loss=2.94882 loss_avg=3.44248 acc=0.73438 acc_top1_avg=0.67839 acc_top5_avg=0.92074 lr=0.00010 gn=31.91137 time=58.92it/s +epoch=82 global_step=32250 loss=3.68854 loss_avg=3.44098 acc=0.65625 acc_top1_avg=0.67873 acc_top5_avg=0.91980 lr=0.00010 gn=35.03141 time=54.05it/s +epoch=82 global_step=32300 loss=3.73677 loss_avg=3.43387 acc=0.64062 acc_top1_avg=0.67949 acc_top5_avg=0.92115 lr=0.00010 gn=38.50875 time=55.28it/s +epoch=82 global_step=32350 loss=3.36676 loss_avg=3.43927 acc=0.67188 acc_top1_avg=0.67863 acc_top5_avg=0.92090 lr=0.00010 gn=32.83449 time=54.84it/s +epoch=82 global_step=32400 loss=3.48411 loss_avg=3.44397 acc=0.67188 acc_top1_avg=0.67779 acc_top5_avg=0.92028 lr=0.00010 gn=34.79080 time=61.01it/s +epoch=82 global_step=32450 loss=3.26921 loss_avg=3.44611 acc=0.71094 acc_top1_avg=0.67771 acc_top5_avg=0.91932 lr=0.00010 gn=40.96382 time=61.83it/s +====================Eval==================== +epoch=82 global_step=32453 loss=2.21116 test_loss_avg=1.21884 acc=0.39844 test_acc_avg=0.64631 test_acc_top5_avg=0.95455 time=239.81it/s +epoch=82 global_step=32453 loss=0.29995 test_loss_avg=1.90250 acc=0.92188 test_acc_avg=0.56272 test_acc_top5_avg=0.90853 time=249.44it/s +epoch=82 global_step=32453 loss=0.04442 test_loss_avg=1.74758 acc=1.00000 test_acc_avg=0.59741 test_acc_top5_avg=0.91624 time=651.49it/s +curr_acc 0.5974 +BEST_ACC 0.6425 +curr_acc_top5 0.9162 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=3.13721 loss_avg=3.41465 acc=0.71094 acc_top1_avg=0.68285 acc_top5_avg=0.92337 lr=0.00010 gn=37.05741 time=56.21it/s +epoch=83 global_step=32550 loss=3.27155 loss_avg=3.42611 acc=0.69531 acc_top1_avg=0.68057 acc_top5_avg=0.92220 lr=0.00010 gn=31.63344 time=54.21it/s +epoch=83 global_step=32600 loss=3.13215 loss_avg=3.41973 acc=0.70312 acc_top1_avg=0.68054 acc_top5_avg=0.92039 lr=0.00010 gn=30.18003 time=55.79it/s +epoch=83 global_step=32650 loss=2.80564 loss_avg=3.39874 acc=0.73438 acc_top1_avg=0.68282 acc_top5_avg=0.92132 lr=0.00010 gn=28.13816 time=60.43it/s +epoch=83 global_step=32700 loss=3.64394 loss_avg=3.43405 acc=0.65625 acc_top1_avg=0.67845 acc_top5_avg=0.92102 lr=0.00010 gn=37.91766 time=62.48it/s +epoch=83 global_step=32750 loss=3.79018 loss_avg=3.44310 acc=0.64062 acc_top1_avg=0.67703 acc_top5_avg=0.92130 lr=0.00010 gn=31.77498 time=54.69it/s +epoch=83 global_step=32800 loss=2.86715 loss_avg=3.44256 acc=0.72656 acc_top1_avg=0.67721 acc_top5_avg=0.92057 lr=0.00010 gn=29.90607 time=60.45it/s +====================Eval==================== +epoch=83 global_step=32844 loss=1.75425 test_loss_avg=1.89118 acc=0.53906 test_acc_avg=0.54651 test_acc_top5_avg=0.88318 time=241.62it/s +epoch=83 global_step=32844 loss=0.07027 test_loss_avg=1.74848 acc=1.00000 test_acc_avg=0.59731 test_acc_top5_avg=0.91911 time=847.51it/s +curr_acc 0.5973 +BEST_ACC 0.6425 +curr_acc_top5 0.9191 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=3.01932 loss_avg=3.60106 acc=0.72656 acc_top1_avg=0.66146 acc_top5_avg=0.91536 lr=0.00010 gn=32.67535 time=53.65it/s +epoch=84 global_step=32900 loss=3.13802 loss_avg=3.48447 acc=0.71875 acc_top1_avg=0.67132 acc_top5_avg=0.91378 lr=0.00010 gn=33.35761 time=61.93it/s +epoch=84 global_step=32950 loss=3.81913 loss_avg=3.45447 acc=0.63281 acc_top1_avg=0.67622 acc_top5_avg=0.91731 lr=0.00010 gn=32.43791 time=48.66it/s +epoch=84 global_step=33000 loss=3.62230 loss_avg=3.43313 acc=0.64844 acc_top1_avg=0.67904 acc_top5_avg=0.91872 lr=0.00010 gn=39.04744 time=60.33it/s +epoch=84 global_step=33050 loss=2.99112 loss_avg=3.42588 acc=0.71875 acc_top1_avg=0.68048 acc_top5_avg=0.91873 lr=0.00010 gn=40.32699 time=57.28it/s +epoch=84 global_step=33100 loss=3.38640 loss_avg=3.41380 acc=0.67969 acc_top1_avg=0.68112 acc_top5_avg=0.91931 lr=0.00010 gn=37.67358 time=57.69it/s +epoch=84 global_step=33150 loss=3.22987 loss_avg=3.41258 acc=0.68750 acc_top1_avg=0.68140 acc_top5_avg=0.91845 lr=0.00010 gn=36.92030 time=56.68it/s +epoch=84 global_step=33200 loss=3.73944 loss_avg=3.41924 acc=0.64062 acc_top1_avg=0.68083 acc_top5_avg=0.91882 lr=0.00010 gn=44.66956 time=60.20it/s +====================Eval==================== +epoch=84 global_step=33235 loss=0.27027 test_loss_avg=0.98400 acc=0.94531 test_acc_avg=0.70982 test_acc_top5_avg=0.97656 time=241.54it/s +epoch=84 global_step=33235 loss=0.23541 test_loss_avg=2.16628 acc=0.92188 test_acc_avg=0.50574 test_acc_top5_avg=0.89917 time=240.83it/s +epoch=84 global_step=33235 loss=0.08991 test_loss_avg=1.79184 acc=0.93750 test_acc_avg=0.58841 test_acc_top5_avg=0.91772 time=768.19it/s +curr_acc 0.5884 +BEST_ACC 0.6425 +curr_acc_top5 0.9177 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=3.68642 loss_avg=3.39259 acc=0.64844 acc_top1_avg=0.68594 acc_top5_avg=0.91146 lr=0.00010 gn=40.02485 time=54.45it/s +epoch=85 global_step=33300 loss=2.95725 loss_avg=3.40652 acc=0.71875 acc_top1_avg=0.68341 acc_top5_avg=0.91466 lr=0.00010 gn=31.35813 time=40.31it/s +epoch=85 global_step=33350 loss=3.74381 loss_avg=3.44507 acc=0.65625 acc_top1_avg=0.67867 acc_top5_avg=0.91637 lr=0.00010 gn=38.00151 time=49.91it/s +epoch=85 global_step=33400 loss=3.21458 loss_avg=3.39887 acc=0.71094 acc_top1_avg=0.68272 acc_top5_avg=0.91818 lr=0.00010 gn=41.45340 time=54.97it/s +epoch=85 global_step=33450 loss=3.77567 loss_avg=3.40291 acc=0.63281 acc_top1_avg=0.68263 acc_top5_avg=0.91951 lr=0.00010 gn=36.82067 time=52.36it/s +epoch=85 global_step=33500 loss=3.43004 loss_avg=3.40019 acc=0.68750 acc_top1_avg=0.68269 acc_top5_avg=0.91863 lr=0.00010 gn=28.68356 time=62.19it/s +epoch=85 global_step=33550 loss=3.08531 loss_avg=3.40587 acc=0.73438 acc_top1_avg=0.68197 acc_top5_avg=0.91749 lr=0.00010 gn=44.25055 time=51.11it/s +epoch=85 global_step=33600 loss=3.34784 loss_avg=3.40832 acc=0.69531 acc_top1_avg=0.68187 acc_top5_avg=0.91762 lr=0.00010 gn=36.10925 time=61.54it/s +====================Eval==================== +epoch=85 global_step=33626 loss=0.34899 test_loss_avg=2.04137 acc=0.90625 test_acc_avg=0.51964 test_acc_top5_avg=0.86138 time=254.20it/s +epoch=85 global_step=33626 loss=0.05558 test_loss_avg=1.74223 acc=1.00000 test_acc_avg=0.59949 test_acc_top5_avg=0.91861 time=848.02it/s +curr_acc 0.5995 +BEST_ACC 0.6425 +curr_acc_top5 0.9186 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=3.91460 loss_avg=3.42089 acc=0.62500 acc_top1_avg=0.67773 acc_top5_avg=0.91732 lr=0.00010 gn=30.62555 time=61.71it/s +epoch=86 global_step=33700 loss=3.03176 loss_avg=3.34899 acc=0.71875 acc_top1_avg=0.68623 acc_top5_avg=0.91755 lr=0.00010 gn=36.61592 time=57.65it/s +epoch=86 global_step=33750 loss=2.75006 loss_avg=3.36780 acc=0.74219 acc_top1_avg=0.68523 acc_top5_avg=0.91935 lr=0.00010 gn=29.93257 time=53.75it/s +epoch=86 global_step=33800 loss=3.57647 loss_avg=3.37146 acc=0.67969 acc_top1_avg=0.68494 acc_top5_avg=0.91855 lr=0.00010 gn=42.83470 time=59.85it/s +epoch=86 global_step=33850 loss=3.31019 loss_avg=3.36146 acc=0.68750 acc_top1_avg=0.68558 acc_top5_avg=0.91992 lr=0.00010 gn=33.39916 time=53.84it/s +epoch=86 global_step=33900 loss=3.64689 loss_avg=3.36920 acc=0.64062 acc_top1_avg=0.68456 acc_top5_avg=0.91974 lr=0.00010 gn=41.37654 time=55.37it/s +epoch=86 global_step=33950 loss=3.43102 loss_avg=3.37285 acc=0.67188 acc_top1_avg=0.68424 acc_top5_avg=0.92055 lr=0.00010 gn=41.01515 time=54.14it/s +epoch=86 global_step=34000 loss=2.78723 loss_avg=3.38528 acc=0.75000 acc_top1_avg=0.68295 acc_top5_avg=0.92020 lr=0.00010 gn=35.12499 time=60.09it/s +====================Eval==================== +epoch=86 global_step=34017 loss=1.76850 test_loss_avg=1.67149 acc=0.49219 test_acc_avg=0.50000 test_acc_top5_avg=0.96484 time=236.33it/s +epoch=86 global_step=34017 loss=0.68141 test_loss_avg=2.37586 acc=0.78906 test_acc_avg=0.46331 test_acc_top5_avg=0.88602 time=121.27it/s +epoch=86 global_step=34017 loss=0.08353 test_loss_avg=1.77481 acc=1.00000 test_acc_avg=0.59019 test_acc_top5_avg=0.91752 time=823.38it/s +curr_acc 0.5902 +BEST_ACC 0.6425 +curr_acc_top5 0.9175 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=4.01843 loss_avg=3.33832 acc=0.61719 acc_top1_avg=0.68987 acc_top5_avg=0.92069 lr=0.00010 gn=39.14195 time=49.29it/s +epoch=87 global_step=34100 loss=3.33799 loss_avg=3.38969 acc=0.68750 acc_top1_avg=0.68524 acc_top5_avg=0.91877 lr=0.00010 gn=39.64046 time=53.61it/s +epoch=87 global_step=34150 loss=3.92128 loss_avg=3.39358 acc=0.64844 acc_top1_avg=0.68492 acc_top5_avg=0.91806 lr=0.00010 gn=42.64793 time=57.92it/s +epoch=87 global_step=34200 loss=3.22945 loss_avg=3.37708 acc=0.69531 acc_top1_avg=0.68613 acc_top5_avg=0.91893 lr=0.00010 gn=35.77037 time=54.17it/s +epoch=87 global_step=34250 loss=3.75794 loss_avg=3.39240 acc=0.64844 acc_top1_avg=0.68425 acc_top5_avg=0.92006 lr=0.00010 gn=47.60157 time=61.34it/s +epoch=87 global_step=34300 loss=3.62195 loss_avg=3.39312 acc=0.65625 acc_top1_avg=0.68408 acc_top5_avg=0.91928 lr=0.00010 gn=35.48741 time=57.06it/s +epoch=87 global_step=34350 loss=3.88956 loss_avg=3.39489 acc=0.64062 acc_top1_avg=0.68396 acc_top5_avg=0.91990 lr=0.00010 gn=43.51462 time=34.81it/s +epoch=87 global_step=34400 loss=3.41778 loss_avg=3.39470 acc=0.67969 acc_top1_avg=0.68387 acc_top5_avg=0.91912 lr=0.00010 gn=43.29426 time=60.30it/s +====================Eval==================== +epoch=87 global_step=34408 loss=5.05494 test_loss_avg=1.81472 acc=0.00000 test_acc_avg=0.54167 test_acc_top5_avg=0.89931 time=241.76it/s +epoch=87 global_step=34408 loss=0.15688 test_loss_avg=1.82790 acc=0.95312 test_acc_avg=0.58127 test_acc_top5_avg=0.91579 time=257.60it/s +epoch=87 global_step=34408 loss=0.09074 test_loss_avg=1.78447 acc=0.93750 test_acc_avg=0.59029 test_acc_top5_avg=0.91792 time=828.91it/s +curr_acc 0.5903 +BEST_ACC 0.6425 +curr_acc_top5 0.9179 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=3.40261 loss_avg=3.36570 acc=0.67969 acc_top1_avg=0.68769 acc_top5_avg=0.91536 lr=0.00010 gn=41.53441 time=61.71it/s +epoch=88 global_step=34500 loss=3.25679 loss_avg=3.35702 acc=0.69531 acc_top1_avg=0.68869 acc_top5_avg=0.91916 lr=0.00010 gn=35.55749 time=57.39it/s +epoch=88 global_step=34550 loss=3.91749 loss_avg=3.35760 acc=0.62500 acc_top1_avg=0.68833 acc_top5_avg=0.91852 lr=0.00010 gn=36.02757 time=61.57it/s +epoch=88 global_step=34600 loss=3.30798 loss_avg=3.34353 acc=0.68750 acc_top1_avg=0.69043 acc_top5_avg=0.91874 lr=0.00010 gn=28.60030 time=52.78it/s +epoch=88 global_step=34650 loss=2.91234 loss_avg=3.34410 acc=0.74219 acc_top1_avg=0.69015 acc_top5_avg=0.91974 lr=0.00010 gn=35.28188 time=52.75it/s +epoch=88 global_step=34700 loss=3.77772 loss_avg=3.35249 acc=0.63281 acc_top1_avg=0.68865 acc_top5_avg=0.91877 lr=0.00010 gn=35.77406 time=60.94it/s +epoch=88 global_step=34750 loss=3.58038 loss_avg=3.35253 acc=0.67188 acc_top1_avg=0.68835 acc_top5_avg=0.91861 lr=0.00010 gn=32.58847 time=58.53it/s +====================Eval==================== +epoch=88 global_step=34799 loss=5.30926 test_loss_avg=2.00603 acc=0.00000 test_acc_avg=0.52523 test_acc_top5_avg=0.88900 time=237.21it/s +epoch=88 global_step=34799 loss=0.10853 test_loss_avg=1.78634 acc=0.93750 test_acc_avg=0.58920 test_acc_top5_avg=0.91653 time=465.93it/s +curr_acc 0.5892 +BEST_ACC 0.6425 +curr_acc_top5 0.9165 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=3.95852 loss_avg=3.95852 acc=0.62500 acc_top1_avg=0.62500 acc_top5_avg=0.90625 lr=0.00010 gn=37.80621 time=42.40it/s +epoch=89 global_step=34850 loss=3.41649 loss_avg=3.36784 acc=0.67969 acc_top1_avg=0.68551 acc_top5_avg=0.91713 lr=0.00010 gn=39.45584 time=57.14it/s +epoch=89 global_step=34900 loss=3.74902 loss_avg=3.34772 acc=0.63281 acc_top1_avg=0.68820 acc_top5_avg=0.91700 lr=0.00010 gn=38.29338 time=49.36it/s +epoch=89 global_step=34950 loss=3.27139 loss_avg=3.35654 acc=0.70312 acc_top1_avg=0.68703 acc_top5_avg=0.91665 lr=0.00010 gn=35.81331 time=63.00it/s +epoch=89 global_step=35000 loss=3.45269 loss_avg=3.35054 acc=0.67188 acc_top1_avg=0.68797 acc_top5_avg=0.91842 lr=0.00010 gn=36.09992 time=60.20it/s +epoch=89 global_step=35050 loss=4.05907 loss_avg=3.36856 acc=0.61719 acc_top1_avg=0.68594 acc_top5_avg=0.91721 lr=0.00010 gn=43.82329 time=55.75it/s +epoch=89 global_step=35100 loss=3.49835 loss_avg=3.36507 acc=0.69531 acc_top1_avg=0.68667 acc_top5_avg=0.91892 lr=0.00010 gn=39.62149 time=60.96it/s +epoch=89 global_step=35150 loss=3.65247 loss_avg=3.37382 acc=0.64844 acc_top1_avg=0.68605 acc_top5_avg=0.91865 lr=0.00010 gn=39.55680 time=55.66it/s +====================Eval==================== +epoch=89 global_step=35190 loss=1.89364 test_loss_avg=1.12645 acc=0.47656 test_acc_avg=0.67146 test_acc_top5_avg=0.96752 time=238.31it/s +epoch=89 global_step=35190 loss=0.26495 test_loss_avg=1.99688 acc=0.92969 test_acc_avg=0.54008 test_acc_top5_avg=0.90557 time=248.39it/s +epoch=89 global_step=35190 loss=0.11621 test_loss_avg=1.76841 acc=0.93750 test_acc_avg=0.58979 test_acc_top5_avg=0.91713 time=481.61it/s +curr_acc 0.5898 +BEST_ACC 0.6425 +curr_acc_top5 0.9171 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=3.57338 loss_avg=3.32099 acc=0.66406 acc_top1_avg=0.69375 acc_top5_avg=0.92891 lr=0.00010 gn=34.74182 time=55.69it/s +epoch=90 global_step=35250 loss=3.42632 loss_avg=3.38441 acc=0.67188 acc_top1_avg=0.68190 acc_top5_avg=0.91732 lr=0.00010 gn=44.01030 time=60.17it/s +epoch=90 global_step=35300 loss=3.46640 loss_avg=3.35187 acc=0.67188 acc_top1_avg=0.68643 acc_top5_avg=0.91768 lr=0.00010 gn=41.51163 time=53.40it/s +epoch=90 global_step=35350 loss=3.56920 loss_avg=3.34153 acc=0.67969 acc_top1_avg=0.68823 acc_top5_avg=0.92002 lr=0.00010 gn=44.41351 time=35.24it/s +epoch=90 global_step=35400 loss=3.50310 loss_avg=3.34904 acc=0.67188 acc_top1_avg=0.68739 acc_top5_avg=0.91946 lr=0.00010 gn=34.13781 time=55.31it/s +epoch=90 global_step=35450 loss=3.35995 loss_avg=3.33898 acc=0.67969 acc_top1_avg=0.68870 acc_top5_avg=0.92016 lr=0.00010 gn=32.08820 time=63.93it/s +epoch=90 global_step=35500 loss=4.18941 loss_avg=3.33630 acc=0.60156 acc_top1_avg=0.68911 acc_top5_avg=0.92011 lr=0.00010 gn=44.28922 time=63.06it/s +epoch=90 global_step=35550 loss=3.53358 loss_avg=3.34526 acc=0.66406 acc_top1_avg=0.68826 acc_top5_avg=0.91962 lr=0.00010 gn=34.07146 time=49.73it/s +====================Eval==================== +epoch=90 global_step=35581 loss=1.79144 test_loss_avg=1.96754 acc=0.56250 test_acc_avg=0.53750 test_acc_top5_avg=0.87500 time=240.91it/s +epoch=90 global_step=35581 loss=0.10185 test_loss_avg=1.79481 acc=0.93750 test_acc_avg=0.58881 test_acc_top5_avg=0.91742 time=499.20it/s +curr_acc 0.5888 +BEST_ACC 0.6425 +curr_acc_top5 0.9174 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=3.50024 loss_avg=3.33068 acc=0.68750 acc_top1_avg=0.69285 acc_top5_avg=0.91982 lr=0.00010 gn=41.72707 time=55.46it/s +epoch=91 global_step=35650 loss=3.13746 loss_avg=3.36722 acc=0.70312 acc_top1_avg=0.68569 acc_top5_avg=0.91916 lr=0.00010 gn=39.23759 time=50.38it/s +epoch=91 global_step=35700 loss=3.28509 loss_avg=3.33141 acc=0.70312 acc_top1_avg=0.68901 acc_top5_avg=0.92122 lr=0.00010 gn=49.23487 time=53.78it/s +epoch=91 global_step=35750 loss=3.01086 loss_avg=3.31845 acc=0.73438 acc_top1_avg=0.69111 acc_top5_avg=0.92021 lr=0.00010 gn=43.71122 time=54.82it/s +epoch=91 global_step=35800 loss=3.45017 loss_avg=3.32353 acc=0.67969 acc_top1_avg=0.69092 acc_top5_avg=0.92063 lr=0.00010 gn=36.03724 time=61.56it/s +epoch=91 global_step=35850 loss=3.28439 loss_avg=3.32974 acc=0.69531 acc_top1_avg=0.69029 acc_top5_avg=0.92045 lr=0.00010 gn=35.84528 time=61.29it/s +epoch=91 global_step=35900 loss=3.29518 loss_avg=3.33109 acc=0.69531 acc_top1_avg=0.69039 acc_top5_avg=0.92041 lr=0.00010 gn=34.88704 time=53.85it/s +epoch=91 global_step=35950 loss=3.97716 loss_avg=3.33905 acc=0.61719 acc_top1_avg=0.68936 acc_top5_avg=0.91925 lr=0.00010 gn=42.29685 time=56.80it/s +====================Eval==================== +epoch=91 global_step=35972 loss=0.20276 test_loss_avg=1.31737 acc=0.94531 test_acc_avg=0.61435 test_acc_top5_avg=0.97230 time=180.90it/s +epoch=91 global_step=35972 loss=0.53070 test_loss_avg=2.22004 acc=0.84375 test_acc_avg=0.48783 test_acc_top5_avg=0.88832 time=235.30it/s +epoch=91 global_step=35972 loss=0.10098 test_loss_avg=1.76516 acc=0.93750 test_acc_avg=0.58742 test_acc_top5_avg=0.91288 time=659.27it/s +curr_acc 0.5874 +BEST_ACC 0.6425 +curr_acc_top5 0.9129 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=3.89017 loss_avg=3.27284 acc=0.60938 acc_top1_avg=0.69727 acc_top5_avg=0.92606 lr=0.00010 gn=33.23571 time=53.34it/s +epoch=92 global_step=36050 loss=3.25291 loss_avg=3.29364 acc=0.69531 acc_top1_avg=0.69371 acc_top5_avg=0.92538 lr=0.00010 gn=36.70328 time=60.01it/s +epoch=92 global_step=36100 loss=2.71033 loss_avg=3.31001 acc=0.74219 acc_top1_avg=0.69202 acc_top5_avg=0.92169 lr=0.00010 gn=33.88404 time=52.10it/s +epoch=92 global_step=36150 loss=3.56975 loss_avg=3.32466 acc=0.65625 acc_top1_avg=0.69084 acc_top5_avg=0.92060 lr=0.00010 gn=34.42270 time=57.02it/s +epoch=92 global_step=36200 loss=3.17447 loss_avg=3.33456 acc=0.69531 acc_top1_avg=0.69017 acc_top5_avg=0.91996 lr=0.00010 gn=36.30774 time=52.32it/s +epoch=92 global_step=36250 loss=4.01345 loss_avg=3.34164 acc=0.60938 acc_top1_avg=0.68888 acc_top5_avg=0.92008 lr=0.00010 gn=43.92712 time=49.91it/s +epoch=92 global_step=36300 loss=3.07470 loss_avg=3.34161 acc=0.72656 acc_top1_avg=0.68926 acc_top5_avg=0.92014 lr=0.00010 gn=38.58126 time=54.29it/s +epoch=92 global_step=36350 loss=3.35849 loss_avg=3.34205 acc=0.68750 acc_top1_avg=0.68909 acc_top5_avg=0.92008 lr=0.00010 gn=30.93679 time=60.11it/s +====================Eval==================== +epoch=92 global_step=36363 loss=1.67606 test_loss_avg=2.29711 acc=0.62500 test_acc_avg=0.46118 test_acc_top5_avg=0.84155 time=197.66it/s +epoch=92 global_step=36363 loss=0.13983 test_loss_avg=1.80944 acc=0.87500 test_acc_avg=0.58169 test_acc_top5_avg=0.91228 time=545.99it/s +curr_acc 0.5817 +BEST_ACC 0.6425 +curr_acc_top5 0.9123 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=2.81965 loss_avg=3.25018 acc=0.75781 acc_top1_avg=0.70228 acc_top5_avg=0.92779 lr=0.00010 gn=36.88842 time=57.11it/s +epoch=93 global_step=36450 loss=3.45767 loss_avg=3.29470 acc=0.65625 acc_top1_avg=0.69558 acc_top5_avg=0.92547 lr=0.00010 gn=38.10447 time=57.74it/s +epoch=93 global_step=36500 loss=3.56142 loss_avg=3.29135 acc=0.64844 acc_top1_avg=0.69662 acc_top5_avg=0.92296 lr=0.00010 gn=26.91106 time=58.31it/s +epoch=93 global_step=36550 loss=3.55834 loss_avg=3.30793 acc=0.67188 acc_top1_avg=0.69360 acc_top5_avg=0.92033 lr=0.00010 gn=46.43415 time=56.99it/s +epoch=93 global_step=36600 loss=3.93094 loss_avg=3.32311 acc=0.63281 acc_top1_avg=0.69188 acc_top5_avg=0.91894 lr=0.00010 gn=38.79681 time=56.61it/s +epoch=93 global_step=36650 loss=2.92045 loss_avg=3.32396 acc=0.71875 acc_top1_avg=0.69169 acc_top5_avg=0.91820 lr=0.00010 gn=36.51586 time=59.44it/s +epoch=93 global_step=36700 loss=3.62506 loss_avg=3.31542 acc=0.65625 acc_top1_avg=0.69239 acc_top5_avg=0.91935 lr=0.00010 gn=32.71302 time=54.08it/s +epoch=93 global_step=36750 loss=3.41532 loss_avg=3.32279 acc=0.69531 acc_top1_avg=0.69158 acc_top5_avg=0.91923 lr=0.00010 gn=37.94079 time=63.60it/s +====================Eval==================== +epoch=93 global_step=36754 loss=1.75026 test_loss_avg=1.80387 acc=0.51562 test_acc_avg=0.47917 test_acc_top5_avg=0.95052 time=240.28it/s +epoch=93 global_step=36754 loss=5.20752 test_loss_avg=2.35798 acc=0.00000 test_acc_avg=0.46536 test_acc_top5_avg=0.87957 time=249.87it/s +epoch=93 global_step=36754 loss=0.12327 test_loss_avg=1.80125 acc=0.87500 test_acc_avg=0.58406 test_acc_top5_avg=0.91317 time=808.46it/s +curr_acc 0.5841 +BEST_ACC 0.6425 +curr_acc_top5 0.9132 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=3.65983 loss_avg=3.35151 acc=0.64844 acc_top1_avg=0.68784 acc_top5_avg=0.91236 lr=0.00010 gn=38.96888 time=54.61it/s +epoch=94 global_step=36850 loss=3.49937 loss_avg=3.31299 acc=0.66406 acc_top1_avg=0.69336 acc_top5_avg=0.91724 lr=0.00010 gn=45.39444 time=52.17it/s +epoch=94 global_step=36900 loss=3.45729 loss_avg=3.31081 acc=0.66406 acc_top1_avg=0.69285 acc_top5_avg=0.91792 lr=0.00010 gn=40.03286 time=63.45it/s +epoch=94 global_step=36950 loss=3.42748 loss_avg=3.32371 acc=0.67969 acc_top1_avg=0.69117 acc_top5_avg=0.91849 lr=0.00010 gn=39.86581 time=63.43it/s +epoch=94 global_step=37000 loss=3.01594 loss_avg=3.32247 acc=0.71875 acc_top1_avg=0.69061 acc_top5_avg=0.91918 lr=0.00010 gn=35.30727 time=55.30it/s +epoch=94 global_step=37050 loss=3.81531 loss_avg=3.31925 acc=0.64062 acc_top1_avg=0.69101 acc_top5_avg=0.91876 lr=0.00010 gn=40.20951 time=57.31it/s +epoch=94 global_step=37100 loss=2.67437 loss_avg=3.31713 acc=0.75781 acc_top1_avg=0.69136 acc_top5_avg=0.91928 lr=0.00010 gn=42.50763 time=61.84it/s +====================Eval==================== +epoch=94 global_step=37145 loss=4.11206 test_loss_avg=1.53539 acc=0.14062 test_acc_avg=0.57943 test_acc_top5_avg=0.93359 time=231.18it/s +epoch=94 global_step=37145 loss=0.28971 test_loss_avg=1.88474 acc=0.92188 test_acc_avg=0.56398 test_acc_top5_avg=0.90583 time=243.68it/s +epoch=94 global_step=37145 loss=0.17538 test_loss_avg=1.77711 acc=0.87500 test_acc_avg=0.58663 test_acc_top5_avg=0.91169 time=647.07it/s +curr_acc 0.5866 +BEST_ACC 0.6425 +curr_acc_top5 0.9117 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=3.47692 loss_avg=3.23279 acc=0.67188 acc_top1_avg=0.70625 acc_top5_avg=0.93281 lr=0.00010 gn=41.02664 time=40.38it/s +epoch=95 global_step=37200 loss=3.11350 loss_avg=3.26986 acc=0.71875 acc_top1_avg=0.69730 acc_top5_avg=0.92074 lr=0.00010 gn=38.50125 time=57.80it/s +epoch=95 global_step=37250 loss=3.06241 loss_avg=3.28508 acc=0.71875 acc_top1_avg=0.69449 acc_top5_avg=0.92076 lr=0.00010 gn=37.67979 time=57.54it/s +epoch=95 global_step=37300 loss=3.40970 loss_avg=3.30783 acc=0.67188 acc_top1_avg=0.69143 acc_top5_avg=0.92193 lr=0.00010 gn=36.07143 time=58.67it/s +epoch=95 global_step=37350 loss=3.54374 loss_avg=3.30208 acc=0.67188 acc_top1_avg=0.69223 acc_top5_avg=0.92176 lr=0.00010 gn=34.69913 time=55.28it/s +epoch=95 global_step=37400 loss=3.53052 loss_avg=3.30768 acc=0.67188 acc_top1_avg=0.69188 acc_top5_avg=0.92016 lr=0.00010 gn=31.92389 time=59.35it/s +epoch=95 global_step=37450 loss=3.05156 loss_avg=3.30099 acc=0.73438 acc_top1_avg=0.69321 acc_top5_avg=0.92057 lr=0.00010 gn=38.37921 time=56.19it/s +epoch=95 global_step=37500 loss=3.32267 loss_avg=3.30586 acc=0.70312 acc_top1_avg=0.69276 acc_top5_avg=0.91987 lr=0.00010 gn=38.87872 time=58.81it/s +====================Eval==================== +epoch=95 global_step=37536 loss=2.11698 test_loss_avg=1.93921 acc=0.45312 test_acc_avg=0.53542 test_acc_top5_avg=0.87969 time=234.66it/s +epoch=95 global_step=37536 loss=0.11363 test_loss_avg=1.78280 acc=0.87500 test_acc_avg=0.58772 test_acc_top5_avg=0.91119 time=541.48it/s +curr_acc 0.5877 +BEST_ACC 0.6425 +curr_acc_top5 0.9112 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=2.83565 loss_avg=3.09208 acc=0.75781 acc_top1_avg=0.71652 acc_top5_avg=0.91071 lr=0.00010 gn=38.04874 time=48.82it/s +epoch=96 global_step=37600 loss=3.65771 loss_avg=3.25322 acc=0.64844 acc_top1_avg=0.69800 acc_top5_avg=0.91956 lr=0.00010 gn=42.77199 time=54.84it/s +epoch=96 global_step=37650 loss=3.84800 loss_avg=3.27408 acc=0.63281 acc_top1_avg=0.69586 acc_top5_avg=0.92112 lr=0.00010 gn=46.07233 time=54.60it/s +epoch=96 global_step=37700 loss=3.55090 loss_avg=3.28580 acc=0.65625 acc_top1_avg=0.69407 acc_top5_avg=0.91959 lr=0.00010 gn=43.49816 time=49.99it/s +epoch=96 global_step=37750 loss=3.68342 loss_avg=3.28437 acc=0.65625 acc_top1_avg=0.69498 acc_top5_avg=0.91950 lr=0.00010 gn=36.90655 time=60.66it/s +epoch=96 global_step=37800 loss=3.45800 loss_avg=3.28826 acc=0.64062 acc_top1_avg=0.69454 acc_top5_avg=0.91933 lr=0.00010 gn=36.78580 time=62.70it/s +epoch=96 global_step=37850 loss=2.38663 loss_avg=3.29588 acc=0.78906 acc_top1_avg=0.69389 acc_top5_avg=0.91899 lr=0.00010 gn=37.87836 time=58.63it/s +epoch=96 global_step=37900 loss=3.19852 loss_avg=3.31094 acc=0.71875 acc_top1_avg=0.69216 acc_top5_avg=0.91924 lr=0.00010 gn=43.54906 time=50.16it/s +====================Eval==================== +epoch=96 global_step=37927 loss=1.15195 test_loss_avg=1.03817 acc=0.68750 test_acc_avg=0.70312 test_acc_top5_avg=0.97363 time=246.77it/s +epoch=96 global_step=37927 loss=0.16995 test_loss_avg=2.09975 acc=0.94531 test_acc_avg=0.51515 test_acc_top5_avg=0.89619 time=223.64it/s +epoch=96 global_step=37927 loss=0.12242 test_loss_avg=1.78570 acc=0.93750 test_acc_avg=0.58495 test_acc_top5_avg=0.91288 time=515.59it/s +curr_acc 0.5849 +BEST_ACC 0.6425 +curr_acc_top5 0.9129 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=3.33688 loss_avg=3.40330 acc=0.68750 acc_top1_avg=0.68071 acc_top5_avg=0.91712 lr=0.00010 gn=35.46886 time=47.01it/s +epoch=97 global_step=38000 loss=3.63776 loss_avg=3.26049 acc=0.65625 acc_top1_avg=0.69649 acc_top5_avg=0.91952 lr=0.00010 gn=39.54898 time=58.25it/s +epoch=97 global_step=38050 loss=3.52247 loss_avg=3.26272 acc=0.66406 acc_top1_avg=0.69652 acc_top5_avg=0.91895 lr=0.00010 gn=38.24124 time=53.84it/s +epoch=97 global_step=38100 loss=2.71608 loss_avg=3.27364 acc=0.75000 acc_top1_avg=0.69581 acc_top5_avg=0.92034 lr=0.00010 gn=31.85443 time=56.73it/s +epoch=97 global_step=38150 loss=3.06196 loss_avg=3.30592 acc=0.71094 acc_top1_avg=0.69237 acc_top5_avg=0.91802 lr=0.00010 gn=41.60320 time=63.49it/s +epoch=97 global_step=38200 loss=3.23043 loss_avg=3.30857 acc=0.71875 acc_top1_avg=0.69291 acc_top5_avg=0.91787 lr=0.00010 gn=44.43010 time=56.73it/s +epoch=97 global_step=38250 loss=3.27668 loss_avg=3.30669 acc=0.70312 acc_top1_avg=0.69330 acc_top5_avg=0.91798 lr=0.00010 gn=49.77848 time=51.56it/s +epoch=97 global_step=38300 loss=3.35241 loss_avg=3.30332 acc=0.67969 acc_top1_avg=0.69357 acc_top5_avg=0.91848 lr=0.00010 gn=39.10650 time=56.25it/s +====================Eval==================== +epoch=97 global_step=38318 loss=0.47342 test_loss_avg=1.98501 acc=0.89062 test_acc_avg=0.52872 test_acc_top5_avg=0.86423 time=221.09it/s +epoch=97 global_step=38318 loss=0.18614 test_loss_avg=1.78536 acc=0.87500 test_acc_avg=0.58416 test_acc_top5_avg=0.91050 time=838.69it/s +curr_acc 0.5842 +BEST_ACC 0.6425 +curr_acc_top5 0.9105 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=3.16000 loss_avg=3.27569 acc=0.69531 acc_top1_avg=0.69775 acc_top5_avg=0.91919 lr=0.00010 gn=45.28187 time=58.54it/s +epoch=98 global_step=38400 loss=3.76773 loss_avg=3.26881 acc=0.61719 acc_top1_avg=0.69569 acc_top5_avg=0.92083 lr=0.00010 gn=44.11532 time=58.31it/s +epoch=98 global_step=38450 loss=3.32908 loss_avg=3.29317 acc=0.68750 acc_top1_avg=0.69377 acc_top5_avg=0.92004 lr=0.00010 gn=29.61004 time=56.88it/s +epoch=98 global_step=38500 loss=2.94716 loss_avg=3.31180 acc=0.75781 acc_top1_avg=0.69166 acc_top5_avg=0.91771 lr=0.00010 gn=49.75796 time=55.44it/s +epoch=98 global_step=38550 loss=2.91958 loss_avg=3.31107 acc=0.74219 acc_top1_avg=0.69248 acc_top5_avg=0.91662 lr=0.00010 gn=42.77795 time=53.25it/s +epoch=98 global_step=38600 loss=3.40441 loss_avg=3.29875 acc=0.67969 acc_top1_avg=0.69348 acc_top5_avg=0.91866 lr=0.00010 gn=35.65227 time=59.00it/s +epoch=98 global_step=38650 loss=3.01389 loss_avg=3.30058 acc=0.71094 acc_top1_avg=0.69329 acc_top5_avg=0.91971 lr=0.00010 gn=38.50635 time=59.88it/s +epoch=98 global_step=38700 loss=3.46500 loss_avg=3.30052 acc=0.66406 acc_top1_avg=0.69343 acc_top5_avg=0.91997 lr=0.00010 gn=43.69308 time=61.09it/s +====================Eval==================== +epoch=98 global_step=38709 loss=1.47349 test_loss_avg=1.76435 acc=0.60938 test_acc_avg=0.49219 test_acc_top5_avg=0.95117 time=51.98it/s +epoch=98 global_step=38709 loss=0.78951 test_loss_avg=2.32842 acc=0.73438 test_acc_avg=0.47522 test_acc_top5_avg=0.88295 time=241.05it/s +epoch=98 global_step=38709 loss=0.11401 test_loss_avg=1.78690 acc=0.93750 test_acc_avg=0.58900 test_acc_top5_avg=0.91248 time=824.68it/s +curr_acc 0.5890 +BEST_ACC 0.6425 +curr_acc_top5 0.9125 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=2.54378 loss_avg=3.30103 acc=0.75781 acc_top1_avg=0.69169 acc_top5_avg=0.91806 lr=0.00010 gn=39.69298 time=56.97it/s +epoch=99 global_step=38800 loss=2.95017 loss_avg=3.26717 acc=0.73438 acc_top1_avg=0.69677 acc_top5_avg=0.92042 lr=0.00010 gn=34.85942 time=61.23it/s +epoch=99 global_step=38850 loss=3.45801 loss_avg=3.27512 acc=0.68750 acc_top1_avg=0.69664 acc_top5_avg=0.91999 lr=0.00010 gn=44.90412 time=56.57it/s +epoch=99 global_step=38900 loss=3.47959 loss_avg=3.28777 acc=0.66406 acc_top1_avg=0.69523 acc_top5_avg=0.91905 lr=0.00010 gn=35.37472 time=55.86it/s +epoch=99 global_step=38950 loss=3.27785 loss_avg=3.27878 acc=0.67969 acc_top1_avg=0.69606 acc_top5_avg=0.92006 lr=0.00010 gn=40.84224 time=57.68it/s +epoch=99 global_step=39000 loss=3.96173 loss_avg=3.29546 acc=0.63281 acc_top1_avg=0.69424 acc_top5_avg=0.92099 lr=0.00010 gn=39.46438 time=59.42it/s +epoch=99 global_step=39050 loss=3.45749 loss_avg=3.29218 acc=0.67969 acc_top1_avg=0.69458 acc_top5_avg=0.91986 lr=0.00010 gn=42.25752 time=60.05it/s +epoch=99 global_step=39100 loss=2.72137 loss_avg=3.29432 acc=0.75000 acc_top1_avg=0.69423 acc_top5_avg=0.91935 lr=0.00010 gn=46.37193 time=78.76it/s +====================Eval==================== +epoch=99 global_step=39100 loss=5.02905 test_loss_avg=2.10447 acc=0.00000 test_acc_avg=0.49057 test_acc_top5_avg=0.85911 time=247.03it/s +epoch=99 global_step=39100 loss=0.17982 test_loss_avg=1.79930 acc=0.87500 test_acc_avg=0.58445 test_acc_top5_avg=0.91258 time=793.92it/s +epoch=99 global_step=39100 loss=0.17982 test_loss_avg=1.79930 acc=0.87500 test_acc_avg=0.58445 test_acc_top5_avg=0.91258 time=793.92it/s +curr_acc 0.5845 +BEST_ACC 0.6425 +curr_acc_top5 0.9126 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=3.67444 loss_avg=3.30106 acc=0.67188 acc_top1_avg=0.69484 acc_top5_avg=0.91906 lr=0.00010 gn=39.20878 time=63.36it/s +epoch=100 global_step=39200 loss=3.73944 loss_avg=3.30166 acc=0.64844 acc_top1_avg=0.69312 acc_top5_avg=0.91953 lr=0.00010 gn=45.54802 time=62.65it/s +epoch=100 global_step=39250 loss=3.30399 loss_avg=3.30405 acc=0.69531 acc_top1_avg=0.69328 acc_top5_avg=0.91859 lr=0.00010 gn=42.54603 time=55.57it/s +epoch=100 global_step=39300 loss=2.86794 loss_avg=3.29528 acc=0.75781 acc_top1_avg=0.69406 acc_top5_avg=0.91863 lr=0.00010 gn=49.60062 time=63.03it/s +epoch=100 global_step=39350 loss=3.20959 loss_avg=3.29674 acc=0.72656 acc_top1_avg=0.69397 acc_top5_avg=0.91963 lr=0.00010 gn=47.74258 time=61.34it/s +epoch=100 global_step=39400 loss=2.82156 loss_avg=3.28644 acc=0.76562 acc_top1_avg=0.69518 acc_top5_avg=0.92000 lr=0.00010 gn=41.18595 time=53.22it/s +epoch=100 global_step=39450 loss=2.96706 loss_avg=3.28872 acc=0.71094 acc_top1_avg=0.69462 acc_top5_avg=0.91984 lr=0.00010 gn=42.02697 time=62.66it/s +====================Eval==================== +epoch=100 global_step=39491 loss=5.37427 test_loss_avg=2.13345 acc=0.00000 test_acc_avg=0.50375 test_acc_top5_avg=0.88406 time=243.84it/s +epoch=100 global_step=39491 loss=0.14801 test_loss_avg=1.76781 acc=0.87500 test_acc_avg=0.59167 test_acc_top5_avg=0.91515 time=548.13it/s +curr_acc 0.5917 +BEST_ACC 0.6425 +curr_acc_top5 0.9152 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=3.08077 loss_avg=3.11263 acc=0.71875 acc_top1_avg=0.71354 acc_top5_avg=0.92448 lr=0.00010 gn=36.72708 time=56.55it/s +epoch=101 global_step=39550 loss=3.23016 loss_avg=3.25897 acc=0.69531 acc_top1_avg=0.69876 acc_top5_avg=0.91618 lr=0.00010 gn=43.55546 time=56.18it/s +epoch=101 global_step=39600 loss=2.96130 loss_avg=3.29570 acc=0.73438 acc_top1_avg=0.69467 acc_top5_avg=0.91707 lr=0.00010 gn=41.34363 time=50.75it/s +epoch=101 global_step=39650 loss=3.08940 loss_avg=3.27798 acc=0.71094 acc_top1_avg=0.69664 acc_top5_avg=0.91927 lr=0.00010 gn=31.60716 time=63.47it/s +epoch=101 global_step=39700 loss=3.68463 loss_avg=3.27946 acc=0.66406 acc_top1_avg=0.69572 acc_top5_avg=0.91971 lr=0.00010 gn=53.08243 time=63.13it/s +epoch=101 global_step=39750 loss=3.57925 loss_avg=3.27706 acc=0.64844 acc_top1_avg=0.69561 acc_top5_avg=0.91898 lr=0.00010 gn=35.93243 time=53.18it/s +epoch=101 global_step=39800 loss=3.12379 loss_avg=3.27134 acc=0.71094 acc_top1_avg=0.69615 acc_top5_avg=0.91935 lr=0.00010 gn=42.37921 time=56.12it/s +epoch=101 global_step=39850 loss=3.54521 loss_avg=3.27862 acc=0.67188 acc_top1_avg=0.69564 acc_top5_avg=0.91931 lr=0.00010 gn=43.72942 time=55.84it/s +====================Eval==================== +epoch=101 global_step=39882 loss=2.15110 test_loss_avg=1.37420 acc=0.39844 test_acc_avg=0.61161 test_acc_top5_avg=0.95164 time=158.43it/s +epoch=101 global_step=39882 loss=0.19472 test_loss_avg=2.01543 acc=0.93750 test_acc_avg=0.53763 test_acc_top5_avg=0.90493 time=235.13it/s +epoch=101 global_step=39882 loss=0.15346 test_loss_avg=1.83349 acc=0.87500 test_acc_avg=0.57654 test_acc_top5_avg=0.91416 time=840.71it/s +curr_acc 0.5765 +BEST_ACC 0.6425 +curr_acc_top5 0.9142 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=2.71549 loss_avg=3.15749 acc=0.75000 acc_top1_avg=0.70920 acc_top5_avg=0.91493 lr=0.00010 gn=32.71310 time=51.97it/s +epoch=102 global_step=39950 loss=4.07849 loss_avg=3.21967 acc=0.58594 acc_top1_avg=0.70244 acc_top5_avg=0.91751 lr=0.00010 gn=40.46990 time=53.09it/s +epoch=102 global_step=40000 loss=3.79142 loss_avg=3.20094 acc=0.64844 acc_top1_avg=0.70405 acc_top5_avg=0.91856 lr=0.00010 gn=50.58651 time=50.75it/s +epoch=102 global_step=40050 loss=3.74142 loss_avg=3.24371 acc=0.65625 acc_top1_avg=0.69950 acc_top5_avg=0.91736 lr=0.00010 gn=45.57546 time=56.91it/s +epoch=102 global_step=40100 loss=3.50156 loss_avg=3.26676 acc=0.69531 acc_top1_avg=0.69707 acc_top5_avg=0.91775 lr=0.00010 gn=55.44582 time=60.03it/s +epoch=102 global_step=40150 loss=3.08259 loss_avg=3.27096 acc=0.71875 acc_top1_avg=0.69645 acc_top5_avg=0.91832 lr=0.00010 gn=39.84490 time=60.18it/s +epoch=102 global_step=40200 loss=3.22073 loss_avg=3.27752 acc=0.71094 acc_top1_avg=0.69625 acc_top5_avg=0.91817 lr=0.00010 gn=44.55549 time=54.33it/s +epoch=102 global_step=40250 loss=3.46947 loss_avg=3.26441 acc=0.67188 acc_top1_avg=0.69744 acc_top5_avg=0.91856 lr=0.00010 gn=37.32818 time=63.33it/s +====================Eval==================== +epoch=102 global_step=40273 loss=2.33941 test_loss_avg=1.97226 acc=0.45312 test_acc_avg=0.53125 test_acc_top5_avg=0.87686 time=242.42it/s +epoch=102 global_step=40273 loss=0.10475 test_loss_avg=1.82556 acc=0.93750 test_acc_avg=0.57971 test_acc_top5_avg=0.91218 time=850.25it/s +curr_acc 0.5797 +BEST_ACC 0.6425 +curr_acc_top5 0.9122 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=3.44905 loss_avg=3.32557 acc=0.67188 acc_top1_avg=0.69126 acc_top5_avg=0.91175 lr=0.00010 gn=35.04311 time=62.52it/s +epoch=103 global_step=40350 loss=3.13146 loss_avg=3.25390 acc=0.69531 acc_top1_avg=0.69795 acc_top5_avg=0.91619 lr=0.00010 gn=35.95097 time=63.18it/s +epoch=103 global_step=40400 loss=3.34354 loss_avg=3.25835 acc=0.67969 acc_top1_avg=0.69759 acc_top5_avg=0.91781 lr=0.00010 gn=38.92646 time=49.13it/s +epoch=103 global_step=40450 loss=2.94190 loss_avg=3.26918 acc=0.72656 acc_top1_avg=0.69681 acc_top5_avg=0.91887 lr=0.00010 gn=45.78363 time=59.82it/s +epoch=103 global_step=40500 loss=3.01857 loss_avg=3.25839 acc=0.71875 acc_top1_avg=0.69786 acc_top5_avg=0.91902 lr=0.00010 gn=41.24677 time=53.82it/s +epoch=103 global_step=40550 loss=3.30734 loss_avg=3.25659 acc=0.67188 acc_top1_avg=0.69881 acc_top5_avg=0.91925 lr=0.00010 gn=39.41356 time=60.23it/s +epoch=103 global_step=40600 loss=3.59107 loss_avg=3.26483 acc=0.67188 acc_top1_avg=0.69813 acc_top5_avg=0.91963 lr=0.00010 gn=43.90533 time=52.56it/s +epoch=103 global_step=40650 loss=2.77653 loss_avg=3.26617 acc=0.75000 acc_top1_avg=0.69807 acc_top5_avg=0.91933 lr=0.00010 gn=40.10882 time=63.55it/s +====================Eval==================== +epoch=103 global_step=40664 loss=0.21272 test_loss_avg=1.07862 acc=0.93750 test_acc_avg=0.67788 test_acc_top5_avg=0.96995 time=194.49it/s +epoch=103 global_step=40664 loss=0.44941 test_loss_avg=2.13777 acc=0.85938 test_acc_avg=0.50657 test_acc_top5_avg=0.88591 time=244.84it/s +epoch=103 global_step=40664 loss=0.13534 test_loss_avg=1.74835 acc=0.87500 test_acc_avg=0.59256 test_acc_top5_avg=0.90793 time=850.43it/s +curr_acc 0.5926 +BEST_ACC 0.6425 +curr_acc_top5 0.9079 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=2.96822 loss_avg=3.19674 acc=0.72656 acc_top1_avg=0.70269 acc_top5_avg=0.92535 lr=0.00010 gn=46.77585 time=54.70it/s +epoch=104 global_step=40750 loss=3.43267 loss_avg=3.22216 acc=0.67188 acc_top1_avg=0.70031 acc_top5_avg=0.92287 lr=0.00010 gn=38.24602 time=54.91it/s +epoch=104 global_step=40800 loss=3.53373 loss_avg=3.22805 acc=0.67969 acc_top1_avg=0.69968 acc_top5_avg=0.92291 lr=0.00010 gn=47.35344 time=55.46it/s +epoch=104 global_step=40850 loss=3.50363 loss_avg=3.25313 acc=0.67188 acc_top1_avg=0.69775 acc_top5_avg=0.92099 lr=0.00010 gn=42.72405 time=61.79it/s +epoch=104 global_step=40900 loss=2.81848 loss_avg=3.25561 acc=0.74219 acc_top1_avg=0.69776 acc_top5_avg=0.91952 lr=0.00010 gn=44.86094 time=62.57it/s +epoch=104 global_step=40950 loss=2.91899 loss_avg=3.26092 acc=0.71875 acc_top1_avg=0.69750 acc_top5_avg=0.91862 lr=0.00010 gn=35.02921 time=60.81it/s +epoch=104 global_step=41000 loss=2.44331 loss_avg=3.25983 acc=0.78906 acc_top1_avg=0.69741 acc_top5_avg=0.91946 lr=0.00010 gn=43.59085 time=47.76it/s +epoch=104 global_step=41050 loss=2.70897 loss_avg=3.25766 acc=0.76562 acc_top1_avg=0.69776 acc_top5_avg=0.91912 lr=0.00010 gn=41.69442 time=54.81it/s +====================Eval==================== +epoch=104 global_step=41055 loss=0.51125 test_loss_avg=2.19554 acc=0.85938 test_acc_avg=0.48024 test_acc_top5_avg=0.84513 time=197.50it/s +epoch=104 global_step=41055 loss=0.15420 test_loss_avg=1.81677 acc=0.87500 test_acc_avg=0.57585 test_acc_top5_avg=0.90892 time=626.02it/s +curr_acc 0.5759 +BEST_ACC 0.6425 +curr_acc_top5 0.9089 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=3.98168 loss_avg=3.27672 acc=0.59375 acc_top1_avg=0.69618 acc_top5_avg=0.91372 lr=0.00010 gn=50.66524 time=57.77it/s +epoch=105 global_step=41150 loss=2.84376 loss_avg=3.24343 acc=0.73438 acc_top1_avg=0.69975 acc_top5_avg=0.91669 lr=0.00010 gn=37.38510 time=57.79it/s +epoch=105 global_step=41200 loss=3.44028 loss_avg=3.26405 acc=0.69531 acc_top1_avg=0.69628 acc_top5_avg=0.91805 lr=0.00010 gn=43.46737 time=49.85it/s +epoch=105 global_step=41250 loss=3.22415 loss_avg=3.24467 acc=0.69531 acc_top1_avg=0.69896 acc_top5_avg=0.91851 lr=0.00010 gn=37.22907 time=54.89it/s +epoch=105 global_step=41300 loss=3.49054 loss_avg=3.22777 acc=0.67969 acc_top1_avg=0.70159 acc_top5_avg=0.91853 lr=0.00010 gn=41.12292 time=55.32it/s +epoch=105 global_step=41350 loss=3.04989 loss_avg=3.23155 acc=0.71875 acc_top1_avg=0.70109 acc_top5_avg=0.91870 lr=0.00010 gn=42.24912 time=53.70it/s +epoch=105 global_step=41400 loss=4.27154 loss_avg=3.24055 acc=0.58594 acc_top1_avg=0.70027 acc_top5_avg=0.91846 lr=0.00010 gn=42.74504 time=54.17it/s +====================Eval==================== +epoch=105 global_step=41446 loss=1.64565 test_loss_avg=1.91576 acc=0.45312 test_acc_avg=0.44531 test_acc_top5_avg=0.95625 time=245.48it/s +epoch=105 global_step=41446 loss=4.12528 test_loss_avg=2.50610 acc=0.19531 test_acc_avg=0.43963 test_acc_top5_avg=0.87571 time=240.98it/s +epoch=105 global_step=41446 loss=0.11600 test_loss_avg=1.85048 acc=0.93750 test_acc_avg=0.57545 test_acc_top5_avg=0.91129 time=650.18it/s +curr_acc 0.5755 +BEST_ACC 0.6425 +curr_acc_top5 0.9113 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=3.38386 loss_avg=3.25746 acc=0.67969 acc_top1_avg=0.69531 acc_top5_avg=0.91211 lr=0.00010 gn=34.87089 time=57.88it/s +epoch=106 global_step=41500 loss=3.06973 loss_avg=3.25777 acc=0.71094 acc_top1_avg=0.69488 acc_top5_avg=0.91609 lr=0.00010 gn=41.03216 time=62.72it/s +epoch=106 global_step=41550 loss=3.31742 loss_avg=3.26341 acc=0.69531 acc_top1_avg=0.69569 acc_top5_avg=0.91774 lr=0.00010 gn=33.62317 time=55.81it/s +epoch=106 global_step=41600 loss=3.77869 loss_avg=3.25039 acc=0.62500 acc_top1_avg=0.69734 acc_top5_avg=0.91650 lr=0.00010 gn=46.56034 time=55.61it/s +epoch=106 global_step=41650 loss=3.51385 loss_avg=3.24888 acc=0.66406 acc_top1_avg=0.69822 acc_top5_avg=0.91701 lr=0.00010 gn=39.11807 time=54.50it/s +epoch=106 global_step=41700 loss=3.56037 loss_avg=3.26070 acc=0.64844 acc_top1_avg=0.69691 acc_top5_avg=0.91794 lr=0.00010 gn=35.56018 time=54.22it/s +epoch=106 global_step=41750 loss=3.54780 loss_avg=3.24550 acc=0.67188 acc_top1_avg=0.69870 acc_top5_avg=0.91887 lr=0.00010 gn=44.27131 time=59.70it/s +epoch=106 global_step=41800 loss=3.48225 loss_avg=3.23980 acc=0.67969 acc_top1_avg=0.69988 acc_top5_avg=0.91872 lr=0.00010 gn=37.64496 time=57.58it/s +====================Eval==================== +epoch=106 global_step=41837 loss=5.29877 test_loss_avg=1.79220 acc=0.00000 test_acc_avg=0.54657 test_acc_top5_avg=0.90144 time=184.61it/s +epoch=106 global_step=41837 loss=0.26122 test_loss_avg=1.91567 acc=0.92188 test_acc_avg=0.56127 test_acc_top5_avg=0.90687 time=243.15it/s +epoch=106 global_step=41837 loss=0.16847 test_loss_avg=1.84926 acc=0.87500 test_acc_avg=0.57486 test_acc_top5_avg=0.91040 time=825.49it/s +curr_acc 0.5749 +BEST_ACC 0.6425 +curr_acc_top5 0.9104 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=3.06568 loss_avg=3.22533 acc=0.71875 acc_top1_avg=0.70433 acc_top5_avg=0.90685 lr=0.00010 gn=45.63659 time=59.43it/s +epoch=107 global_step=41900 loss=3.50658 loss_avg=3.22935 acc=0.67188 acc_top1_avg=0.70325 acc_top5_avg=0.91704 lr=0.00010 gn=40.22863 time=63.58it/s +epoch=107 global_step=41950 loss=3.47276 loss_avg=3.21551 acc=0.67188 acc_top1_avg=0.70472 acc_top5_avg=0.91821 lr=0.00010 gn=41.54606 time=53.65it/s +epoch=107 global_step=42000 loss=3.69818 loss_avg=3.23092 acc=0.65625 acc_top1_avg=0.70308 acc_top5_avg=0.91866 lr=0.00010 gn=47.81854 time=61.03it/s +epoch=107 global_step=42050 loss=2.96098 loss_avg=3.24387 acc=0.75000 acc_top1_avg=0.70100 acc_top5_avg=0.91711 lr=0.00010 gn=40.51951 time=57.06it/s +epoch=107 global_step=42100 loss=3.24371 loss_avg=3.25534 acc=0.70312 acc_top1_avg=0.69935 acc_top5_avg=0.91742 lr=0.00010 gn=36.45650 time=62.22it/s +epoch=107 global_step=42150 loss=3.10890 loss_avg=3.24949 acc=0.72656 acc_top1_avg=0.69988 acc_top5_avg=0.91811 lr=0.00010 gn=42.29436 time=63.00it/s +epoch=107 global_step=42200 loss=3.25228 loss_avg=3.23608 acc=0.69531 acc_top1_avg=0.70138 acc_top5_avg=0.91884 lr=0.00010 gn=40.47287 time=59.16it/s +====================Eval==================== +epoch=107 global_step=42228 loss=2.70733 test_loss_avg=2.00425 acc=0.39062 test_acc_avg=0.52194 test_acc_top5_avg=0.88531 time=234.87it/s +epoch=107 global_step=42228 loss=0.15935 test_loss_avg=1.81773 acc=0.87500 test_acc_avg=0.57902 test_acc_top5_avg=0.91100 time=843.92it/s +curr_acc 0.5790 +BEST_ACC 0.6425 +curr_acc_top5 0.9110 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=3.41811 loss_avg=3.28954 acc=0.67188 acc_top1_avg=0.69922 acc_top5_avg=0.91832 lr=0.00010 gn=36.34556 time=53.92it/s +epoch=108 global_step=42300 loss=2.98613 loss_avg=3.26369 acc=0.71875 acc_top1_avg=0.69694 acc_top5_avg=0.92057 lr=0.00010 gn=38.19080 time=60.29it/s +epoch=108 global_step=42350 loss=3.34567 loss_avg=3.28418 acc=0.67969 acc_top1_avg=0.69448 acc_top5_avg=0.91906 lr=0.00010 gn=39.55082 time=51.19it/s +epoch=108 global_step=42400 loss=3.17757 loss_avg=3.25751 acc=0.69531 acc_top1_avg=0.69736 acc_top5_avg=0.92069 lr=0.00010 gn=44.50495 time=53.35it/s +epoch=108 global_step=42450 loss=3.49028 loss_avg=3.24130 acc=0.66406 acc_top1_avg=0.69901 acc_top5_avg=0.91969 lr=0.00010 gn=48.32054 time=56.66it/s +epoch=108 global_step=42500 loss=3.69726 loss_avg=3.23969 acc=0.64062 acc_top1_avg=0.69902 acc_top5_avg=0.91920 lr=0.00010 gn=37.01857 time=56.58it/s +epoch=108 global_step=42550 loss=3.28662 loss_avg=3.24066 acc=0.69531 acc_top1_avg=0.69917 acc_top5_avg=0.91879 lr=0.00010 gn=38.50104 time=46.95it/s +epoch=108 global_step=42600 loss=3.03395 loss_avg=3.23584 acc=0.72656 acc_top1_avg=0.70006 acc_top5_avg=0.91923 lr=0.00010 gn=36.65590 time=54.11it/s +====================Eval==================== +epoch=108 global_step=42619 loss=2.15538 test_loss_avg=1.16301 acc=0.42969 test_acc_avg=0.66189 test_acc_top5_avg=0.96528 time=220.82it/s +epoch=108 global_step=42619 loss=0.26713 test_loss_avg=2.05895 acc=0.92188 test_acc_avg=0.52321 test_acc_top5_avg=0.89430 time=242.32it/s +epoch=108 global_step=42619 loss=0.14825 test_loss_avg=1.80203 acc=0.87500 test_acc_avg=0.57951 test_acc_top5_avg=0.90843 time=526.26it/s +curr_acc 0.5795 +BEST_ACC 0.6425 +curr_acc_top5 0.9084 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=3.29559 loss_avg=3.25506 acc=0.69531 acc_top1_avg=0.69783 acc_top5_avg=0.91961 lr=0.00010 gn=33.29393 time=55.07it/s +epoch=109 global_step=42700 loss=3.16176 loss_avg=3.23616 acc=0.70312 acc_top1_avg=0.70062 acc_top5_avg=0.91946 lr=0.00010 gn=40.64231 time=55.19it/s +epoch=109 global_step=42750 loss=2.35456 loss_avg=3.21680 acc=0.78125 acc_top1_avg=0.70289 acc_top5_avg=0.91746 lr=0.00010 gn=35.72296 time=51.36it/s +epoch=109 global_step=42800 loss=3.36849 loss_avg=3.19994 acc=0.68750 acc_top1_avg=0.70559 acc_top5_avg=0.91782 lr=0.00010 gn=41.61550 time=54.56it/s +epoch=109 global_step=42850 loss=3.03589 loss_avg=3.20092 acc=0.71094 acc_top1_avg=0.70553 acc_top5_avg=0.91846 lr=0.00010 gn=37.71270 time=61.53it/s +epoch=109 global_step=42900 loss=3.50028 loss_avg=3.20950 acc=0.66406 acc_top1_avg=0.70382 acc_top5_avg=0.91829 lr=0.00010 gn=40.13867 time=51.41it/s +epoch=109 global_step=42950 loss=3.66813 loss_avg=3.21172 acc=0.64844 acc_top1_avg=0.70343 acc_top5_avg=0.91862 lr=0.00010 gn=32.75455 time=43.64it/s +epoch=109 global_step=43000 loss=3.38943 loss_avg=3.21476 acc=0.69531 acc_top1_avg=0.70360 acc_top5_avg=0.91853 lr=0.00010 gn=36.06726 time=56.59it/s +====================Eval==================== +epoch=109 global_step=43010 loss=0.43420 test_loss_avg=2.02697 acc=0.86719 test_acc_avg=0.52163 test_acc_top5_avg=0.87039 time=238.12it/s +epoch=109 global_step=43010 loss=0.09558 test_loss_avg=1.85645 acc=0.93750 test_acc_avg=0.57318 test_acc_top5_avg=0.91317 time=645.28it/s +curr_acc 0.5732 +BEST_ACC 0.6425 +curr_acc_top5 0.9132 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=2.96050 loss_avg=3.08635 acc=0.72656 acc_top1_avg=0.71699 acc_top5_avg=0.92461 lr=0.00010 gn=38.09496 time=61.69it/s +epoch=110 global_step=43100 loss=2.98237 loss_avg=3.19605 acc=0.74219 acc_top1_avg=0.70469 acc_top5_avg=0.92231 lr=0.00010 gn=45.65304 time=52.37it/s +epoch=110 global_step=43150 loss=3.39824 loss_avg=3.22626 acc=0.67188 acc_top1_avg=0.70117 acc_top5_avg=0.91936 lr=0.00010 gn=36.72555 time=61.63it/s +epoch=110 global_step=43200 loss=3.15567 loss_avg=3.21293 acc=0.71875 acc_top1_avg=0.70280 acc_top5_avg=0.91970 lr=0.00010 gn=44.03246 time=56.56it/s +epoch=110 global_step=43250 loss=3.83466 loss_avg=3.23829 acc=0.64062 acc_top1_avg=0.70042 acc_top5_avg=0.91865 lr=0.00010 gn=52.15672 time=53.12it/s +epoch=110 global_step=43300 loss=2.98326 loss_avg=3.23028 acc=0.72656 acc_top1_avg=0.70078 acc_top5_avg=0.91918 lr=0.00010 gn=43.39665 time=55.19it/s +epoch=110 global_step=43350 loss=3.31121 loss_avg=3.22057 acc=0.68750 acc_top1_avg=0.70211 acc_top5_avg=0.91994 lr=0.00010 gn=43.69319 time=53.63it/s +epoch=110 global_step=43400 loss=3.41491 loss_avg=3.22238 acc=0.69531 acc_top1_avg=0.70194 acc_top5_avg=0.91971 lr=0.00010 gn=42.59303 time=63.03it/s +====================Eval==================== +epoch=110 global_step=43401 loss=0.24521 test_loss_avg=1.36907 acc=0.92969 test_acc_avg=0.59141 test_acc_top5_avg=0.95781 time=240.36it/s +epoch=110 global_step=43401 loss=0.70317 test_loss_avg=2.31834 acc=0.77344 test_acc_avg=0.47031 test_acc_top5_avg=0.87930 time=226.32it/s +epoch=110 global_step=43401 loss=0.15389 test_loss_avg=1.82125 acc=0.87500 test_acc_avg=0.57852 test_acc_top5_avg=0.90773 time=509.26it/s +curr_acc 0.5785 +BEST_ACC 0.6425 +curr_acc_top5 0.9077 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=3.40511 loss_avg=3.21074 acc=0.67969 acc_top1_avg=0.70472 acc_top5_avg=0.91422 lr=0.00010 gn=47.84809 time=45.94it/s +epoch=111 global_step=43500 loss=2.79994 loss_avg=3.23235 acc=0.75781 acc_top1_avg=0.70186 acc_top5_avg=0.91604 lr=0.00010 gn=37.67991 time=60.72it/s +epoch=111 global_step=43550 loss=3.37762 loss_avg=3.22676 acc=0.67969 acc_top1_avg=0.70255 acc_top5_avg=0.91710 lr=0.00010 gn=32.23241 time=63.16it/s +epoch=111 global_step=43600 loss=2.51269 loss_avg=3.19928 acc=0.78906 acc_top1_avg=0.70532 acc_top5_avg=0.91885 lr=0.00010 gn=48.79996 time=54.76it/s +epoch=111 global_step=43650 loss=2.51119 loss_avg=3.18538 acc=0.80469 acc_top1_avg=0.70714 acc_top5_avg=0.91871 lr=0.00010 gn=56.60587 time=56.22it/s +epoch=111 global_step=43700 loss=3.27916 loss_avg=3.20118 acc=0.70312 acc_top1_avg=0.70537 acc_top5_avg=0.91858 lr=0.00010 gn=38.66443 time=61.32it/s +epoch=111 global_step=43750 loss=3.70351 loss_avg=3.21382 acc=0.64062 acc_top1_avg=0.70400 acc_top5_avg=0.91912 lr=0.00010 gn=45.77013 time=53.96it/s +====================Eval==================== +epoch=111 global_step=43792 loss=5.14606 test_loss_avg=2.40607 acc=0.00000 test_acc_avg=0.43700 test_acc_top5_avg=0.83619 time=182.69it/s +epoch=111 global_step=43792 loss=0.11900 test_loss_avg=1.86709 acc=0.87500 test_acc_avg=0.57110 test_acc_top5_avg=0.90892 time=626.67it/s +curr_acc 0.5711 +BEST_ACC 0.6425 +curr_acc_top5 0.9089 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=3.46598 loss_avg=3.25960 acc=0.66406 acc_top1_avg=0.69727 acc_top5_avg=0.91406 lr=0.00010 gn=40.37156 time=51.94it/s +epoch=112 global_step=43850 loss=2.95905 loss_avg=3.16399 acc=0.72656 acc_top1_avg=0.70784 acc_top5_avg=0.92322 lr=0.00010 gn=34.86700 time=61.35it/s +epoch=112 global_step=43900 loss=3.32467 loss_avg=3.19333 acc=0.67969 acc_top1_avg=0.70443 acc_top5_avg=0.91811 lr=0.00010 gn=40.16495 time=62.17it/s +epoch=112 global_step=43950 loss=3.67354 loss_avg=3.20765 acc=0.64844 acc_top1_avg=0.70347 acc_top5_avg=0.91772 lr=0.00010 gn=36.08099 time=47.65it/s +epoch=112 global_step=44000 loss=3.24722 loss_avg=3.19714 acc=0.70312 acc_top1_avg=0.70523 acc_top5_avg=0.91906 lr=0.00010 gn=43.85881 time=49.29it/s +epoch=112 global_step=44050 loss=3.36633 loss_avg=3.20478 acc=0.68750 acc_top1_avg=0.70440 acc_top5_avg=0.91948 lr=0.00010 gn=39.85233 time=51.28it/s +epoch=112 global_step=44100 loss=3.53072 loss_avg=3.20603 acc=0.68750 acc_top1_avg=0.70482 acc_top5_avg=0.91835 lr=0.00010 gn=46.58195 time=62.05it/s +epoch=112 global_step=44150 loss=2.32459 loss_avg=3.21455 acc=0.80469 acc_top1_avg=0.70413 acc_top5_avg=0.91937 lr=0.00010 gn=41.83831 time=56.21it/s +====================Eval==================== +epoch=112 global_step=44183 loss=2.05018 test_loss_avg=1.94770 acc=0.40625 test_acc_avg=0.44531 test_acc_top5_avg=0.93750 time=238.54it/s +epoch=112 global_step=44183 loss=5.26231 test_loss_avg=2.32144 acc=0.00000 test_acc_avg=0.46890 test_acc_top5_avg=0.87094 time=237.84it/s +epoch=112 global_step=44183 loss=0.13335 test_loss_avg=1.82486 acc=0.87500 test_acc_avg=0.57644 test_acc_top5_avg=0.90645 time=632.24it/s +curr_acc 0.5764 +BEST_ACC 0.6425 +curr_acc_top5 0.9064 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=3.26826 loss_avg=3.13607 acc=0.70312 acc_top1_avg=0.71415 acc_top5_avg=0.91406 lr=0.00010 gn=35.83150 time=50.87it/s +epoch=113 global_step=44250 loss=2.67342 loss_avg=3.22896 acc=0.77344 acc_top1_avg=0.70278 acc_top5_avg=0.91569 lr=0.00010 gn=35.29814 time=56.17it/s +epoch=113 global_step=44300 loss=2.59297 loss_avg=3.20403 acc=0.77344 acc_top1_avg=0.70493 acc_top5_avg=0.91540 lr=0.00010 gn=48.09920 time=44.31it/s +epoch=113 global_step=44350 loss=2.99268 loss_avg=3.19503 acc=0.71875 acc_top1_avg=0.70500 acc_top5_avg=0.91720 lr=0.00010 gn=46.03414 time=56.75it/s +epoch=113 global_step=44400 loss=3.04054 loss_avg=3.21694 acc=0.73438 acc_top1_avg=0.70312 acc_top5_avg=0.91820 lr=0.00010 gn=51.25023 time=58.04it/s +epoch=113 global_step=44450 loss=2.60451 loss_avg=3.21884 acc=0.76562 acc_top1_avg=0.70324 acc_top5_avg=0.91904 lr=0.00010 gn=42.71089 time=53.27it/s +epoch=113 global_step=44500 loss=2.89225 loss_avg=3.20787 acc=0.74219 acc_top1_avg=0.70448 acc_top5_avg=0.91889 lr=0.00010 gn=42.13176 time=46.67it/s +epoch=113 global_step=44550 loss=3.03161 loss_avg=3.21360 acc=0.71094 acc_top1_avg=0.70336 acc_top5_avg=0.91915 lr=0.00010 gn=28.79521 time=61.25it/s +====================Eval==================== +epoch=113 global_step=44574 loss=2.08164 test_loss_avg=1.45068 acc=0.41406 test_acc_avg=0.59613 test_acc_top5_avg=0.94260 time=126.91it/s +epoch=113 global_step=44574 loss=0.24259 test_loss_avg=1.94628 acc=0.92188 test_acc_avg=0.54944 test_acc_top5_avg=0.90186 time=258.13it/s +epoch=113 global_step=44574 loss=0.09438 test_loss_avg=1.81077 acc=0.93750 test_acc_avg=0.57921 test_acc_top5_avg=0.90902 time=505.16it/s +curr_acc 0.5792 +BEST_ACC 0.6425 +curr_acc_top5 0.9090 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=3.12240 loss_avg=3.16601 acc=0.71094 acc_top1_avg=0.71064 acc_top5_avg=0.91707 lr=0.00010 gn=40.25934 time=53.14it/s +epoch=114 global_step=44650 loss=2.73263 loss_avg=3.18184 acc=0.76562 acc_top1_avg=0.70847 acc_top5_avg=0.91910 lr=0.00010 gn=45.05281 time=58.27it/s +epoch=114 global_step=44700 loss=3.41858 loss_avg=3.20140 acc=0.66406 acc_top1_avg=0.70616 acc_top5_avg=0.91735 lr=0.00010 gn=31.72184 time=53.04it/s +epoch=114 global_step=44750 loss=4.00771 loss_avg=3.20910 acc=0.63281 acc_top1_avg=0.70521 acc_top5_avg=0.91713 lr=0.00010 gn=38.90622 time=54.42it/s +epoch=114 global_step=44800 loss=3.19194 loss_avg=3.20426 acc=0.71094 acc_top1_avg=0.70534 acc_top5_avg=0.91762 lr=0.00010 gn=49.93402 time=52.05it/s +epoch=114 global_step=44850 loss=2.83121 loss_avg=3.19549 acc=0.75000 acc_top1_avg=0.70641 acc_top5_avg=0.91853 lr=0.00010 gn=38.11296 time=52.82it/s +epoch=114 global_step=44900 loss=3.27291 loss_avg=3.19497 acc=0.69531 acc_top1_avg=0.70655 acc_top5_avg=0.91859 lr=0.00010 gn=42.68165 time=56.16it/s +epoch=114 global_step=44950 loss=3.47572 loss_avg=3.20321 acc=0.67188 acc_top1_avg=0.70545 acc_top5_avg=0.91824 lr=0.00010 gn=43.27859 time=52.75it/s +====================Eval==================== +epoch=114 global_step=44965 loss=1.99414 test_loss_avg=1.96108 acc=0.49219 test_acc_avg=0.52841 test_acc_top5_avg=0.87536 time=240.97it/s +epoch=114 global_step=44965 loss=0.12167 test_loss_avg=1.80407 acc=0.93750 test_acc_avg=0.58238 test_acc_top5_avg=0.90961 time=809.71it/s +curr_acc 0.5824 +BEST_ACC 0.6425 +curr_acc_top5 0.9096 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=3.18674 loss_avg=3.12663 acc=0.71875 acc_top1_avg=0.71317 acc_top5_avg=0.92031 lr=0.00010 gn=45.87502 time=54.68it/s +epoch=115 global_step=45050 loss=3.05017 loss_avg=3.16648 acc=0.72656 acc_top1_avg=0.70901 acc_top5_avg=0.92270 lr=0.00010 gn=52.64193 time=58.50it/s +epoch=115 global_step=45100 loss=3.95693 loss_avg=3.17251 acc=0.63281 acc_top1_avg=0.70856 acc_top5_avg=0.92135 lr=0.00010 gn=47.68711 time=53.94it/s +epoch=115 global_step=45150 loss=3.43819 loss_avg=3.19371 acc=0.67969 acc_top1_avg=0.70650 acc_top5_avg=0.92031 lr=0.00010 gn=48.14574 time=52.77it/s +epoch=115 global_step=45200 loss=3.92447 loss_avg=3.19581 acc=0.64844 acc_top1_avg=0.70575 acc_top5_avg=0.91965 lr=0.00010 gn=54.11967 time=57.33it/s +epoch=115 global_step=45250 loss=3.46667 loss_avg=3.19642 acc=0.67969 acc_top1_avg=0.70581 acc_top5_avg=0.91870 lr=0.00010 gn=47.89740 time=57.48it/s +epoch=115 global_step=45300 loss=2.66990 loss_avg=3.18827 acc=0.77344 acc_top1_avg=0.70681 acc_top5_avg=0.91877 lr=0.00010 gn=41.85535 time=53.30it/s +epoch=115 global_step=45350 loss=3.38830 loss_avg=3.19306 acc=0.68750 acc_top1_avg=0.70643 acc_top5_avg=0.91875 lr=0.00010 gn=37.79425 time=56.46it/s +====================Eval==================== +epoch=115 global_step=45356 loss=0.36497 test_loss_avg=1.02533 acc=0.92969 test_acc_avg=0.70260 test_acc_top5_avg=0.97656 time=158.42it/s +epoch=115 global_step=45356 loss=0.05651 test_loss_avg=2.18515 acc=0.98438 test_acc_avg=0.50048 test_acc_top5_avg=0.89026 time=132.67it/s +epoch=115 global_step=45356 loss=0.17736 test_loss_avg=1.83555 acc=0.87500 test_acc_avg=0.57674 test_acc_top5_avg=0.90882 time=839.70it/s +curr_acc 0.5767 +BEST_ACC 0.6425 +curr_acc_top5 0.9088 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=4.10778 loss_avg=3.15661 acc=0.60156 acc_top1_avg=0.71094 acc_top5_avg=0.91673 lr=0.00010 gn=32.56168 time=54.31it/s +epoch=116 global_step=45450 loss=3.44607 loss_avg=3.16142 acc=0.67188 acc_top1_avg=0.70952 acc_top5_avg=0.92079 lr=0.00010 gn=43.32787 time=45.41it/s +epoch=116 global_step=45500 loss=2.89777 loss_avg=3.16343 acc=0.72656 acc_top1_avg=0.70850 acc_top5_avg=0.92036 lr=0.00010 gn=42.32755 time=58.21it/s +epoch=116 global_step=45550 loss=2.82858 loss_avg=3.18606 acc=0.75000 acc_top1_avg=0.70606 acc_top5_avg=0.91873 lr=0.00010 gn=42.23526 time=53.74it/s +epoch=116 global_step=45600 loss=3.67505 loss_avg=3.20319 acc=0.64844 acc_top1_avg=0.70409 acc_top5_avg=0.91861 lr=0.00010 gn=31.74917 time=58.91it/s +epoch=116 global_step=45650 loss=2.68503 loss_avg=3.20380 acc=0.75000 acc_top1_avg=0.70398 acc_top5_avg=0.91831 lr=0.00010 gn=41.71324 time=57.43it/s +epoch=116 global_step=45700 loss=3.07479 loss_avg=3.20125 acc=0.72656 acc_top1_avg=0.70403 acc_top5_avg=0.91913 lr=0.00010 gn=42.20946 time=47.78it/s +====================Eval==================== +epoch=116 global_step=45747 loss=0.36738 test_loss_avg=2.10448 acc=0.91406 test_acc_avg=0.50412 test_acc_top5_avg=0.85503 time=100.82it/s +epoch=116 global_step=45747 loss=0.15656 test_loss_avg=1.84940 acc=0.87500 test_acc_avg=0.57288 test_acc_top5_avg=0.90843 time=561.26it/s +curr_acc 0.5729 +BEST_ACC 0.6425 +curr_acc_top5 0.9084 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=3.61581 loss_avg=2.94149 acc=0.66406 acc_top1_avg=0.73698 acc_top5_avg=0.91406 lr=0.00010 gn=40.40648 time=54.44it/s +epoch=117 global_step=45800 loss=3.22873 loss_avg=3.21347 acc=0.71094 acc_top1_avg=0.70475 acc_top5_avg=0.91524 lr=0.00010 gn=34.78738 time=54.60it/s +epoch=117 global_step=45850 loss=2.99699 loss_avg=3.17378 acc=0.71875 acc_top1_avg=0.70843 acc_top5_avg=0.91937 lr=0.00010 gn=43.74598 time=54.98it/s +epoch=117 global_step=45900 loss=2.86350 loss_avg=3.18053 acc=0.75000 acc_top1_avg=0.70741 acc_top5_avg=0.91912 lr=0.00010 gn=43.80100 time=52.74it/s +epoch=117 global_step=45950 loss=2.78219 loss_avg=3.17930 acc=0.75781 acc_top1_avg=0.70751 acc_top5_avg=0.91922 lr=0.00010 gn=45.00471 time=54.88it/s +epoch=117 global_step=46000 loss=2.71512 loss_avg=3.18076 acc=0.74219 acc_top1_avg=0.70751 acc_top5_avg=0.91922 lr=0.00010 gn=40.59543 time=55.42it/s +epoch=117 global_step=46050 loss=3.07290 loss_avg=3.18314 acc=0.73438 acc_top1_avg=0.70676 acc_top5_avg=0.91837 lr=0.00010 gn=48.52222 time=54.69it/s +epoch=117 global_step=46100 loss=3.04471 loss_avg=3.18352 acc=0.72656 acc_top1_avg=0.70667 acc_top5_avg=0.91875 lr=0.00010 gn=38.48603 time=59.51it/s +====================Eval==================== +epoch=117 global_step=46138 loss=1.88897 test_loss_avg=1.98595 acc=0.40625 test_acc_avg=0.42522 test_acc_top5_avg=0.94531 time=225.11it/s +epoch=117 global_step=46138 loss=0.90434 test_loss_avg=2.40804 acc=0.70312 test_acc_avg=0.45491 test_acc_top5_avg=0.87431 time=238.30it/s +epoch=117 global_step=46138 loss=0.16476 test_loss_avg=1.82898 acc=0.87500 test_acc_avg=0.57595 test_acc_top5_avg=0.90803 time=589.83it/s +curr_acc 0.5759 +BEST_ACC 0.6425 +curr_acc_top5 0.9080 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=3.18452 loss_avg=3.23847 acc=0.70312 acc_top1_avg=0.70117 acc_top5_avg=0.90299 lr=0.00010 gn=38.62166 time=49.17it/s +epoch=118 global_step=46200 loss=3.16640 loss_avg=3.17457 acc=0.71094 acc_top1_avg=0.70640 acc_top5_avg=0.92049 lr=0.00010 gn=45.94862 time=58.21it/s +epoch=118 global_step=46250 loss=3.22196 loss_avg=3.17363 acc=0.70312 acc_top1_avg=0.70647 acc_top5_avg=0.92076 lr=0.00010 gn=46.56717 time=55.75it/s +epoch=118 global_step=46300 loss=2.98978 loss_avg=3.19230 acc=0.73438 acc_top1_avg=0.70510 acc_top5_avg=0.92052 lr=0.00010 gn=38.16798 time=57.93it/s +epoch=118 global_step=46350 loss=3.58296 loss_avg=3.18008 acc=0.66406 acc_top1_avg=0.70685 acc_top5_avg=0.92011 lr=0.00010 gn=43.54866 time=60.22it/s +epoch=118 global_step=46400 loss=3.59292 loss_avg=3.18499 acc=0.65625 acc_top1_avg=0.70635 acc_top5_avg=0.92029 lr=0.00010 gn=40.21900 time=55.41it/s +epoch=118 global_step=46450 loss=2.39288 loss_avg=3.17484 acc=0.81250 acc_top1_avg=0.70746 acc_top5_avg=0.91967 lr=0.00010 gn=44.28480 time=54.68it/s +epoch=118 global_step=46500 loss=3.29252 loss_avg=3.17971 acc=0.70312 acc_top1_avg=0.70703 acc_top5_avg=0.91954 lr=0.00010 gn=55.89120 time=50.78it/s +====================Eval==================== +epoch=118 global_step=46529 loss=4.94984 test_loss_avg=1.96807 acc=0.00000 test_acc_avg=0.51646 test_acc_top5_avg=0.87751 time=236.31it/s +epoch=118 global_step=46529 loss=0.14921 test_loss_avg=1.83758 acc=0.94531 test_acc_avg=0.57843 test_acc_top5_avg=0.90795 time=246.81it/s +epoch=118 global_step=46529 loss=0.14327 test_loss_avg=1.81614 acc=0.87500 test_acc_avg=0.58218 test_acc_top5_avg=0.90912 time=499.38it/s +curr_acc 0.5822 +BEST_ACC 0.6425 +curr_acc_top5 0.9091 +BEST_ACC_top5 0.9426 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=3.39738 loss_avg=3.23556 acc=0.67969 acc_top1_avg=0.70015 acc_top5_avg=0.90811 lr=0.00010 gn=36.30381 time=53.57it/s +epoch=119 global_step=46600 loss=3.25221 loss_avg=3.26762 acc=0.72656 acc_top1_avg=0.69927 acc_top5_avg=0.90889 lr=0.00010 gn=52.20328 time=60.41it/s +epoch=119 global_step=46650 loss=2.68465 loss_avg=3.19986 acc=0.77344 acc_top1_avg=0.70622 acc_top5_avg=0.91348 lr=0.00010 gn=53.36059 time=55.97it/s +epoch=119 global_step=46700 loss=3.36694 loss_avg=3.17670 acc=0.68750 acc_top1_avg=0.70788 acc_top5_avg=0.91772 lr=0.00010 gn=37.46547 time=54.97it/s +epoch=119 global_step=46750 loss=2.68883 loss_avg=3.18777 acc=0.75781 acc_top1_avg=0.70599 acc_top5_avg=0.91820 lr=0.00010 gn=45.18680 time=57.27it/s +epoch=119 global_step=46800 loss=2.26110 loss_avg=3.18281 acc=0.80469 acc_top1_avg=0.70661 acc_top5_avg=0.91844 lr=0.00010 gn=41.78040 time=55.32it/s +epoch=119 global_step=46850 loss=3.42751 loss_avg=3.18164 acc=0.69531 acc_top1_avg=0.70668 acc_top5_avg=0.91825 lr=0.00010 gn=47.01571 time=52.86it/s +epoch=119 global_step=46900 loss=3.24618 loss_avg=3.17713 acc=0.69531 acc_top1_avg=0.70687 acc_top5_avg=0.91834 lr=0.00010 gn=47.53740 time=55.25it/s +====================Eval==================== +epoch=119 global_step=46920 loss=5.16950 test_loss_avg=2.12020 acc=0.00000 test_acc_avg=0.50287 test_acc_top5_avg=0.87612 time=237.36it/s +epoch=119 global_step=46920 loss=0.14645 test_loss_avg=1.80580 acc=0.87500 test_acc_avg=0.58267 test_acc_top5_avg=0.90892 time=589.17it/s +curr_acc 0.5827 +BEST_ACC 0.6425 +curr_acc_top5 0.9089 +BEST_ACC_top5 0.9426 +Model Saved! + diff --git a/other_methods/sceloss/sceloss_results/out_6_6.log b/other_methods/sceloss/sceloss_results/out_6_6.log new file mode 100644 index 0000000..e50145a --- /dev/null +++ b/other_methods/sceloss/sceloss_results/out_6_6.log @@ -0,0 +1,2227 @@ +lr 0.01 +l2_reg 0.005 +grad_bound 5.0 +train_log_every 50 +resume False +batch_size 128 +data_path data +checkpoint_path checkpoints +data_nums_workers 8 +epoch 120 +nr 0.4 +loss SCE +version SCE0.0 +train_cifar100 False +fn /home/cgn/Dropbox (MIT)/cgn/cleanlab/examples/cifar10/cifar10/cifar10_noisy_labels/cifar10_noisy_labels__frac_zero_noise_rates__0.6__noise_amount__0.6.json +Using CUDA! +Num of train 50000 +Num of test 10000 +SCELoss +Number of Trainable Parameters 1.6398 +====================Training==================== +epoch=0 global_step=50 loss=7.01118 loss_avg=7.53365 acc=0.24219 acc_top1_avg=0.23312 acc_top5_avg=0.69406 lr=0.01000 gn=8.61007 time=63.28it/s +epoch=0 global_step=100 loss=7.11009 loss_avg=7.27814 acc=0.26562 acc_top1_avg=0.26156 acc_top5_avg=0.73617 lr=0.01000 gn=5.95384 time=62.25it/s +epoch=0 global_step=150 loss=7.31956 loss_avg=7.14990 acc=0.25781 acc_top1_avg=0.27484 acc_top5_avg=0.75208 lr=0.01000 gn=5.58647 time=56.90it/s +epoch=0 global_step=200 loss=6.44664 loss_avg=7.06433 acc=0.35938 acc_top1_avg=0.28441 acc_top5_avg=0.76133 lr=0.01000 gn=4.86322 time=63.40it/s +epoch=0 global_step=250 loss=6.35591 loss_avg=6.98009 acc=0.35156 acc_top1_avg=0.29350 acc_top5_avg=0.76709 lr=0.01000 gn=4.03626 time=63.24it/s +epoch=0 global_step=300 loss=6.93842 loss_avg=6.93042 acc=0.29688 acc_top1_avg=0.29792 acc_top5_avg=0.77310 lr=0.01000 gn=4.72906 time=60.61it/s +epoch=0 global_step=350 loss=6.92483 loss_avg=6.87458 acc=0.29688 acc_top1_avg=0.30346 acc_top5_avg=0.77915 lr=0.01000 gn=2.84342 time=63.30it/s +====================Eval==================== +epoch=0 global_step=391 loss=6.24124 test_loss_avg=5.46476 acc=0.00000 test_acc_avg=0.17906 test_acc_top5_avg=0.67172 time=225.84it/s +epoch=0 global_step=391 loss=8.21015 test_loss_avg=5.04058 acc=0.00000 test_acc_avg=0.22231 test_acc_top5_avg=0.67207 time=32.64it/s +curr_acc 0.2223 +BEST_ACC 0.0000 +curr_acc_top5 0.6721 +BEST_ACC_top5 0.0000 +Model Saved! + +====================Training==================== +epoch=1 global_step=400 loss=6.13326 loss_avg=6.35949 acc=0.39062 acc_top1_avg=0.35417 acc_top5_avg=0.83420 lr=0.01000 gn=4.36406 time=66.38it/s +epoch=1 global_step=450 loss=5.86535 loss_avg=6.41857 acc=0.39844 acc_top1_avg=0.35130 acc_top5_avg=0.82044 lr=0.01000 gn=4.11744 time=61.14it/s +epoch=1 global_step=500 loss=5.99964 loss_avg=6.35160 acc=0.40625 acc_top1_avg=0.36009 acc_top5_avg=0.82239 lr=0.01000 gn=4.48077 time=62.49it/s +epoch=1 global_step=550 loss=6.06113 loss_avg=6.33708 acc=0.38281 acc_top1_avg=0.36119 acc_top5_avg=0.82385 lr=0.01000 gn=4.64343 time=67.36it/s +epoch=1 global_step=600 loss=6.20613 loss_avg=6.32510 acc=0.35156 acc_top1_avg=0.36195 acc_top5_avg=0.82360 lr=0.01000 gn=3.75395 time=67.33it/s +epoch=1 global_step=650 loss=6.34442 loss_avg=6.32643 acc=0.32812 acc_top1_avg=0.36185 acc_top5_avg=0.82390 lr=0.01000 gn=3.60247 time=62.56it/s +epoch=1 global_step=700 loss=5.73753 loss_avg=6.32679 acc=0.40625 acc_top1_avg=0.36157 acc_top5_avg=0.82370 lr=0.01000 gn=3.93437 time=67.25it/s +epoch=1 global_step=750 loss=6.15409 loss_avg=6.32243 acc=0.35156 acc_top1_avg=0.36188 acc_top5_avg=0.82458 lr=0.01000 gn=3.85086 time=67.51it/s +====================Eval==================== +epoch=1 global_step=782 loss=5.43368 test_loss_avg=4.31865 acc=0.00000 test_acc_avg=0.22991 test_acc_top5_avg=0.91518 time=246.26it/s +epoch=1 global_step=782 loss=4.48356 test_loss_avg=4.47643 acc=0.30469 test_acc_avg=0.25165 test_acc_top5_avg=0.76981 time=252.18it/s +epoch=1 global_step=782 loss=6.99377 test_loss_avg=4.68572 acc=0.00000 test_acc_avg=0.22735 test_acc_top5_avg=0.74417 time=871.09it/s +curr_acc 0.2274 +BEST_ACC 0.2223 +curr_acc_top5 0.7442 +BEST_ACC_top5 0.6721 +Model Saved! + +====================Training==================== +epoch=2 global_step=800 loss=5.92616 loss_avg=6.26591 acc=0.40625 acc_top1_avg=0.36849 acc_top5_avg=0.82161 lr=0.01000 gn=4.52025 time=63.05it/s +epoch=2 global_step=850 loss=6.08540 loss_avg=6.17593 acc=0.39062 acc_top1_avg=0.37569 acc_top5_avg=0.83134 lr=0.01000 gn=5.04172 time=61.84it/s +epoch=2 global_step=900 loss=6.00742 loss_avg=6.17316 acc=0.39844 acc_top1_avg=0.37659 acc_top5_avg=0.83561 lr=0.01000 gn=3.45914 time=61.19it/s +epoch=2 global_step=950 loss=6.07192 loss_avg=6.13730 acc=0.37500 acc_top1_avg=0.37932 acc_top5_avg=0.83794 lr=0.01000 gn=4.57048 time=63.18it/s +epoch=2 global_step=1000 loss=6.19509 loss_avg=6.11443 acc=0.39062 acc_top1_avg=0.38238 acc_top5_avg=0.83923 lr=0.01000 gn=5.37581 time=57.93it/s +epoch=2 global_step=1050 loss=6.30163 loss_avg=6.10834 acc=0.35156 acc_top1_avg=0.38351 acc_top5_avg=0.83882 lr=0.01000 gn=4.21358 time=63.27it/s +epoch=2 global_step=1100 loss=6.66822 loss_avg=6.10594 acc=0.35156 acc_top1_avg=0.38392 acc_top5_avg=0.83893 lr=0.01000 gn=4.06975 time=62.77it/s +epoch=2 global_step=1150 loss=6.26431 loss_avg=6.09343 acc=0.35938 acc_top1_avg=0.38536 acc_top5_avg=0.83933 lr=0.01000 gn=4.84730 time=61.07it/s +====================Eval==================== +epoch=2 global_step=1173 loss=2.98084 test_loss_avg=4.93022 acc=0.21875 test_acc_avg=0.16164 test_acc_top5_avg=0.61533 time=244.59it/s +epoch=2 global_step=1173 loss=6.07684 test_loss_avg=4.41384 acc=0.00000 test_acc_avg=0.23695 test_acc_top5_avg=0.76038 time=894.50it/s +curr_acc 0.2369 +BEST_ACC 0.2274 +curr_acc_top5 0.7604 +BEST_ACC_top5 0.7442 +Model Saved! + +====================Training==================== +epoch=3 global_step=1200 loss=6.19258 loss_avg=6.10694 acc=0.35938 acc_top1_avg=0.38628 acc_top5_avg=0.83883 lr=0.01000 gn=4.33462 time=62.95it/s +epoch=3 global_step=1250 loss=6.05563 loss_avg=5.95337 acc=0.40625 acc_top1_avg=0.40321 acc_top5_avg=0.84375 lr=0.01000 gn=4.27036 time=62.36it/s +epoch=3 global_step=1300 loss=6.51168 loss_avg=5.96078 acc=0.34375 acc_top1_avg=0.40157 acc_top5_avg=0.84437 lr=0.01000 gn=2.31776 time=63.07it/s +epoch=3 global_step=1350 loss=5.70162 loss_avg=5.97699 acc=0.42188 acc_top1_avg=0.39967 acc_top5_avg=0.84459 lr=0.01000 gn=3.63901 time=63.02it/s +epoch=3 global_step=1400 loss=6.19475 loss_avg=5.98408 acc=0.39844 acc_top1_avg=0.39868 acc_top5_avg=0.84595 lr=0.01000 gn=5.37524 time=58.16it/s +epoch=3 global_step=1450 loss=6.13468 loss_avg=5.97557 acc=0.38281 acc_top1_avg=0.39934 acc_top5_avg=0.84651 lr=0.01000 gn=4.19624 time=63.06it/s +epoch=3 global_step=1500 loss=6.22747 loss_avg=5.98537 acc=0.37500 acc_top1_avg=0.39801 acc_top5_avg=0.84674 lr=0.01000 gn=3.57434 time=62.89it/s +epoch=3 global_step=1550 loss=5.64631 loss_avg=5.98205 acc=0.44531 acc_top1_avg=0.39848 acc_top5_avg=0.84709 lr=0.01000 gn=4.12996 time=62.67it/s +====================Eval==================== +epoch=3 global_step=1564 loss=2.84493 test_loss_avg=5.46329 acc=0.18750 test_acc_avg=0.06671 test_acc_top5_avg=0.80108 time=262.88it/s +epoch=3 global_step=1564 loss=0.78269 test_loss_avg=4.64114 acc=0.77344 test_acc_avg=0.16009 test_acc_top5_avg=0.72036 time=255.78it/s +epoch=3 global_step=1564 loss=6.37674 test_loss_avg=4.51675 acc=0.00000 test_acc_avg=0.18928 test_acc_top5_avg=0.72122 time=920.21it/s +curr_acc 0.1893 +BEST_ACC 0.2369 +curr_acc_top5 0.7212 +BEST_ACC_top5 0.7604 +Model Saved! + +====================Training==================== +epoch=4 global_step=1600 loss=5.88657 loss_avg=5.92756 acc=0.42969 acc_top1_avg=0.40690 acc_top5_avg=0.85894 lr=0.01000 gn=5.10912 time=56.20it/s +epoch=4 global_step=1650 loss=6.15334 loss_avg=5.91635 acc=0.37500 acc_top1_avg=0.40725 acc_top5_avg=0.85574 lr=0.01000 gn=3.59916 time=65.51it/s +epoch=4 global_step=1700 loss=5.48124 loss_avg=5.90850 acc=0.45312 acc_top1_avg=0.40769 acc_top5_avg=0.85478 lr=0.01000 gn=3.59200 time=62.74it/s +epoch=4 global_step=1750 loss=5.65181 loss_avg=5.88942 acc=0.42969 acc_top1_avg=0.40961 acc_top5_avg=0.85555 lr=0.01000 gn=3.81335 time=63.81it/s +epoch=4 global_step=1800 loss=5.81572 loss_avg=5.87790 acc=0.42188 acc_top1_avg=0.41108 acc_top5_avg=0.85378 lr=0.01000 gn=4.93130 time=63.92it/s +epoch=4 global_step=1850 loss=5.97907 loss_avg=5.87778 acc=0.40625 acc_top1_avg=0.41013 acc_top5_avg=0.85402 lr=0.01000 gn=4.75185 time=62.90it/s +epoch=4 global_step=1900 loss=6.16541 loss_avg=5.88169 acc=0.36719 acc_top1_avg=0.40951 acc_top5_avg=0.85361 lr=0.01000 gn=4.21485 time=60.60it/s +epoch=4 global_step=1950 loss=5.57595 loss_avg=5.88712 acc=0.43750 acc_top1_avg=0.40852 acc_top5_avg=0.85288 lr=0.01000 gn=4.33180 time=63.00it/s +====================Eval==================== +epoch=4 global_step=1955 loss=7.84307 test_loss_avg=5.56160 acc=0.00000 test_acc_avg=0.04802 test_acc_top5_avg=0.71553 time=260.81it/s +epoch=4 global_step=1955 loss=4.77655 test_loss_avg=4.57328 acc=0.00000 test_acc_avg=0.22201 test_acc_top5_avg=0.80024 time=911.21it/s +curr_acc 0.2220 +BEST_ACC 0.2369 +curr_acc_top5 0.8002 +BEST_ACC_top5 0.7604 +Model Saved! + +====================Training==================== +epoch=5 global_step=2000 loss=5.92179 loss_avg=5.87132 acc=0.40625 acc_top1_avg=0.41233 acc_top5_avg=0.85451 lr=0.01000 gn=4.08055 time=66.27it/s +epoch=5 global_step=2050 loss=5.54664 loss_avg=5.85029 acc=0.45312 acc_top1_avg=0.41456 acc_top5_avg=0.85633 lr=0.01000 gn=5.05184 time=62.04it/s +epoch=5 global_step=2100 loss=5.94778 loss_avg=5.84304 acc=0.39062 acc_top1_avg=0.41439 acc_top5_avg=0.85679 lr=0.01000 gn=4.13535 time=61.54it/s +epoch=5 global_step=2150 loss=5.26270 loss_avg=5.85933 acc=0.47656 acc_top1_avg=0.41222 acc_top5_avg=0.85545 lr=0.01000 gn=4.14927 time=62.62it/s +epoch=5 global_step=2200 loss=5.90548 loss_avg=5.82312 acc=0.40625 acc_top1_avg=0.41629 acc_top5_avg=0.85654 lr=0.01000 gn=4.92363 time=61.27it/s +epoch=5 global_step=2250 loss=6.35914 loss_avg=5.83100 acc=0.33594 acc_top1_avg=0.41488 acc_top5_avg=0.85728 lr=0.01000 gn=5.02873 time=59.88it/s +epoch=5 global_step=2300 loss=5.58672 loss_avg=5.83841 acc=0.45312 acc_top1_avg=0.41388 acc_top5_avg=0.85632 lr=0.01000 gn=4.58514 time=63.42it/s +====================Eval==================== +epoch=5 global_step=2346 loss=5.73556 test_loss_avg=5.78233 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.80469 time=247.82it/s +epoch=5 global_step=2346 loss=4.31586 test_loss_avg=4.99195 acc=0.20312 test_acc_avg=0.09318 test_acc_top5_avg=0.70469 time=258.94it/s +epoch=5 global_step=2346 loss=6.30164 test_loss_avg=4.28913 acc=0.00000 test_acc_avg=0.21262 test_acc_top5_avg=0.76464 time=921.42it/s +curr_acc 0.2126 +BEST_ACC 0.2369 +curr_acc_top5 0.7646 +BEST_ACC_top5 0.8002 +Model Saved! + +====================Training==================== +epoch=6 global_step=2350 loss=6.18784 loss_avg=5.50211 acc=0.39844 acc_top1_avg=0.44727 acc_top5_avg=0.87500 lr=0.01000 gn=4.48687 time=57.12it/s +epoch=6 global_step=2400 loss=5.93264 loss_avg=5.72658 acc=0.40625 acc_top1_avg=0.42578 acc_top5_avg=0.86560 lr=0.01000 gn=5.11916 time=64.61it/s +epoch=6 global_step=2450 loss=5.76718 loss_avg=5.76633 acc=0.42188 acc_top1_avg=0.42090 acc_top5_avg=0.86185 lr=0.01000 gn=4.97782 time=65.02it/s +epoch=6 global_step=2500 loss=5.70606 loss_avg=5.76967 acc=0.39844 acc_top1_avg=0.42152 acc_top5_avg=0.85988 lr=0.01000 gn=4.51756 time=66.18it/s +epoch=6 global_step=2550 loss=6.11725 loss_avg=5.78945 acc=0.37500 acc_top1_avg=0.41847 acc_top5_avg=0.86026 lr=0.01000 gn=4.39158 time=62.27it/s +epoch=6 global_step=2600 loss=5.73412 loss_avg=5.79205 acc=0.42969 acc_top1_avg=0.41834 acc_top5_avg=0.86002 lr=0.01000 gn=4.04039 time=67.36it/s +epoch=6 global_step=2650 loss=5.32173 loss_avg=5.80232 acc=0.48438 acc_top1_avg=0.41733 acc_top5_avg=0.85871 lr=0.01000 gn=4.65231 time=67.38it/s +epoch=6 global_step=2700 loss=6.14399 loss_avg=5.79625 acc=0.38281 acc_top1_avg=0.41810 acc_top5_avg=0.85851 lr=0.01000 gn=4.88862 time=67.14it/s +====================Eval==================== +epoch=6 global_step=2737 loss=7.08060 test_loss_avg=4.54339 acc=0.00000 test_acc_avg=0.15715 test_acc_top5_avg=0.86809 time=240.90it/s +epoch=6 global_step=2737 loss=6.16477 test_loss_avg=4.52871 acc=0.02344 test_acc_avg=0.22615 test_acc_top5_avg=0.79688 time=259.63it/s +epoch=6 global_step=2737 loss=5.84486 test_loss_avg=4.58332 acc=0.00000 test_acc_avg=0.21786 test_acc_top5_avg=0.79668 time=892.98it/s +curr_acc 0.2179 +BEST_ACC 0.2369 +curr_acc_top5 0.7967 +BEST_ACC_top5 0.8002 +Model Saved! + +====================Training==================== +epoch=7 global_step=2750 loss=6.02859 loss_avg=5.69259 acc=0.39844 acc_top1_avg=0.43450 acc_top5_avg=0.86418 lr=0.01000 gn=4.92160 time=62.85it/s +epoch=7 global_step=2800 loss=6.12602 loss_avg=5.78946 acc=0.39062 acc_top1_avg=0.41964 acc_top5_avg=0.85466 lr=0.01000 gn=4.78692 time=62.76it/s +epoch=7 global_step=2850 loss=5.93402 loss_avg=5.73411 acc=0.39062 acc_top1_avg=0.42478 acc_top5_avg=0.85654 lr=0.01000 gn=3.38532 time=61.36it/s +epoch=7 global_step=2900 loss=6.58066 loss_avg=5.75056 acc=0.35156 acc_top1_avg=0.42259 acc_top5_avg=0.85842 lr=0.01000 gn=4.69313 time=58.07it/s +epoch=7 global_step=2950 loss=5.92224 loss_avg=5.74742 acc=0.41406 acc_top1_avg=0.42265 acc_top5_avg=0.85978 lr=0.01000 gn=5.20323 time=56.82it/s +epoch=7 global_step=3000 loss=5.42492 loss_avg=5.74443 acc=0.45312 acc_top1_avg=0.42309 acc_top5_avg=0.85982 lr=0.01000 gn=5.95843 time=61.79it/s +epoch=7 global_step=3050 loss=5.89293 loss_avg=5.73036 acc=0.39844 acc_top1_avg=0.42457 acc_top5_avg=0.86020 lr=0.01000 gn=5.44551 time=61.65it/s +epoch=7 global_step=3100 loss=5.32487 loss_avg=5.74612 acc=0.46875 acc_top1_avg=0.42332 acc_top5_avg=0.86060 lr=0.01000 gn=5.10331 time=56.25it/s +====================Eval==================== +epoch=7 global_step=3128 loss=3.72905 test_loss_avg=4.81300 acc=0.21094 test_acc_avg=0.15209 test_acc_top5_avg=0.75199 time=262.88it/s +epoch=7 global_step=3128 loss=5.33264 test_loss_avg=4.48456 acc=0.00000 test_acc_avg=0.23527 test_acc_top5_avg=0.81833 time=926.30it/s +curr_acc 0.2353 +BEST_ACC 0.2369 +curr_acc_top5 0.8183 +BEST_ACC_top5 0.8002 +Model Saved! + +====================Training==================== +epoch=8 global_step=3150 loss=5.07166 loss_avg=5.71745 acc=0.49219 acc_top1_avg=0.42685 acc_top5_avg=0.85902 lr=0.01000 gn=4.55830 time=62.77it/s +epoch=8 global_step=3200 loss=5.20631 loss_avg=5.80496 acc=0.48438 acc_top1_avg=0.41688 acc_top5_avg=0.86100 lr=0.01000 gn=4.56309 time=64.62it/s +epoch=8 global_step=3250 loss=5.19416 loss_avg=5.77668 acc=0.47656 acc_top1_avg=0.41887 acc_top5_avg=0.86155 lr=0.01000 gn=4.47342 time=58.00it/s +epoch=8 global_step=3300 loss=6.28652 loss_avg=5.75521 acc=0.36719 acc_top1_avg=0.42110 acc_top5_avg=0.86251 lr=0.01000 gn=6.01911 time=56.50it/s +epoch=8 global_step=3350 loss=5.45450 loss_avg=5.74041 acc=0.46094 acc_top1_avg=0.42251 acc_top5_avg=0.86402 lr=0.01000 gn=4.95874 time=60.56it/s +epoch=8 global_step=3400 loss=5.72958 loss_avg=5.75218 acc=0.42188 acc_top1_avg=0.42133 acc_top5_avg=0.86302 lr=0.01000 gn=4.52873 time=63.11it/s +epoch=8 global_step=3450 loss=5.38102 loss_avg=5.74988 acc=0.46094 acc_top1_avg=0.42144 acc_top5_avg=0.86221 lr=0.01000 gn=5.53237 time=61.33it/s +epoch=8 global_step=3500 loss=6.00988 loss_avg=5.74805 acc=0.39062 acc_top1_avg=0.42194 acc_top5_avg=0.86229 lr=0.01000 gn=4.68599 time=50.61it/s +====================Eval==================== +epoch=8 global_step=3519 loss=5.21443 test_loss_avg=4.08600 acc=0.00000 test_acc_avg=0.21832 test_acc_top5_avg=0.89931 time=259.81it/s +epoch=8 global_step=3519 loss=0.43610 test_loss_avg=4.33391 acc=0.87500 test_acc_avg=0.24724 test_acc_top5_avg=0.79366 time=233.41it/s +epoch=8 global_step=3519 loss=5.77936 test_loss_avg=4.39623 acc=0.00000 test_acc_avg=0.24080 test_acc_top5_avg=0.78570 time=912.40it/s +curr_acc 0.2408 +BEST_ACC 0.2369 +curr_acc_top5 0.7857 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=9 global_step=3550 loss=5.39809 loss_avg=5.64474 acc=0.46094 acc_top1_avg=0.43775 acc_top5_avg=0.86542 lr=0.01000 gn=5.07148 time=61.25it/s +epoch=9 global_step=3600 loss=6.13000 loss_avg=5.67736 acc=0.35156 acc_top1_avg=0.43027 acc_top5_avg=0.86246 lr=0.01000 gn=4.53372 time=48.75it/s +epoch=9 global_step=3650 loss=6.11319 loss_avg=5.69145 acc=0.35938 acc_top1_avg=0.42951 acc_top5_avg=0.86367 lr=0.01000 gn=5.05176 time=65.31it/s +epoch=9 global_step=3700 loss=6.13034 loss_avg=5.67620 acc=0.36719 acc_top1_avg=0.43033 acc_top5_avg=0.86369 lr=0.01000 gn=6.31450 time=61.75it/s +epoch=9 global_step=3750 loss=5.85958 loss_avg=5.68919 acc=0.42188 acc_top1_avg=0.42871 acc_top5_avg=0.86333 lr=0.01000 gn=5.45209 time=62.41it/s +epoch=9 global_step=3800 loss=5.09909 loss_avg=5.68286 acc=0.49219 acc_top1_avg=0.42947 acc_top5_avg=0.86421 lr=0.01000 gn=4.85542 time=63.19it/s +epoch=9 global_step=3850 loss=5.80504 loss_avg=5.68495 acc=0.39844 acc_top1_avg=0.42943 acc_top5_avg=0.86360 lr=0.01000 gn=5.63874 time=59.91it/s +epoch=9 global_step=3900 loss=6.34126 loss_avg=5.69337 acc=0.37500 acc_top1_avg=0.42827 acc_top5_avg=0.86292 lr=0.01000 gn=5.41517 time=60.11it/s +====================Eval==================== +epoch=9 global_step=3910 loss=6.40195 test_loss_avg=5.45228 acc=0.00000 test_acc_avg=0.09976 test_acc_top5_avg=0.72596 time=251.93it/s +epoch=9 global_step=3910 loss=7.47723 test_loss_avg=4.71849 acc=0.00000 test_acc_avg=0.21450 test_acc_top5_avg=0.79658 time=898.91it/s +curr_acc 0.2145 +BEST_ACC 0.2408 +curr_acc_top5 0.7966 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=10 global_step=3950 loss=5.35874 loss_avg=5.62237 acc=0.47656 acc_top1_avg=0.43672 acc_top5_avg=0.86602 lr=0.01000 gn=5.73342 time=59.02it/s +epoch=10 global_step=4000 loss=5.93234 loss_avg=5.61520 acc=0.40625 acc_top1_avg=0.43594 acc_top5_avg=0.86510 lr=0.01000 gn=4.88108 time=66.42it/s +epoch=10 global_step=4050 loss=5.85392 loss_avg=5.65955 acc=0.39844 acc_top1_avg=0.43075 acc_top5_avg=0.86663 lr=0.01000 gn=5.07444 time=63.63it/s +epoch=10 global_step=4100 loss=5.81330 loss_avg=5.67635 acc=0.44531 acc_top1_avg=0.42915 acc_top5_avg=0.86604 lr=0.01000 gn=5.16280 time=64.97it/s +epoch=10 global_step=4150 loss=5.12536 loss_avg=5.67569 acc=0.50000 acc_top1_avg=0.42946 acc_top5_avg=0.86549 lr=0.01000 gn=6.15090 time=65.80it/s +epoch=10 global_step=4200 loss=5.96537 loss_avg=5.68308 acc=0.39844 acc_top1_avg=0.42880 acc_top5_avg=0.86546 lr=0.01000 gn=5.85020 time=60.38it/s +epoch=10 global_step=4250 loss=5.53699 loss_avg=5.69871 acc=0.44531 acc_top1_avg=0.42714 acc_top5_avg=0.86553 lr=0.01000 gn=5.11289 time=65.82it/s +epoch=10 global_step=4300 loss=5.27079 loss_avg=5.69880 acc=0.49219 acc_top1_avg=0.42682 acc_top5_avg=0.86542 lr=0.01000 gn=6.38888 time=65.88it/s +====================Eval==================== +epoch=10 global_step=4301 loss=1.27560 test_loss_avg=6.23664 acc=0.46094 test_acc_avg=0.08672 test_acc_top5_avg=0.99062 time=256.80it/s +epoch=10 global_step=4301 loss=2.56309 test_loss_avg=5.20470 acc=0.39844 test_acc_avg=0.15313 test_acc_top5_avg=0.76549 time=257.04it/s +epoch=10 global_step=4301 loss=6.03859 test_loss_avg=4.71730 acc=0.00000 test_acc_avg=0.21766 test_acc_top5_avg=0.77453 time=905.12it/s +curr_acc 0.2177 +BEST_ACC 0.2408 +curr_acc_top5 0.7745 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=11 global_step=4350 loss=6.59252 loss_avg=5.63846 acc=0.31250 acc_top1_avg=0.43670 acc_top5_avg=0.86703 lr=0.01000 gn=5.68729 time=56.12it/s +epoch=11 global_step=4400 loss=6.42361 loss_avg=5.70258 acc=0.33594 acc_top1_avg=0.42677 acc_top5_avg=0.86293 lr=0.01000 gn=4.92372 time=62.90it/s +epoch=11 global_step=4450 loss=4.85942 loss_avg=5.70451 acc=0.52344 acc_top1_avg=0.42743 acc_top5_avg=0.86336 lr=0.01000 gn=5.60697 time=48.48it/s +epoch=11 global_step=4500 loss=5.02731 loss_avg=5.70801 acc=0.47656 acc_top1_avg=0.42635 acc_top5_avg=0.86397 lr=0.01000 gn=5.30262 time=60.17it/s +epoch=11 global_step=4550 loss=5.61216 loss_avg=5.69191 acc=0.42969 acc_top1_avg=0.42824 acc_top5_avg=0.86345 lr=0.01000 gn=6.85721 time=60.85it/s +epoch=11 global_step=4600 loss=5.15089 loss_avg=5.68500 acc=0.49219 acc_top1_avg=0.42849 acc_top5_avg=0.86403 lr=0.01000 gn=5.54175 time=60.94it/s +epoch=11 global_step=4650 loss=5.32912 loss_avg=5.67312 acc=0.46094 acc_top1_avg=0.42993 acc_top5_avg=0.86437 lr=0.01000 gn=5.38028 time=48.09it/s +====================Eval==================== +epoch=11 global_step=4692 loss=6.27284 test_loss_avg=5.19847 acc=0.00000 test_acc_avg=0.05872 test_acc_top5_avg=0.76789 time=241.70it/s +epoch=11 global_step=4692 loss=5.15380 test_loss_avg=4.45683 acc=0.00000 test_acc_avg=0.20985 test_acc_top5_avg=0.79559 time=905.90it/s +curr_acc 0.2098 +BEST_ACC 0.2408 +curr_acc_top5 0.7956 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=12 global_step=4700 loss=5.32636 loss_avg=5.54170 acc=0.48438 acc_top1_avg=0.44727 acc_top5_avg=0.86719 lr=0.01000 gn=5.48536 time=64.62it/s +epoch=12 global_step=4750 loss=5.70147 loss_avg=5.60093 acc=0.43750 acc_top1_avg=0.44046 acc_top5_avg=0.86274 lr=0.01000 gn=4.71899 time=63.08it/s +epoch=12 global_step=4800 loss=5.59446 loss_avg=5.61575 acc=0.42969 acc_top1_avg=0.43605 acc_top5_avg=0.86437 lr=0.01000 gn=5.72530 time=61.69it/s +epoch=12 global_step=4850 loss=5.14814 loss_avg=5.64324 acc=0.49219 acc_top1_avg=0.43359 acc_top5_avg=0.86556 lr=0.01000 gn=5.39789 time=61.50it/s +epoch=12 global_step=4900 loss=5.60045 loss_avg=5.64691 acc=0.42969 acc_top1_avg=0.43292 acc_top5_avg=0.86711 lr=0.01000 gn=5.74234 time=56.62it/s +epoch=12 global_step=4950 loss=6.12316 loss_avg=5.66244 acc=0.39062 acc_top1_avg=0.43160 acc_top5_avg=0.86631 lr=0.01000 gn=5.82488 time=58.20it/s +epoch=12 global_step=5000 loss=5.87298 loss_avg=5.67533 acc=0.39062 acc_top1_avg=0.43022 acc_top5_avg=0.86638 lr=0.01000 gn=5.94590 time=56.18it/s +epoch=12 global_step=5050 loss=5.84920 loss_avg=5.67212 acc=0.40625 acc_top1_avg=0.43078 acc_top5_avg=0.86688 lr=0.01000 gn=4.78200 time=62.86it/s +====================Eval==================== +epoch=12 global_step=5083 loss=7.11108 test_loss_avg=7.13571 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.90234 time=232.63it/s +epoch=12 global_step=5083 loss=9.58411 test_loss_avg=5.92116 acc=0.00000 test_acc_avg=0.08924 test_acc_top5_avg=0.78125 time=259.48it/s +epoch=12 global_step=5083 loss=6.68970 test_loss_avg=5.10501 acc=0.00000 test_acc_avg=0.20599 test_acc_top5_avg=0.78125 time=907.66it/s +curr_acc 0.2060 +BEST_ACC 0.2408 +curr_acc_top5 0.7812 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=13 global_step=5100 loss=5.60049 loss_avg=5.59063 acc=0.43750 acc_top1_avg=0.44347 acc_top5_avg=0.86535 lr=0.01000 gn=5.52543 time=61.47it/s +epoch=13 global_step=5150 loss=6.42004 loss_avg=5.64244 acc=0.35156 acc_top1_avg=0.43762 acc_top5_avg=0.86567 lr=0.01000 gn=4.73587 time=59.63it/s +epoch=13 global_step=5200 loss=5.94593 loss_avg=5.65888 acc=0.40625 acc_top1_avg=0.43483 acc_top5_avg=0.86278 lr=0.01000 gn=5.02783 time=60.15it/s +epoch=13 global_step=5250 loss=4.61426 loss_avg=5.64158 acc=0.53906 acc_top1_avg=0.43582 acc_top5_avg=0.86111 lr=0.01000 gn=4.93269 time=64.49it/s +epoch=13 global_step=5300 loss=5.07783 loss_avg=5.63730 acc=0.50000 acc_top1_avg=0.43649 acc_top5_avg=0.86100 lr=0.01000 gn=5.74314 time=58.72it/s +epoch=13 global_step=5350 loss=5.21156 loss_avg=5.64795 acc=0.49219 acc_top1_avg=0.43487 acc_top5_avg=0.86119 lr=0.01000 gn=5.48409 time=57.83it/s +epoch=13 global_step=5400 loss=5.25379 loss_avg=5.64522 acc=0.45312 acc_top1_avg=0.43484 acc_top5_avg=0.86260 lr=0.01000 gn=4.55550 time=65.51it/s +epoch=13 global_step=5450 loss=5.38333 loss_avg=5.64404 acc=0.47656 acc_top1_avg=0.43458 acc_top5_avg=0.86338 lr=0.01000 gn=5.74092 time=61.83it/s +====================Eval==================== +epoch=13 global_step=5474 loss=5.19618 test_loss_avg=4.97226 acc=0.00000 test_acc_avg=0.04823 test_acc_top5_avg=0.92833 time=230.90it/s +epoch=13 global_step=5474 loss=5.82553 test_loss_avg=4.63264 acc=0.01562 test_acc_avg=0.20837 test_acc_top5_avg=0.82427 time=262.88it/s +epoch=13 global_step=5474 loss=6.21368 test_loss_avg=4.73154 acc=0.00000 test_acc_avg=0.19403 test_acc_top5_avg=0.82783 time=922.84it/s +curr_acc 0.1940 +BEST_ACC 0.2408 +curr_acc_top5 0.8278 +BEST_ACC_top5 0.8183 +Model Saved! + +====================Training==================== +epoch=14 global_step=5500 loss=4.82790 loss_avg=5.64425 acc=0.53906 acc_top1_avg=0.43480 acc_top5_avg=0.85998 lr=0.01000 gn=5.97102 time=65.49it/s +epoch=14 global_step=5550 loss=5.75122 loss_avg=5.58823 acc=0.42188 acc_top1_avg=0.43925 acc_top5_avg=0.86739 lr=0.01000 gn=5.43237 time=55.94it/s +epoch=14 global_step=5600 loss=5.60032 loss_avg=5.57628 acc=0.42969 acc_top1_avg=0.44085 acc_top5_avg=0.86967 lr=0.01000 gn=5.64964 time=58.01it/s +epoch=14 global_step=5650 loss=5.96314 loss_avg=5.60711 acc=0.39062 acc_top1_avg=0.43777 acc_top5_avg=0.86932 lr=0.01000 gn=4.96141 time=58.84it/s +epoch=14 global_step=5700 loss=5.63067 loss_avg=5.64394 acc=0.42969 acc_top1_avg=0.43328 acc_top5_avg=0.86812 lr=0.01000 gn=4.86203 time=62.99it/s +epoch=14 global_step=5750 loss=5.41214 loss_avg=5.64586 acc=0.45312 acc_top1_avg=0.43342 acc_top5_avg=0.86877 lr=0.01000 gn=4.16152 time=55.68it/s +epoch=14 global_step=5800 loss=5.31844 loss_avg=5.64312 acc=0.45312 acc_top1_avg=0.43386 acc_top5_avg=0.86870 lr=0.01000 gn=6.20619 time=54.60it/s +epoch=14 global_step=5850 loss=5.87544 loss_avg=5.64469 acc=0.39844 acc_top1_avg=0.43380 acc_top5_avg=0.86852 lr=0.01000 gn=5.51544 time=58.29it/s +====================Eval==================== +epoch=14 global_step=5865 loss=2.89876 test_loss_avg=4.86845 acc=0.23438 test_acc_avg=0.15341 test_acc_top5_avg=0.71058 time=262.26it/s +epoch=14 global_step=5865 loss=5.36515 test_loss_avg=4.46636 acc=0.00000 test_acc_avg=0.22992 test_acc_top5_avg=0.79955 time=916.19it/s +curr_acc 0.2299 +BEST_ACC 0.2408 +curr_acc_top5 0.7995 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=15 global_step=5900 loss=6.32813 loss_avg=5.51308 acc=0.35938 acc_top1_avg=0.44665 acc_top5_avg=0.87835 lr=0.01000 gn=6.06502 time=63.81it/s +epoch=15 global_step=5950 loss=5.59870 loss_avg=5.64109 acc=0.42969 acc_top1_avg=0.43364 acc_top5_avg=0.86939 lr=0.01000 gn=5.44828 time=51.94it/s +epoch=15 global_step=6000 loss=5.27035 loss_avg=5.64123 acc=0.47656 acc_top1_avg=0.43339 acc_top5_avg=0.86649 lr=0.01000 gn=4.85795 time=61.09it/s +epoch=15 global_step=6050 loss=6.03216 loss_avg=5.64465 acc=0.39844 acc_top1_avg=0.43281 acc_top5_avg=0.86723 lr=0.01000 gn=4.84728 time=58.47it/s +epoch=15 global_step=6100 loss=5.85690 loss_avg=5.63942 acc=0.40625 acc_top1_avg=0.43388 acc_top5_avg=0.86659 lr=0.01000 gn=5.56842 time=61.59it/s +epoch=15 global_step=6150 loss=5.21837 loss_avg=5.63170 acc=0.46875 acc_top1_avg=0.43465 acc_top5_avg=0.86678 lr=0.01000 gn=5.14244 time=56.19it/s +epoch=15 global_step=6200 loss=5.50855 loss_avg=5.63110 acc=0.42188 acc_top1_avg=0.43479 acc_top5_avg=0.86740 lr=0.01000 gn=3.68075 time=61.00it/s +epoch=15 global_step=6250 loss=4.99929 loss_avg=5.64020 acc=0.52344 acc_top1_avg=0.43379 acc_top5_avg=0.86727 lr=0.01000 gn=5.22828 time=61.77it/s +====================Eval==================== +epoch=15 global_step=6256 loss=3.19062 test_loss_avg=5.26705 acc=0.10156 test_acc_avg=0.04948 test_acc_top5_avg=0.97813 time=262.06it/s +epoch=15 global_step=6256 loss=0.17256 test_loss_avg=4.95923 acc=0.94531 test_acc_avg=0.14315 test_acc_top5_avg=0.79387 time=262.65it/s +epoch=15 global_step=6256 loss=5.98548 test_loss_avg=4.73286 acc=0.00000 test_acc_avg=0.18176 test_acc_top5_avg=0.80370 time=910.82it/s +curr_acc 0.1818 +BEST_ACC 0.2408 +curr_acc_top5 0.8037 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=16 global_step=6300 loss=6.33974 loss_avg=5.74213 acc=0.36719 acc_top1_avg=0.42596 acc_top5_avg=0.85423 lr=0.01000 gn=6.10312 time=56.26it/s +epoch=16 global_step=6350 loss=6.08297 loss_avg=5.67095 acc=0.38281 acc_top1_avg=0.43301 acc_top5_avg=0.86104 lr=0.01000 gn=5.06976 time=64.71it/s +epoch=16 global_step=6400 loss=5.68387 loss_avg=5.64249 acc=0.42188 acc_top1_avg=0.43533 acc_top5_avg=0.86589 lr=0.01000 gn=4.94100 time=65.96it/s +epoch=16 global_step=6450 loss=5.48399 loss_avg=5.63322 acc=0.43750 acc_top1_avg=0.43569 acc_top5_avg=0.86674 lr=0.01000 gn=6.08917 time=61.84it/s +epoch=16 global_step=6500 loss=5.68339 loss_avg=5.62836 acc=0.42188 acc_top1_avg=0.43567 acc_top5_avg=0.86587 lr=0.01000 gn=5.54263 time=56.96it/s +epoch=16 global_step=6550 loss=5.30138 loss_avg=5.63231 acc=0.46875 acc_top1_avg=0.43508 acc_top5_avg=0.86527 lr=0.01000 gn=4.77906 time=65.96it/s +epoch=16 global_step=6600 loss=5.23595 loss_avg=5.63336 acc=0.46094 acc_top1_avg=0.43489 acc_top5_avg=0.86510 lr=0.01000 gn=6.23833 time=62.21it/s +====================Eval==================== +epoch=16 global_step=6647 loss=7.09562 test_loss_avg=5.67167 acc=0.00000 test_acc_avg=0.08898 test_acc_top5_avg=0.69878 time=256.82it/s +epoch=16 global_step=6647 loss=6.43991 test_loss_avg=4.90553 acc=0.00000 test_acc_avg=0.21529 test_acc_top5_avg=0.73586 time=914.99it/s +curr_acc 0.2153 +BEST_ACC 0.2408 +curr_acc_top5 0.7359 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=17 global_step=6650 loss=5.43327 loss_avg=5.85187 acc=0.47656 acc_top1_avg=0.41667 acc_top5_avg=0.85677 lr=0.01000 gn=5.74435 time=59.73it/s +epoch=17 global_step=6700 loss=5.82609 loss_avg=5.67302 acc=0.42969 acc_top1_avg=0.42925 acc_top5_avg=0.85584 lr=0.01000 gn=6.96572 time=61.68it/s +epoch=17 global_step=6750 loss=5.27979 loss_avg=5.64712 acc=0.49219 acc_top1_avg=0.43348 acc_top5_avg=0.86438 lr=0.01000 gn=6.18603 time=58.96it/s +epoch=17 global_step=6800 loss=5.35955 loss_avg=5.60100 acc=0.44531 acc_top1_avg=0.43827 acc_top5_avg=0.86770 lr=0.01000 gn=5.83901 time=60.71it/s +epoch=17 global_step=6850 loss=5.71769 loss_avg=5.62562 acc=0.40625 acc_top1_avg=0.43461 acc_top5_avg=0.86757 lr=0.01000 gn=4.31669 time=59.37it/s +epoch=17 global_step=6900 loss=5.43284 loss_avg=5.62545 acc=0.43750 acc_top1_avg=0.43450 acc_top5_avg=0.86676 lr=0.01000 gn=4.67170 time=64.26it/s +epoch=17 global_step=6950 loss=5.70772 loss_avg=5.62696 acc=0.41406 acc_top1_avg=0.43497 acc_top5_avg=0.86794 lr=0.01000 gn=5.93252 time=65.40it/s +epoch=17 global_step=7000 loss=5.41491 loss_avg=5.63213 acc=0.45312 acc_top1_avg=0.43434 acc_top5_avg=0.86703 lr=0.01000 gn=5.83591 time=62.93it/s +====================Eval==================== +epoch=17 global_step=7038 loss=6.44800 test_loss_avg=6.65522 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.81920 time=256.28it/s +epoch=17 global_step=7038 loss=1.29302 test_loss_avg=5.24768 acc=0.67969 test_acc_avg=0.11911 test_acc_top5_avg=0.74342 time=259.34it/s +epoch=17 global_step=7038 loss=5.01397 test_loss_avg=4.49802 acc=0.00000 test_acc_avg=0.21677 test_acc_top5_avg=0.79727 time=918.80it/s +curr_acc 0.2168 +BEST_ACC 0.2408 +curr_acc_top5 0.7973 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=18 global_step=7050 loss=5.57289 loss_avg=5.46415 acc=0.43750 acc_top1_avg=0.44857 acc_top5_avg=0.87565 lr=0.01000 gn=4.64297 time=62.87it/s +epoch=18 global_step=7100 loss=5.36657 loss_avg=5.58157 acc=0.46094 acc_top1_avg=0.43914 acc_top5_avg=0.87223 lr=0.01000 gn=6.47142 time=61.63it/s +epoch=18 global_step=7150 loss=5.86860 loss_avg=5.60293 acc=0.40625 acc_top1_avg=0.43785 acc_top5_avg=0.86914 lr=0.01000 gn=4.58147 time=59.51it/s +epoch=18 global_step=7200 loss=5.36519 loss_avg=5.61137 acc=0.48438 acc_top1_avg=0.43740 acc_top5_avg=0.86907 lr=0.01000 gn=5.61931 time=54.38it/s +epoch=18 global_step=7250 loss=6.41046 loss_avg=5.59851 acc=0.35156 acc_top1_avg=0.43901 acc_top5_avg=0.86785 lr=0.01000 gn=5.47825 time=61.68it/s +epoch=18 global_step=7300 loss=5.17969 loss_avg=5.58780 acc=0.46094 acc_top1_avg=0.43992 acc_top5_avg=0.86978 lr=0.01000 gn=5.58966 time=56.22it/s +epoch=18 global_step=7350 loss=5.54466 loss_avg=5.60259 acc=0.42969 acc_top1_avg=0.43815 acc_top5_avg=0.86871 lr=0.01000 gn=6.65161 time=59.35it/s +epoch=18 global_step=7400 loss=5.93069 loss_avg=5.61126 acc=0.40625 acc_top1_avg=0.43713 acc_top5_avg=0.86814 lr=0.01000 gn=5.99897 time=61.57it/s +====================Eval==================== +epoch=18 global_step=7429 loss=6.33038 test_loss_avg=5.29280 acc=0.00000 test_acc_avg=0.02985 test_acc_top5_avg=0.82533 time=248.27it/s +epoch=18 global_step=7429 loss=5.31947 test_loss_avg=4.35781 acc=0.01562 test_acc_avg=0.22035 test_acc_top5_avg=0.78015 time=263.93it/s +epoch=18 global_step=7429 loss=5.38141 test_loss_avg=4.37077 acc=0.00000 test_acc_avg=0.21756 test_acc_top5_avg=0.78135 time=891.08it/s +curr_acc 0.2176 +BEST_ACC 0.2408 +curr_acc_top5 0.7813 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=19 global_step=7450 loss=5.12570 loss_avg=5.61554 acc=0.50781 acc_top1_avg=0.43750 acc_top5_avg=0.85677 lr=0.01000 gn=5.87513 time=65.09it/s +epoch=19 global_step=7500 loss=5.81090 loss_avg=5.63313 acc=0.41406 acc_top1_avg=0.43354 acc_top5_avg=0.86323 lr=0.01000 gn=6.50967 time=58.88it/s +epoch=19 global_step=7550 loss=5.36730 loss_avg=5.60668 acc=0.48438 acc_top1_avg=0.43647 acc_top5_avg=0.86661 lr=0.01000 gn=5.25735 time=51.31it/s +epoch=19 global_step=7600 loss=6.61636 loss_avg=5.59843 acc=0.30469 acc_top1_avg=0.43741 acc_top5_avg=0.86700 lr=0.01000 gn=6.21367 time=57.64it/s +epoch=19 global_step=7650 loss=5.60207 loss_avg=5.60654 acc=0.44531 acc_top1_avg=0.43722 acc_top5_avg=0.86839 lr=0.01000 gn=5.41834 time=63.12it/s +epoch=19 global_step=7700 loss=5.81244 loss_avg=5.59970 acc=0.42969 acc_top1_avg=0.43785 acc_top5_avg=0.86750 lr=0.01000 gn=6.85684 time=56.61it/s +epoch=19 global_step=7750 loss=5.42860 loss_avg=5.60224 acc=0.46875 acc_top1_avg=0.43794 acc_top5_avg=0.86711 lr=0.01000 gn=4.91274 time=64.61it/s +epoch=19 global_step=7800 loss=5.46223 loss_avg=5.59880 acc=0.45312 acc_top1_avg=0.43763 acc_top5_avg=0.86738 lr=0.01000 gn=5.01305 time=60.53it/s +====================Eval==================== +epoch=19 global_step=7820 loss=7.43620 test_loss_avg=4.86711 acc=0.00000 test_acc_avg=0.12149 test_acc_top5_avg=0.65673 time=248.96it/s +epoch=19 global_step=7820 loss=5.76172 test_loss_avg=4.30888 acc=0.00000 test_acc_avg=0.24347 test_acc_top5_avg=0.73497 time=898.33it/s +curr_acc 0.2435 +BEST_ACC 0.2408 +curr_acc_top5 0.7350 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=20 global_step=7850 loss=5.82417 loss_avg=5.54791 acc=0.39844 acc_top1_avg=0.44818 acc_top5_avg=0.87813 lr=0.01000 gn=7.09022 time=61.26it/s +epoch=20 global_step=7900 loss=6.81123 loss_avg=5.58595 acc=0.32031 acc_top1_avg=0.44277 acc_top5_avg=0.87061 lr=0.01000 gn=5.83034 time=61.76it/s +epoch=20 global_step=7950 loss=6.48439 loss_avg=5.58686 acc=0.35938 acc_top1_avg=0.43996 acc_top5_avg=0.86905 lr=0.01000 gn=5.71444 time=56.77it/s +epoch=20 global_step=8000 loss=5.05255 loss_avg=5.58945 acc=0.50781 acc_top1_avg=0.43967 acc_top5_avg=0.86970 lr=0.01000 gn=5.34250 time=52.88it/s +epoch=20 global_step=8050 loss=5.45214 loss_avg=5.59069 acc=0.46094 acc_top1_avg=0.43933 acc_top5_avg=0.87055 lr=0.01000 gn=6.19778 time=61.67it/s +epoch=20 global_step=8100 loss=5.57305 loss_avg=5.58981 acc=0.43750 acc_top1_avg=0.43993 acc_top5_avg=0.86964 lr=0.01000 gn=5.61022 time=62.46it/s +epoch=20 global_step=8150 loss=5.76571 loss_avg=5.58993 acc=0.45312 acc_top1_avg=0.44006 acc_top5_avg=0.86955 lr=0.01000 gn=6.12483 time=61.60it/s +epoch=20 global_step=8200 loss=5.91187 loss_avg=5.60179 acc=0.41406 acc_top1_avg=0.43873 acc_top5_avg=0.86846 lr=0.01000 gn=5.59539 time=61.78it/s +====================Eval==================== +epoch=20 global_step=8211 loss=5.53548 test_loss_avg=4.63617 acc=0.00000 test_acc_avg=0.21328 test_acc_top5_avg=0.91836 time=232.20it/s +epoch=20 global_step=8211 loss=1.23620 test_loss_avg=4.63116 acc=0.61719 test_acc_avg=0.25056 test_acc_top5_avg=0.81038 time=262.59it/s +epoch=20 global_step=8211 loss=4.92608 test_loss_avg=4.68212 acc=0.00000 test_acc_avg=0.22854 test_acc_top5_avg=0.80489 time=882.45it/s +curr_acc 0.2285 +BEST_ACC 0.2435 +curr_acc_top5 0.8049 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=21 global_step=8250 loss=5.34703 loss_avg=5.54281 acc=0.44531 acc_top1_avg=0.44251 acc_top5_avg=0.86879 lr=0.01000 gn=5.24201 time=58.39it/s +epoch=21 global_step=8300 loss=5.82615 loss_avg=5.56251 acc=0.42188 acc_top1_avg=0.44127 acc_top5_avg=0.86692 lr=0.01000 gn=5.86515 time=60.35it/s +epoch=21 global_step=8350 loss=5.59405 loss_avg=5.55782 acc=0.44531 acc_top1_avg=0.44228 acc_top5_avg=0.86589 lr=0.01000 gn=5.83647 time=60.71it/s +epoch=21 global_step=8400 loss=5.88925 loss_avg=5.56087 acc=0.40625 acc_top1_avg=0.44267 acc_top5_avg=0.86739 lr=0.01000 gn=5.92833 time=61.65it/s +epoch=21 global_step=8450 loss=5.96945 loss_avg=5.56698 acc=0.39062 acc_top1_avg=0.44208 acc_top5_avg=0.86804 lr=0.01000 gn=5.87842 time=56.48it/s +epoch=21 global_step=8500 loss=5.46630 loss_avg=5.56614 acc=0.46094 acc_top1_avg=0.44245 acc_top5_avg=0.86840 lr=0.01000 gn=3.73063 time=62.92it/s +epoch=21 global_step=8550 loss=5.12628 loss_avg=5.56204 acc=0.48438 acc_top1_avg=0.44294 acc_top5_avg=0.86804 lr=0.01000 gn=5.63883 time=56.54it/s +epoch=21 global_step=8600 loss=6.11363 loss_avg=5.57388 acc=0.39844 acc_top1_avg=0.44144 acc_top5_avg=0.86717 lr=0.01000 gn=6.08439 time=62.04it/s +====================Eval==================== +epoch=21 global_step=8602 loss=4.04004 test_loss_avg=6.07052 acc=0.20312 test_acc_avg=0.03811 test_acc_top5_avg=0.64977 time=250.68it/s +epoch=21 global_step=8602 loss=2.30065 test_loss_avg=5.17252 acc=0.43750 test_acc_avg=0.23269 test_acc_top5_avg=0.80202 time=912.40it/s +curr_acc 0.2327 +BEST_ACC 0.2435 +curr_acc_top5 0.8020 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=22 global_step=8650 loss=4.94279 loss_avg=5.54271 acc=0.50781 acc_top1_avg=0.44320 acc_top5_avg=0.87093 lr=0.01000 gn=4.97382 time=64.50it/s +epoch=22 global_step=8700 loss=6.13614 loss_avg=5.61885 acc=0.37500 acc_top1_avg=0.43519 acc_top5_avg=0.86695 lr=0.01000 gn=4.51442 time=63.35it/s +epoch=22 global_step=8750 loss=6.11847 loss_avg=5.61388 acc=0.38281 acc_top1_avg=0.43586 acc_top5_avg=0.86824 lr=0.01000 gn=5.13623 time=61.28it/s +epoch=22 global_step=8800 loss=5.28075 loss_avg=5.58242 acc=0.47656 acc_top1_avg=0.43979 acc_top5_avg=0.86790 lr=0.01000 gn=5.37191 time=57.66it/s +epoch=22 global_step=8850 loss=5.86723 loss_avg=5.58303 acc=0.39062 acc_top1_avg=0.43945 acc_top5_avg=0.86879 lr=0.01000 gn=4.49509 time=63.12it/s +epoch=22 global_step=8900 loss=5.77255 loss_avg=5.56285 acc=0.41406 acc_top1_avg=0.44107 acc_top5_avg=0.86952 lr=0.01000 gn=5.92255 time=58.16it/s +epoch=22 global_step=8950 loss=5.57968 loss_avg=5.57886 acc=0.43750 acc_top1_avg=0.43966 acc_top5_avg=0.86997 lr=0.01000 gn=5.71286 time=61.43it/s +====================Eval==================== +epoch=22 global_step=8993 loss=2.50589 test_loss_avg=5.55867 acc=0.10156 test_acc_avg=0.03060 test_acc_top5_avg=0.95964 time=259.79it/s +epoch=22 global_step=8993 loss=1.41225 test_loss_avg=4.75663 acc=0.58594 test_acc_avg=0.15625 test_acc_top5_avg=0.80557 time=259.98it/s +epoch=22 global_step=8993 loss=4.98132 test_loss_avg=4.31371 acc=0.00000 test_acc_avg=0.21826 test_acc_top5_avg=0.81299 time=884.13it/s +curr_acc 0.2183 +BEST_ACC 0.2435 +curr_acc_top5 0.8130 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=23 global_step=9000 loss=5.08656 loss_avg=5.46725 acc=0.50781 acc_top1_avg=0.45536 acc_top5_avg=0.85938 lr=0.01000 gn=5.66111 time=62.84it/s +epoch=23 global_step=9050 loss=4.85383 loss_avg=5.53125 acc=0.52344 acc_top1_avg=0.44668 acc_top5_avg=0.87185 lr=0.01000 gn=4.74057 time=62.80it/s +epoch=23 global_step=9100 loss=5.44859 loss_avg=5.57410 acc=0.46875 acc_top1_avg=0.44152 acc_top5_avg=0.87390 lr=0.01000 gn=6.86565 time=61.70it/s +epoch=23 global_step=9150 loss=6.27209 loss_avg=5.57514 acc=0.36719 acc_top1_avg=0.44118 acc_top5_avg=0.87107 lr=0.01000 gn=5.81955 time=61.07it/s +epoch=23 global_step=9200 loss=5.74232 loss_avg=5.56628 acc=0.42188 acc_top1_avg=0.44203 acc_top5_avg=0.86911 lr=0.01000 gn=5.66522 time=56.17it/s +epoch=23 global_step=9250 loss=5.59451 loss_avg=5.56376 acc=0.42969 acc_top1_avg=0.44221 acc_top5_avg=0.86959 lr=0.01000 gn=5.06707 time=64.28it/s +epoch=23 global_step=9300 loss=5.32762 loss_avg=5.57292 acc=0.47656 acc_top1_avg=0.44155 acc_top5_avg=0.87057 lr=0.01000 gn=7.27110 time=65.42it/s +epoch=23 global_step=9350 loss=5.98434 loss_avg=5.56958 acc=0.42188 acc_top1_avg=0.44194 acc_top5_avg=0.87029 lr=0.01000 gn=8.00241 time=65.61it/s +====================Eval==================== +epoch=23 global_step=9384 loss=7.23166 test_loss_avg=5.83204 acc=0.00000 test_acc_avg=0.03243 test_acc_top5_avg=0.69508 time=246.85it/s +epoch=23 global_step=9384 loss=5.62985 test_loss_avg=4.85589 acc=0.00000 test_acc_avg=0.22587 test_acc_top5_avg=0.77670 time=917.59it/s +curr_acc 0.2259 +BEST_ACC 0.2435 +curr_acc_top5 0.7767 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=24 global_step=9400 loss=5.38399 loss_avg=5.45712 acc=0.42969 acc_top1_avg=0.45898 acc_top5_avg=0.86572 lr=0.01000 gn=6.21626 time=60.13it/s +epoch=24 global_step=9450 loss=5.54977 loss_avg=5.48521 acc=0.46094 acc_top1_avg=0.45170 acc_top5_avg=0.87275 lr=0.01000 gn=5.81871 time=64.75it/s +epoch=24 global_step=9500 loss=4.98299 loss_avg=5.50184 acc=0.49219 acc_top1_avg=0.44982 acc_top5_avg=0.87675 lr=0.01000 gn=4.49079 time=61.84it/s +epoch=24 global_step=9550 loss=5.64852 loss_avg=5.51786 acc=0.40625 acc_top1_avg=0.44658 acc_top5_avg=0.87481 lr=0.01000 gn=7.28652 time=54.12it/s +epoch=24 global_step=9600 loss=5.63666 loss_avg=5.54605 acc=0.43750 acc_top1_avg=0.44383 acc_top5_avg=0.87366 lr=0.01000 gn=5.29319 time=64.56it/s +epoch=24 global_step=9650 loss=5.38954 loss_avg=5.54351 acc=0.44531 acc_top1_avg=0.44428 acc_top5_avg=0.87197 lr=0.01000 gn=6.16933 time=64.57it/s +epoch=24 global_step=9700 loss=5.07880 loss_avg=5.54914 acc=0.48438 acc_top1_avg=0.44373 acc_top5_avg=0.87201 lr=0.01000 gn=5.13076 time=60.31it/s +epoch=24 global_step=9750 loss=5.53297 loss_avg=5.54444 acc=0.43750 acc_top1_avg=0.44427 acc_top5_avg=0.87124 lr=0.01000 gn=5.52096 time=61.59it/s +====================Eval==================== +epoch=24 global_step=9775 loss=5.92499 test_loss_avg=6.04579 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.67383 time=235.69it/s +epoch=24 global_step=9775 loss=8.80060 test_loss_avg=5.62095 acc=0.00000 test_acc_avg=0.08984 test_acc_top5_avg=0.71571 time=262.31it/s +epoch=24 global_step=9775 loss=5.97936 test_loss_avg=4.75159 acc=0.00000 test_acc_avg=0.21608 test_acc_top5_avg=0.78817 time=897.18it/s +curr_acc 0.2161 +BEST_ACC 0.2435 +curr_acc_top5 0.7882 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=25 global_step=9800 loss=5.75883 loss_avg=5.41911 acc=0.43750 acc_top1_avg=0.45781 acc_top5_avg=0.87844 lr=0.01000 gn=6.08876 time=61.64it/s +epoch=25 global_step=9850 loss=5.26557 loss_avg=5.52878 acc=0.46094 acc_top1_avg=0.44812 acc_top5_avg=0.87052 lr=0.01000 gn=5.27377 time=53.96it/s +epoch=25 global_step=9900 loss=6.11360 loss_avg=5.54385 acc=0.37500 acc_top1_avg=0.44538 acc_top5_avg=0.87144 lr=0.01000 gn=5.96807 time=58.58it/s +epoch=25 global_step=9950 loss=5.48394 loss_avg=5.54113 acc=0.43750 acc_top1_avg=0.44580 acc_top5_avg=0.87103 lr=0.01000 gn=6.02108 time=61.50it/s +epoch=25 global_step=10000 loss=5.06825 loss_avg=5.54487 acc=0.50000 acc_top1_avg=0.44542 acc_top5_avg=0.87021 lr=0.01000 gn=4.16309 time=58.71it/s +epoch=25 global_step=10050 loss=6.12278 loss_avg=5.55589 acc=0.39062 acc_top1_avg=0.44378 acc_top5_avg=0.86994 lr=0.01000 gn=5.10754 time=57.48it/s +epoch=25 global_step=10100 loss=5.99463 loss_avg=5.57698 acc=0.39844 acc_top1_avg=0.44173 acc_top5_avg=0.86851 lr=0.01000 gn=4.92027 time=57.47it/s +epoch=25 global_step=10150 loss=5.76392 loss_avg=5.54696 acc=0.42969 acc_top1_avg=0.44490 acc_top5_avg=0.87010 lr=0.01000 gn=4.27171 time=61.72it/s +====================Eval==================== +epoch=25 global_step=10166 loss=6.79343 test_loss_avg=5.02874 acc=0.00000 test_acc_avg=0.15500 test_acc_top5_avg=0.91656 time=262.88it/s +epoch=25 global_step=10166 loss=5.63139 test_loss_avg=4.53456 acc=0.00781 test_acc_avg=0.24229 test_acc_top5_avg=0.79271 time=263.26it/s +epoch=25 global_step=10166 loss=5.74697 test_loss_avg=4.59685 acc=0.00000 test_acc_avg=0.23012 test_acc_top5_avg=0.78283 time=894.69it/s +curr_acc 0.2301 +BEST_ACC 0.2435 +curr_acc_top5 0.7828 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=26 global_step=10200 loss=5.54999 loss_avg=5.59092 acc=0.45312 acc_top1_avg=0.43980 acc_top5_avg=0.87224 lr=0.01000 gn=6.53549 time=63.01it/s +epoch=26 global_step=10250 loss=6.04690 loss_avg=5.58669 acc=0.39062 acc_top1_avg=0.43927 acc_top5_avg=0.86970 lr=0.01000 gn=5.93654 time=61.42it/s +epoch=26 global_step=10300 loss=5.81658 loss_avg=5.59480 acc=0.42969 acc_top1_avg=0.43907 acc_top5_avg=0.86690 lr=0.01000 gn=6.18986 time=63.16it/s +epoch=26 global_step=10350 loss=6.06014 loss_avg=5.60505 acc=0.38281 acc_top1_avg=0.43814 acc_top5_avg=0.86765 lr=0.01000 gn=7.87969 time=61.41it/s +epoch=26 global_step=10400 loss=6.11671 loss_avg=5.58126 acc=0.37500 acc_top1_avg=0.44037 acc_top5_avg=0.86829 lr=0.01000 gn=6.01013 time=62.10it/s +epoch=26 global_step=10450 loss=5.73178 loss_avg=5.58744 acc=0.42969 acc_top1_avg=0.43904 acc_top5_avg=0.86898 lr=0.01000 gn=4.99140 time=62.82it/s +epoch=26 global_step=10500 loss=5.33132 loss_avg=5.57906 acc=0.48438 acc_top1_avg=0.44017 acc_top5_avg=0.86969 lr=0.01000 gn=5.28137 time=57.31it/s +epoch=26 global_step=10550 loss=5.50576 loss_avg=5.57200 acc=0.44531 acc_top1_avg=0.44076 acc_top5_avg=0.87020 lr=0.01000 gn=5.56486 time=54.77it/s +====================Eval==================== +epoch=26 global_step=10557 loss=2.24439 test_loss_avg=4.70614 acc=0.40625 test_acc_avg=0.18852 test_acc_top5_avg=0.68988 time=257.68it/s +epoch=26 global_step=10557 loss=7.35540 test_loss_avg=4.41467 acc=0.00000 test_acc_avg=0.25999 test_acc_top5_avg=0.79203 time=924.26it/s +curr_acc 0.2600 +BEST_ACC 0.2435 +curr_acc_top5 0.7920 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=27 global_step=10600 loss=6.04170 loss_avg=5.55860 acc=0.38281 acc_top1_avg=0.44386 acc_top5_avg=0.86955 lr=0.01000 gn=6.54733 time=61.57it/s +epoch=27 global_step=10650 loss=4.76695 loss_avg=5.52371 acc=0.52344 acc_top1_avg=0.44682 acc_top5_avg=0.87290 lr=0.01000 gn=5.32024 time=52.61it/s +epoch=27 global_step=10700 loss=5.94790 loss_avg=5.52401 acc=0.42188 acc_top1_avg=0.44591 acc_top5_avg=0.87243 lr=0.01000 gn=4.80914 time=66.38it/s +epoch=27 global_step=10750 loss=5.58966 loss_avg=5.53759 acc=0.41406 acc_top1_avg=0.44406 acc_top5_avg=0.87043 lr=0.01000 gn=6.23910 time=56.17it/s +epoch=27 global_step=10800 loss=6.15248 loss_avg=5.55498 acc=0.37500 acc_top1_avg=0.44229 acc_top5_avg=0.87076 lr=0.01000 gn=5.13465 time=62.94it/s +epoch=27 global_step=10850 loss=5.80065 loss_avg=5.55534 acc=0.42969 acc_top1_avg=0.44254 acc_top5_avg=0.87153 lr=0.01000 gn=7.42216 time=65.15it/s +epoch=27 global_step=10900 loss=5.74817 loss_avg=5.54831 acc=0.41406 acc_top1_avg=0.44335 acc_top5_avg=0.87097 lr=0.01000 gn=6.28683 time=64.57it/s +====================Eval==================== +epoch=27 global_step=10948 loss=5.24047 test_loss_avg=3.62755 acc=0.00000 test_acc_avg=0.42647 test_acc_top5_avg=0.95221 time=256.80it/s +epoch=27 global_step=10948 loss=0.67483 test_loss_avg=4.40141 acc=0.84375 test_acc_avg=0.26796 test_acc_top5_avg=0.78825 time=255.08it/s +epoch=27 global_step=10948 loss=7.09313 test_loss_avg=4.50486 acc=0.00000 test_acc_avg=0.26434 test_acc_top5_avg=0.80667 time=893.93it/s +curr_acc 0.2643 +BEST_ACC 0.2600 +curr_acc_top5 0.8067 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=28 global_step=10950 loss=5.44746 loss_avg=5.28594 acc=0.46094 acc_top1_avg=0.48438 acc_top5_avg=0.87500 lr=0.01000 gn=6.13632 time=57.76it/s +epoch=28 global_step=11000 loss=5.63389 loss_avg=5.50244 acc=0.44531 acc_top1_avg=0.44967 acc_top5_avg=0.86538 lr=0.01000 gn=5.92073 time=64.67it/s +epoch=28 global_step=11050 loss=6.05373 loss_avg=5.51016 acc=0.39844 acc_top1_avg=0.44891 acc_top5_avg=0.86642 lr=0.01000 gn=6.08013 time=64.90it/s +epoch=28 global_step=11100 loss=5.38332 loss_avg=5.51646 acc=0.46094 acc_top1_avg=0.44773 acc_top5_avg=0.86729 lr=0.01000 gn=5.58655 time=60.71it/s +epoch=28 global_step=11150 loss=5.21790 loss_avg=5.52487 acc=0.48438 acc_top1_avg=0.44674 acc_top5_avg=0.86699 lr=0.01000 gn=5.20629 time=54.02it/s +epoch=28 global_step=11200 loss=5.97111 loss_avg=5.52951 acc=0.39844 acc_top1_avg=0.44584 acc_top5_avg=0.86837 lr=0.01000 gn=4.85552 time=65.98it/s +epoch=28 global_step=11250 loss=5.48193 loss_avg=5.52955 acc=0.44531 acc_top1_avg=0.44578 acc_top5_avg=0.86887 lr=0.01000 gn=6.05571 time=61.88it/s +epoch=28 global_step=11300 loss=6.42777 loss_avg=5.53250 acc=0.35156 acc_top1_avg=0.44540 acc_top5_avg=0.86985 lr=0.01000 gn=5.46284 time=56.23it/s +====================Eval==================== +epoch=28 global_step=11339 loss=7.04263 test_loss_avg=5.61708 acc=0.00000 test_acc_avg=0.18627 test_acc_top5_avg=0.74856 time=245.28it/s +epoch=28 global_step=11339 loss=6.61166 test_loss_avg=4.97590 acc=0.06250 test_acc_avg=0.23289 test_acc_top5_avg=0.80320 time=876.92it/s +curr_acc 0.2329 +BEST_ACC 0.2643 +curr_acc_top5 0.8032 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=29 global_step=11350 loss=5.32531 loss_avg=5.69956 acc=0.48438 acc_top1_avg=0.43040 acc_top5_avg=0.86222 lr=0.01000 gn=6.05477 time=59.17it/s +epoch=29 global_step=11400 loss=5.09520 loss_avg=5.50448 acc=0.49219 acc_top1_avg=0.45018 acc_top5_avg=0.86847 lr=0.01000 gn=5.40077 time=57.57it/s +epoch=29 global_step=11450 loss=5.75478 loss_avg=5.52448 acc=0.42188 acc_top1_avg=0.44771 acc_top5_avg=0.87120 lr=0.01000 gn=5.77945 time=59.42it/s +epoch=29 global_step=11500 loss=5.46199 loss_avg=5.52958 acc=0.46094 acc_top1_avg=0.44580 acc_top5_avg=0.87325 lr=0.01000 gn=5.32407 time=58.46it/s +epoch=29 global_step=11550 loss=6.01418 loss_avg=5.54768 acc=0.40625 acc_top1_avg=0.44387 acc_top5_avg=0.87004 lr=0.01000 gn=6.03604 time=62.00it/s +epoch=29 global_step=11600 loss=5.54409 loss_avg=5.55694 acc=0.44531 acc_top1_avg=0.44295 acc_top5_avg=0.87150 lr=0.01000 gn=4.77833 time=57.13it/s +epoch=29 global_step=11650 loss=6.04178 loss_avg=5.54269 acc=0.36719 acc_top1_avg=0.44423 acc_top5_avg=0.87068 lr=0.01000 gn=4.24496 time=59.27it/s +epoch=29 global_step=11700 loss=5.40379 loss_avg=5.55665 acc=0.45312 acc_top1_avg=0.44254 acc_top5_avg=0.87033 lr=0.01000 gn=4.89489 time=60.36it/s +====================Eval==================== +epoch=29 global_step=11730 loss=3.19572 test_loss_avg=6.46851 acc=0.07031 test_acc_avg=0.01215 test_acc_top5_avg=0.92795 time=243.32it/s +epoch=29 global_step=11730 loss=0.64942 test_loss_avg=4.88290 acc=0.79688 test_acc_avg=0.11917 test_acc_top5_avg=0.74378 time=261.51it/s +epoch=29 global_step=11730 loss=5.89242 test_loss_avg=4.31793 acc=0.00000 test_acc_avg=0.21717 test_acc_top5_avg=0.79045 time=897.56it/s +curr_acc 0.2172 +BEST_ACC 0.2643 +curr_acc_top5 0.7904 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=30 global_step=11750 loss=5.20971 loss_avg=5.57052 acc=0.48438 acc_top1_avg=0.44297 acc_top5_avg=0.86953 lr=0.01000 gn=5.20554 time=60.46it/s +epoch=30 global_step=11800 loss=5.50564 loss_avg=5.53872 acc=0.45312 acc_top1_avg=0.44565 acc_top5_avg=0.87455 lr=0.01000 gn=4.19393 time=62.99it/s +epoch=30 global_step=11850 loss=5.28850 loss_avg=5.51832 acc=0.48438 acc_top1_avg=0.44922 acc_top5_avg=0.87318 lr=0.01000 gn=5.22187 time=62.04it/s +epoch=30 global_step=11900 loss=6.04627 loss_avg=5.51025 acc=0.38281 acc_top1_avg=0.44940 acc_top5_avg=0.87523 lr=0.01000 gn=6.46576 time=61.88it/s +epoch=30 global_step=11950 loss=5.18243 loss_avg=5.50304 acc=0.49219 acc_top1_avg=0.44950 acc_top5_avg=0.87411 lr=0.01000 gn=6.16237 time=58.74it/s +epoch=30 global_step=12000 loss=5.85505 loss_avg=5.50285 acc=0.40625 acc_top1_avg=0.44939 acc_top5_avg=0.87387 lr=0.01000 gn=4.76264 time=63.21it/s +epoch=30 global_step=12050 loss=6.01251 loss_avg=5.51003 acc=0.39062 acc_top1_avg=0.44829 acc_top5_avg=0.87278 lr=0.01000 gn=6.00580 time=61.61it/s +epoch=30 global_step=12100 loss=5.59413 loss_avg=5.52943 acc=0.44531 acc_top1_avg=0.44601 acc_top5_avg=0.87171 lr=0.01000 gn=5.46949 time=59.44it/s +====================Eval==================== +epoch=30 global_step=12121 loss=6.53107 test_loss_avg=5.10166 acc=0.00000 test_acc_avg=0.08047 test_acc_top5_avg=0.74870 time=261.21it/s +epoch=30 global_step=12121 loss=5.92545 test_loss_avg=4.35365 acc=0.00000 test_acc_avg=0.21005 test_acc_top5_avg=0.74980 time=896.60it/s +curr_acc 0.2100 +BEST_ACC 0.2643 +curr_acc_top5 0.7498 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=31 global_step=12150 loss=5.62558 loss_avg=5.50474 acc=0.45312 acc_top1_avg=0.45097 acc_top5_avg=0.85911 lr=0.01000 gn=4.99416 time=62.90it/s +epoch=31 global_step=12200 loss=5.37504 loss_avg=5.50956 acc=0.43750 acc_top1_avg=0.44858 acc_top5_avg=0.86748 lr=0.01000 gn=5.03471 time=58.99it/s +epoch=31 global_step=12250 loss=5.90223 loss_avg=5.52982 acc=0.39062 acc_top1_avg=0.44513 acc_top5_avg=0.86652 lr=0.01000 gn=5.39390 time=54.43it/s +epoch=31 global_step=12300 loss=5.82486 loss_avg=5.51805 acc=0.41406 acc_top1_avg=0.44619 acc_top5_avg=0.86959 lr=0.01000 gn=6.16332 time=61.64it/s +epoch=31 global_step=12350 loss=5.83749 loss_avg=5.54693 acc=0.39844 acc_top1_avg=0.44313 acc_top5_avg=0.86985 lr=0.01000 gn=6.34571 time=59.95it/s +epoch=31 global_step=12400 loss=5.55042 loss_avg=5.55558 acc=0.43750 acc_top1_avg=0.44209 acc_top5_avg=0.86957 lr=0.01000 gn=6.35394 time=61.77it/s +epoch=31 global_step=12450 loss=5.54716 loss_avg=5.54458 acc=0.43750 acc_top1_avg=0.44332 acc_top5_avg=0.86899 lr=0.01000 gn=4.64557 time=61.18it/s +epoch=31 global_step=12500 loss=5.56621 loss_avg=5.54270 acc=0.43750 acc_top1_avg=0.44346 acc_top5_avg=0.87018 lr=0.01000 gn=4.58998 time=63.19it/s +====================Eval==================== +epoch=31 global_step=12512 loss=7.60912 test_loss_avg=7.60912 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.96094 time=113.39it/s +epoch=31 global_step=12512 loss=7.09812 test_loss_avg=5.29469 acc=0.00000 test_acc_avg=0.15564 test_acc_top5_avg=0.74249 time=259.32it/s +epoch=31 global_step=12512 loss=5.60350 test_loss_avg=4.59332 acc=0.06250 test_acc_avg=0.24426 test_acc_top5_avg=0.76038 time=923.45it/s +curr_acc 0.2443 +BEST_ACC 0.2643 +curr_acc_top5 0.7604 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=32 global_step=12550 loss=6.06335 loss_avg=5.53647 acc=0.38281 acc_top1_avg=0.44428 acc_top5_avg=0.86698 lr=0.01000 gn=5.15990 time=57.03it/s +epoch=32 global_step=12600 loss=6.08864 loss_avg=5.55771 acc=0.39062 acc_top1_avg=0.44194 acc_top5_avg=0.86808 lr=0.01000 gn=6.13358 time=62.11it/s +epoch=32 global_step=12650 loss=5.26433 loss_avg=5.54191 acc=0.49219 acc_top1_avg=0.44401 acc_top5_avg=0.86838 lr=0.01000 gn=6.44966 time=64.17it/s +epoch=32 global_step=12700 loss=6.14126 loss_avg=5.54455 acc=0.39844 acc_top1_avg=0.44402 acc_top5_avg=0.86835 lr=0.01000 gn=5.55185 time=62.09it/s +epoch=32 global_step=12750 loss=6.29251 loss_avg=5.55698 acc=0.36719 acc_top1_avg=0.44259 acc_top5_avg=0.87090 lr=0.01000 gn=5.56060 time=56.87it/s +epoch=32 global_step=12800 loss=5.06081 loss_avg=5.55584 acc=0.49219 acc_top1_avg=0.44336 acc_top5_avg=0.86990 lr=0.01000 gn=5.63542 time=61.67it/s +epoch=32 global_step=12850 loss=5.74575 loss_avg=5.55911 acc=0.40625 acc_top1_avg=0.44261 acc_top5_avg=0.87096 lr=0.01000 gn=5.43239 time=63.22it/s +epoch=32 global_step=12900 loss=5.07238 loss_avg=5.55551 acc=0.50781 acc_top1_avg=0.44312 acc_top5_avg=0.87148 lr=0.01000 gn=6.17057 time=62.11it/s +====================Eval==================== +epoch=32 global_step=12903 loss=5.12062 test_loss_avg=4.54772 acc=0.00000 test_acc_avg=0.12429 test_acc_top5_avg=0.93253 time=252.68it/s +epoch=32 global_step=12903 loss=5.56779 test_loss_avg=4.51110 acc=0.00781 test_acc_avg=0.22396 test_acc_top5_avg=0.81977 time=262.39it/s +epoch=32 global_step=12903 loss=5.46237 test_loss_avg=4.59293 acc=0.00000 test_acc_avg=0.20461 test_acc_top5_avg=0.81636 time=903.94it/s +curr_acc 0.2046 +BEST_ACC 0.2643 +curr_acc_top5 0.8164 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=33 global_step=12950 loss=4.80758 loss_avg=5.40516 acc=0.53125 acc_top1_avg=0.46210 acc_top5_avg=0.87683 lr=0.01000 gn=6.13326 time=52.03it/s +epoch=33 global_step=13000 loss=5.14630 loss_avg=5.45203 acc=0.50000 acc_top1_avg=0.45490 acc_top5_avg=0.87146 lr=0.01000 gn=5.74587 time=59.79it/s +epoch=33 global_step=13050 loss=5.39644 loss_avg=5.48493 acc=0.46094 acc_top1_avg=0.45015 acc_top5_avg=0.87149 lr=0.01000 gn=4.88652 time=59.48it/s +epoch=33 global_step=13100 loss=6.20604 loss_avg=5.50416 acc=0.37500 acc_top1_avg=0.44852 acc_top5_avg=0.87036 lr=0.01000 gn=6.10956 time=61.63it/s +epoch=33 global_step=13150 loss=5.90540 loss_avg=5.51372 acc=0.41406 acc_top1_avg=0.44727 acc_top5_avg=0.86969 lr=0.01000 gn=4.93670 time=57.71it/s +epoch=33 global_step=13200 loss=5.71325 loss_avg=5.54164 acc=0.42969 acc_top1_avg=0.44476 acc_top5_avg=0.86948 lr=0.01000 gn=5.42695 time=57.15it/s +epoch=33 global_step=13250 loss=5.83382 loss_avg=5.54876 acc=0.42969 acc_top1_avg=0.44396 acc_top5_avg=0.86944 lr=0.01000 gn=5.85323 time=63.17it/s +====================Eval==================== +epoch=33 global_step=13294 loss=2.76814 test_loss_avg=5.33485 acc=0.28125 test_acc_avg=0.06541 test_acc_top5_avg=0.63517 time=228.22it/s +epoch=33 global_step=13294 loss=6.85188 test_loss_avg=4.81521 acc=0.00000 test_acc_avg=0.21529 test_acc_top5_avg=0.71084 time=902.00it/s +curr_acc 0.2153 +BEST_ACC 0.2643 +curr_acc_top5 0.7108 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=34 global_step=13300 loss=5.23340 loss_avg=5.35850 acc=0.47656 acc_top1_avg=0.46094 acc_top5_avg=0.86719 lr=0.01000 gn=4.54811 time=64.21it/s +epoch=34 global_step=13350 loss=5.80305 loss_avg=5.50005 acc=0.42188 acc_top1_avg=0.44880 acc_top5_avg=0.87430 lr=0.01000 gn=4.94902 time=57.57it/s +epoch=34 global_step=13400 loss=5.05907 loss_avg=5.46450 acc=0.47656 acc_top1_avg=0.45172 acc_top5_avg=0.87552 lr=0.01000 gn=5.46972 time=59.05it/s +epoch=34 global_step=13450 loss=5.75924 loss_avg=5.48847 acc=0.42969 acc_top1_avg=0.44957 acc_top5_avg=0.87420 lr=0.01000 gn=5.31275 time=59.43it/s +epoch=34 global_step=13500 loss=5.09021 loss_avg=5.48670 acc=0.47656 acc_top1_avg=0.44979 acc_top5_avg=0.87117 lr=0.01000 gn=6.75177 time=53.72it/s +epoch=34 global_step=13550 loss=5.13433 loss_avg=5.49912 acc=0.48438 acc_top1_avg=0.44882 acc_top5_avg=0.87122 lr=0.01000 gn=4.58880 time=59.25it/s +epoch=34 global_step=13600 loss=5.30304 loss_avg=5.50437 acc=0.47656 acc_top1_avg=0.44820 acc_top5_avg=0.87234 lr=0.01000 gn=6.62143 time=61.09it/s +epoch=34 global_step=13650 loss=5.20290 loss_avg=5.51243 acc=0.49219 acc_top1_avg=0.44753 acc_top5_avg=0.87254 lr=0.01000 gn=5.28473 time=58.85it/s +====================Eval==================== +epoch=34 global_step=13685 loss=0.81031 test_loss_avg=4.28318 acc=0.73438 test_acc_avg=0.33371 test_acc_top5_avg=0.93694 time=260.08it/s +epoch=34 global_step=13685 loss=0.60710 test_loss_avg=4.34829 acc=0.84375 test_acc_avg=0.22937 test_acc_top5_avg=0.79321 time=238.81it/s +epoch=34 global_step=13685 loss=5.92968 test_loss_avg=4.23561 acc=0.00000 test_acc_avg=0.25633 test_acc_top5_avg=0.77383 time=907.66it/s +curr_acc 0.2563 +BEST_ACC 0.2643 +curr_acc_top5 0.7738 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=35 global_step=13700 loss=5.83850 loss_avg=5.70055 acc=0.42188 acc_top1_avg=0.42292 acc_top5_avg=0.85990 lr=0.01000 gn=4.86899 time=63.04it/s +epoch=35 global_step=13750 loss=4.90708 loss_avg=5.52514 acc=0.50781 acc_top1_avg=0.44651 acc_top5_avg=0.87440 lr=0.01000 gn=5.12182 time=64.61it/s +epoch=35 global_step=13800 loss=5.15867 loss_avg=5.56063 acc=0.50000 acc_top1_avg=0.44327 acc_top5_avg=0.86929 lr=0.01000 gn=5.72931 time=65.85it/s +epoch=35 global_step=13850 loss=5.33748 loss_avg=5.54867 acc=0.44531 acc_top1_avg=0.44493 acc_top5_avg=0.86989 lr=0.01000 gn=5.47580 time=64.92it/s +epoch=35 global_step=13900 loss=5.55223 loss_avg=5.54694 acc=0.46094 acc_top1_avg=0.44462 acc_top5_avg=0.86948 lr=0.01000 gn=5.41023 time=65.96it/s +epoch=35 global_step=13950 loss=5.29376 loss_avg=5.53653 acc=0.46094 acc_top1_avg=0.44525 acc_top5_avg=0.86952 lr=0.01000 gn=4.92028 time=64.25it/s +epoch=35 global_step=14000 loss=5.76392 loss_avg=5.52956 acc=0.43750 acc_top1_avg=0.44568 acc_top5_avg=0.86972 lr=0.01000 gn=6.46681 time=65.72it/s +epoch=35 global_step=14050 loss=5.84697 loss_avg=5.54614 acc=0.39844 acc_top1_avg=0.44409 acc_top5_avg=0.86988 lr=0.01000 gn=6.00345 time=64.73it/s +====================Eval==================== +epoch=35 global_step=14076 loss=6.39061 test_loss_avg=5.86741 acc=0.00000 test_acc_avg=0.02478 test_acc_top5_avg=0.74621 time=261.07it/s +epoch=35 global_step=14076 loss=5.02226 test_loss_avg=4.82376 acc=0.00000 test_acc_avg=0.17504 test_acc_top5_avg=0.78699 time=892.60it/s +curr_acc 0.1750 +BEST_ACC 0.2643 +curr_acc_top5 0.7870 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=36 global_step=14100 loss=5.22665 loss_avg=5.60084 acc=0.48438 acc_top1_avg=0.43392 acc_top5_avg=0.86751 lr=0.01000 gn=6.07295 time=62.66it/s +epoch=36 global_step=14150 loss=5.01731 loss_avg=5.55379 acc=0.50781 acc_top1_avg=0.44046 acc_top5_avg=0.86455 lr=0.01000 gn=4.96458 time=65.94it/s +epoch=36 global_step=14200 loss=5.43308 loss_avg=5.53718 acc=0.45312 acc_top1_avg=0.44279 acc_top5_avg=0.86511 lr=0.01000 gn=5.60632 time=61.48it/s +epoch=36 global_step=14250 loss=5.06254 loss_avg=5.53081 acc=0.49219 acc_top1_avg=0.44361 acc_top5_avg=0.86800 lr=0.01000 gn=5.74484 time=61.84it/s +epoch=36 global_step=14300 loss=5.79651 loss_avg=5.52417 acc=0.41406 acc_top1_avg=0.44465 acc_top5_avg=0.86904 lr=0.01000 gn=5.47250 time=58.38it/s +epoch=36 global_step=14350 loss=5.33293 loss_avg=5.53253 acc=0.46875 acc_top1_avg=0.44400 acc_top5_avg=0.86722 lr=0.01000 gn=5.63444 time=56.32it/s +epoch=36 global_step=14400 loss=4.99593 loss_avg=5.53764 acc=0.51562 acc_top1_avg=0.44372 acc_top5_avg=0.86721 lr=0.01000 gn=6.69608 time=60.60it/s +epoch=36 global_step=14450 loss=5.31285 loss_avg=5.53147 acc=0.45312 acc_top1_avg=0.44460 acc_top5_avg=0.86763 lr=0.01000 gn=5.38193 time=65.57it/s +====================Eval==================== +epoch=36 global_step=14467 loss=7.47880 test_loss_avg=7.58064 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.94792 time=250.15it/s +epoch=36 global_step=14467 loss=2.07609 test_loss_avg=5.57079 acc=0.57031 test_acc_avg=0.11175 test_acc_top5_avg=0.75809 time=259.98it/s +epoch=36 global_step=14467 loss=5.11365 test_loss_avg=4.70303 acc=0.00000 test_acc_avg=0.21608 test_acc_top5_avg=0.78600 time=879.12it/s +curr_acc 0.2161 +BEST_ACC 0.2643 +curr_acc_top5 0.7860 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=37 global_step=14500 loss=5.39914 loss_avg=5.53912 acc=0.46875 acc_top1_avg=0.44484 acc_top5_avg=0.86222 lr=0.01000 gn=5.28604 time=61.45it/s +epoch=37 global_step=14550 loss=5.25852 loss_avg=5.53638 acc=0.46875 acc_top1_avg=0.44578 acc_top5_avg=0.86822 lr=0.01000 gn=5.96515 time=58.20it/s +epoch=37 global_step=14600 loss=5.87782 loss_avg=5.52600 acc=0.39062 acc_top1_avg=0.44613 acc_top5_avg=0.86777 lr=0.01000 gn=6.38872 time=61.00it/s +epoch=37 global_step=14650 loss=5.52838 loss_avg=5.51654 acc=0.45312 acc_top1_avg=0.44770 acc_top5_avg=0.86868 lr=0.01000 gn=5.62594 time=61.61it/s +epoch=37 global_step=14700 loss=5.90062 loss_avg=5.51926 acc=0.38281 acc_top1_avg=0.44696 acc_top5_avg=0.86685 lr=0.01000 gn=5.06454 time=62.12it/s +epoch=37 global_step=14750 loss=5.39292 loss_avg=5.51944 acc=0.48438 acc_top1_avg=0.44738 acc_top5_avg=0.86658 lr=0.01000 gn=6.54106 time=61.48it/s +epoch=37 global_step=14800 loss=5.33050 loss_avg=5.52946 acc=0.45312 acc_top1_avg=0.44552 acc_top5_avg=0.86754 lr=0.01000 gn=5.62313 time=61.95it/s +epoch=37 global_step=14850 loss=6.01951 loss_avg=5.52641 acc=0.41406 acc_top1_avg=0.44637 acc_top5_avg=0.86831 lr=0.01000 gn=7.66594 time=61.77it/s +====================Eval==================== +epoch=37 global_step=14858 loss=7.06965 test_loss_avg=4.89808 acc=0.00000 test_acc_avg=0.14439 test_acc_top5_avg=0.88744 time=262.67it/s +epoch=37 global_step=14858 loss=5.69090 test_loss_avg=4.61907 acc=0.00000 test_acc_avg=0.20759 test_acc_top5_avg=0.79647 time=263.31it/s +epoch=37 global_step=14858 loss=5.53096 test_loss_avg=4.64324 acc=0.00000 test_acc_avg=0.20243 test_acc_top5_avg=0.78886 time=918.80it/s +curr_acc 0.2024 +BEST_ACC 0.2643 +curr_acc_top5 0.7889 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=38 global_step=14900 loss=5.40789 loss_avg=5.50585 acc=0.45312 acc_top1_avg=0.44773 acc_top5_avg=0.86440 lr=0.01000 gn=5.61101 time=65.70it/s +epoch=38 global_step=14950 loss=5.90286 loss_avg=5.48238 acc=0.41406 acc_top1_avg=0.45312 acc_top5_avg=0.87118 lr=0.01000 gn=5.31069 time=65.19it/s +epoch=38 global_step=15000 loss=5.95770 loss_avg=5.50228 acc=0.39844 acc_top1_avg=0.45004 acc_top5_avg=0.86961 lr=0.01000 gn=7.56436 time=65.98it/s +epoch=38 global_step=15050 loss=5.65966 loss_avg=5.48408 acc=0.42188 acc_top1_avg=0.45300 acc_top5_avg=0.86934 lr=0.01000 gn=5.30491 time=65.71it/s +epoch=38 global_step=15100 loss=5.73253 loss_avg=5.47424 acc=0.40625 acc_top1_avg=0.45406 acc_top5_avg=0.87019 lr=0.01000 gn=7.18590 time=65.59it/s +epoch=38 global_step=15150 loss=5.70817 loss_avg=5.49988 acc=0.43750 acc_top1_avg=0.45085 acc_top5_avg=0.87034 lr=0.01000 gn=5.63024 time=64.50it/s +epoch=38 global_step=15200 loss=5.64021 loss_avg=5.49897 acc=0.41406 acc_top1_avg=0.45084 acc_top5_avg=0.87032 lr=0.01000 gn=5.11000 time=65.95it/s +====================Eval==================== +epoch=38 global_step=15249 loss=7.56765 test_loss_avg=5.34914 acc=0.00000 test_acc_avg=0.10303 test_acc_top5_avg=0.74512 time=263.31it/s +epoch=38 global_step=15249 loss=6.04111 test_loss_avg=4.79899 acc=0.00000 test_acc_avg=0.20174 test_acc_top5_avg=0.82160 time=920.01it/s +curr_acc 0.2017 +BEST_ACC 0.2643 +curr_acc_top5 0.8216 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=39 global_step=15250 loss=5.86520 loss_avg=5.86520 acc=0.38281 acc_top1_avg=0.38281 acc_top5_avg=0.85938 lr=0.01000 gn=5.77976 time=54.61it/s +epoch=39 global_step=15300 loss=4.75699 loss_avg=5.52188 acc=0.51562 acc_top1_avg=0.44225 acc_top5_avg=0.87010 lr=0.01000 gn=5.64960 time=60.68it/s +epoch=39 global_step=15350 loss=5.82946 loss_avg=5.49890 acc=0.40625 acc_top1_avg=0.44694 acc_top5_avg=0.87036 lr=0.01000 gn=4.41026 time=62.14it/s +epoch=39 global_step=15400 loss=5.67740 loss_avg=5.49309 acc=0.42188 acc_top1_avg=0.44785 acc_top5_avg=0.87236 lr=0.01000 gn=6.79224 time=56.71it/s +epoch=39 global_step=15450 loss=6.03698 loss_avg=5.49900 acc=0.39062 acc_top1_avg=0.44757 acc_top5_avg=0.87251 lr=0.01000 gn=6.14148 time=56.92it/s +epoch=39 global_step=15500 loss=6.31383 loss_avg=5.51294 acc=0.36719 acc_top1_avg=0.44622 acc_top5_avg=0.87273 lr=0.01000 gn=6.63998 time=61.74it/s +epoch=39 global_step=15550 loss=5.46178 loss_avg=5.50585 acc=0.46875 acc_top1_avg=0.44744 acc_top5_avg=0.87240 lr=0.01000 gn=5.76905 time=58.76it/s +epoch=39 global_step=15600 loss=5.65610 loss_avg=5.51124 acc=0.41406 acc_top1_avg=0.44707 acc_top5_avg=0.87200 lr=0.01000 gn=6.69112 time=57.81it/s +====================Eval==================== +epoch=39 global_step=15640 loss=4.94769 test_loss_avg=3.69916 acc=0.00000 test_acc_avg=0.35526 test_acc_top5_avg=0.88980 time=239.10it/s +epoch=39 global_step=15640 loss=0.08919 test_loss_avg=4.14226 acc=0.96875 test_acc_avg=0.29031 test_acc_top5_avg=0.75113 time=262.50it/s +epoch=39 global_step=15640 loss=6.46787 test_loss_avg=4.32209 acc=0.00000 test_acc_avg=0.27047 test_acc_top5_avg=0.76938 time=893.74it/s +curr_acc 0.2705 +BEST_ACC 0.2643 +curr_acc_top5 0.7694 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=40 global_step=15650 loss=5.61618 loss_avg=5.58006 acc=0.44531 acc_top1_avg=0.44141 acc_top5_avg=0.87578 lr=0.00100 gn=6.74735 time=63.94it/s +epoch=40 global_step=15700 loss=5.05749 loss_avg=5.42590 acc=0.48438 acc_top1_avg=0.45560 acc_top5_avg=0.88307 lr=0.00100 gn=5.14544 time=58.95it/s +epoch=40 global_step=15750 loss=5.23330 loss_avg=5.29064 acc=0.46094 acc_top1_avg=0.46974 acc_top5_avg=0.88352 lr=0.00100 gn=6.19755 time=60.79it/s +epoch=40 global_step=15800 loss=4.43774 loss_avg=5.29830 acc=0.54688 acc_top1_avg=0.46919 acc_top5_avg=0.88145 lr=0.00100 gn=5.15283 time=61.45it/s +epoch=40 global_step=15850 loss=4.71275 loss_avg=5.27057 acc=0.52344 acc_top1_avg=0.47281 acc_top5_avg=0.88278 lr=0.00100 gn=5.45755 time=60.34it/s +epoch=40 global_step=15900 loss=5.69253 loss_avg=5.26815 acc=0.42969 acc_top1_avg=0.47260 acc_top5_avg=0.88269 lr=0.00100 gn=5.09055 time=61.52it/s +epoch=40 global_step=15950 loss=5.09194 loss_avg=5.26250 acc=0.50781 acc_top1_avg=0.47356 acc_top5_avg=0.88236 lr=0.00100 gn=5.92076 time=61.47it/s +epoch=40 global_step=16000 loss=5.25691 loss_avg=5.24962 acc=0.46875 acc_top1_avg=0.47491 acc_top5_avg=0.88290 lr=0.00100 gn=5.44063 time=61.44it/s +====================Eval==================== +epoch=40 global_step=16031 loss=3.34398 test_loss_avg=5.41773 acc=0.22656 test_acc_avg=0.08340 test_acc_top5_avg=0.72383 time=261.98it/s +epoch=40 global_step=16031 loss=5.09495 test_loss_avg=4.57996 acc=0.00000 test_acc_avg=0.23517 test_acc_top5_avg=0.83900 time=904.53it/s +curr_acc 0.2352 +BEST_ACC 0.2705 +curr_acc_top5 0.8390 +BEST_ACC_top5 0.8278 +Model Saved! + +====================Training==================== +epoch=41 global_step=16050 loss=5.48155 loss_avg=4.99670 acc=0.42969 acc_top1_avg=0.50206 acc_top5_avg=0.89145 lr=0.00100 gn=5.26646 time=58.43it/s +epoch=41 global_step=16100 loss=5.78849 loss_avg=5.16798 acc=0.42969 acc_top1_avg=0.48426 acc_top5_avg=0.88440 lr=0.00100 gn=5.84440 time=61.48it/s +epoch=41 global_step=16150 loss=4.94866 loss_avg=5.15735 acc=0.51562 acc_top1_avg=0.48516 acc_top5_avg=0.88754 lr=0.00100 gn=5.02857 time=61.54it/s +epoch=41 global_step=16200 loss=5.18620 loss_avg=5.16423 acc=0.47656 acc_top1_avg=0.48474 acc_top5_avg=0.88757 lr=0.00100 gn=5.67789 time=58.20it/s +epoch=41 global_step=16250 loss=5.14366 loss_avg=5.14451 acc=0.48438 acc_top1_avg=0.48684 acc_top5_avg=0.88681 lr=0.00100 gn=4.75149 time=62.47it/s +epoch=41 global_step=16300 loss=5.07673 loss_avg=5.15162 acc=0.50000 acc_top1_avg=0.48565 acc_top5_avg=0.88848 lr=0.00100 gn=7.33989 time=58.73it/s +epoch=41 global_step=16350 loss=5.03091 loss_avg=5.14320 acc=0.49219 acc_top1_avg=0.48648 acc_top5_avg=0.88849 lr=0.00100 gn=7.53228 time=61.55it/s +epoch=41 global_step=16400 loss=4.96965 loss_avg=5.14740 acc=0.50781 acc_top1_avg=0.48592 acc_top5_avg=0.88751 lr=0.00100 gn=5.53687 time=61.58it/s +====================Eval==================== +epoch=41 global_step=16422 loss=1.71152 test_loss_avg=5.47776 acc=0.35938 test_acc_avg=0.10866 test_acc_top5_avg=0.93253 time=248.99it/s +epoch=41 global_step=16422 loss=0.45925 test_loss_avg=5.06356 acc=0.86719 test_acc_avg=0.16778 test_acc_top5_avg=0.81634 time=262.34it/s +epoch=41 global_step=16422 loss=5.25731 test_loss_avg=4.54444 acc=0.00000 test_acc_avg=0.23853 test_acc_top5_avg=0.83999 time=903.75it/s +curr_acc 0.2385 +BEST_ACC 0.2705 +curr_acc_top5 0.8400 +BEST_ACC_top5 0.8390 +Model Saved! + +====================Training==================== +epoch=42 global_step=16450 loss=5.21968 loss_avg=4.95985 acc=0.47656 acc_top1_avg=0.50614 acc_top5_avg=0.88951 lr=0.00100 gn=6.63285 time=58.16it/s +epoch=42 global_step=16500 loss=4.93853 loss_avg=5.03811 acc=0.51562 acc_top1_avg=0.49800 acc_top5_avg=0.88662 lr=0.00100 gn=6.38102 time=57.33it/s +epoch=42 global_step=16550 loss=4.33116 loss_avg=5.07162 acc=0.58594 acc_top1_avg=0.49445 acc_top5_avg=0.88757 lr=0.00100 gn=6.39339 time=59.09it/s +epoch=42 global_step=16600 loss=5.60004 loss_avg=5.09736 acc=0.42969 acc_top1_avg=0.49214 acc_top5_avg=0.88760 lr=0.00100 gn=6.83867 time=55.40it/s +epoch=42 global_step=16650 loss=5.68679 loss_avg=5.10924 acc=0.42188 acc_top1_avg=0.49099 acc_top5_avg=0.88871 lr=0.00100 gn=4.99501 time=60.94it/s +epoch=42 global_step=16700 loss=5.56977 loss_avg=5.10582 acc=0.43750 acc_top1_avg=0.49112 acc_top5_avg=0.88922 lr=0.00100 gn=8.52361 time=56.38it/s +epoch=42 global_step=16750 loss=5.16545 loss_avg=5.09527 acc=0.48438 acc_top1_avg=0.49185 acc_top5_avg=0.89029 lr=0.00100 gn=4.80630 time=62.97it/s +epoch=42 global_step=16800 loss=4.81789 loss_avg=5.10218 acc=0.51562 acc_top1_avg=0.49105 acc_top5_avg=0.88961 lr=0.00100 gn=6.31638 time=61.92it/s +====================Eval==================== +epoch=42 global_step=16813 loss=6.55004 test_loss_avg=5.18621 acc=0.00000 test_acc_avg=0.06250 test_acc_top5_avg=0.81006 time=238.10it/s +epoch=42 global_step=16813 loss=4.94132 test_loss_avg=4.48631 acc=0.00000 test_acc_avg=0.23190 test_acc_top5_avg=0.83534 time=914.19it/s +curr_acc 0.2319 +BEST_ACC 0.2705 +curr_acc_top5 0.8353 +BEST_ACC_top5 0.8400 +Model Saved! + +====================Training==================== +epoch=43 global_step=16850 loss=5.44808 loss_avg=5.00880 acc=0.42969 acc_top1_avg=0.50148 acc_top5_avg=0.89506 lr=0.00100 gn=5.45585 time=61.96it/s +epoch=43 global_step=16900 loss=5.43364 loss_avg=5.03055 acc=0.45312 acc_top1_avg=0.49910 acc_top5_avg=0.89565 lr=0.00100 gn=5.20528 time=61.60it/s +epoch=43 global_step=16950 loss=4.49864 loss_avg=5.02588 acc=0.57031 acc_top1_avg=0.49920 acc_top5_avg=0.89376 lr=0.00100 gn=5.73063 time=61.25it/s +epoch=43 global_step=17000 loss=4.66670 loss_avg=5.03769 acc=0.55469 acc_top1_avg=0.49812 acc_top5_avg=0.89205 lr=0.00100 gn=6.37660 time=60.28it/s +epoch=43 global_step=17050 loss=4.26162 loss_avg=5.04544 acc=0.58594 acc_top1_avg=0.49703 acc_top5_avg=0.89194 lr=0.00100 gn=5.98650 time=55.68it/s +epoch=43 global_step=17100 loss=5.40628 loss_avg=5.04969 acc=0.45312 acc_top1_avg=0.49703 acc_top5_avg=0.89131 lr=0.00100 gn=7.97289 time=65.47it/s +epoch=43 global_step=17150 loss=4.94673 loss_avg=5.05760 acc=0.50781 acc_top1_avg=0.49659 acc_top5_avg=0.89011 lr=0.00100 gn=6.84739 time=62.70it/s +epoch=43 global_step=17200 loss=5.05920 loss_avg=5.05439 acc=0.48438 acc_top1_avg=0.49673 acc_top5_avg=0.88986 lr=0.00100 gn=6.23944 time=61.91it/s +====================Eval==================== +epoch=43 global_step=17204 loss=7.25371 test_loss_avg=7.25385 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.92448 time=218.02it/s +epoch=43 global_step=17204 loss=8.37830 test_loss_avg=5.46828 acc=0.00000 test_acc_avg=0.11203 test_acc_top5_avg=0.81840 time=261.62it/s +epoch=43 global_step=17204 loss=5.65768 test_loss_avg=4.56163 acc=0.00000 test_acc_avg=0.24684 test_acc_top5_avg=0.85403 time=880.42it/s +curr_acc 0.2468 +BEST_ACC 0.2705 +curr_acc_top5 0.8540 +BEST_ACC_top5 0.8400 +Model Saved! + +====================Training==================== +epoch=44 global_step=17250 loss=5.36917 loss_avg=4.97579 acc=0.44531 acc_top1_avg=0.50306 acc_top5_avg=0.89997 lr=0.00100 gn=7.05017 time=62.66it/s +epoch=44 global_step=17300 loss=4.22210 loss_avg=4.97294 acc=0.59375 acc_top1_avg=0.50553 acc_top5_avg=0.89437 lr=0.00100 gn=6.19823 time=65.43it/s +epoch=44 global_step=17350 loss=5.07388 loss_avg=4.98923 acc=0.50000 acc_top1_avg=0.50455 acc_top5_avg=0.89309 lr=0.00100 gn=8.54716 time=61.85it/s +epoch=44 global_step=17400 loss=5.41511 loss_avg=5.01927 acc=0.45312 acc_top1_avg=0.50120 acc_top5_avg=0.88919 lr=0.00100 gn=7.45629 time=61.78it/s +epoch=44 global_step=17450 loss=5.27938 loss_avg=5.03044 acc=0.49219 acc_top1_avg=0.49962 acc_top5_avg=0.88907 lr=0.00100 gn=8.64700 time=61.63it/s +epoch=44 global_step=17500 loss=5.82244 loss_avg=5.03021 acc=0.41406 acc_top1_avg=0.49902 acc_top5_avg=0.88983 lr=0.00100 gn=7.00265 time=61.86it/s +epoch=44 global_step=17550 loss=4.63070 loss_avg=5.02814 acc=0.53125 acc_top1_avg=0.49896 acc_top5_avg=0.89087 lr=0.00100 gn=8.28022 time=54.47it/s +====================Eval==================== +epoch=44 global_step=17595 loss=5.91048 test_loss_avg=4.78532 acc=0.00000 test_acc_avg=0.08626 test_acc_top5_avg=0.94173 time=256.69it/s +epoch=44 global_step=17595 loss=5.34545 test_loss_avg=4.49504 acc=0.02344 test_acc_avg=0.23543 test_acc_top5_avg=0.85452 time=262.14it/s +epoch=44 global_step=17595 loss=4.90511 test_loss_avg=4.53836 acc=0.00000 test_acc_avg=0.22102 test_acc_top5_avg=0.84751 time=892.03it/s +curr_acc 0.2210 +BEST_ACC 0.2705 +curr_acc_top5 0.8475 +BEST_ACC_top5 0.8540 +Model Saved! + +====================Training==================== +epoch=45 global_step=17600 loss=4.80692 loss_avg=4.73413 acc=0.53906 acc_top1_avg=0.53906 acc_top5_avg=0.90000 lr=0.00100 gn=8.24301 time=56.76it/s +epoch=45 global_step=17650 loss=4.86783 loss_avg=4.98949 acc=0.53906 acc_top1_avg=0.50426 acc_top5_avg=0.89062 lr=0.00100 gn=7.63329 time=56.97it/s +epoch=45 global_step=17700 loss=5.02549 loss_avg=4.94733 acc=0.50000 acc_top1_avg=0.50908 acc_top5_avg=0.89204 lr=0.00100 gn=6.46498 time=59.80it/s +epoch=45 global_step=17750 loss=5.70843 loss_avg=4.98535 acc=0.43750 acc_top1_avg=0.50383 acc_top5_avg=0.89153 lr=0.00100 gn=7.48660 time=56.31it/s +epoch=45 global_step=17800 loss=4.97219 loss_avg=4.98535 acc=0.48438 acc_top1_avg=0.50373 acc_top5_avg=0.89002 lr=0.00100 gn=6.23573 time=61.71it/s +epoch=45 global_step=17850 loss=5.40139 loss_avg=5.00291 acc=0.46094 acc_top1_avg=0.50181 acc_top5_avg=0.88992 lr=0.00100 gn=7.84238 time=56.22it/s +epoch=45 global_step=17900 loss=4.40625 loss_avg=5.00353 acc=0.57031 acc_top1_avg=0.50192 acc_top5_avg=0.89119 lr=0.00100 gn=7.44312 time=60.31it/s +epoch=45 global_step=17950 loss=5.81954 loss_avg=5.00388 acc=0.40625 acc_top1_avg=0.50187 acc_top5_avg=0.89074 lr=0.00100 gn=8.44308 time=58.15it/s +====================Eval==================== +epoch=45 global_step=17986 loss=3.37950 test_loss_avg=5.09698 acc=0.24219 test_acc_avg=0.09670 test_acc_top5_avg=0.77830 time=241.87it/s +epoch=45 global_step=17986 loss=5.37048 test_loss_avg=4.44008 acc=0.00000 test_acc_avg=0.24130 test_acc_top5_avg=0.85314 time=919.20it/s +curr_acc 0.2413 +BEST_ACC 0.2705 +curr_acc_top5 0.8531 +BEST_ACC_top5 0.8540 +Model Saved! + +====================Training==================== +epoch=46 global_step=18000 loss=4.81263 loss_avg=4.97822 acc=0.51562 acc_top1_avg=0.50781 acc_top5_avg=0.89230 lr=0.00100 gn=8.34981 time=57.82it/s +epoch=46 global_step=18050 loss=5.23905 loss_avg=4.98634 acc=0.46875 acc_top1_avg=0.50598 acc_top5_avg=0.89392 lr=0.00100 gn=7.33335 time=55.10it/s +epoch=46 global_step=18100 loss=4.53534 loss_avg=4.95374 acc=0.56250 acc_top1_avg=0.50809 acc_top5_avg=0.89206 lr=0.00100 gn=8.64994 time=53.48it/s +epoch=46 global_step=18150 loss=5.71581 loss_avg=4.99059 acc=0.42969 acc_top1_avg=0.50362 acc_top5_avg=0.89005 lr=0.00100 gn=6.62352 time=62.84it/s +epoch=46 global_step=18200 loss=4.06290 loss_avg=4.98584 acc=0.60938 acc_top1_avg=0.50515 acc_top5_avg=0.89146 lr=0.00100 gn=7.93437 time=63.44it/s +epoch=46 global_step=18250 loss=5.37565 loss_avg=4.96116 acc=0.46094 acc_top1_avg=0.50728 acc_top5_avg=0.89202 lr=0.00100 gn=9.57576 time=61.72it/s +epoch=46 global_step=18300 loss=5.39047 loss_avg=4.96871 acc=0.44531 acc_top1_avg=0.50615 acc_top5_avg=0.89247 lr=0.00100 gn=6.61414 time=56.94it/s +epoch=46 global_step=18350 loss=4.98959 loss_avg=4.97206 acc=0.50781 acc_top1_avg=0.50552 acc_top5_avg=0.89208 lr=0.00100 gn=8.85511 time=61.93it/s +====================Eval==================== +epoch=46 global_step=18377 loss=2.80753 test_loss_avg=4.15461 acc=0.30469 test_acc_avg=0.24268 test_acc_top5_avg=0.94775 time=260.56it/s +epoch=46 global_step=18377 loss=0.24383 test_loss_avg=4.51950 acc=0.89844 test_acc_avg=0.24586 test_acc_top5_avg=0.84801 time=260.94it/s +epoch=46 global_step=18377 loss=5.58737 test_loss_avg=4.41599 acc=0.00000 test_acc_avg=0.25752 test_acc_top5_avg=0.85483 time=880.97it/s +curr_acc 0.2575 +BEST_ACC 0.2705 +curr_acc_top5 0.8548 +BEST_ACC_top5 0.8540 +Model Saved! + +====================Training==================== +epoch=47 global_step=18400 loss=4.77415 loss_avg=4.87090 acc=0.51562 acc_top1_avg=0.51529 acc_top5_avg=0.89742 lr=0.00100 gn=8.62477 time=61.71it/s +epoch=47 global_step=18450 loss=4.67514 loss_avg=4.99789 acc=0.53906 acc_top1_avg=0.50225 acc_top5_avg=0.88945 lr=0.00100 gn=7.52739 time=57.05it/s +epoch=47 global_step=18500 loss=5.31848 loss_avg=4.97895 acc=0.46875 acc_top1_avg=0.50476 acc_top5_avg=0.89126 lr=0.00100 gn=7.33638 time=65.53it/s +epoch=47 global_step=18550 loss=4.71023 loss_avg=4.95945 acc=0.53125 acc_top1_avg=0.50695 acc_top5_avg=0.88941 lr=0.00100 gn=8.78865 time=63.17it/s +epoch=47 global_step=18600 loss=5.43792 loss_avg=4.97702 acc=0.45312 acc_top1_avg=0.50501 acc_top5_avg=0.88824 lr=0.00100 gn=6.94020 time=61.69it/s +epoch=47 global_step=18650 loss=5.29372 loss_avg=4.97075 acc=0.48438 acc_top1_avg=0.50581 acc_top5_avg=0.88908 lr=0.00100 gn=8.51755 time=56.64it/s +epoch=47 global_step=18700 loss=4.71638 loss_avg=4.96986 acc=0.52344 acc_top1_avg=0.50535 acc_top5_avg=0.88968 lr=0.00100 gn=6.66515 time=59.67it/s +epoch=47 global_step=18750 loss=4.71498 loss_avg=4.95636 acc=0.51562 acc_top1_avg=0.50653 acc_top5_avg=0.89000 lr=0.00100 gn=10.04086 time=56.39it/s +====================Eval==================== +epoch=47 global_step=18768 loss=6.64534 test_loss_avg=5.29299 acc=0.00000 test_acc_avg=0.06546 test_acc_top5_avg=0.76626 time=262.47it/s +epoch=47 global_step=18768 loss=5.26381 test_loss_avg=4.43615 acc=0.00000 test_acc_avg=0.23101 test_acc_top5_avg=0.85216 time=926.30it/s +curr_acc 0.2310 +BEST_ACC 0.2705 +curr_acc_top5 0.8522 +BEST_ACC_top5 0.8548 +Model Saved! + +====================Training==================== +epoch=48 global_step=18800 loss=4.19694 loss_avg=4.93043 acc=0.60156 acc_top1_avg=0.51343 acc_top5_avg=0.88940 lr=0.00100 gn=8.51810 time=65.22it/s +epoch=48 global_step=18850 loss=4.98567 loss_avg=4.92407 acc=0.50781 acc_top1_avg=0.51248 acc_top5_avg=0.89129 lr=0.00100 gn=8.57290 time=63.43it/s +epoch=48 global_step=18900 loss=4.75110 loss_avg=4.90999 acc=0.51562 acc_top1_avg=0.51314 acc_top5_avg=0.89205 lr=0.00100 gn=8.79190 time=58.25it/s +epoch=48 global_step=18950 loss=4.04801 loss_avg=4.88123 acc=0.58594 acc_top1_avg=0.51575 acc_top5_avg=0.89414 lr=0.00100 gn=8.96015 time=58.84it/s +epoch=48 global_step=19000 loss=4.21420 loss_avg=4.92520 acc=0.58594 acc_top1_avg=0.51074 acc_top5_avg=0.89278 lr=0.00100 gn=7.90797 time=58.57it/s +epoch=48 global_step=19050 loss=5.11622 loss_avg=4.93424 acc=0.48438 acc_top1_avg=0.50986 acc_top5_avg=0.89176 lr=0.00100 gn=8.83767 time=59.41it/s +epoch=48 global_step=19100 loss=5.30481 loss_avg=4.93987 acc=0.48438 acc_top1_avg=0.50918 acc_top5_avg=0.89218 lr=0.00100 gn=10.42753 time=60.05it/s +epoch=48 global_step=19150 loss=4.82523 loss_avg=4.93388 acc=0.52344 acc_top1_avg=0.51021 acc_top5_avg=0.89173 lr=0.00100 gn=9.47999 time=62.01it/s +====================Eval==================== +epoch=48 global_step=19159 loss=5.89472 test_loss_avg=6.80359 acc=0.07812 test_acc_avg=0.00977 test_acc_top5_avg=0.90430 time=259.12it/s +epoch=48 global_step=19159 loss=0.34471 test_loss_avg=5.21905 acc=0.89844 test_acc_avg=0.11921 test_acc_top5_avg=0.83217 time=262.90it/s +epoch=48 global_step=19159 loss=5.05979 test_loss_avg=4.45976 acc=0.00000 test_acc_avg=0.23002 test_acc_top5_avg=0.85572 time=874.36it/s +curr_acc 0.2300 +BEST_ACC 0.2705 +curr_acc_top5 0.8557 +BEST_ACC_top5 0.8548 +Model Saved! + +====================Training==================== +epoch=49 global_step=19200 loss=4.79973 loss_avg=4.95301 acc=0.50000 acc_top1_avg=0.50896 acc_top5_avg=0.88929 lr=0.00100 gn=10.29440 time=55.97it/s +epoch=49 global_step=19250 loss=4.45256 loss_avg=4.93577 acc=0.55469 acc_top1_avg=0.50987 acc_top5_avg=0.88994 lr=0.00100 gn=8.32195 time=58.10it/s +epoch=49 global_step=19300 loss=4.68703 loss_avg=4.94712 acc=0.53906 acc_top1_avg=0.50848 acc_top5_avg=0.89007 lr=0.00100 gn=9.13338 time=56.59it/s +epoch=49 global_step=19350 loss=4.93355 loss_avg=4.91258 acc=0.50000 acc_top1_avg=0.51186 acc_top5_avg=0.89103 lr=0.00100 gn=9.42751 time=54.58it/s +epoch=49 global_step=19400 loss=5.36702 loss_avg=4.91515 acc=0.47656 acc_top1_avg=0.51161 acc_top5_avg=0.88975 lr=0.00100 gn=11.10298 time=64.57it/s +epoch=49 global_step=19450 loss=4.57541 loss_avg=4.90929 acc=0.55469 acc_top1_avg=0.51230 acc_top5_avg=0.88995 lr=0.00100 gn=10.89061 time=65.52it/s +epoch=49 global_step=19500 loss=4.60691 loss_avg=4.90336 acc=0.55469 acc_top1_avg=0.51278 acc_top5_avg=0.89104 lr=0.00100 gn=10.10509 time=65.83it/s +epoch=49 global_step=19550 loss=5.02045 loss_avg=4.90224 acc=0.50000 acc_top1_avg=0.51259 acc_top5_avg=0.89055 lr=0.00100 gn=11.62197 time=89.85it/s +====================Eval==================== +epoch=49 global_step=19550 loss=6.40449 test_loss_avg=5.06026 acc=0.00000 test_acc_avg=0.08728 test_acc_top5_avg=0.88093 time=253.97it/s +epoch=49 global_step=19550 loss=5.18431 test_loss_avg=4.45631 acc=0.00000 test_acc_avg=0.23358 test_acc_top5_avg=0.84691 time=902.19it/s +epoch=49 global_step=19550 loss=5.18431 test_loss_avg=4.45631 acc=0.00000 test_acc_avg=0.23358 test_acc_top5_avg=0.84691 time=902.19it/s +curr_acc 0.2336 +BEST_ACC 0.2705 +curr_acc_top5 0.8469 +BEST_ACC_top5 0.8557 +Model Saved! + +====================Training==================== +epoch=50 global_step=19600 loss=5.07175 loss_avg=4.91371 acc=0.49219 acc_top1_avg=0.51203 acc_top5_avg=0.89172 lr=0.00100 gn=10.31535 time=60.03it/s +epoch=50 global_step=19650 loss=4.58608 loss_avg=4.82213 acc=0.53906 acc_top1_avg=0.52109 acc_top5_avg=0.89320 lr=0.00100 gn=9.50851 time=61.93it/s +epoch=50 global_step=19700 loss=4.43174 loss_avg=4.80369 acc=0.53906 acc_top1_avg=0.52333 acc_top5_avg=0.89521 lr=0.00100 gn=9.55271 time=57.95it/s +epoch=50 global_step=19750 loss=5.15722 loss_avg=4.83502 acc=0.49219 acc_top1_avg=0.51992 acc_top5_avg=0.89359 lr=0.00100 gn=10.92313 time=60.80it/s +epoch=50 global_step=19800 loss=5.33546 loss_avg=4.85549 acc=0.47656 acc_top1_avg=0.51756 acc_top5_avg=0.89338 lr=0.00100 gn=10.59651 time=56.08it/s +epoch=50 global_step=19850 loss=5.15331 loss_avg=4.87742 acc=0.47656 acc_top1_avg=0.51557 acc_top5_avg=0.89190 lr=0.00100 gn=6.61801 time=60.34it/s +epoch=50 global_step=19900 loss=4.38579 loss_avg=4.87881 acc=0.56250 acc_top1_avg=0.51520 acc_top5_avg=0.89094 lr=0.00100 gn=10.24293 time=60.50it/s +====================Eval==================== +epoch=50 global_step=19941 loss=7.92187 test_loss_avg=5.14607 acc=0.00000 test_acc_avg=0.09781 test_acc_top5_avg=0.81078 time=261.77it/s +epoch=50 global_step=19941 loss=5.30055 test_loss_avg=4.35761 acc=0.00000 test_acc_avg=0.24219 test_acc_top5_avg=0.85443 time=897.56it/s +curr_acc 0.2422 +BEST_ACC 0.2705 +curr_acc_top5 0.8544 +BEST_ACC_top5 0.8557 +Model Saved! + +====================Training==================== +epoch=51 global_step=19950 loss=5.46896 loss_avg=4.76317 acc=0.46875 acc_top1_avg=0.53299 acc_top5_avg=0.90017 lr=0.00100 gn=9.91337 time=64.03it/s +epoch=51 global_step=20000 loss=5.01238 loss_avg=4.79114 acc=0.50781 acc_top1_avg=0.52542 acc_top5_avg=0.88771 lr=0.00100 gn=10.39508 time=62.29it/s +epoch=51 global_step=20050 loss=4.94954 loss_avg=4.78867 acc=0.50781 acc_top1_avg=0.52659 acc_top5_avg=0.89170 lr=0.00100 gn=8.30221 time=60.27it/s +epoch=51 global_step=20100 loss=4.84104 loss_avg=4.81336 acc=0.53125 acc_top1_avg=0.52403 acc_top5_avg=0.89225 lr=0.00100 gn=11.38871 time=55.41it/s +epoch=51 global_step=20150 loss=5.03689 loss_avg=4.83109 acc=0.49219 acc_top1_avg=0.52205 acc_top5_avg=0.89205 lr=0.00100 gn=9.72933 time=59.78it/s +epoch=51 global_step=20200 loss=5.02541 loss_avg=4.84668 acc=0.49219 acc_top1_avg=0.51931 acc_top5_avg=0.89078 lr=0.00100 gn=12.91892 time=54.70it/s +epoch=51 global_step=20250 loss=4.61466 loss_avg=4.85696 acc=0.54688 acc_top1_avg=0.51767 acc_top5_avg=0.89050 lr=0.00100 gn=10.50142 time=59.84it/s +epoch=51 global_step=20300 loss=5.08778 loss_avg=4.85288 acc=0.49219 acc_top1_avg=0.51821 acc_top5_avg=0.89017 lr=0.00100 gn=9.53345 time=60.03it/s +====================Eval==================== +epoch=51 global_step=20332 loss=4.76341 test_loss_avg=4.60582 acc=0.00000 test_acc_avg=0.16183 test_acc_top5_avg=0.95015 time=249.08it/s +epoch=51 global_step=20332 loss=3.77957 test_loss_avg=4.25752 acc=0.30469 test_acc_avg=0.27894 test_acc_top5_avg=0.85849 time=262.34it/s +epoch=51 global_step=20332 loss=5.18164 test_loss_avg=4.37134 acc=0.00000 test_acc_avg=0.25178 test_acc_top5_avg=0.85334 time=893.93it/s +curr_acc 0.2518 +BEST_ACC 0.2705 +curr_acc_top5 0.8533 +BEST_ACC_top5 0.8557 +Model Saved! + +====================Training==================== +epoch=52 global_step=20350 loss=5.17562 loss_avg=4.87886 acc=0.47656 acc_top1_avg=0.51519 acc_top5_avg=0.88542 lr=0.00100 gn=11.23046 time=62.23it/s +epoch=52 global_step=20400 loss=5.02503 loss_avg=4.80615 acc=0.49219 acc_top1_avg=0.52585 acc_top5_avg=0.89131 lr=0.00100 gn=10.81148 time=60.86it/s +epoch=52 global_step=20450 loss=4.52026 loss_avg=4.79250 acc=0.53906 acc_top1_avg=0.52562 acc_top5_avg=0.89102 lr=0.00100 gn=12.25453 time=59.07it/s +epoch=52 global_step=20500 loss=5.13183 loss_avg=4.81978 acc=0.48438 acc_top1_avg=0.52195 acc_top5_avg=0.89049 lr=0.00100 gn=12.74666 time=65.46it/s +epoch=52 global_step=20550 loss=4.97839 loss_avg=4.81934 acc=0.50781 acc_top1_avg=0.52265 acc_top5_avg=0.88991 lr=0.00100 gn=8.15370 time=64.45it/s +epoch=52 global_step=20600 loss=4.98668 loss_avg=4.85599 acc=0.50000 acc_top1_avg=0.51877 acc_top5_avg=0.89022 lr=0.00100 gn=12.73252 time=63.03it/s +epoch=52 global_step=20650 loss=4.94133 loss_avg=4.83852 acc=0.52344 acc_top1_avg=0.52059 acc_top5_avg=0.89028 lr=0.00100 gn=11.54639 time=59.03it/s +epoch=52 global_step=20700 loss=4.95714 loss_avg=4.85102 acc=0.51562 acc_top1_avg=0.51875 acc_top5_avg=0.88975 lr=0.00100 gn=11.70661 time=63.26it/s +====================Eval==================== +epoch=52 global_step=20723 loss=3.49349 test_loss_avg=5.29247 acc=0.21875 test_acc_avg=0.08631 test_acc_top5_avg=0.79371 time=262.00it/s +epoch=52 global_step=20723 loss=5.39102 test_loss_avg=4.41517 acc=0.00000 test_acc_avg=0.23744 test_acc_top5_avg=0.86788 time=883.01it/s +curr_acc 0.2374 +BEST_ACC 0.2705 +curr_acc_top5 0.8679 +BEST_ACC_top5 0.8557 +Model Saved! + +====================Training==================== +epoch=53 global_step=20750 loss=4.51417 loss_avg=4.71950 acc=0.57031 acc_top1_avg=0.53443 acc_top5_avg=0.89612 lr=0.00100 gn=13.59602 time=55.81it/s +epoch=53 global_step=20800 loss=4.47913 loss_avg=4.78442 acc=0.55469 acc_top1_avg=0.52729 acc_top5_avg=0.89276 lr=0.00100 gn=12.03708 time=59.11it/s +epoch=53 global_step=20850 loss=5.64716 loss_avg=4.80038 acc=0.42969 acc_top1_avg=0.52479 acc_top5_avg=0.89192 lr=0.00100 gn=9.97840 time=59.25it/s +epoch=53 global_step=20900 loss=4.86202 loss_avg=4.78973 acc=0.53125 acc_top1_avg=0.52494 acc_top5_avg=0.89186 lr=0.00100 gn=12.21894 time=58.20it/s +epoch=53 global_step=20950 loss=4.40122 loss_avg=4.77958 acc=0.57031 acc_top1_avg=0.52633 acc_top5_avg=0.89190 lr=0.00100 gn=12.77612 time=57.89it/s +epoch=53 global_step=21000 loss=4.90155 loss_avg=4.80544 acc=0.50000 acc_top1_avg=0.52347 acc_top5_avg=0.89119 lr=0.00100 gn=10.39133 time=61.51it/s +epoch=53 global_step=21050 loss=4.36937 loss_avg=4.82517 acc=0.57031 acc_top1_avg=0.52107 acc_top5_avg=0.89067 lr=0.00100 gn=11.15054 time=59.30it/s +epoch=53 global_step=21100 loss=4.52252 loss_avg=4.82036 acc=0.55469 acc_top1_avg=0.52195 acc_top5_avg=0.89044 lr=0.00100 gn=10.23819 time=61.58it/s +====================Eval==================== +epoch=53 global_step=21114 loss=2.15442 test_loss_avg=5.13243 acc=0.32812 test_acc_avg=0.13401 test_acc_top5_avg=0.92488 time=257.37it/s +epoch=53 global_step=21114 loss=0.35279 test_loss_avg=4.87096 acc=0.89062 test_acc_avg=0.18366 test_acc_top5_avg=0.84958 time=261.20it/s +epoch=53 global_step=21114 loss=4.88194 test_loss_avg=4.46629 acc=0.00000 test_acc_avg=0.23497 test_acc_top5_avg=0.87104 time=877.47it/s +curr_acc 0.2350 +BEST_ACC 0.2705 +curr_acc_top5 0.8710 +BEST_ACC_top5 0.8679 +Model Saved! + +====================Training==================== +epoch=54 global_step=21150 loss=5.23194 loss_avg=4.72025 acc=0.48438 acc_top1_avg=0.53255 acc_top5_avg=0.88759 lr=0.00100 gn=12.89937 time=59.47it/s +epoch=54 global_step=21200 loss=4.86215 loss_avg=4.80674 acc=0.50781 acc_top1_avg=0.52380 acc_top5_avg=0.88708 lr=0.00100 gn=16.27081 time=54.71it/s +epoch=54 global_step=21250 loss=4.83918 loss_avg=4.79864 acc=0.51562 acc_top1_avg=0.52401 acc_top5_avg=0.88879 lr=0.00100 gn=10.86749 time=56.31it/s +epoch=54 global_step=21300 loss=5.03720 loss_avg=4.79952 acc=0.50000 acc_top1_avg=0.52428 acc_top5_avg=0.88861 lr=0.00100 gn=14.66933 time=56.41it/s +epoch=54 global_step=21350 loss=4.78283 loss_avg=4.81349 acc=0.53906 acc_top1_avg=0.52268 acc_top5_avg=0.88758 lr=0.00100 gn=13.94069 time=58.68it/s +epoch=54 global_step=21400 loss=4.92310 loss_avg=4.80935 acc=0.51562 acc_top1_avg=0.52336 acc_top5_avg=0.88841 lr=0.00100 gn=12.46170 time=60.95it/s +epoch=54 global_step=21450 loss=4.67477 loss_avg=4.80404 acc=0.53906 acc_top1_avg=0.52411 acc_top5_avg=0.88893 lr=0.00100 gn=11.14386 time=61.02it/s +epoch=54 global_step=21500 loss=4.48808 loss_avg=4.78440 acc=0.55469 acc_top1_avg=0.52615 acc_top5_avg=0.88937 lr=0.00100 gn=10.30724 time=61.92it/s +====================Eval==================== +epoch=54 global_step=21505 loss=6.48676 test_loss_avg=5.23133 acc=0.00000 test_acc_avg=0.07973 test_acc_top5_avg=0.81756 time=259.63it/s +epoch=54 global_step=21505 loss=5.25622 test_loss_avg=4.36363 acc=0.00000 test_acc_avg=0.23616 test_acc_top5_avg=0.85354 time=889.00it/s +curr_acc 0.2362 +BEST_ACC 0.2705 +curr_acc_top5 0.8535 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=55 global_step=21550 loss=4.25969 loss_avg=4.63606 acc=0.59375 acc_top1_avg=0.54219 acc_top5_avg=0.89323 lr=0.00100 gn=13.65019 time=65.75it/s +epoch=55 global_step=21600 loss=4.63801 loss_avg=4.66626 acc=0.53906 acc_top1_avg=0.53849 acc_top5_avg=0.89367 lr=0.00100 gn=11.35434 time=65.53it/s +epoch=55 global_step=21650 loss=4.21299 loss_avg=4.71068 acc=0.59375 acc_top1_avg=0.53475 acc_top5_avg=0.89316 lr=0.00100 gn=13.27215 time=65.79it/s +epoch=55 global_step=21700 loss=4.52234 loss_avg=4.73156 acc=0.54688 acc_top1_avg=0.53209 acc_top5_avg=0.89375 lr=0.00100 gn=11.54533 time=63.64it/s +epoch=55 global_step=21750 loss=5.74545 loss_avg=4.76413 acc=0.42969 acc_top1_avg=0.52851 acc_top5_avg=0.89104 lr=0.00100 gn=13.86255 time=64.88it/s +epoch=55 global_step=21800 loss=5.43731 loss_avg=4.78007 acc=0.46875 acc_top1_avg=0.52685 acc_top5_avg=0.88967 lr=0.00100 gn=13.45658 time=61.12it/s +epoch=55 global_step=21850 loss=4.88549 loss_avg=4.77972 acc=0.52344 acc_top1_avg=0.52692 acc_top5_avg=0.89019 lr=0.00100 gn=13.91309 time=65.92it/s +====================Eval==================== +epoch=55 global_step=21896 loss=7.13409 test_loss_avg=7.03430 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.89062 time=229.56it/s +epoch=55 global_step=21896 loss=5.19429 test_loss_avg=5.25806 acc=0.23438 test_acc_avg=0.10710 test_acc_top5_avg=0.82869 time=229.94it/s +epoch=55 global_step=21896 loss=5.48907 test_loss_avg=4.34909 acc=0.00000 test_acc_avg=0.24436 test_acc_top5_avg=0.86264 time=869.83it/s +curr_acc 0.2444 +BEST_ACC 0.2705 +curr_acc_top5 0.8626 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=56 global_step=21900 loss=4.84976 loss_avg=4.83240 acc=0.51562 acc_top1_avg=0.52148 acc_top5_avg=0.90820 lr=0.00100 gn=14.01051 time=57.82it/s +epoch=56 global_step=21950 loss=5.26719 loss_avg=4.75529 acc=0.47656 acc_top1_avg=0.52980 acc_top5_avg=0.88759 lr=0.00100 gn=14.20676 time=58.90it/s +epoch=56 global_step=22000 loss=4.99451 loss_avg=4.75569 acc=0.50781 acc_top1_avg=0.52855 acc_top5_avg=0.88882 lr=0.00100 gn=15.27045 time=56.12it/s +epoch=56 global_step=22050 loss=5.63409 loss_avg=4.75512 acc=0.45312 acc_top1_avg=0.52958 acc_top5_avg=0.88860 lr=0.00100 gn=13.71339 time=61.90it/s +epoch=56 global_step=22100 loss=4.68852 loss_avg=4.74506 acc=0.52344 acc_top1_avg=0.53064 acc_top5_avg=0.88856 lr=0.00100 gn=12.03476 time=55.53it/s +epoch=56 global_step=22150 loss=4.68876 loss_avg=4.73646 acc=0.53906 acc_top1_avg=0.53165 acc_top5_avg=0.88752 lr=0.00100 gn=14.61110 time=61.61it/s +epoch=56 global_step=22200 loss=4.71023 loss_avg=4.75397 acc=0.52344 acc_top1_avg=0.52989 acc_top5_avg=0.88764 lr=0.00100 gn=12.97314 time=60.08it/s +epoch=56 global_step=22250 loss=5.66926 loss_avg=4.75359 acc=0.42969 acc_top1_avg=0.53010 acc_top5_avg=0.88826 lr=0.00100 gn=15.47693 time=57.80it/s +====================Eval==================== +epoch=56 global_step=22287 loss=6.34238 test_loss_avg=4.50536 acc=0.00000 test_acc_avg=0.15595 test_acc_top5_avg=0.90144 time=250.39it/s +epoch=56 global_step=22287 loss=5.43392 test_loss_avg=4.23451 acc=0.00781 test_acc_avg=0.26306 test_acc_top5_avg=0.84889 time=262.60it/s +epoch=56 global_step=22287 loss=5.10032 test_loss_avg=4.27564 acc=0.00000 test_acc_avg=0.25316 test_acc_top5_avg=0.84573 time=903.56it/s +curr_acc 0.2532 +BEST_ACC 0.2705 +curr_acc_top5 0.8457 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=57 global_step=22300 loss=4.74492 loss_avg=4.86221 acc=0.52344 acc_top1_avg=0.52163 acc_top5_avg=0.89303 lr=0.00100 gn=12.50319 time=62.25it/s +epoch=57 global_step=22350 loss=5.35147 loss_avg=4.74746 acc=0.45312 acc_top1_avg=0.53212 acc_top5_avg=0.88740 lr=0.00100 gn=13.82355 time=51.68it/s +epoch=57 global_step=22400 loss=4.59653 loss_avg=4.75294 acc=0.54688 acc_top1_avg=0.53042 acc_top5_avg=0.88682 lr=0.00100 gn=15.70885 time=61.83it/s +epoch=57 global_step=22450 loss=5.33654 loss_avg=4.74282 acc=0.46094 acc_top1_avg=0.53187 acc_top5_avg=0.88727 lr=0.00100 gn=11.48425 time=60.39it/s +epoch=57 global_step=22500 loss=4.32878 loss_avg=4.72049 acc=0.60156 acc_top1_avg=0.53389 acc_top5_avg=0.88685 lr=0.00100 gn=14.97792 time=65.11it/s +epoch=57 global_step=22550 loss=4.87701 loss_avg=4.73865 acc=0.50000 acc_top1_avg=0.53196 acc_top5_avg=0.88792 lr=0.00100 gn=15.73807 time=63.22it/s +epoch=57 global_step=22600 loss=4.99306 loss_avg=4.73315 acc=0.50781 acc_top1_avg=0.53245 acc_top5_avg=0.88805 lr=0.00100 gn=14.05117 time=65.79it/s +epoch=57 global_step=22650 loss=4.96366 loss_avg=4.74116 acc=0.49219 acc_top1_avg=0.53097 acc_top5_avg=0.88899 lr=0.00100 gn=14.45439 time=65.80it/s +====================Eval==================== +epoch=57 global_step=22678 loss=3.42357 test_loss_avg=5.05335 acc=0.28125 test_acc_avg=0.09525 test_acc_top5_avg=0.78923 time=261.70it/s +epoch=57 global_step=22678 loss=4.86150 test_loss_avg=4.35790 acc=0.00000 test_acc_avg=0.23576 test_acc_top5_avg=0.84998 time=876.19it/s +curr_acc 0.2358 +BEST_ACC 0.2705 +curr_acc_top5 0.8500 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=58 global_step=22700 loss=5.06432 loss_avg=4.73658 acc=0.50781 acc_top1_avg=0.53196 acc_top5_avg=0.87962 lr=0.00100 gn=14.88770 time=61.12it/s +epoch=58 global_step=22750 loss=5.31984 loss_avg=4.74304 acc=0.46875 acc_top1_avg=0.53103 acc_top5_avg=0.88650 lr=0.00100 gn=14.68831 time=61.46it/s +epoch=58 global_step=22800 loss=5.01969 loss_avg=4.74560 acc=0.53125 acc_top1_avg=0.53176 acc_top5_avg=0.88678 lr=0.00100 gn=21.18947 time=65.48it/s +epoch=58 global_step=22850 loss=4.08497 loss_avg=4.72090 acc=0.60156 acc_top1_avg=0.53370 acc_top5_avg=0.88636 lr=0.00100 gn=13.22597 time=61.68it/s +epoch=58 global_step=22900 loss=4.59851 loss_avg=4.71617 acc=0.56250 acc_top1_avg=0.53428 acc_top5_avg=0.88619 lr=0.00100 gn=20.66579 time=61.57it/s +epoch=58 global_step=22950 loss=4.23671 loss_avg=4.72067 acc=0.57812 acc_top1_avg=0.53323 acc_top5_avg=0.88686 lr=0.00100 gn=14.29264 time=62.82it/s +epoch=58 global_step=23000 loss=4.33070 loss_avg=4.72278 acc=0.54688 acc_top1_avg=0.53324 acc_top5_avg=0.88839 lr=0.00100 gn=18.92611 time=65.34it/s +epoch=58 global_step=23050 loss=4.15421 loss_avg=4.72070 acc=0.62500 acc_top1_avg=0.53383 acc_top5_avg=0.88884 lr=0.00100 gn=17.66500 time=61.64it/s +====================Eval==================== +epoch=58 global_step=23069 loss=4.79872 test_loss_avg=4.41561 acc=0.00000 test_acc_avg=0.20399 test_acc_top5_avg=0.95009 time=259.08it/s +epoch=58 global_step=23069 loss=0.36827 test_loss_avg=4.26177 acc=0.88281 test_acc_avg=0.25954 test_acc_top5_avg=0.85420 time=244.20it/s +epoch=58 global_step=23069 loss=5.10081 test_loss_avg=4.25187 acc=0.00000 test_acc_avg=0.25287 test_acc_top5_avg=0.85394 time=887.50it/s +curr_acc 0.2529 +BEST_ACC 0.2705 +curr_acc_top5 0.8539 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=59 global_step=23100 loss=4.64368 loss_avg=4.67080 acc=0.53125 acc_top1_avg=0.53957 acc_top5_avg=0.89189 lr=0.00100 gn=16.09974 time=61.52it/s +epoch=59 global_step=23150 loss=4.29239 loss_avg=4.65913 acc=0.57812 acc_top1_avg=0.54003 acc_top5_avg=0.89005 lr=0.00100 gn=13.28245 time=60.81it/s +epoch=59 global_step=23200 loss=4.83280 loss_avg=4.67518 acc=0.52344 acc_top1_avg=0.53811 acc_top5_avg=0.88907 lr=0.00100 gn=15.87256 time=65.55it/s +epoch=59 global_step=23250 loss=5.00046 loss_avg=4.71100 acc=0.50781 acc_top1_avg=0.53423 acc_top5_avg=0.88886 lr=0.00100 gn=16.28039 time=61.66it/s +epoch=59 global_step=23300 loss=4.42983 loss_avg=4.70545 acc=0.57031 acc_top1_avg=0.53517 acc_top5_avg=0.89029 lr=0.00100 gn=18.75643 time=61.61it/s +epoch=59 global_step=23350 loss=4.56344 loss_avg=4.70502 acc=0.55469 acc_top1_avg=0.53539 acc_top5_avg=0.89021 lr=0.00100 gn=15.55919 time=60.60it/s +epoch=59 global_step=23400 loss=4.60048 loss_avg=4.69984 acc=0.54688 acc_top1_avg=0.53566 acc_top5_avg=0.88992 lr=0.00100 gn=15.18013 time=55.01it/s +epoch=59 global_step=23450 loss=4.87897 loss_avg=4.69603 acc=0.52344 acc_top1_avg=0.53599 acc_top5_avg=0.88880 lr=0.00100 gn=15.00964 time=61.98it/s +====================Eval==================== +epoch=59 global_step=23460 loss=6.97285 test_loss_avg=5.74435 acc=0.00000 test_acc_avg=0.04768 test_acc_top5_avg=0.75060 time=262.00it/s +epoch=59 global_step=23460 loss=4.73365 test_loss_avg=4.45058 acc=0.00000 test_acc_avg=0.23774 test_acc_top5_avg=0.84949 time=914.59it/s +curr_acc 0.2377 +BEST_ACC 0.2705 +curr_acc_top5 0.8495 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=60 global_step=23500 loss=5.47199 loss_avg=4.65272 acc=0.45312 acc_top1_avg=0.54141 acc_top5_avg=0.88691 lr=0.00100 gn=17.54891 time=62.64it/s +epoch=60 global_step=23550 loss=4.61493 loss_avg=4.67442 acc=0.55469 acc_top1_avg=0.53889 acc_top5_avg=0.88568 lr=0.00100 gn=21.73548 time=60.04it/s +epoch=60 global_step=23600 loss=4.34043 loss_avg=4.64432 acc=0.57812 acc_top1_avg=0.54202 acc_top5_avg=0.88823 lr=0.00100 gn=18.75744 time=56.09it/s +epoch=60 global_step=23650 loss=4.39563 loss_avg=4.66166 acc=0.56250 acc_top1_avg=0.54054 acc_top5_avg=0.88812 lr=0.00100 gn=15.46287 time=52.57it/s +epoch=60 global_step=23700 loss=4.58636 loss_avg=4.67144 acc=0.56250 acc_top1_avg=0.53978 acc_top5_avg=0.88848 lr=0.00100 gn=19.13834 time=58.06it/s +epoch=60 global_step=23750 loss=4.99825 loss_avg=4.67520 acc=0.52344 acc_top1_avg=0.53901 acc_top5_avg=0.88844 lr=0.00100 gn=20.13191 time=58.11it/s +epoch=60 global_step=23800 loss=4.89414 loss_avg=4.67472 acc=0.52344 acc_top1_avg=0.53909 acc_top5_avg=0.88766 lr=0.00100 gn=20.34169 time=56.94it/s +epoch=60 global_step=23850 loss=4.92816 loss_avg=4.67852 acc=0.49219 acc_top1_avg=0.53844 acc_top5_avg=0.88786 lr=0.00100 gn=15.85164 time=61.93it/s +====================Eval==================== +epoch=60 global_step=23851 loss=1.26911 test_loss_avg=5.53285 acc=0.56250 test_acc_avg=0.12656 test_acc_top5_avg=0.89531 time=248.61it/s +epoch=60 global_step=23851 loss=0.52108 test_loss_avg=4.70947 acc=0.85938 test_acc_avg=0.19193 test_acc_top5_avg=0.82253 time=261.60it/s +epoch=60 global_step=23851 loss=5.51284 test_loss_avg=4.21823 acc=0.00000 test_acc_avg=0.26513 test_acc_top5_avg=0.84879 time=920.21it/s +curr_acc 0.2651 +BEST_ACC 0.2705 +curr_acc_top5 0.8488 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=61 global_step=23900 loss=4.56921 loss_avg=4.61330 acc=0.55469 acc_top1_avg=0.54672 acc_top5_avg=0.89078 lr=0.00100 gn=16.72365 time=65.63it/s +epoch=61 global_step=23950 loss=4.51172 loss_avg=4.64323 acc=0.54688 acc_top1_avg=0.54309 acc_top5_avg=0.88952 lr=0.00100 gn=13.85193 time=65.90it/s +epoch=61 global_step=24000 loss=4.56641 loss_avg=4.64973 acc=0.56250 acc_top1_avg=0.54242 acc_top5_avg=0.88764 lr=0.00100 gn=15.87770 time=65.79it/s +epoch=61 global_step=24050 loss=4.82947 loss_avg=4.62882 acc=0.53906 acc_top1_avg=0.54483 acc_top5_avg=0.88760 lr=0.00100 gn=20.78681 time=65.71it/s +epoch=61 global_step=24100 loss=4.45538 loss_avg=4.63373 acc=0.57031 acc_top1_avg=0.54418 acc_top5_avg=0.88736 lr=0.00100 gn=19.17201 time=65.72it/s +epoch=61 global_step=24150 loss=4.67204 loss_avg=4.63244 acc=0.55469 acc_top1_avg=0.54411 acc_top5_avg=0.88665 lr=0.00100 gn=18.95516 time=65.98it/s +epoch=61 global_step=24200 loss=5.15052 loss_avg=4.64642 acc=0.47656 acc_top1_avg=0.54276 acc_top5_avg=0.88608 lr=0.00100 gn=17.08731 time=64.47it/s +====================Eval==================== +epoch=61 global_step=24242 loss=6.51630 test_loss_avg=5.51012 acc=0.00000 test_acc_avg=0.05771 test_acc_top5_avg=0.87651 time=262.87it/s +epoch=61 global_step=24242 loss=5.03915 test_loss_avg=4.57027 acc=0.00000 test_acc_avg=0.21924 test_acc_top5_avg=0.85275 time=893.74it/s +curr_acc 0.2192 +BEST_ACC 0.2705 +curr_acc_top5 0.8527 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=62 global_step=24250 loss=4.38389 loss_avg=4.43125 acc=0.57031 acc_top1_avg=0.56836 acc_top5_avg=0.89648 lr=0.00100 gn=16.64855 time=58.27it/s +epoch=62 global_step=24300 loss=4.32110 loss_avg=4.59944 acc=0.59375 acc_top1_avg=0.54836 acc_top5_avg=0.89265 lr=0.00100 gn=18.75378 time=56.34it/s +epoch=62 global_step=24350 loss=4.68919 loss_avg=4.59436 acc=0.53906 acc_top1_avg=0.54854 acc_top5_avg=0.89106 lr=0.00100 gn=19.84907 time=62.01it/s +epoch=62 global_step=24400 loss=3.67457 loss_avg=4.62948 acc=0.64844 acc_top1_avg=0.54490 acc_top5_avg=0.88875 lr=0.00100 gn=18.34102 time=55.24it/s +epoch=62 global_step=24450 loss=4.42907 loss_avg=4.62164 acc=0.57812 acc_top1_avg=0.54545 acc_top5_avg=0.88912 lr=0.00100 gn=22.24341 time=56.13it/s +epoch=62 global_step=24500 loss=4.23844 loss_avg=4.63882 acc=0.60156 acc_top1_avg=0.54363 acc_top5_avg=0.88757 lr=0.00100 gn=17.46411 time=64.90it/s +epoch=62 global_step=24550 loss=4.22639 loss_avg=4.62123 acc=0.57031 acc_top1_avg=0.54558 acc_top5_avg=0.88809 lr=0.00100 gn=14.35750 time=63.01it/s +epoch=62 global_step=24600 loss=5.08715 loss_avg=4.63810 acc=0.50781 acc_top1_avg=0.54354 acc_top5_avg=0.88678 lr=0.00100 gn=21.31936 time=57.98it/s +====================Eval==================== +epoch=62 global_step=24633 loss=6.89905 test_loss_avg=6.94463 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.87500 time=224.29it/s +epoch=62 global_step=24633 loss=7.39220 test_loss_avg=5.14788 acc=0.00000 test_acc_avg=0.12891 test_acc_top5_avg=0.81220 time=261.21it/s +epoch=62 global_step=24633 loss=5.17350 test_loss_avg=4.31964 acc=0.00000 test_acc_avg=0.25564 test_acc_top5_avg=0.84850 time=922.84it/s +curr_acc 0.2556 +BEST_ACC 0.2705 +curr_acc_top5 0.8485 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=63 global_step=24650 loss=4.82760 loss_avg=4.47265 acc=0.53125 acc_top1_avg=0.56250 acc_top5_avg=0.90028 lr=0.00100 gn=22.63916 time=65.58it/s +epoch=63 global_step=24700 loss=4.78382 loss_avg=4.50935 acc=0.53906 acc_top1_avg=0.55819 acc_top5_avg=0.89191 lr=0.00100 gn=18.90076 time=61.18it/s +epoch=63 global_step=24750 loss=4.49900 loss_avg=4.56234 acc=0.55469 acc_top1_avg=0.55255 acc_top5_avg=0.88709 lr=0.00100 gn=21.95427 time=65.83it/s +epoch=63 global_step=24800 loss=4.84234 loss_avg=4.57071 acc=0.51562 acc_top1_avg=0.55174 acc_top5_avg=0.88478 lr=0.00100 gn=17.86562 time=65.77it/s +epoch=63 global_step=24850 loss=4.32096 loss_avg=4.59537 acc=0.56250 acc_top1_avg=0.54889 acc_top5_avg=0.88454 lr=0.00100 gn=15.87595 time=65.67it/s +epoch=63 global_step=24900 loss=4.63197 loss_avg=4.60751 acc=0.54688 acc_top1_avg=0.54758 acc_top5_avg=0.88469 lr=0.00100 gn=20.17656 time=65.85it/s +epoch=63 global_step=24950 loss=4.70470 loss_avg=4.61555 acc=0.53906 acc_top1_avg=0.54648 acc_top5_avg=0.88486 lr=0.00100 gn=18.00769 time=64.80it/s +epoch=63 global_step=25000 loss=4.68006 loss_avg=4.60942 acc=0.53906 acc_top1_avg=0.54702 acc_top5_avg=0.88503 lr=0.00100 gn=17.87690 time=65.70it/s +====================Eval==================== +epoch=63 global_step=25024 loss=4.61049 test_loss_avg=4.89524 acc=0.00000 test_acc_avg=0.09545 test_acc_top5_avg=0.93750 time=254.54it/s +epoch=63 global_step=25024 loss=5.24338 test_loss_avg=4.26177 acc=0.00781 test_acc_avg=0.25578 test_acc_top5_avg=0.85445 time=263.41it/s +epoch=63 global_step=25024 loss=5.46879 test_loss_avg=4.34847 acc=0.00000 test_acc_avg=0.23675 test_acc_top5_avg=0.85502 time=913.59it/s +curr_acc 0.2367 +BEST_ACC 0.2705 +curr_acc_top5 0.8550 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=64 global_step=25050 loss=4.60426 loss_avg=4.59484 acc=0.55469 acc_top1_avg=0.54838 acc_top5_avg=0.88762 lr=0.00100 gn=18.57406 time=65.79it/s +epoch=64 global_step=25100 loss=4.87123 loss_avg=4.51422 acc=0.50000 acc_top1_avg=0.55705 acc_top5_avg=0.88579 lr=0.00100 gn=19.32231 time=65.53it/s +epoch=64 global_step=25150 loss=4.17039 loss_avg=4.53688 acc=0.60938 acc_top1_avg=0.55432 acc_top5_avg=0.88597 lr=0.00100 gn=20.53714 time=65.71it/s +epoch=64 global_step=25200 loss=4.78687 loss_avg=4.53984 acc=0.53906 acc_top1_avg=0.55433 acc_top5_avg=0.88663 lr=0.00100 gn=17.26615 time=51.06it/s +epoch=64 global_step=25250 loss=4.59526 loss_avg=4.53611 acc=0.54688 acc_top1_avg=0.55448 acc_top5_avg=0.88630 lr=0.00100 gn=17.09144 time=64.98it/s +epoch=64 global_step=25300 loss=4.56606 loss_avg=4.57327 acc=0.53906 acc_top1_avg=0.55033 acc_top5_avg=0.88627 lr=0.00100 gn=16.91350 time=61.68it/s +epoch=64 global_step=25350 loss=4.98006 loss_avg=4.59401 acc=0.50781 acc_top1_avg=0.54815 acc_top5_avg=0.88545 lr=0.00100 gn=21.17318 time=64.30it/s +epoch=64 global_step=25400 loss=4.86560 loss_avg=4.59711 acc=0.51562 acc_top1_avg=0.54827 acc_top5_avg=0.88529 lr=0.00100 gn=20.68055 time=59.25it/s +====================Eval==================== +epoch=64 global_step=25415 loss=2.42284 test_loss_avg=5.23773 acc=0.41406 test_acc_avg=0.09428 test_acc_top5_avg=0.77628 time=260.65it/s +epoch=64 global_step=25415 loss=4.93845 test_loss_avg=4.35665 acc=0.00000 test_acc_avg=0.23873 test_acc_top5_avg=0.85957 time=920.21it/s +curr_acc 0.2387 +BEST_ACC 0.2705 +curr_acc_top5 0.8596 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=65 global_step=25450 loss=4.66807 loss_avg=4.59778 acc=0.54688 acc_top1_avg=0.54531 acc_top5_avg=0.88192 lr=0.00100 gn=18.80093 time=64.08it/s +epoch=65 global_step=25500 loss=4.96994 loss_avg=4.60825 acc=0.51562 acc_top1_avg=0.54522 acc_top5_avg=0.88061 lr=0.00100 gn=17.48819 time=62.05it/s +epoch=65 global_step=25550 loss=5.10456 loss_avg=4.59460 acc=0.48438 acc_top1_avg=0.54699 acc_top5_avg=0.88131 lr=0.00100 gn=15.18319 time=63.50it/s +epoch=65 global_step=25600 loss=4.42700 loss_avg=4.61025 acc=0.57031 acc_top1_avg=0.54531 acc_top5_avg=0.88184 lr=0.00100 gn=20.33535 time=59.07it/s +epoch=65 global_step=25650 loss=4.23953 loss_avg=4.60024 acc=0.57031 acc_top1_avg=0.54727 acc_top5_avg=0.88118 lr=0.00100 gn=22.09320 time=59.34it/s +epoch=65 global_step=25700 loss=4.50857 loss_avg=4.58846 acc=0.55469 acc_top1_avg=0.54833 acc_top5_avg=0.88322 lr=0.00100 gn=17.21991 time=56.88it/s +epoch=65 global_step=25750 loss=5.29128 loss_avg=4.58884 acc=0.47656 acc_top1_avg=0.54839 acc_top5_avg=0.88465 lr=0.00100 gn=20.18107 time=62.10it/s +epoch=65 global_step=25800 loss=5.18752 loss_avg=4.59191 acc=0.49219 acc_top1_avg=0.54817 acc_top5_avg=0.88452 lr=0.00100 gn=20.31845 time=61.71it/s +====================Eval==================== +epoch=65 global_step=25806 loss=2.31422 test_loss_avg=4.66270 acc=0.45312 test_acc_avg=0.21615 test_acc_top5_avg=0.91771 time=248.96it/s +epoch=65 global_step=25806 loss=0.15865 test_loss_avg=4.49335 acc=0.96094 test_acc_avg=0.22043 test_acc_top5_avg=0.84976 time=256.56it/s +epoch=65 global_step=25806 loss=5.33514 test_loss_avg=4.33685 acc=0.00000 test_acc_avg=0.24545 test_acc_top5_avg=0.85483 time=902.78it/s +curr_acc 0.2455 +BEST_ACC 0.2705 +curr_acc_top5 0.8548 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=66 global_step=25850 loss=4.17502 loss_avg=4.51231 acc=0.59375 acc_top1_avg=0.55611 acc_top5_avg=0.88956 lr=0.00100 gn=18.16734 time=59.93it/s +epoch=66 global_step=25900 loss=4.42434 loss_avg=4.49877 acc=0.57812 acc_top1_avg=0.55934 acc_top5_avg=0.88431 lr=0.00100 gn=19.31968 time=58.19it/s +epoch=66 global_step=25950 loss=4.59436 loss_avg=4.53140 acc=0.52344 acc_top1_avg=0.55577 acc_top5_avg=0.88422 lr=0.00100 gn=21.09987 time=61.61it/s +epoch=66 global_step=26000 loss=4.45300 loss_avg=4.55259 acc=0.57812 acc_top1_avg=0.55360 acc_top5_avg=0.88382 lr=0.00100 gn=22.34800 time=51.59it/s +epoch=66 global_step=26050 loss=4.68632 loss_avg=4.57064 acc=0.53125 acc_top1_avg=0.55133 acc_top5_avg=0.88214 lr=0.00100 gn=23.23705 time=64.99it/s +epoch=66 global_step=26100 loss=4.43000 loss_avg=4.58796 acc=0.56250 acc_top1_avg=0.54940 acc_top5_avg=0.88233 lr=0.00100 gn=22.16573 time=63.57it/s +epoch=66 global_step=26150 loss=4.56704 loss_avg=4.57874 acc=0.54688 acc_top1_avg=0.55040 acc_top5_avg=0.88320 lr=0.00100 gn=19.10479 time=65.50it/s +====================Eval==================== +epoch=66 global_step=26197 loss=6.58922 test_loss_avg=5.09860 acc=0.00000 test_acc_avg=0.09440 test_acc_top5_avg=0.77344 time=254.93it/s +epoch=66 global_step=26197 loss=5.09170 test_loss_avg=4.15965 acc=0.00000 test_acc_avg=0.25485 test_acc_top5_avg=0.84573 time=913.39it/s +curr_acc 0.2548 +BEST_ACC 0.2705 +curr_acc_top5 0.8457 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=67 global_step=26200 loss=4.49079 loss_avg=4.72441 acc=0.55469 acc_top1_avg=0.53385 acc_top5_avg=0.88542 lr=0.00100 gn=23.12096 time=59.72it/s +epoch=67 global_step=26250 loss=4.32915 loss_avg=4.50259 acc=0.57031 acc_top1_avg=0.55911 acc_top5_avg=0.88723 lr=0.00100 gn=25.44015 time=60.19it/s +epoch=67 global_step=26300 loss=4.96062 loss_avg=4.52799 acc=0.50781 acc_top1_avg=0.55651 acc_top5_avg=0.88562 lr=0.00100 gn=20.78530 time=56.72it/s +epoch=67 global_step=26350 loss=4.55836 loss_avg=4.53726 acc=0.53906 acc_top1_avg=0.55556 acc_top5_avg=0.88271 lr=0.00100 gn=14.63356 time=51.70it/s +epoch=67 global_step=26400 loss=4.43922 loss_avg=4.54971 acc=0.55469 acc_top1_avg=0.55373 acc_top5_avg=0.88293 lr=0.00100 gn=20.42217 time=57.08it/s +epoch=67 global_step=26450 loss=4.95658 loss_avg=4.54152 acc=0.53125 acc_top1_avg=0.55497 acc_top5_avg=0.88164 lr=0.00100 gn=21.16676 time=65.45it/s +epoch=67 global_step=26500 loss=5.01558 loss_avg=4.54547 acc=0.50000 acc_top1_avg=0.55446 acc_top5_avg=0.88188 lr=0.00100 gn=23.33525 time=61.85it/s +epoch=67 global_step=26550 loss=4.21740 loss_avg=4.55067 acc=0.58594 acc_top1_avg=0.55349 acc_top5_avg=0.88255 lr=0.00100 gn=22.27402 time=63.14it/s +====================Eval==================== +epoch=67 global_step=26588 loss=6.89051 test_loss_avg=7.11699 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.91295 time=254.40it/s +epoch=67 global_step=26588 loss=0.79582 test_loss_avg=5.19488 acc=0.78125 test_acc_avg=0.12226 test_acc_top5_avg=0.83991 time=260.55it/s +epoch=67 global_step=26588 loss=5.11653 test_loss_avg=4.39678 acc=0.00000 test_acc_avg=0.23348 test_acc_top5_avg=0.85295 time=916.79it/s +curr_acc 0.2335 +BEST_ACC 0.2705 +curr_acc_top5 0.8529 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=68 global_step=26600 loss=4.42820 loss_avg=4.59178 acc=0.56250 acc_top1_avg=0.54622 acc_top5_avg=0.88411 lr=0.00100 gn=22.01102 time=65.49it/s +epoch=68 global_step=26650 loss=4.96133 loss_avg=4.56372 acc=0.50000 acc_top1_avg=0.55078 acc_top5_avg=0.87525 lr=0.00100 gn=20.91836 time=55.95it/s +epoch=68 global_step=26700 loss=4.32070 loss_avg=4.53943 acc=0.57031 acc_top1_avg=0.55392 acc_top5_avg=0.88121 lr=0.00100 gn=23.96619 time=56.36it/s +epoch=68 global_step=26750 loss=4.05507 loss_avg=4.55957 acc=0.60156 acc_top1_avg=0.55242 acc_top5_avg=0.88339 lr=0.00100 gn=24.96323 time=61.63it/s +epoch=68 global_step=26800 loss=4.15336 loss_avg=4.56992 acc=0.60156 acc_top1_avg=0.55189 acc_top5_avg=0.88359 lr=0.00100 gn=18.80311 time=62.34it/s +epoch=68 global_step=26850 loss=4.38113 loss_avg=4.54461 acc=0.57812 acc_top1_avg=0.55475 acc_top5_avg=0.88451 lr=0.00100 gn=24.31127 time=62.91it/s +epoch=68 global_step=26900 loss=4.63637 loss_avg=4.53490 acc=0.53906 acc_top1_avg=0.55564 acc_top5_avg=0.88479 lr=0.00100 gn=24.81221 time=62.90it/s +epoch=68 global_step=26950 loss=4.17586 loss_avg=4.54108 acc=0.59375 acc_top1_avg=0.55495 acc_top5_avg=0.88311 lr=0.00100 gn=18.60618 time=52.72it/s +====================Eval==================== +epoch=68 global_step=26979 loss=6.52332 test_loss_avg=4.81407 acc=0.00000 test_acc_avg=0.14062 test_acc_top5_avg=0.88365 time=250.98it/s +epoch=68 global_step=26979 loss=5.49983 test_loss_avg=4.25020 acc=0.00781 test_acc_avg=0.25321 test_acc_top5_avg=0.83804 time=263.64it/s +epoch=68 global_step=26979 loss=5.30360 test_loss_avg=4.26353 acc=0.00000 test_acc_avg=0.25000 test_acc_top5_avg=0.83930 time=883.94it/s +curr_acc 0.2500 +BEST_ACC 0.2705 +curr_acc_top5 0.8393 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=69 global_step=27000 loss=4.45137 loss_avg=4.46001 acc=0.57031 acc_top1_avg=0.56362 acc_top5_avg=0.88988 lr=0.00100 gn=20.84560 time=57.37it/s +epoch=69 global_step=27050 loss=5.09968 loss_avg=4.55403 acc=0.48438 acc_top1_avg=0.55414 acc_top5_avg=0.88138 lr=0.00100 gn=18.04011 time=61.09it/s +epoch=69 global_step=27100 loss=4.10532 loss_avg=4.51355 acc=0.60156 acc_top1_avg=0.55830 acc_top5_avg=0.88333 lr=0.00100 gn=15.14238 time=59.08it/s +epoch=69 global_step=27150 loss=4.66550 loss_avg=4.52048 acc=0.53906 acc_top1_avg=0.55720 acc_top5_avg=0.88373 lr=0.00100 gn=23.63378 time=60.89it/s +epoch=69 global_step=27200 loss=4.30294 loss_avg=4.52008 acc=0.57031 acc_top1_avg=0.55713 acc_top5_avg=0.88225 lr=0.00100 gn=21.78460 time=60.13it/s +epoch=69 global_step=27250 loss=4.74453 loss_avg=4.53193 acc=0.53906 acc_top1_avg=0.55558 acc_top5_avg=0.88281 lr=0.00100 gn=19.14141 time=51.25it/s +epoch=69 global_step=27300 loss=3.85901 loss_avg=4.52356 acc=0.62500 acc_top1_avg=0.55634 acc_top5_avg=0.88325 lr=0.00100 gn=24.91859 time=64.95it/s +epoch=69 global_step=27350 loss=3.75797 loss_avg=4.52438 acc=0.63281 acc_top1_avg=0.55644 acc_top5_avg=0.88279 lr=0.00100 gn=22.18150 time=64.08it/s +====================Eval==================== +epoch=69 global_step=27370 loss=6.83267 test_loss_avg=5.01067 acc=0.00000 test_acc_avg=0.11368 test_acc_top5_avg=0.78157 time=251.25it/s +epoch=69 global_step=27370 loss=5.67152 test_loss_avg=4.29012 acc=0.00000 test_acc_avg=0.24515 test_acc_top5_avg=0.84988 time=898.33it/s +curr_acc 0.2452 +BEST_ACC 0.2705 +curr_acc_top5 0.8499 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=70 global_step=27400 loss=4.66605 loss_avg=4.41691 acc=0.55469 acc_top1_avg=0.57005 acc_top5_avg=0.88229 lr=0.00100 gn=20.81949 time=61.71it/s +epoch=70 global_step=27450 loss=4.35694 loss_avg=4.47910 acc=0.55469 acc_top1_avg=0.56182 acc_top5_avg=0.88271 lr=0.00100 gn=26.38276 time=53.77it/s +epoch=70 global_step=27500 loss=4.10116 loss_avg=4.45507 acc=0.60938 acc_top1_avg=0.56424 acc_top5_avg=0.88221 lr=0.00100 gn=24.19010 time=59.85it/s +epoch=70 global_step=27550 loss=3.92871 loss_avg=4.47019 acc=0.61719 acc_top1_avg=0.56228 acc_top5_avg=0.88294 lr=0.00100 gn=24.48129 time=57.54it/s +epoch=70 global_step=27600 loss=3.77525 loss_avg=4.47022 acc=0.65625 acc_top1_avg=0.56274 acc_top5_avg=0.88336 lr=0.00100 gn=25.76668 time=56.05it/s +epoch=70 global_step=27650 loss=4.67888 loss_avg=4.47673 acc=0.55469 acc_top1_avg=0.56189 acc_top5_avg=0.88267 lr=0.00100 gn=23.43336 time=60.02it/s +epoch=70 global_step=27700 loss=4.19082 loss_avg=4.48662 acc=0.60156 acc_top1_avg=0.56122 acc_top5_avg=0.88274 lr=0.00100 gn=21.42131 time=59.69it/s +epoch=70 global_step=27750 loss=5.26466 loss_avg=4.49743 acc=0.47656 acc_top1_avg=0.55964 acc_top5_avg=0.88197 lr=0.00100 gn=22.39352 time=62.17it/s +====================Eval==================== +epoch=70 global_step=27761 loss=5.02798 test_loss_avg=4.72909 acc=0.00000 test_acc_avg=0.15273 test_acc_top5_avg=0.92539 time=238.95it/s +epoch=70 global_step=27761 loss=0.20823 test_loss_avg=4.26730 acc=0.95312 test_acc_avg=0.26350 test_acc_top5_avg=0.83817 time=263.68it/s +epoch=70 global_step=27761 loss=5.29433 test_loss_avg=4.35388 acc=0.00000 test_acc_avg=0.23892 test_acc_top5_avg=0.83732 time=920.41it/s +curr_acc 0.2389 +BEST_ACC 0.2705 +curr_acc_top5 0.8373 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=71 global_step=27800 loss=4.75082 loss_avg=4.52869 acc=0.53125 acc_top1_avg=0.55649 acc_top5_avg=0.87580 lr=0.00100 gn=26.56545 time=60.43it/s +epoch=71 global_step=27850 loss=4.15936 loss_avg=4.48352 acc=0.59375 acc_top1_avg=0.56110 acc_top5_avg=0.88272 lr=0.00100 gn=27.56172 time=61.77it/s +epoch=71 global_step=27900 loss=4.01969 loss_avg=4.45686 acc=0.60156 acc_top1_avg=0.56447 acc_top5_avg=0.88225 lr=0.00100 gn=24.03763 time=62.08it/s +epoch=71 global_step=27950 loss=4.46041 loss_avg=4.45171 acc=0.55469 acc_top1_avg=0.56498 acc_top5_avg=0.88215 lr=0.00100 gn=25.42974 time=61.59it/s +epoch=71 global_step=28000 loss=4.80108 loss_avg=4.47598 acc=0.53125 acc_top1_avg=0.56230 acc_top5_avg=0.88200 lr=0.00100 gn=23.28737 time=57.47it/s +epoch=71 global_step=28050 loss=4.58990 loss_avg=4.48216 acc=0.54688 acc_top1_avg=0.56196 acc_top5_avg=0.88208 lr=0.00100 gn=24.42877 time=64.12it/s +epoch=71 global_step=28100 loss=4.58182 loss_avg=4.48698 acc=0.55469 acc_top1_avg=0.56176 acc_top5_avg=0.88187 lr=0.00100 gn=25.04009 time=63.57it/s +epoch=71 global_step=28150 loss=4.37261 loss_avg=4.49101 acc=0.57031 acc_top1_avg=0.56119 acc_top5_avg=0.88181 lr=0.00100 gn=22.02445 time=61.98it/s +====================Eval==================== +epoch=71 global_step=28152 loss=2.93558 test_loss_avg=5.25558 acc=0.33594 test_acc_avg=0.06898 test_acc_top5_avg=0.73571 time=259.05it/s +epoch=71 global_step=28152 loss=4.89176 test_loss_avg=4.23292 acc=0.00000 test_acc_avg=0.23111 test_acc_top5_avg=0.83070 time=899.49it/s +curr_acc 0.2311 +BEST_ACC 0.2705 +curr_acc_top5 0.8307 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=72 global_step=28200 loss=4.78389 loss_avg=4.44789 acc=0.51562 acc_top1_avg=0.56787 acc_top5_avg=0.88395 lr=0.00100 gn=26.67453 time=60.57it/s +epoch=72 global_step=28250 loss=5.22008 loss_avg=4.46111 acc=0.48438 acc_top1_avg=0.56649 acc_top5_avg=0.87875 lr=0.00100 gn=30.96281 time=61.67it/s +epoch=72 global_step=28300 loss=4.29578 loss_avg=4.45333 acc=0.57812 acc_top1_avg=0.56678 acc_top5_avg=0.87965 lr=0.00100 gn=30.21061 time=52.61it/s +epoch=72 global_step=28350 loss=3.55719 loss_avg=4.47115 acc=0.65625 acc_top1_avg=0.56396 acc_top5_avg=0.87918 lr=0.00100 gn=27.41764 time=58.89it/s +epoch=72 global_step=28400 loss=3.87254 loss_avg=4.47280 acc=0.63281 acc_top1_avg=0.56354 acc_top5_avg=0.88032 lr=0.00100 gn=19.57493 time=58.96it/s +epoch=72 global_step=28450 loss=4.71190 loss_avg=4.47238 acc=0.53906 acc_top1_avg=0.56326 acc_top5_avg=0.88192 lr=0.00100 gn=22.95505 time=61.78it/s +epoch=72 global_step=28500 loss=3.77579 loss_avg=4.46733 acc=0.64844 acc_top1_avg=0.56342 acc_top5_avg=0.88214 lr=0.00100 gn=30.88242 time=56.40it/s +====================Eval==================== +epoch=72 global_step=28543 loss=1.95827 test_loss_avg=5.07939 acc=0.47656 test_acc_avg=0.15755 test_acc_top5_avg=0.89193 time=250.60it/s +epoch=72 global_step=28543 loss=0.77748 test_loss_avg=4.56543 acc=0.79688 test_acc_avg=0.20804 test_acc_top5_avg=0.82397 time=258.08it/s +epoch=72 global_step=28543 loss=5.25315 test_loss_avg=4.17540 acc=0.00000 test_acc_avg=0.26315 test_acc_top5_avg=0.84059 time=917.99it/s +curr_acc 0.2632 +BEST_ACC 0.2705 +curr_acc_top5 0.8406 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=73 global_step=28550 loss=4.15832 loss_avg=4.36984 acc=0.59375 acc_top1_avg=0.56473 acc_top5_avg=0.88281 lr=0.00100 gn=16.13399 time=65.05it/s +epoch=73 global_step=28600 loss=4.36366 loss_avg=4.42174 acc=0.57812 acc_top1_avg=0.56716 acc_top5_avg=0.87829 lr=0.00100 gn=25.96436 time=58.29it/s +epoch=73 global_step=28650 loss=4.91403 loss_avg=4.42775 acc=0.51562 acc_top1_avg=0.56637 acc_top5_avg=0.87814 lr=0.00100 gn=21.81760 time=61.34it/s +epoch=73 global_step=28700 loss=4.69024 loss_avg=4.42808 acc=0.54688 acc_top1_avg=0.56658 acc_top5_avg=0.87863 lr=0.00100 gn=22.62966 time=58.33it/s +epoch=73 global_step=28750 loss=4.52580 loss_avg=4.42310 acc=0.55469 acc_top1_avg=0.56771 acc_top5_avg=0.88051 lr=0.00100 gn=26.59402 time=52.74it/s +epoch=73 global_step=28800 loss=4.13381 loss_avg=4.41971 acc=0.58594 acc_top1_avg=0.56770 acc_top5_avg=0.88123 lr=0.00100 gn=19.88444 time=61.87it/s +epoch=73 global_step=28850 loss=4.46467 loss_avg=4.42621 acc=0.54688 acc_top1_avg=0.56741 acc_top5_avg=0.88159 lr=0.00100 gn=22.24562 time=55.95it/s +epoch=73 global_step=28900 loss=4.86423 loss_avg=4.43540 acc=0.52344 acc_top1_avg=0.56668 acc_top5_avg=0.88076 lr=0.00100 gn=26.76416 time=55.57it/s +====================Eval==================== +epoch=73 global_step=28934 loss=6.16727 test_loss_avg=5.30997 acc=0.00000 test_acc_avg=0.08262 test_acc_top5_avg=0.82292 time=262.21it/s +epoch=73 global_step=28934 loss=5.53597 test_loss_avg=4.30930 acc=0.00000 test_acc_avg=0.23447 test_acc_top5_avg=0.84227 time=890.89it/s +curr_acc 0.2345 +BEST_ACC 0.2705 +curr_acc_top5 0.8423 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=74 global_step=28950 loss=4.68265 loss_avg=4.40852 acc=0.53906 acc_top1_avg=0.57178 acc_top5_avg=0.87305 lr=0.00100 gn=25.38665 time=55.40it/s +epoch=74 global_step=29000 loss=4.18081 loss_avg=4.44124 acc=0.60156 acc_top1_avg=0.56558 acc_top5_avg=0.88045 lr=0.00100 gn=20.89318 time=58.77it/s +epoch=74 global_step=29050 loss=4.37221 loss_avg=4.41667 acc=0.57031 acc_top1_avg=0.56917 acc_top5_avg=0.88416 lr=0.00100 gn=21.05804 time=61.38it/s +epoch=74 global_step=29100 loss=4.68357 loss_avg=4.42941 acc=0.53906 acc_top1_avg=0.56772 acc_top5_avg=0.88380 lr=0.00100 gn=31.27968 time=63.25it/s +epoch=74 global_step=29150 loss=4.60543 loss_avg=4.43379 acc=0.54688 acc_top1_avg=0.56764 acc_top5_avg=0.88205 lr=0.00100 gn=25.70447 time=62.35it/s +epoch=74 global_step=29200 loss=4.98509 loss_avg=4.42483 acc=0.51562 acc_top1_avg=0.56870 acc_top5_avg=0.88284 lr=0.00100 gn=29.54982 time=57.27it/s +epoch=74 global_step=29250 loss=3.97312 loss_avg=4.43468 acc=0.60156 acc_top1_avg=0.56735 acc_top5_avg=0.88252 lr=0.00100 gn=31.29459 time=62.73it/s +epoch=74 global_step=29300 loss=3.97215 loss_avg=4.44306 acc=0.61719 acc_top1_avg=0.56662 acc_top5_avg=0.88140 lr=0.00100 gn=28.51997 time=65.38it/s +====================Eval==================== +epoch=74 global_step=29325 loss=6.71290 test_loss_avg=6.61798 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.83203 time=260.22it/s +epoch=74 global_step=29325 loss=6.69694 test_loss_avg=5.22199 acc=0.00000 test_acc_avg=0.09404 test_acc_top5_avg=0.77749 time=259.10it/s +epoch=74 global_step=29325 loss=4.92671 test_loss_avg=4.23971 acc=0.00000 test_acc_avg=0.24496 test_acc_top5_avg=0.82704 time=885.06it/s +curr_acc 0.2450 +BEST_ACC 0.2705 +curr_acc_top5 0.8270 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=75 global_step=29350 loss=3.80904 loss_avg=4.29687 acc=0.63281 acc_top1_avg=0.58313 acc_top5_avg=0.89344 lr=0.00100 gn=27.97623 time=56.41it/s +epoch=75 global_step=29400 loss=4.65965 loss_avg=4.37279 acc=0.53125 acc_top1_avg=0.57635 acc_top5_avg=0.88250 lr=0.00100 gn=32.57146 time=64.81it/s +epoch=75 global_step=29450 loss=4.10713 loss_avg=4.35533 acc=0.60156 acc_top1_avg=0.57712 acc_top5_avg=0.88238 lr=0.00100 gn=26.71635 time=62.51it/s +epoch=75 global_step=29500 loss=4.57185 loss_avg=4.38452 acc=0.53906 acc_top1_avg=0.57335 acc_top5_avg=0.88098 lr=0.00100 gn=30.95038 time=56.40it/s +epoch=75 global_step=29550 loss=3.97974 loss_avg=4.39210 acc=0.62500 acc_top1_avg=0.57247 acc_top5_avg=0.88021 lr=0.00100 gn=33.49480 time=59.59it/s +epoch=75 global_step=29600 loss=4.73801 loss_avg=4.41571 acc=0.53125 acc_top1_avg=0.57006 acc_top5_avg=0.87935 lr=0.00100 gn=20.72843 time=56.38it/s +epoch=75 global_step=29650 loss=3.68780 loss_avg=4.42432 acc=0.67188 acc_top1_avg=0.56909 acc_top5_avg=0.87938 lr=0.00100 gn=27.95403 time=61.54it/s +epoch=75 global_step=29700 loss=3.97146 loss_avg=4.43330 acc=0.62500 acc_top1_avg=0.56779 acc_top5_avg=0.87994 lr=0.00100 gn=28.70622 time=60.88it/s +====================Eval==================== +epoch=75 global_step=29716 loss=6.47471 test_loss_avg=4.56242 acc=0.00000 test_acc_avg=0.15812 test_acc_top5_avg=0.91312 time=241.02it/s +epoch=75 global_step=29716 loss=5.19454 test_loss_avg=4.12849 acc=0.00000 test_acc_avg=0.26740 test_acc_top5_avg=0.84760 time=264.34it/s +epoch=75 global_step=29716 loss=5.58624 test_loss_avg=4.18780 acc=0.00000 test_acc_avg=0.25425 test_acc_top5_avg=0.84850 time=913.39it/s +curr_acc 0.2543 +BEST_ACC 0.2705 +curr_acc_top5 0.8485 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=76 global_step=29750 loss=4.79405 loss_avg=4.45832 acc=0.53125 acc_top1_avg=0.56457 acc_top5_avg=0.87592 lr=0.00100 gn=21.82518 time=59.24it/s +epoch=76 global_step=29800 loss=4.52581 loss_avg=4.43692 acc=0.55469 acc_top1_avg=0.56743 acc_top5_avg=0.88132 lr=0.00100 gn=24.17717 time=61.48it/s +epoch=76 global_step=29850 loss=4.83365 loss_avg=4.39560 acc=0.53125 acc_top1_avg=0.57212 acc_top5_avg=0.88252 lr=0.00100 gn=24.14213 time=61.45it/s +epoch=76 global_step=29900 loss=4.29234 loss_avg=4.39791 acc=0.57812 acc_top1_avg=0.57269 acc_top5_avg=0.88290 lr=0.00100 gn=33.57017 time=58.10it/s +epoch=76 global_step=29950 loss=3.40025 loss_avg=4.41432 acc=0.68750 acc_top1_avg=0.57051 acc_top5_avg=0.88078 lr=0.00100 gn=23.76767 time=54.41it/s +epoch=76 global_step=30000 loss=4.04497 loss_avg=4.41660 acc=0.60938 acc_top1_avg=0.57042 acc_top5_avg=0.88045 lr=0.00100 gn=25.28838 time=61.72it/s +epoch=76 global_step=30050 loss=4.61462 loss_avg=4.42984 acc=0.54688 acc_top1_avg=0.56889 acc_top5_avg=0.87930 lr=0.00100 gn=26.35123 time=59.52it/s +epoch=76 global_step=30100 loss=4.72538 loss_avg=4.42369 acc=0.54688 acc_top1_avg=0.56974 acc_top5_avg=0.87984 lr=0.00100 gn=25.92263 time=62.07it/s +====================Eval==================== +epoch=76 global_step=30107 loss=2.95674 test_loss_avg=4.86212 acc=0.27344 test_acc_avg=0.12942 test_acc_top5_avg=0.76715 time=252.76it/s +epoch=76 global_step=30107 loss=5.04010 test_loss_avg=4.17783 acc=0.00000 test_acc_avg=0.25336 test_acc_top5_avg=0.84751 time=897.37it/s +curr_acc 0.2534 +BEST_ACC 0.2705 +curr_acc_top5 0.8475 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=77 global_step=30150 loss=4.37915 loss_avg=4.34434 acc=0.55469 acc_top1_avg=0.57540 acc_top5_avg=0.88663 lr=0.00100 gn=28.86928 time=64.72it/s +epoch=77 global_step=30200 loss=3.83284 loss_avg=4.33615 acc=0.63281 acc_top1_avg=0.57594 acc_top5_avg=0.88096 lr=0.00100 gn=26.30862 time=59.46it/s +epoch=77 global_step=30250 loss=4.60084 loss_avg=4.37342 acc=0.56250 acc_top1_avg=0.57272 acc_top5_avg=0.88134 lr=0.00100 gn=28.92529 time=61.69it/s +epoch=77 global_step=30300 loss=4.33837 loss_avg=4.37105 acc=0.58594 acc_top1_avg=0.57266 acc_top5_avg=0.88119 lr=0.00100 gn=32.49352 time=57.63it/s +epoch=77 global_step=30350 loss=3.91279 loss_avg=4.39828 acc=0.63281 acc_top1_avg=0.57022 acc_top5_avg=0.88111 lr=0.00100 gn=26.14452 time=61.93it/s +epoch=77 global_step=30400 loss=3.79842 loss_avg=4.41915 acc=0.63281 acc_top1_avg=0.56807 acc_top5_avg=0.87972 lr=0.00100 gn=29.98304 time=62.89it/s +epoch=77 global_step=30450 loss=4.52772 loss_avg=4.41462 acc=0.55469 acc_top1_avg=0.56901 acc_top5_avg=0.87933 lr=0.00100 gn=23.81296 time=65.33it/s +====================Eval==================== +epoch=77 global_step=30498 loss=4.73943 test_loss_avg=4.76928 acc=0.00000 test_acc_avg=0.16866 test_acc_top5_avg=0.91728 time=243.30it/s +epoch=77 global_step=30498 loss=0.25829 test_loss_avg=4.33375 acc=0.87500 test_acc_avg=0.23099 test_acc_top5_avg=0.82812 time=242.91it/s +epoch=77 global_step=30498 loss=4.96238 test_loss_avg=4.24239 acc=0.00000 test_acc_avg=0.23685 test_acc_top5_avg=0.83248 time=892.98it/s +curr_acc 0.2368 +BEST_ACC 0.2705 +curr_acc_top5 0.8325 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=78 global_step=30500 loss=4.42455 loss_avg=4.14186 acc=0.57031 acc_top1_avg=0.59375 acc_top5_avg=0.89062 lr=0.00100 gn=31.40786 time=62.63it/s +epoch=78 global_step=30550 loss=4.10956 loss_avg=4.29326 acc=0.62500 acc_top1_avg=0.58323 acc_top5_avg=0.87800 lr=0.00100 gn=30.35574 time=61.22it/s +epoch=78 global_step=30600 loss=4.03626 loss_avg=4.31996 acc=0.60938 acc_top1_avg=0.57958 acc_top5_avg=0.87891 lr=0.00100 gn=23.41027 time=61.99it/s +epoch=78 global_step=30650 loss=4.32924 loss_avg=4.34794 acc=0.58594 acc_top1_avg=0.57643 acc_top5_avg=0.88194 lr=0.00100 gn=21.04688 time=56.22it/s +epoch=78 global_step=30700 loss=4.25440 loss_avg=4.35753 acc=0.59375 acc_top1_avg=0.57584 acc_top5_avg=0.88231 lr=0.00100 gn=29.60162 time=58.47it/s +epoch=78 global_step=30750 loss=3.92330 loss_avg=4.36741 acc=0.62500 acc_top1_avg=0.57527 acc_top5_avg=0.88157 lr=0.00100 gn=30.73049 time=61.74it/s +epoch=78 global_step=30800 loss=4.36465 loss_avg=4.38474 acc=0.57812 acc_top1_avg=0.57360 acc_top5_avg=0.88103 lr=0.00100 gn=37.23775 time=58.77it/s +epoch=78 global_step=30850 loss=4.48999 loss_avg=4.39449 acc=0.55469 acc_top1_avg=0.57273 acc_top5_avg=0.88010 lr=0.00100 gn=23.23824 time=58.77it/s +====================Eval==================== +epoch=78 global_step=30889 loss=6.18482 test_loss_avg=5.46021 acc=0.00000 test_acc_avg=0.06188 test_acc_top5_avg=0.72944 time=250.00it/s +epoch=78 global_step=30889 loss=5.30656 test_loss_avg=4.29147 acc=0.00000 test_acc_avg=0.23863 test_acc_top5_avg=0.83515 time=876.74it/s +curr_acc 0.2386 +BEST_ACC 0.2705 +curr_acc_top5 0.8351 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=79 global_step=30900 loss=4.43226 loss_avg=4.44604 acc=0.57031 acc_top1_avg=0.56250 acc_top5_avg=0.87287 lr=0.00100 gn=30.18399 time=61.65it/s +epoch=79 global_step=30950 loss=3.84930 loss_avg=4.39332 acc=0.62500 acc_top1_avg=0.57057 acc_top5_avg=0.87731 lr=0.00100 gn=25.85504 time=65.93it/s +epoch=79 global_step=31000 loss=4.65815 loss_avg=4.37469 acc=0.53125 acc_top1_avg=0.57390 acc_top5_avg=0.87627 lr=0.00100 gn=23.97588 time=61.71it/s +epoch=79 global_step=31050 loss=5.01725 loss_avg=4.39199 acc=0.50000 acc_top1_avg=0.57225 acc_top5_avg=0.87680 lr=0.00100 gn=27.27770 time=65.87it/s +epoch=79 global_step=31100 loss=3.68700 loss_avg=4.38304 acc=0.63281 acc_top1_avg=0.57346 acc_top5_avg=0.87793 lr=0.00100 gn=23.90882 time=64.64it/s +epoch=79 global_step=31150 loss=4.34138 loss_avg=4.38562 acc=0.57812 acc_top1_avg=0.57268 acc_top5_avg=0.87850 lr=0.00100 gn=29.40720 time=65.89it/s +epoch=79 global_step=31200 loss=4.23852 loss_avg=4.38458 acc=0.59375 acc_top1_avg=0.57293 acc_top5_avg=0.87827 lr=0.00100 gn=30.14361 time=65.96it/s +epoch=79 global_step=31250 loss=4.35047 loss_avg=4.38659 acc=0.56250 acc_top1_avg=0.57258 acc_top5_avg=0.87827 lr=0.00100 gn=29.26348 time=66.04it/s +====================Eval==================== +epoch=79 global_step=31280 loss=2.61313 test_loss_avg=6.17931 acc=0.36719 test_acc_avg=0.04948 test_acc_top5_avg=0.84809 time=259.64it/s +epoch=79 global_step=31280 loss=0.77655 test_loss_avg=4.89533 acc=0.76562 test_acc_avg=0.13864 test_acc_top5_avg=0.79793 time=261.44it/s +epoch=79 global_step=31280 loss=4.83413 test_loss_avg=4.26175 acc=0.00000 test_acc_avg=0.23348 test_acc_top5_avg=0.81903 time=902.78it/s +curr_acc 0.2335 +BEST_ACC 0.2705 +curr_acc_top5 0.8190 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=80 global_step=31300 loss=4.95451 loss_avg=4.29767 acc=0.50000 acc_top1_avg=0.58242 acc_top5_avg=0.87852 lr=0.00010 gn=36.55971 time=56.66it/s +epoch=80 global_step=31350 loss=4.61252 loss_avg=4.28931 acc=0.55469 acc_top1_avg=0.58371 acc_top5_avg=0.87824 lr=0.00010 gn=28.18265 time=60.35it/s +epoch=80 global_step=31400 loss=4.61946 loss_avg=4.30766 acc=0.53906 acc_top1_avg=0.58171 acc_top5_avg=0.87585 lr=0.00010 gn=21.28196 time=60.07it/s +epoch=80 global_step=31450 loss=4.55429 loss_avg=4.27937 acc=0.56250 acc_top1_avg=0.58401 acc_top5_avg=0.87790 lr=0.00010 gn=31.67566 time=56.91it/s +epoch=80 global_step=31500 loss=4.34234 loss_avg=4.28119 acc=0.57812 acc_top1_avg=0.58299 acc_top5_avg=0.87813 lr=0.00010 gn=29.95438 time=57.21it/s +epoch=80 global_step=31550 loss=4.70591 loss_avg=4.27329 acc=0.55469 acc_top1_avg=0.58371 acc_top5_avg=0.87801 lr=0.00010 gn=29.25739 time=61.90it/s +epoch=80 global_step=31600 loss=4.10155 loss_avg=4.27343 acc=0.60938 acc_top1_avg=0.58406 acc_top5_avg=0.87820 lr=0.00010 gn=36.55012 time=58.75it/s +epoch=80 global_step=31650 loss=4.11601 loss_avg=4.27437 acc=0.60938 acc_top1_avg=0.58414 acc_top5_avg=0.87874 lr=0.00010 gn=26.39667 time=55.32it/s +====================Eval==================== +epoch=80 global_step=31671 loss=6.38983 test_loss_avg=4.95123 acc=0.00000 test_acc_avg=0.10339 test_acc_top5_avg=0.84740 time=259.04it/s +epoch=80 global_step=31671 loss=5.10949 test_loss_avg=4.17922 acc=0.00000 test_acc_avg=0.24011 test_acc_top5_avg=0.83327 time=892.22it/s +curr_acc 0.2401 +BEST_ACC 0.2705 +curr_acc_top5 0.8333 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=81 global_step=31700 loss=3.87313 loss_avg=4.21890 acc=0.63281 acc_top1_avg=0.59079 acc_top5_avg=0.87608 lr=0.00010 gn=33.62614 time=61.52it/s +epoch=81 global_step=31750 loss=3.96516 loss_avg=4.20989 acc=0.61719 acc_top1_avg=0.58979 acc_top5_avg=0.87965 lr=0.00010 gn=22.66118 time=63.07it/s +epoch=81 global_step=31800 loss=4.17066 loss_avg=4.21951 acc=0.58594 acc_top1_avg=0.58812 acc_top5_avg=0.88021 lr=0.00010 gn=26.31324 time=64.98it/s +epoch=81 global_step=31850 loss=4.59149 loss_avg=4.20937 acc=0.57031 acc_top1_avg=0.58991 acc_top5_avg=0.88128 lr=0.00010 gn=31.18438 time=60.19it/s +epoch=81 global_step=31900 loss=4.17542 loss_avg=4.20528 acc=0.58594 acc_top1_avg=0.58983 acc_top5_avg=0.87889 lr=0.00010 gn=28.77863 time=58.20it/s +epoch=81 global_step=31950 loss=3.94126 loss_avg=4.20788 acc=0.61719 acc_top1_avg=0.58958 acc_top5_avg=0.87875 lr=0.00010 gn=28.67107 time=55.74it/s +epoch=81 global_step=32000 loss=4.29719 loss_avg=4.21249 acc=0.57812 acc_top1_avg=0.58898 acc_top5_avg=0.87885 lr=0.00010 gn=25.28601 time=56.27it/s +epoch=81 global_step=32050 loss=4.22119 loss_avg=4.21632 acc=0.60156 acc_top1_avg=0.58891 acc_top5_avg=0.87865 lr=0.00010 gn=33.93195 time=61.52it/s +====================Eval==================== +epoch=81 global_step=32062 loss=6.79604 test_loss_avg=6.79604 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.87500 time=104.53it/s +epoch=81 global_step=32062 loss=6.88941 test_loss_avg=4.97296 acc=0.00000 test_acc_avg=0.11366 test_acc_top5_avg=0.77543 time=257.67it/s +epoch=81 global_step=32062 loss=5.08271 test_loss_avg=4.16139 acc=0.00000 test_acc_avg=0.24614 test_acc_top5_avg=0.83208 time=885.81it/s +curr_acc 0.2461 +BEST_ACC 0.2705 +curr_acc_top5 0.8321 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=82 global_step=32100 loss=3.87619 loss_avg=4.10373 acc=0.63281 acc_top1_avg=0.60382 acc_top5_avg=0.88507 lr=0.00010 gn=29.89233 time=65.41it/s +epoch=82 global_step=32150 loss=3.67782 loss_avg=4.15093 acc=0.64062 acc_top1_avg=0.59766 acc_top5_avg=0.88050 lr=0.00010 gn=28.50258 time=65.60it/s +epoch=82 global_step=32200 loss=4.33835 loss_avg=4.14038 acc=0.58594 acc_top1_avg=0.59845 acc_top5_avg=0.88043 lr=0.00010 gn=37.21313 time=60.09it/s +epoch=82 global_step=32250 loss=4.18720 loss_avg=4.16891 acc=0.60156 acc_top1_avg=0.59520 acc_top5_avg=0.88011 lr=0.00010 gn=25.05095 time=65.64it/s +epoch=82 global_step=32300 loss=4.21884 loss_avg=4.17977 acc=0.59375 acc_top1_avg=0.59372 acc_top5_avg=0.87937 lr=0.00010 gn=30.05075 time=65.61it/s +epoch=82 global_step=32350 loss=4.52742 loss_avg=4.19632 acc=0.56250 acc_top1_avg=0.59212 acc_top5_avg=0.87977 lr=0.00010 gn=27.11754 time=65.79it/s +epoch=82 global_step=32400 loss=4.16078 loss_avg=4.20090 acc=0.61719 acc_top1_avg=0.59167 acc_top5_avg=0.88004 lr=0.00010 gn=32.19927 time=65.41it/s +epoch=82 global_step=32450 loss=4.85607 loss_avg=4.20196 acc=0.51562 acc_top1_avg=0.59149 acc_top5_avg=0.87965 lr=0.00010 gn=26.39415 time=65.61it/s +====================Eval==================== +epoch=82 global_step=32453 loss=4.78170 test_loss_avg=4.60747 acc=0.00000 test_acc_avg=0.15341 test_acc_top5_avg=0.91903 time=261.25it/s +epoch=82 global_step=32453 loss=5.10539 test_loss_avg=4.10075 acc=0.01562 test_acc_avg=0.26769 test_acc_top5_avg=0.83073 time=262.49it/s +epoch=82 global_step=32453 loss=5.16387 test_loss_avg=4.19586 acc=0.00000 test_acc_avg=0.24496 test_acc_top5_avg=0.83208 time=919.00it/s +curr_acc 0.2450 +BEST_ACC 0.2705 +curr_acc_top5 0.8321 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=83 global_step=32500 loss=4.60214 loss_avg=4.24781 acc=0.53906 acc_top1_avg=0.58727 acc_top5_avg=0.87683 lr=0.00010 gn=17.98349 time=65.67it/s +epoch=83 global_step=32550 loss=4.19467 loss_avg=4.19476 acc=0.59375 acc_top1_avg=0.59222 acc_top5_avg=0.88136 lr=0.00010 gn=28.66702 time=65.80it/s +epoch=83 global_step=32600 loss=4.57182 loss_avg=4.14512 acc=0.55469 acc_top1_avg=0.59805 acc_top5_avg=0.88196 lr=0.00010 gn=27.84466 time=65.80it/s +epoch=83 global_step=32650 loss=4.99365 loss_avg=4.17244 acc=0.50000 acc_top1_avg=0.59542 acc_top5_avg=0.88115 lr=0.00010 gn=35.03237 time=65.69it/s +epoch=83 global_step=32700 loss=3.61431 loss_avg=4.17931 acc=0.66406 acc_top1_avg=0.59464 acc_top5_avg=0.88123 lr=0.00010 gn=28.32079 time=65.37it/s +epoch=83 global_step=32750 loss=4.11141 loss_avg=4.17138 acc=0.59375 acc_top1_avg=0.59546 acc_top5_avg=0.88029 lr=0.00010 gn=24.81294 time=65.48it/s +epoch=83 global_step=32800 loss=4.03855 loss_avg=4.18546 acc=0.61719 acc_top1_avg=0.59413 acc_top5_avg=0.87953 lr=0.00010 gn=32.47224 time=65.83it/s +====================Eval==================== +epoch=83 global_step=32844 loss=3.21637 test_loss_avg=5.02766 acc=0.24219 test_acc_avg=0.10647 test_acc_top5_avg=0.74582 time=236.23it/s +epoch=83 global_step=32844 loss=5.17548 test_loss_avg=4.18480 acc=0.00000 test_acc_avg=0.24288 test_acc_top5_avg=0.83752 time=875.82it/s +curr_acc 0.2429 +BEST_ACC 0.2705 +curr_acc_top5 0.8375 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=84 global_step=32850 loss=4.35721 loss_avg=4.24591 acc=0.57031 acc_top1_avg=0.58073 acc_top5_avg=0.85547 lr=0.00010 gn=22.95508 time=65.67it/s +epoch=84 global_step=32900 loss=3.16246 loss_avg=4.18655 acc=0.70312 acc_top1_avg=0.59319 acc_top5_avg=0.87807 lr=0.00010 gn=25.56083 time=59.78it/s +epoch=84 global_step=32950 loss=4.21637 loss_avg=4.18938 acc=0.60156 acc_top1_avg=0.59250 acc_top5_avg=0.87647 lr=0.00010 gn=34.91604 time=53.47it/s +epoch=84 global_step=33000 loss=3.40635 loss_avg=4.19043 acc=0.66406 acc_top1_avg=0.59220 acc_top5_avg=0.87535 lr=0.00010 gn=20.81205 time=61.80it/s +epoch=84 global_step=33050 loss=4.39974 loss_avg=4.19373 acc=0.57812 acc_top1_avg=0.59197 acc_top5_avg=0.87561 lr=0.00010 gn=30.12952 time=58.39it/s +epoch=84 global_step=33100 loss=3.54332 loss_avg=4.18833 acc=0.65625 acc_top1_avg=0.59238 acc_top5_avg=0.87613 lr=0.00010 gn=30.88846 time=60.14it/s +epoch=84 global_step=33150 loss=4.37500 loss_avg=4.19194 acc=0.57812 acc_top1_avg=0.59224 acc_top5_avg=0.87679 lr=0.00010 gn=21.70357 time=54.05it/s +epoch=84 global_step=33200 loss=3.97576 loss_avg=4.19265 acc=0.61719 acc_top1_avg=0.59246 acc_top5_avg=0.87693 lr=0.00010 gn=33.50769 time=59.15it/s +====================Eval==================== +epoch=84 global_step=33235 loss=3.02417 test_loss_avg=4.95938 acc=0.26562 test_acc_avg=0.15513 test_acc_top5_avg=0.88449 time=261.38it/s +epoch=84 global_step=33235 loss=0.38440 test_loss_avg=4.45444 acc=0.89844 test_acc_avg=0.20630 test_acc_top5_avg=0.80786 time=257.56it/s +epoch=84 global_step=33235 loss=5.04077 test_loss_avg=4.17950 acc=0.00000 test_acc_avg=0.24258 test_acc_top5_avg=0.82437 time=893.74it/s +curr_acc 0.2426 +BEST_ACC 0.2705 +curr_acc_top5 0.8244 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=85 global_step=33250 loss=4.10422 loss_avg=4.14680 acc=0.58594 acc_top1_avg=0.59688 acc_top5_avg=0.87760 lr=0.00010 gn=31.32587 time=58.72it/s +epoch=85 global_step=33300 loss=4.39710 loss_avg=4.17561 acc=0.57812 acc_top1_avg=0.59579 acc_top5_avg=0.87692 lr=0.00010 gn=27.63343 time=51.72it/s +epoch=85 global_step=33350 loss=3.75128 loss_avg=4.18926 acc=0.64062 acc_top1_avg=0.59395 acc_top5_avg=0.87846 lr=0.00010 gn=26.95407 time=56.62it/s +epoch=85 global_step=33400 loss=4.06990 loss_avg=4.17709 acc=0.60156 acc_top1_avg=0.59489 acc_top5_avg=0.87765 lr=0.00010 gn=30.69584 time=54.71it/s +epoch=85 global_step=33450 loss=4.18331 loss_avg=4.18022 acc=0.60156 acc_top1_avg=0.59440 acc_top5_avg=0.87754 lr=0.00010 gn=26.82103 time=61.92it/s +epoch=85 global_step=33500 loss=3.45903 loss_avg=4.15792 acc=0.67969 acc_top1_avg=0.59643 acc_top5_avg=0.87936 lr=0.00010 gn=22.30539 time=55.99it/s +epoch=85 global_step=33550 loss=4.39662 loss_avg=4.15332 acc=0.58594 acc_top1_avg=0.59678 acc_top5_avg=0.87996 lr=0.00010 gn=32.19556 time=59.84it/s +epoch=85 global_step=33600 loss=4.69865 loss_avg=4.15431 acc=0.53906 acc_top1_avg=0.59628 acc_top5_avg=0.87922 lr=0.00010 gn=26.57336 time=56.33it/s +====================Eval==================== +epoch=85 global_step=33626 loss=5.87144 test_loss_avg=5.07706 acc=0.00000 test_acc_avg=0.09308 test_acc_top5_avg=0.76027 time=244.30it/s +epoch=85 global_step=33626 loss=5.01834 test_loss_avg=4.14189 acc=0.00000 test_acc_avg=0.24545 test_acc_top5_avg=0.83099 time=919.00it/s +curr_acc 0.2455 +BEST_ACC 0.2705 +curr_acc_top5 0.8310 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=86 global_step=33650 loss=4.11281 loss_avg=4.20542 acc=0.60156 acc_top1_avg=0.58984 acc_top5_avg=0.87044 lr=0.00010 gn=40.68373 time=57.60it/s +epoch=86 global_step=33700 loss=4.05932 loss_avg=4.18447 acc=0.60156 acc_top1_avg=0.59291 acc_top5_avg=0.87838 lr=0.00010 gn=24.05046 time=61.79it/s +epoch=86 global_step=33750 loss=3.92782 loss_avg=4.14994 acc=0.61719 acc_top1_avg=0.59665 acc_top5_avg=0.88036 lr=0.00010 gn=22.67150 time=56.54it/s +epoch=86 global_step=33800 loss=3.79747 loss_avg=4.14270 acc=0.62500 acc_top1_avg=0.59703 acc_top5_avg=0.88079 lr=0.00010 gn=29.18007 time=53.00it/s +epoch=86 global_step=33850 loss=4.31752 loss_avg=4.16145 acc=0.57812 acc_top1_avg=0.59490 acc_top5_avg=0.88002 lr=0.00010 gn=25.23777 time=61.88it/s +epoch=86 global_step=33900 loss=4.90518 loss_avg=4.15247 acc=0.52344 acc_top1_avg=0.59617 acc_top5_avg=0.87939 lr=0.00010 gn=27.38491 time=57.19it/s +epoch=86 global_step=33950 loss=4.85241 loss_avg=4.15501 acc=0.54688 acc_top1_avg=0.59628 acc_top5_avg=0.87857 lr=0.00010 gn=36.83252 time=56.49it/s +epoch=86 global_step=34000 loss=4.25181 loss_avg=4.15739 acc=0.57812 acc_top1_avg=0.59601 acc_top5_avg=0.87859 lr=0.00010 gn=28.94146 time=53.47it/s +====================Eval==================== +epoch=86 global_step=34017 loss=6.67317 test_loss_avg=6.74071 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.83984 time=242.99it/s +epoch=86 global_step=34017 loss=0.69513 test_loss_avg=4.93875 acc=0.82031 test_acc_avg=0.12402 test_acc_top5_avg=0.79060 time=261.31it/s +epoch=86 global_step=34017 loss=5.06541 test_loss_avg=4.13212 acc=0.00000 test_acc_avg=0.24763 test_acc_top5_avg=0.83356 time=879.49it/s +curr_acc 0.2476 +BEST_ACC 0.2705 +curr_acc_top5 0.8336 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=87 global_step=34050 loss=4.17954 loss_avg=4.19898 acc=0.59375 acc_top1_avg=0.59138 acc_top5_avg=0.87476 lr=0.00010 gn=36.42775 time=60.06it/s +epoch=87 global_step=34100 loss=4.44765 loss_avg=4.19381 acc=0.55469 acc_top1_avg=0.59441 acc_top5_avg=0.87773 lr=0.00010 gn=32.32884 time=47.75it/s +epoch=87 global_step=34150 loss=3.58172 loss_avg=4.18946 acc=0.65625 acc_top1_avg=0.59422 acc_top5_avg=0.87747 lr=0.00010 gn=27.27251 time=65.01it/s +epoch=87 global_step=34200 loss=4.07612 loss_avg=4.15430 acc=0.60938 acc_top1_avg=0.59746 acc_top5_avg=0.87901 lr=0.00010 gn=21.33775 time=51.60it/s +epoch=87 global_step=34250 loss=3.98308 loss_avg=4.14042 acc=0.60938 acc_top1_avg=0.59844 acc_top5_avg=0.87969 lr=0.00010 gn=32.91125 time=59.59it/s +epoch=87 global_step=34300 loss=4.00410 loss_avg=4.14620 acc=0.61719 acc_top1_avg=0.59770 acc_top5_avg=0.87931 lr=0.00010 gn=29.23798 time=60.08it/s +epoch=87 global_step=34350 loss=4.68597 loss_avg=4.14609 acc=0.54688 acc_top1_avg=0.59757 acc_top5_avg=0.87960 lr=0.00010 gn=28.75694 time=61.64it/s +epoch=87 global_step=34400 loss=3.99459 loss_avg=4.14195 acc=0.60156 acc_top1_avg=0.59789 acc_top5_avg=0.87943 lr=0.00010 gn=17.19937 time=61.94it/s +====================Eval==================== +epoch=87 global_step=34408 loss=6.10876 test_loss_avg=4.89320 acc=0.00000 test_acc_avg=0.10359 test_acc_top5_avg=0.86314 time=250.92it/s +epoch=87 global_step=34408 loss=4.90396 test_loss_avg=4.14158 acc=0.01562 test_acc_avg=0.24878 test_acc_top5_avg=0.82528 time=262.46it/s +epoch=87 global_step=34408 loss=4.89427 test_loss_avg=4.16105 acc=0.00000 test_acc_avg=0.24268 test_acc_top5_avg=0.82555 time=885.81it/s +curr_acc 0.2427 +BEST_ACC 0.2705 +curr_acc_top5 0.8256 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=88 global_step=34450 loss=4.27608 loss_avg=3.99623 acc=0.59375 acc_top1_avg=0.61272 acc_top5_avg=0.88318 lr=0.00010 gn=36.36368 time=59.31it/s +epoch=88 global_step=34500 loss=3.45484 loss_avg=4.07673 acc=0.66406 acc_top1_avg=0.60496 acc_top5_avg=0.88162 lr=0.00010 gn=31.64703 time=60.83it/s +epoch=88 global_step=34550 loss=4.41766 loss_avg=4.11541 acc=0.57812 acc_top1_avg=0.60107 acc_top5_avg=0.87830 lr=0.00010 gn=36.32372 time=64.72it/s +epoch=88 global_step=34600 loss=3.82052 loss_avg=4.11475 acc=0.63281 acc_top1_avg=0.60107 acc_top5_avg=0.87976 lr=0.00010 gn=24.86213 time=64.53it/s +epoch=88 global_step=34650 loss=3.57751 loss_avg=4.12415 acc=0.64844 acc_top1_avg=0.60027 acc_top5_avg=0.87836 lr=0.00010 gn=18.63281 time=65.69it/s +epoch=88 global_step=34700 loss=4.37027 loss_avg=4.13576 acc=0.57812 acc_top1_avg=0.59913 acc_top5_avg=0.87714 lr=0.00010 gn=30.86342 time=64.85it/s +epoch=88 global_step=34750 loss=4.19442 loss_avg=4.13948 acc=0.58594 acc_top1_avg=0.59887 acc_top5_avg=0.87706 lr=0.00010 gn=32.01657 time=64.99it/s +====================Eval==================== +epoch=88 global_step=34799 loss=6.52532 test_loss_avg=4.89406 acc=0.00000 test_acc_avg=0.11572 test_acc_top5_avg=0.76237 time=251.88it/s +epoch=88 global_step=34799 loss=4.95910 test_loss_avg=4.17341 acc=0.00000 test_acc_avg=0.24031 test_acc_top5_avg=0.83060 time=912.00it/s +curr_acc 0.2403 +BEST_ACC 0.2705 +curr_acc_top5 0.8306 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=89 global_step=34800 loss=4.61592 loss_avg=4.61592 acc=0.54688 acc_top1_avg=0.54688 acc_top5_avg=0.87500 lr=0.00010 gn=24.95161 time=54.09it/s +epoch=89 global_step=34850 loss=4.21541 loss_avg=4.10627 acc=0.59375 acc_top1_avg=0.60218 acc_top5_avg=0.87806 lr=0.00010 gn=32.31297 time=56.69it/s +epoch=89 global_step=34900 loss=3.63479 loss_avg=4.12349 acc=0.67188 acc_top1_avg=0.60040 acc_top5_avg=0.87833 lr=0.00010 gn=34.25978 time=61.87it/s +epoch=89 global_step=34950 loss=4.11257 loss_avg=4.12485 acc=0.61719 acc_top1_avg=0.60105 acc_top5_avg=0.87847 lr=0.00010 gn=32.27071 time=54.49it/s +epoch=89 global_step=35000 loss=3.70367 loss_avg=4.10807 acc=0.64844 acc_top1_avg=0.60207 acc_top5_avg=0.88126 lr=0.00010 gn=34.89593 time=59.11it/s +epoch=89 global_step=35050 loss=4.22902 loss_avg=4.11180 acc=0.57812 acc_top1_avg=0.60128 acc_top5_avg=0.87998 lr=0.00010 gn=31.35154 time=58.53it/s +epoch=89 global_step=35100 loss=4.12359 loss_avg=4.11978 acc=0.58594 acc_top1_avg=0.60047 acc_top5_avg=0.87975 lr=0.00010 gn=24.58957 time=61.71it/s +epoch=89 global_step=35150 loss=4.58051 loss_avg=4.12229 acc=0.55469 acc_top1_avg=0.60016 acc_top5_avg=0.87934 lr=0.00010 gn=31.57441 time=61.86it/s +====================Eval==================== +epoch=89 global_step=35190 loss=4.62142 test_loss_avg=4.75935 acc=0.00000 test_acc_avg=0.13980 test_acc_top5_avg=0.90666 time=262.59it/s +epoch=89 global_step=35190 loss=0.19912 test_loss_avg=4.11276 acc=0.93750 test_acc_avg=0.25532 test_acc_top5_avg=0.82258 time=261.57it/s +epoch=89 global_step=35190 loss=5.18205 test_loss_avg=4.15091 acc=0.00000 test_acc_avg=0.24001 test_acc_top5_avg=0.82219 time=905.31it/s +curr_acc 0.2400 +BEST_ACC 0.2705 +curr_acc_top5 0.8222 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=90 global_step=35200 loss=3.51287 loss_avg=4.03578 acc=0.67188 acc_top1_avg=0.61719 acc_top5_avg=0.87578 lr=0.00010 gn=29.03795 time=62.11it/s +epoch=90 global_step=35250 loss=4.20232 loss_avg=4.09011 acc=0.60156 acc_top1_avg=0.60391 acc_top5_avg=0.87708 lr=0.00010 gn=35.10678 time=63.25it/s +epoch=90 global_step=35300 loss=4.34501 loss_avg=4.09897 acc=0.58594 acc_top1_avg=0.60241 acc_top5_avg=0.87955 lr=0.00010 gn=31.35513 time=61.88it/s +epoch=90 global_step=35350 loss=4.19233 loss_avg=4.12975 acc=0.60156 acc_top1_avg=0.59917 acc_top5_avg=0.87920 lr=0.00010 gn=26.17857 time=59.14it/s +epoch=90 global_step=35400 loss=4.33624 loss_avg=4.12620 acc=0.56250 acc_top1_avg=0.59940 acc_top5_avg=0.87958 lr=0.00010 gn=32.57243 time=64.82it/s +epoch=90 global_step=35450 loss=4.19055 loss_avg=4.13182 acc=0.59375 acc_top1_avg=0.59826 acc_top5_avg=0.87945 lr=0.00010 gn=30.50007 time=61.52it/s +epoch=90 global_step=35500 loss=4.97904 loss_avg=4.12431 acc=0.50000 acc_top1_avg=0.59922 acc_top5_avg=0.87941 lr=0.00010 gn=33.93148 time=60.30it/s +epoch=90 global_step=35550 loss=4.22605 loss_avg=4.11971 acc=0.58594 acc_top1_avg=0.59963 acc_top5_avg=0.87910 lr=0.00010 gn=26.11751 time=59.18it/s +====================Eval==================== +epoch=90 global_step=35581 loss=2.94703 test_loss_avg=5.18934 acc=0.29688 test_acc_avg=0.08613 test_acc_top5_avg=0.71367 time=258.08it/s +epoch=90 global_step=35581 loss=5.04721 test_loss_avg=4.19263 acc=0.00000 test_acc_avg=0.24061 test_acc_top5_avg=0.82892 time=907.07it/s +curr_acc 0.2406 +BEST_ACC 0.2705 +curr_acc_top5 0.8289 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=91 global_step=35600 loss=3.59155 loss_avg=4.20827 acc=0.64844 acc_top1_avg=0.59663 acc_top5_avg=0.88569 lr=0.00010 gn=25.84736 time=61.61it/s +epoch=91 global_step=35650 loss=4.44913 loss_avg=4.18367 acc=0.57031 acc_top1_avg=0.59488 acc_top5_avg=0.88383 lr=0.00010 gn=28.26010 time=57.16it/s +epoch=91 global_step=35700 loss=4.28124 loss_avg=4.15798 acc=0.58594 acc_top1_avg=0.59703 acc_top5_avg=0.87861 lr=0.00010 gn=34.78691 time=61.72it/s +epoch=91 global_step=35750 loss=4.26727 loss_avg=4.17211 acc=0.57031 acc_top1_avg=0.59509 acc_top5_avg=0.87759 lr=0.00010 gn=25.80419 time=61.87it/s +epoch=91 global_step=35800 loss=3.77155 loss_avg=4.13981 acc=0.64062 acc_top1_avg=0.59860 acc_top5_avg=0.87896 lr=0.00010 gn=33.71156 time=62.96it/s +epoch=91 global_step=35850 loss=3.94662 loss_avg=4.13116 acc=0.63281 acc_top1_avg=0.59970 acc_top5_avg=0.87849 lr=0.00010 gn=35.14989 time=59.97it/s +epoch=91 global_step=35900 loss=4.21349 loss_avg=4.11454 acc=0.59375 acc_top1_avg=0.60112 acc_top5_avg=0.87904 lr=0.00010 gn=30.10427 time=61.57it/s +epoch=91 global_step=35950 loss=4.53927 loss_avg=4.11147 acc=0.53906 acc_top1_avg=0.60118 acc_top5_avg=0.87985 lr=0.00010 gn=24.79519 time=59.73it/s +====================Eval==================== +epoch=91 global_step=35972 loss=2.62822 test_loss_avg=5.52395 acc=0.40625 test_acc_avg=0.11506 test_acc_top5_avg=0.87287 time=243.54it/s +epoch=91 global_step=35972 loss=0.57899 test_loss_avg=4.67479 acc=0.80469 test_acc_avg=0.17508 test_acc_top5_avg=0.80251 time=261.41it/s +epoch=91 global_step=35972 loss=5.09471 test_loss_avg=4.20213 acc=0.00000 test_acc_avg=0.24278 test_acc_top5_avg=0.82822 time=922.84it/s +curr_acc 0.2428 +BEST_ACC 0.2705 +curr_acc_top5 0.8282 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=92 global_step=36000 loss=3.60202 loss_avg=4.03097 acc=0.64844 acc_top1_avg=0.60742 acc_top5_avg=0.88281 lr=0.00010 gn=35.90087 time=61.87it/s +epoch=92 global_step=36050 loss=4.65422 loss_avg=4.10618 acc=0.53906 acc_top1_avg=0.60076 acc_top5_avg=0.88011 lr=0.00010 gn=21.67040 time=56.70it/s +epoch=92 global_step=36100 loss=3.92683 loss_avg=4.09472 acc=0.62500 acc_top1_avg=0.60187 acc_top5_avg=0.88226 lr=0.00010 gn=27.72423 time=61.69it/s +epoch=92 global_step=36150 loss=3.79556 loss_avg=4.08236 acc=0.63281 acc_top1_avg=0.60336 acc_top5_avg=0.88093 lr=0.00010 gn=22.90133 time=61.92it/s +epoch=92 global_step=36200 loss=3.78125 loss_avg=4.10144 acc=0.63281 acc_top1_avg=0.60143 acc_top5_avg=0.87959 lr=0.00010 gn=25.35859 time=61.47it/s +epoch=92 global_step=36250 loss=3.21080 loss_avg=4.10144 acc=0.69531 acc_top1_avg=0.60173 acc_top5_avg=0.87896 lr=0.00010 gn=31.69590 time=54.17it/s +epoch=92 global_step=36300 loss=4.25177 loss_avg=4.09759 acc=0.60156 acc_top1_avg=0.60199 acc_top5_avg=0.87967 lr=0.00010 gn=40.49857 time=61.55it/s +epoch=92 global_step=36350 loss=4.45069 loss_avg=4.10156 acc=0.57031 acc_top1_avg=0.60162 acc_top5_avg=0.87969 lr=0.00010 gn=28.68906 time=61.74it/s +====================Eval==================== +epoch=92 global_step=36363 loss=6.08397 test_loss_avg=5.10906 acc=0.00000 test_acc_avg=0.08716 test_acc_top5_avg=0.80811 time=256.38it/s +epoch=92 global_step=36363 loss=4.98751 test_loss_avg=4.16538 acc=0.00000 test_acc_avg=0.23912 test_acc_top5_avg=0.82852 time=891.84it/s +curr_acc 0.2391 +BEST_ACC 0.2705 +curr_acc_top5 0.8285 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=93 global_step=36400 loss=3.71844 loss_avg=3.98850 acc=0.64844 acc_top1_avg=0.61296 acc_top5_avg=0.88133 lr=0.00010 gn=34.28818 time=59.25it/s +epoch=93 global_step=36450 loss=4.53986 loss_avg=4.03982 acc=0.54688 acc_top1_avg=0.60785 acc_top5_avg=0.87850 lr=0.00010 gn=30.70677 time=52.32it/s +epoch=93 global_step=36500 loss=4.03106 loss_avg=4.02868 acc=0.58594 acc_top1_avg=0.60972 acc_top5_avg=0.87871 lr=0.00010 gn=27.78387 time=56.25it/s +epoch=93 global_step=36550 loss=4.54900 loss_avg=4.05240 acc=0.54688 acc_top1_avg=0.60699 acc_top5_avg=0.87914 lr=0.00010 gn=28.23806 time=53.80it/s +epoch=93 global_step=36600 loss=4.01522 loss_avg=4.06648 acc=0.60938 acc_top1_avg=0.60519 acc_top5_avg=0.87823 lr=0.00010 gn=33.45949 time=59.81it/s +epoch=93 global_step=36650 loss=4.29240 loss_avg=4.07012 acc=0.59375 acc_top1_avg=0.60445 acc_top5_avg=0.87857 lr=0.00010 gn=31.05633 time=56.36it/s +epoch=93 global_step=36700 loss=3.90638 loss_avg=4.08870 acc=0.63281 acc_top1_avg=0.60268 acc_top5_avg=0.87848 lr=0.00010 gn=29.78848 time=58.69it/s +epoch=93 global_step=36750 loss=3.65288 loss_avg=4.09779 acc=0.65625 acc_top1_avg=0.60180 acc_top5_avg=0.87815 lr=0.00010 gn=25.25519 time=62.12it/s +====================Eval==================== +epoch=93 global_step=36754 loss=6.66447 test_loss_avg=6.67900 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.84375 time=239.54it/s +epoch=93 global_step=36754 loss=6.92755 test_loss_avg=5.06746 acc=0.00000 test_acc_avg=0.10938 test_acc_top5_avg=0.77668 time=261.23it/s +epoch=93 global_step=36754 loss=5.00134 test_loss_avg=4.18190 acc=0.00000 test_acc_avg=0.24278 test_acc_top5_avg=0.83129 time=895.45it/s +curr_acc 0.2428 +BEST_ACC 0.2705 +curr_acc_top5 0.8313 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=94 global_step=36800 loss=4.07432 loss_avg=4.06806 acc=0.60938 acc_top1_avg=0.60343 acc_top5_avg=0.88145 lr=0.00010 gn=36.24508 time=60.61it/s +epoch=94 global_step=36850 loss=3.59980 loss_avg=4.04436 acc=0.68750 acc_top1_avg=0.60718 acc_top5_avg=0.88005 lr=0.00010 gn=42.90480 time=61.81it/s +epoch=94 global_step=36900 loss=3.59810 loss_avg=4.07636 acc=0.65625 acc_top1_avg=0.60418 acc_top5_avg=0.87933 lr=0.00010 gn=35.31669 time=59.72it/s +epoch=94 global_step=36950 loss=3.63139 loss_avg=4.10304 acc=0.65625 acc_top1_avg=0.60200 acc_top5_avg=0.87930 lr=0.00010 gn=32.46324 time=62.08it/s +epoch=94 global_step=37000 loss=4.09575 loss_avg=4.09356 acc=0.60156 acc_top1_avg=0.60302 acc_top5_avg=0.88107 lr=0.00010 gn=27.51414 time=61.03it/s +epoch=94 global_step=37050 loss=3.88569 loss_avg=4.10763 acc=0.63281 acc_top1_avg=0.60167 acc_top5_avg=0.87965 lr=0.00010 gn=38.23198 time=59.37it/s +epoch=94 global_step=37100 loss=4.20196 loss_avg=4.10477 acc=0.57812 acc_top1_avg=0.60201 acc_top5_avg=0.87956 lr=0.00010 gn=28.16313 time=59.46it/s +====================Eval==================== +epoch=94 global_step=37145 loss=5.57825 test_loss_avg=4.70837 acc=0.00000 test_acc_avg=0.12956 test_acc_top5_avg=0.90788 time=261.57it/s +epoch=94 global_step=37145 loss=5.01581 test_loss_avg=4.14747 acc=0.01562 test_acc_avg=0.25370 test_acc_top5_avg=0.82939 time=262.37it/s +epoch=94 global_step=37145 loss=5.06521 test_loss_avg=4.19817 acc=0.00000 test_acc_avg=0.23833 test_acc_top5_avg=0.82921 time=908.84it/s +curr_acc 0.2383 +BEST_ACC 0.2705 +curr_acc_top5 0.8292 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=95 global_step=37150 loss=4.25877 loss_avg=4.24763 acc=0.57812 acc_top1_avg=0.58281 acc_top5_avg=0.86250 lr=0.00010 gn=20.50414 time=58.60it/s +epoch=95 global_step=37200 loss=4.25632 loss_avg=4.09531 acc=0.59375 acc_top1_avg=0.60085 acc_top5_avg=0.87784 lr=0.00010 gn=25.76123 time=56.23it/s +epoch=95 global_step=37250 loss=3.86548 loss_avg=4.07512 acc=0.61719 acc_top1_avg=0.60335 acc_top5_avg=0.87820 lr=0.00010 gn=31.80965 time=61.85it/s +epoch=95 global_step=37300 loss=4.38835 loss_avg=4.05307 acc=0.57031 acc_top1_avg=0.60696 acc_top5_avg=0.88100 lr=0.00010 gn=32.16024 time=54.96it/s +epoch=95 global_step=37350 loss=4.06696 loss_avg=4.05584 acc=0.60156 acc_top1_avg=0.60625 acc_top5_avg=0.88075 lr=0.00010 gn=28.50921 time=61.70it/s +epoch=95 global_step=37400 loss=4.05503 loss_avg=4.08578 acc=0.60156 acc_top1_avg=0.60328 acc_top5_avg=0.87932 lr=0.00010 gn=29.12676 time=65.50it/s +epoch=95 global_step=37450 loss=3.96989 loss_avg=4.09164 acc=0.62500 acc_top1_avg=0.60272 acc_top5_avg=0.87874 lr=0.00010 gn=30.42621 time=54.21it/s +epoch=95 global_step=37500 loss=3.74007 loss_avg=4.08573 acc=0.63281 acc_top1_avg=0.60317 acc_top5_avg=0.87940 lr=0.00010 gn=29.19468 time=59.34it/s +====================Eval==================== +epoch=95 global_step=37536 loss=2.67657 test_loss_avg=4.98238 acc=0.30469 test_acc_avg=0.10764 test_acc_top5_avg=0.74566 time=258.70it/s +epoch=95 global_step=37536 loss=5.11080 test_loss_avg=4.18608 acc=0.00000 test_acc_avg=0.23952 test_acc_top5_avg=0.83109 time=913.59it/s +curr_acc 0.2395 +BEST_ACC 0.2705 +curr_acc_top5 0.8311 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=96 global_step=37550 loss=4.33697 loss_avg=3.91727 acc=0.57812 acc_top1_avg=0.62165 acc_top5_avg=0.88002 lr=0.00010 gn=25.89437 time=63.28it/s +epoch=96 global_step=37600 loss=3.98570 loss_avg=4.08423 acc=0.63281 acc_top1_avg=0.60291 acc_top5_avg=0.87610 lr=0.00010 gn=35.73316 time=63.02it/s +epoch=96 global_step=37650 loss=4.03015 loss_avg=4.09446 acc=0.60938 acc_top1_avg=0.60163 acc_top5_avg=0.87788 lr=0.00010 gn=38.11181 time=59.27it/s +epoch=96 global_step=37700 loss=3.77996 loss_avg=4.10037 acc=0.64062 acc_top1_avg=0.60085 acc_top5_avg=0.87776 lr=0.00010 gn=34.82139 time=57.57it/s +epoch=96 global_step=37750 loss=3.95077 loss_avg=4.10917 acc=0.61719 acc_top1_avg=0.60036 acc_top5_avg=0.87756 lr=0.00010 gn=28.24638 time=61.88it/s +epoch=96 global_step=37800 loss=4.04854 loss_avg=4.10491 acc=0.61719 acc_top1_avg=0.60130 acc_top5_avg=0.87766 lr=0.00010 gn=36.30260 time=64.76it/s +epoch=96 global_step=37850 loss=3.84629 loss_avg=4.10094 acc=0.63281 acc_top1_avg=0.60184 acc_top5_avg=0.87781 lr=0.00010 gn=28.12130 time=61.64it/s +epoch=96 global_step=37900 loss=3.31766 loss_avg=4.08845 acc=0.68750 acc_top1_avg=0.60345 acc_top5_avg=0.87929 lr=0.00010 gn=36.96833 time=62.02it/s +====================Eval==================== +epoch=96 global_step=37927 loss=3.26101 test_loss_avg=4.83077 acc=0.25781 test_acc_avg=0.17822 test_acc_top5_avg=0.90039 time=261.69it/s +epoch=96 global_step=37927 loss=0.19085 test_loss_avg=4.33978 acc=0.93750 test_acc_avg=0.22597 test_acc_top5_avg=0.81747 time=254.42it/s +epoch=96 global_step=37927 loss=5.01528 test_loss_avg=4.18745 acc=0.00000 test_acc_avg=0.24110 test_acc_top5_avg=0.82763 time=887.68it/s +curr_acc 0.2411 +BEST_ACC 0.2705 +curr_acc_top5 0.8276 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=97 global_step=37950 loss=3.23275 loss_avg=3.93372 acc=0.69531 acc_top1_avg=0.61923 acc_top5_avg=0.88281 lr=0.00010 gn=31.52271 time=58.65it/s +epoch=97 global_step=38000 loss=3.67087 loss_avg=4.02343 acc=0.63281 acc_top1_avg=0.61012 acc_top5_avg=0.87939 lr=0.00010 gn=22.47274 time=62.05it/s +epoch=97 global_step=38050 loss=4.00027 loss_avg=4.05147 acc=0.60156 acc_top1_avg=0.60645 acc_top5_avg=0.87983 lr=0.00010 gn=30.05945 time=59.47it/s +epoch=97 global_step=38100 loss=4.00346 loss_avg=4.05129 acc=0.59375 acc_top1_avg=0.60676 acc_top5_avg=0.87938 lr=0.00010 gn=31.35639 time=63.40it/s +epoch=97 global_step=38150 loss=3.93731 loss_avg=4.05075 acc=0.60938 acc_top1_avg=0.60738 acc_top5_avg=0.87812 lr=0.00010 gn=31.83714 time=61.61it/s +epoch=97 global_step=38200 loss=4.42309 loss_avg=4.04790 acc=0.56250 acc_top1_avg=0.60760 acc_top5_avg=0.87829 lr=0.00010 gn=29.84839 time=58.80it/s +epoch=97 global_step=38250 loss=4.52895 loss_avg=4.06837 acc=0.56250 acc_top1_avg=0.60548 acc_top5_avg=0.87732 lr=0.00010 gn=32.05395 time=61.64it/s +epoch=97 global_step=38300 loss=4.04096 loss_avg=4.06700 acc=0.60938 acc_top1_avg=0.60531 acc_top5_avg=0.87764 lr=0.00010 gn=30.41237 time=61.04it/s +====================Eval==================== +epoch=97 global_step=38318 loss=6.14463 test_loss_avg=5.21390 acc=0.00000 test_acc_avg=0.08932 test_acc_top5_avg=0.74113 time=258.62it/s +epoch=97 global_step=38318 loss=5.09171 test_loss_avg=4.17097 acc=0.00000 test_acc_avg=0.24446 test_acc_top5_avg=0.83198 time=915.79it/s +curr_acc 0.2445 +BEST_ACC 0.2705 +curr_acc_top5 0.8320 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=98 global_step=38350 loss=3.64554 loss_avg=3.97067 acc=0.65625 acc_top1_avg=0.61548 acc_top5_avg=0.87622 lr=0.00010 gn=29.02288 time=61.81it/s +epoch=98 global_step=38400 loss=3.98465 loss_avg=3.99523 acc=0.61719 acc_top1_avg=0.61347 acc_top5_avg=0.87614 lr=0.00010 gn=36.48115 time=59.67it/s +epoch=98 global_step=38450 loss=3.95171 loss_avg=4.03959 acc=0.62500 acc_top1_avg=0.60849 acc_top5_avg=0.87636 lr=0.00010 gn=38.33961 time=63.53it/s +epoch=98 global_step=38500 loss=4.44823 loss_avg=4.04577 acc=0.57812 acc_top1_avg=0.60792 acc_top5_avg=0.87891 lr=0.00010 gn=36.35644 time=64.51it/s +epoch=98 global_step=38550 loss=3.47083 loss_avg=4.03951 acc=0.68750 acc_top1_avg=0.60870 acc_top5_avg=0.87887 lr=0.00010 gn=40.54948 time=61.73it/s +epoch=98 global_step=38600 loss=3.96630 loss_avg=4.05966 acc=0.62500 acc_top1_avg=0.60674 acc_top5_avg=0.87816 lr=0.00010 gn=22.68270 time=55.04it/s +epoch=98 global_step=38650 loss=4.30072 loss_avg=4.06337 acc=0.58594 acc_top1_avg=0.60608 acc_top5_avg=0.87837 lr=0.00010 gn=30.09290 time=64.35it/s +epoch=98 global_step=38700 loss=4.19475 loss_avg=4.06353 acc=0.60156 acc_top1_avg=0.60616 acc_top5_avg=0.87811 lr=0.00010 gn=36.94973 time=62.28it/s +====================Eval==================== +epoch=98 global_step=38709 loss=5.87883 test_loss_avg=6.54104 acc=0.07031 test_acc_avg=0.00879 test_acc_top5_avg=0.83887 time=259.32it/s +epoch=98 global_step=38709 loss=0.55263 test_loss_avg=4.77189 acc=0.79688 test_acc_avg=0.14547 test_acc_top5_avg=0.79432 time=218.93it/s +epoch=98 global_step=38709 loss=5.12889 test_loss_avg=4.11748 acc=0.00000 test_acc_avg=0.24525 test_acc_top5_avg=0.82793 time=880.97it/s +curr_acc 0.2453 +BEST_ACC 0.2705 +curr_acc_top5 0.8279 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=99 global_step=38750 loss=4.54581 loss_avg=4.08274 acc=0.54688 acc_top1_avg=0.60328 acc_top5_avg=0.87995 lr=0.00010 gn=32.04314 time=65.49it/s +epoch=99 global_step=38800 loss=4.70711 loss_avg=4.06541 acc=0.53125 acc_top1_avg=0.60491 acc_top5_avg=0.87998 lr=0.00010 gn=25.53181 time=64.69it/s +epoch=99 global_step=38850 loss=4.01511 loss_avg=4.07097 acc=0.59375 acc_top1_avg=0.60383 acc_top5_avg=0.88010 lr=0.00010 gn=38.10567 time=65.52it/s +epoch=99 global_step=38900 loss=4.00616 loss_avg=4.07552 acc=0.63281 acc_top1_avg=0.60398 acc_top5_avg=0.87962 lr=0.00010 gn=35.14187 time=65.52it/s +epoch=99 global_step=38950 loss=3.78912 loss_avg=4.07528 acc=0.63281 acc_top1_avg=0.60406 acc_top5_avg=0.87967 lr=0.00010 gn=34.72814 time=65.11it/s +epoch=99 global_step=39000 loss=4.18778 loss_avg=4.07413 acc=0.58594 acc_top1_avg=0.60398 acc_top5_avg=0.87946 lr=0.00010 gn=30.13594 time=65.64it/s +epoch=99 global_step=39050 loss=3.75861 loss_avg=4.07549 acc=0.64062 acc_top1_avg=0.60406 acc_top5_avg=0.87862 lr=0.00010 gn=39.74211 time=65.56it/s +epoch=99 global_step=39100 loss=3.28074 loss_avg=4.06637 acc=0.70000 acc_top1_avg=0.60529 acc_top5_avg=0.87889 lr=0.00010 gn=35.93914 time=89.60it/s +====================Eval==================== +epoch=99 global_step=39100 loss=5.99413 test_loss_avg=4.95925 acc=0.00000 test_acc_avg=0.10480 test_acc_top5_avg=0.84644 time=221.53it/s +epoch=99 global_step=39100 loss=5.12190 test_loss_avg=4.17401 acc=0.00000 test_acc_avg=0.23784 test_acc_top5_avg=0.82447 time=886.75it/s +epoch=99 global_step=39100 loss=5.12190 test_loss_avg=4.17401 acc=0.00000 test_acc_avg=0.23784 test_acc_top5_avg=0.82447 time=886.75it/s +curr_acc 0.2378 +BEST_ACC 0.2705 +curr_acc_top5 0.8245 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=100 global_step=39150 loss=3.87728 loss_avg=4.05479 acc=0.63281 acc_top1_avg=0.60547 acc_top5_avg=0.87609 lr=0.00010 gn=34.94539 time=64.07it/s +epoch=100 global_step=39200 loss=4.49129 loss_avg=4.07434 acc=0.56250 acc_top1_avg=0.60437 acc_top5_avg=0.87547 lr=0.00010 gn=30.67795 time=61.81it/s +epoch=100 global_step=39250 loss=3.96088 loss_avg=4.06408 acc=0.61719 acc_top1_avg=0.60609 acc_top5_avg=0.87641 lr=0.00010 gn=26.78165 time=57.42it/s +epoch=100 global_step=39300 loss=4.39724 loss_avg=4.05110 acc=0.57031 acc_top1_avg=0.60699 acc_top5_avg=0.87719 lr=0.00010 gn=31.49041 time=65.73it/s +epoch=100 global_step=39350 loss=4.64129 loss_avg=4.05842 acc=0.54688 acc_top1_avg=0.60625 acc_top5_avg=0.87697 lr=0.00010 gn=29.87741 time=64.28it/s +epoch=100 global_step=39400 loss=4.39698 loss_avg=4.06260 acc=0.56250 acc_top1_avg=0.60542 acc_top5_avg=0.87753 lr=0.00010 gn=28.23099 time=63.68it/s +epoch=100 global_step=39450 loss=4.10188 loss_avg=4.06479 acc=0.60156 acc_top1_avg=0.60507 acc_top5_avg=0.87828 lr=0.00010 gn=25.68098 time=64.33it/s +====================Eval==================== +epoch=100 global_step=39491 loss=6.71451 test_loss_avg=4.91943 acc=0.00000 test_acc_avg=0.11047 test_acc_top5_avg=0.75766 time=247.77it/s +epoch=100 global_step=39491 loss=5.13534 test_loss_avg=4.14359 acc=0.00000 test_acc_avg=0.24446 test_acc_top5_avg=0.82476 time=883.94it/s +curr_acc 0.2445 +BEST_ACC 0.2705 +curr_acc_top5 0.8248 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=101 global_step=39500 loss=3.90339 loss_avg=4.03147 acc=0.62500 acc_top1_avg=0.60677 acc_top5_avg=0.87760 lr=0.00010 gn=32.87751 time=60.15it/s +epoch=101 global_step=39550 loss=4.01315 loss_avg=4.01349 acc=0.60156 acc_top1_avg=0.61057 acc_top5_avg=0.88003 lr=0.00010 gn=35.07616 time=59.03it/s +epoch=101 global_step=39600 loss=4.33036 loss_avg=4.02769 acc=0.56250 acc_top1_avg=0.60801 acc_top5_avg=0.87973 lr=0.00010 gn=28.58286 time=59.57it/s +epoch=101 global_step=39650 loss=4.18307 loss_avg=4.09334 acc=0.60156 acc_top1_avg=0.60161 acc_top5_avg=0.87726 lr=0.00010 gn=34.42903 time=61.85it/s +epoch=101 global_step=39700 loss=4.21324 loss_avg=4.10286 acc=0.58594 acc_top1_avg=0.60104 acc_top5_avg=0.87732 lr=0.00010 gn=27.65655 time=60.61it/s +epoch=101 global_step=39750 loss=3.31435 loss_avg=4.08581 acc=0.69531 acc_top1_avg=0.60313 acc_top5_avg=0.87717 lr=0.00010 gn=30.38677 time=62.11it/s +epoch=101 global_step=39800 loss=3.73172 loss_avg=4.07089 acc=0.64844 acc_top1_avg=0.60505 acc_top5_avg=0.87781 lr=0.00010 gn=39.12324 time=62.01it/s +epoch=101 global_step=39850 loss=3.61371 loss_avg=4.06385 acc=0.65625 acc_top1_avg=0.60587 acc_top5_avg=0.87731 lr=0.00010 gn=36.91340 time=61.74it/s +====================Eval==================== +epoch=101 global_step=39882 loss=4.54532 test_loss_avg=4.63736 acc=0.00000 test_acc_avg=0.15179 test_acc_top5_avg=0.91071 time=245.22it/s +epoch=101 global_step=39882 loss=3.45154 test_loss_avg=4.04762 acc=0.31250 test_acc_avg=0.26893 test_acc_top5_avg=0.82592 time=261.80it/s +epoch=101 global_step=39882 loss=5.10062 test_loss_avg=4.13928 acc=0.00000 test_acc_avg=0.24308 test_acc_top5_avg=0.82545 time=908.84it/s +curr_acc 0.2431 +BEST_ACC 0.2705 +curr_acc_top5 0.8255 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=102 global_step=39900 loss=4.57611 loss_avg=4.12140 acc=0.54688 acc_top1_avg=0.59896 acc_top5_avg=0.87370 lr=0.00010 gn=29.35732 time=65.26it/s +epoch=102 global_step=39950 loss=4.23940 loss_avg=4.15021 acc=0.59375 acc_top1_avg=0.59674 acc_top5_avg=0.87477 lr=0.00010 gn=30.99129 time=56.92it/s +epoch=102 global_step=40000 loss=3.77290 loss_avg=4.09469 acc=0.64062 acc_top1_avg=0.60295 acc_top5_avg=0.87593 lr=0.00010 gn=36.06816 time=64.40it/s +epoch=102 global_step=40050 loss=3.73801 loss_avg=4.08040 acc=0.64844 acc_top1_avg=0.60412 acc_top5_avg=0.87756 lr=0.00010 gn=36.45470 time=60.29it/s +epoch=102 global_step=40100 loss=3.62837 loss_avg=4.06551 acc=0.65625 acc_top1_avg=0.60561 acc_top5_avg=0.87694 lr=0.00010 gn=31.08935 time=62.10it/s +epoch=102 global_step=40150 loss=4.46021 loss_avg=4.05165 acc=0.56250 acc_top1_avg=0.60684 acc_top5_avg=0.87832 lr=0.00010 gn=27.13370 time=57.33it/s +epoch=102 global_step=40200 loss=4.09479 loss_avg=4.05263 acc=0.59375 acc_top1_avg=0.60675 acc_top5_avg=0.87706 lr=0.00010 gn=33.51749 time=64.84it/s +epoch=102 global_step=40250 loss=4.14988 loss_avg=4.05658 acc=0.60938 acc_top1_avg=0.60623 acc_top5_avg=0.87704 lr=0.00010 gn=33.73740 time=64.79it/s +====================Eval==================== +epoch=102 global_step=40273 loss=3.25693 test_loss_avg=5.08643 acc=0.23438 test_acc_avg=0.09487 test_acc_top5_avg=0.72154 time=250.71it/s +epoch=102 global_step=40273 loss=5.18406 test_loss_avg=4.16803 acc=0.00000 test_acc_avg=0.23981 test_acc_top5_avg=0.82783 time=878.39it/s +curr_acc 0.2398 +BEST_ACC 0.2705 +curr_acc_top5 0.8278 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=103 global_step=40300 loss=4.17904 loss_avg=4.08486 acc=0.59375 acc_top1_avg=0.60561 acc_top5_avg=0.87587 lr=0.00010 gn=35.27662 time=65.50it/s +epoch=103 global_step=40350 loss=4.36507 loss_avg=4.08426 acc=0.56250 acc_top1_avg=0.60390 acc_top5_avg=0.87622 lr=0.00010 gn=35.68361 time=64.64it/s +epoch=103 global_step=40400 loss=3.51286 loss_avg=4.06362 acc=0.68750 acc_top1_avg=0.60618 acc_top5_avg=0.87820 lr=0.00010 gn=29.55278 time=65.88it/s +epoch=103 global_step=40450 loss=4.85053 loss_avg=4.04965 acc=0.53125 acc_top1_avg=0.60761 acc_top5_avg=0.87999 lr=0.00010 gn=40.46146 time=65.11it/s +epoch=103 global_step=40500 loss=3.34843 loss_avg=4.03794 acc=0.70312 acc_top1_avg=0.60862 acc_top5_avg=0.88006 lr=0.00010 gn=37.87530 time=65.69it/s +epoch=103 global_step=40550 loss=3.15540 loss_avg=4.04045 acc=0.70312 acc_top1_avg=0.60870 acc_top5_avg=0.88036 lr=0.00010 gn=29.40515 time=64.45it/s +epoch=103 global_step=40600 loss=4.31618 loss_avg=4.04559 acc=0.57812 acc_top1_avg=0.60792 acc_top5_avg=0.87937 lr=0.00010 gn=27.24695 time=65.81it/s +epoch=103 global_step=40650 loss=3.87531 loss_avg=4.04821 acc=0.61719 acc_top1_avg=0.60759 acc_top5_avg=0.87927 lr=0.00010 gn=38.25586 time=65.35it/s +====================Eval==================== +epoch=103 global_step=40664 loss=2.57993 test_loss_avg=4.99755 acc=0.46875 test_acc_avg=0.15986 test_acc_top5_avg=0.87260 time=259.23it/s +epoch=103 global_step=40664 loss=0.30868 test_loss_avg=4.51782 acc=0.91406 test_acc_avg=0.19358 test_acc_top5_avg=0.80332 time=254.77it/s +epoch=103 global_step=40664 loss=5.01760 test_loss_avg=4.15915 acc=0.00000 test_acc_avg=0.24120 test_acc_top5_avg=0.82783 time=884.31it/s +curr_acc 0.2412 +BEST_ACC 0.2705 +curr_acc_top5 0.8278 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=104 global_step=40700 loss=4.12568 loss_avg=4.12789 acc=0.60156 acc_top1_avg=0.59831 acc_top5_avg=0.87674 lr=0.00010 gn=28.20527 time=56.42it/s +epoch=104 global_step=40750 loss=4.11198 loss_avg=4.08192 acc=0.59375 acc_top1_avg=0.60338 acc_top5_avg=0.87500 lr=0.00010 gn=33.07167 time=56.32it/s +epoch=104 global_step=40800 loss=4.07285 loss_avg=4.06264 acc=0.60938 acc_top1_avg=0.60633 acc_top5_avg=0.87615 lr=0.00010 gn=31.12793 time=55.30it/s +epoch=104 global_step=40850 loss=4.11584 loss_avg=4.03115 acc=0.59375 acc_top1_avg=0.60946 acc_top5_avg=0.87643 lr=0.00010 gn=31.10423 time=56.73it/s +epoch=104 global_step=40900 loss=3.98276 loss_avg=4.04002 acc=0.61719 acc_top1_avg=0.60848 acc_top5_avg=0.87593 lr=0.00010 gn=32.02750 time=59.83it/s +epoch=104 global_step=40950 loss=3.96999 loss_avg=4.04031 acc=0.60938 acc_top1_avg=0.60836 acc_top5_avg=0.87688 lr=0.00010 gn=32.66873 time=56.23it/s +epoch=104 global_step=41000 loss=4.53688 loss_avg=4.03979 acc=0.54688 acc_top1_avg=0.60842 acc_top5_avg=0.87660 lr=0.00010 gn=27.70172 time=58.09it/s +epoch=104 global_step=41050 loss=4.42675 loss_avg=4.04016 acc=0.57031 acc_top1_avg=0.60818 acc_top5_avg=0.87688 lr=0.00010 gn=32.85953 time=61.96it/s +====================Eval==================== +epoch=104 global_step=41055 loss=6.05612 test_loss_avg=5.06974 acc=0.00000 test_acc_avg=0.09191 test_acc_top5_avg=0.77688 time=250.84it/s +epoch=104 global_step=41055 loss=5.28406 test_loss_avg=4.16590 acc=0.00000 test_acc_avg=0.24130 test_acc_top5_avg=0.82654 time=883.01it/s +curr_acc 0.2413 +BEST_ACC 0.2705 +curr_acc_top5 0.8265 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=105 global_step=41100 loss=4.28777 loss_avg=3.94407 acc=0.58594 acc_top1_avg=0.61771 acc_top5_avg=0.88212 lr=0.00010 gn=35.79270 time=61.14it/s +epoch=105 global_step=41150 loss=3.81804 loss_avg=3.95549 acc=0.62500 acc_top1_avg=0.61694 acc_top5_avg=0.88043 lr=0.00010 gn=31.33904 time=55.89it/s +epoch=105 global_step=41200 loss=4.21677 loss_avg=4.03021 acc=0.58594 acc_top1_avg=0.60824 acc_top5_avg=0.87780 lr=0.00010 gn=33.03018 time=61.20it/s +epoch=105 global_step=41250 loss=4.12114 loss_avg=4.04606 acc=0.60156 acc_top1_avg=0.60669 acc_top5_avg=0.87740 lr=0.00010 gn=28.60989 time=61.83it/s +epoch=105 global_step=41300 loss=3.63128 loss_avg=4.05500 acc=0.64844 acc_top1_avg=0.60631 acc_top5_avg=0.87698 lr=0.00010 gn=32.58871 time=61.64it/s +epoch=105 global_step=41350 loss=3.62961 loss_avg=4.05152 acc=0.64844 acc_top1_avg=0.60654 acc_top5_avg=0.87696 lr=0.00010 gn=36.62830 time=56.33it/s +epoch=105 global_step=41400 loss=3.04428 loss_avg=4.03532 acc=0.71875 acc_top1_avg=0.60840 acc_top5_avg=0.87767 lr=0.00010 gn=32.18012 time=61.34it/s +====================Eval==================== +epoch=105 global_step=41446 loss=6.73797 test_loss_avg=6.62762 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.85000 time=220.74it/s +epoch=105 global_step=41446 loss=4.43876 test_loss_avg=5.00405 acc=0.21875 test_acc_avg=0.10781 test_acc_top5_avg=0.77997 time=240.91it/s +epoch=105 global_step=41446 loss=5.26705 test_loss_avg=4.12653 acc=0.00000 test_acc_avg=0.24110 test_acc_top5_avg=0.82555 time=913.79it/s +curr_acc 0.2411 +BEST_ACC 0.2705 +curr_acc_top5 0.8256 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=106 global_step=41450 loss=4.75684 loss_avg=4.29509 acc=0.53906 acc_top1_avg=0.58203 acc_top5_avg=0.87305 lr=0.00010 gn=28.43688 time=64.89it/s +epoch=106 global_step=41500 loss=4.07708 loss_avg=4.07908 acc=0.61719 acc_top1_avg=0.60489 acc_top5_avg=0.87717 lr=0.00010 gn=31.38975 time=65.47it/s +epoch=106 global_step=41550 loss=3.37778 loss_avg=3.99505 acc=0.67969 acc_top1_avg=0.61373 acc_top5_avg=0.87725 lr=0.00010 gn=40.30707 time=65.89it/s +epoch=106 global_step=41600 loss=3.90668 loss_avg=3.97814 acc=0.62500 acc_top1_avg=0.61516 acc_top5_avg=0.87733 lr=0.00010 gn=32.24541 time=64.57it/s +epoch=106 global_step=41650 loss=3.84789 loss_avg=3.98606 acc=0.63281 acc_top1_avg=0.61420 acc_top5_avg=0.87749 lr=0.00010 gn=32.39350 time=61.62it/s +epoch=106 global_step=41700 loss=3.15389 loss_avg=4.01123 acc=0.70312 acc_top1_avg=0.61097 acc_top5_avg=0.87666 lr=0.00010 gn=31.94259 time=61.00it/s +epoch=106 global_step=41750 loss=3.36487 loss_avg=4.00430 acc=0.67969 acc_top1_avg=0.61177 acc_top5_avg=0.87783 lr=0.00010 gn=37.51013 time=62.04it/s +epoch=106 global_step=41800 loss=4.22364 loss_avg=4.00737 acc=0.59375 acc_top1_avg=0.61198 acc_top5_avg=0.87714 lr=0.00010 gn=38.07280 time=58.12it/s +====================Eval==================== +epoch=106 global_step=41837 loss=6.20398 test_loss_avg=4.80057 acc=0.00000 test_acc_avg=0.12470 test_acc_top5_avg=0.87981 time=259.95it/s +epoch=106 global_step=41837 loss=5.00485 test_loss_avg=4.16236 acc=0.02344 test_acc_avg=0.24887 test_acc_top5_avg=0.82607 time=262.52it/s +epoch=106 global_step=41837 loss=5.14469 test_loss_avg=4.19602 acc=0.00000 test_acc_avg=0.23981 test_acc_top5_avg=0.82605 time=903.94it/s +curr_acc 0.2398 +BEST_ACC 0.2705 +curr_acc_top5 0.8260 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=107 global_step=41850 loss=4.32808 loss_avg=4.08952 acc=0.58594 acc_top1_avg=0.60156 acc_top5_avg=0.87921 lr=0.00010 gn=31.88889 time=59.52it/s +epoch=107 global_step=41900 loss=4.32724 loss_avg=4.04949 acc=0.57812 acc_top1_avg=0.60714 acc_top5_avg=0.88232 lr=0.00010 gn=37.32278 time=58.94it/s +epoch=107 global_step=41950 loss=4.03131 loss_avg=4.05228 acc=0.61719 acc_top1_avg=0.60716 acc_top5_avg=0.87714 lr=0.00010 gn=36.05799 time=61.36it/s +epoch=107 global_step=42000 loss=4.02239 loss_avg=4.03379 acc=0.61719 acc_top1_avg=0.60885 acc_top5_avg=0.87797 lr=0.00010 gn=35.10973 time=61.63it/s +epoch=107 global_step=42050 loss=5.05795 loss_avg=4.02249 acc=0.49219 acc_top1_avg=0.60993 acc_top5_avg=0.87698 lr=0.00010 gn=24.17893 time=59.68it/s +epoch=107 global_step=42100 loss=3.83516 loss_avg=4.02614 acc=0.64062 acc_top1_avg=0.60961 acc_top5_avg=0.87619 lr=0.00010 gn=31.82972 time=60.54it/s +epoch=107 global_step=42150 loss=3.56305 loss_avg=4.02953 acc=0.65625 acc_top1_avg=0.60915 acc_top5_avg=0.87610 lr=0.00010 gn=29.60204 time=61.82it/s +epoch=107 global_step=42200 loss=3.94955 loss_avg=4.03263 acc=0.61719 acc_top1_avg=0.60892 acc_top5_avg=0.87599 lr=0.00010 gn=35.39709 time=57.91it/s +====================Eval==================== +epoch=107 global_step=42228 loss=3.15292 test_loss_avg=4.78139 acc=0.27344 test_acc_avg=0.12566 test_acc_top5_avg=0.74435 time=260.53it/s +epoch=107 global_step=42228 loss=5.06831 test_loss_avg=4.12059 acc=0.00000 test_acc_avg=0.24733 test_acc_top5_avg=0.82496 time=913.00it/s +curr_acc 0.2473 +BEST_ACC 0.2705 +curr_acc_top5 0.8250 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=108 global_step=42250 loss=3.91632 loss_avg=4.02598 acc=0.61719 acc_top1_avg=0.61115 acc_top5_avg=0.87855 lr=0.00010 gn=26.30362 time=64.98it/s +epoch=108 global_step=42300 loss=4.19121 loss_avg=4.03334 acc=0.60156 acc_top1_avg=0.60916 acc_top5_avg=0.88151 lr=0.00010 gn=33.72786 time=63.16it/s +epoch=108 global_step=42350 loss=4.88639 loss_avg=4.04681 acc=0.50781 acc_top1_avg=0.60854 acc_top5_avg=0.87974 lr=0.00010 gn=21.24457 time=61.64it/s +epoch=108 global_step=42400 loss=4.19270 loss_avg=4.03619 acc=0.58594 acc_top1_avg=0.60888 acc_top5_avg=0.87895 lr=0.00010 gn=25.72857 time=57.86it/s +epoch=108 global_step=42450 loss=4.10465 loss_avg=4.04171 acc=0.59375 acc_top1_avg=0.60888 acc_top5_avg=0.87827 lr=0.00010 gn=31.82989 time=64.50it/s +epoch=108 global_step=42500 loss=4.00043 loss_avg=4.03054 acc=0.60938 acc_top1_avg=0.60986 acc_top5_avg=0.87839 lr=0.00010 gn=32.34702 time=61.47it/s +epoch=108 global_step=42550 loss=3.83149 loss_avg=4.02545 acc=0.63281 acc_top1_avg=0.61022 acc_top5_avg=0.87886 lr=0.00010 gn=26.78717 time=63.49it/s +epoch=108 global_step=42600 loss=3.76965 loss_avg=4.02025 acc=0.62500 acc_top1_avg=0.61080 acc_top5_avg=0.87842 lr=0.00010 gn=21.50880 time=62.04it/s +====================Eval==================== +epoch=108 global_step=42619 loss=4.59048 test_loss_avg=4.60552 acc=0.00000 test_acc_avg=0.17752 test_acc_top5_avg=0.90104 time=256.25it/s +epoch=108 global_step=42619 loss=0.34733 test_loss_avg=4.16013 acc=0.92188 test_acc_avg=0.25149 test_acc_top5_avg=0.81250 time=249.84it/s +epoch=108 global_step=42619 loss=5.09173 test_loss_avg=4.12271 acc=0.00000 test_acc_avg=0.24555 test_acc_top5_avg=0.82160 time=910.22it/s +curr_acc 0.2455 +BEST_ACC 0.2705 +curr_acc_top5 0.8216 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=109 global_step=42650 loss=3.78352 loss_avg=3.84782 acc=0.63281 acc_top1_avg=0.62626 acc_top5_avg=0.88281 lr=0.00010 gn=27.34600 time=61.62it/s +epoch=109 global_step=42700 loss=4.67603 loss_avg=3.94873 acc=0.53906 acc_top1_avg=0.61728 acc_top5_avg=0.87944 lr=0.00010 gn=32.30504 time=51.48it/s +epoch=109 global_step=42750 loss=4.33615 loss_avg=3.96510 acc=0.57031 acc_top1_avg=0.61599 acc_top5_avg=0.87959 lr=0.00010 gn=29.65469 time=61.71it/s +epoch=109 global_step=42800 loss=4.36059 loss_avg=3.98375 acc=0.57031 acc_top1_avg=0.61386 acc_top5_avg=0.88040 lr=0.00010 gn=39.82736 time=62.08it/s +epoch=109 global_step=42850 loss=3.86460 loss_avg=4.00693 acc=0.62500 acc_top1_avg=0.61181 acc_top5_avg=0.87909 lr=0.00010 gn=35.45650 time=60.17it/s +epoch=109 global_step=42900 loss=4.04694 loss_avg=4.00851 acc=0.61719 acc_top1_avg=0.61177 acc_top5_avg=0.87942 lr=0.00010 gn=40.00796 time=61.63it/s +epoch=109 global_step=42950 loss=4.16559 loss_avg=4.00694 acc=0.60156 acc_top1_avg=0.61209 acc_top5_avg=0.87878 lr=0.00010 gn=32.04210 time=59.48it/s +epoch=109 global_step=43000 loss=4.05611 loss_avg=4.01920 acc=0.60156 acc_top1_avg=0.61089 acc_top5_avg=0.87849 lr=0.00010 gn=26.77399 time=61.97it/s +====================Eval==================== +epoch=109 global_step=43010 loss=6.05071 test_loss_avg=5.24982 acc=0.00000 test_acc_avg=0.07151 test_acc_top5_avg=0.69591 time=262.72it/s +epoch=109 global_step=43010 loss=5.13456 test_loss_avg=4.15824 acc=0.00000 test_acc_avg=0.23922 test_acc_top5_avg=0.81833 time=853.54it/s +curr_acc 0.2392 +BEST_ACC 0.2705 +curr_acc_top5 0.8183 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=110 global_step=43050 loss=4.15341 loss_avg=3.95912 acc=0.60156 acc_top1_avg=0.61465 acc_top5_avg=0.87461 lr=0.00010 gn=34.83604 time=55.27it/s +epoch=110 global_step=43100 loss=4.24004 loss_avg=3.98476 acc=0.58594 acc_top1_avg=0.61337 acc_top5_avg=0.87387 lr=0.00010 gn=30.85826 time=60.30it/s +epoch=110 global_step=43150 loss=4.29640 loss_avg=4.02093 acc=0.57812 acc_top1_avg=0.61060 acc_top5_avg=0.87500 lr=0.00010 gn=37.02561 time=56.39it/s +epoch=110 global_step=43200 loss=4.21643 loss_avg=4.00128 acc=0.58594 acc_top1_avg=0.61283 acc_top5_avg=0.87693 lr=0.00010 gn=30.57361 time=60.74it/s +epoch=110 global_step=43250 loss=4.37166 loss_avg=3.99502 acc=0.57031 acc_top1_avg=0.61367 acc_top5_avg=0.87832 lr=0.00010 gn=27.50130 time=59.09it/s +epoch=110 global_step=43300 loss=4.35332 loss_avg=4.00970 acc=0.56250 acc_top1_avg=0.61196 acc_top5_avg=0.87899 lr=0.00010 gn=25.43157 time=58.36it/s +epoch=110 global_step=43350 loss=4.46816 loss_avg=4.02448 acc=0.56250 acc_top1_avg=0.61034 acc_top5_avg=0.87849 lr=0.00010 gn=32.06224 time=61.65it/s +epoch=110 global_step=43400 loss=4.38773 loss_avg=4.01540 acc=0.57031 acc_top1_avg=0.61122 acc_top5_avg=0.87897 lr=0.00010 gn=32.75766 time=61.92it/s +====================Eval==================== +epoch=110 global_step=43401 loss=2.72107 test_loss_avg=5.76838 acc=0.37500 test_acc_avg=0.07656 test_acc_top5_avg=0.87109 time=251.67it/s +epoch=110 global_step=43401 loss=0.59979 test_loss_avg=4.66938 acc=0.82031 test_acc_avg=0.15924 test_acc_top5_avg=0.80312 time=260.45it/s +epoch=110 global_step=43401 loss=5.13087 test_loss_avg=4.14555 acc=0.00000 test_acc_avg=0.23803 test_acc_top5_avg=0.82981 time=899.10it/s +curr_acc 0.2380 +BEST_ACC 0.2705 +curr_acc_top5 0.8298 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=111 global_step=43450 loss=3.74220 loss_avg=4.08765 acc=0.64062 acc_top1_avg=0.60029 acc_top5_avg=0.87803 lr=0.00010 gn=33.30295 time=65.00it/s +epoch=111 global_step=43500 loss=4.24176 loss_avg=4.07945 acc=0.59375 acc_top1_avg=0.60290 acc_top5_avg=0.87792 lr=0.00010 gn=38.33849 time=65.60it/s +epoch=111 global_step=43550 loss=3.93175 loss_avg=4.03255 acc=0.60938 acc_top1_avg=0.60796 acc_top5_avg=0.87898 lr=0.00010 gn=34.92866 time=57.90it/s +epoch=111 global_step=43600 loss=4.08262 loss_avg=4.04336 acc=0.60156 acc_top1_avg=0.60714 acc_top5_avg=0.87806 lr=0.00010 gn=33.67587 time=60.71it/s +epoch=111 global_step=43650 loss=3.52438 loss_avg=4.01813 acc=0.66406 acc_top1_avg=0.61003 acc_top5_avg=0.87864 lr=0.00010 gn=30.12318 time=60.80it/s +epoch=111 global_step=43700 loss=3.37986 loss_avg=4.01412 acc=0.67188 acc_top1_avg=0.61073 acc_top5_avg=0.87834 lr=0.00010 gn=25.24638 time=52.47it/s +epoch=111 global_step=43750 loss=4.53864 loss_avg=4.01307 acc=0.55469 acc_top1_avg=0.61092 acc_top5_avg=0.87791 lr=0.00010 gn=37.44604 time=56.46it/s +====================Eval==================== +epoch=111 global_step=43792 loss=6.00079 test_loss_avg=5.02244 acc=0.00000 test_acc_avg=0.09526 test_acc_top5_avg=0.82434 time=251.70it/s +epoch=111 global_step=43792 loss=4.97874 test_loss_avg=4.17141 acc=0.00000 test_acc_avg=0.23784 test_acc_top5_avg=0.82130 time=892.79it/s +curr_acc 0.2378 +BEST_ACC 0.2705 +curr_acc_top5 0.8213 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=112 global_step=43800 loss=4.28934 loss_avg=3.94775 acc=0.58594 acc_top1_avg=0.61914 acc_top5_avg=0.88281 lr=0.00010 gn=33.99658 time=64.06it/s +epoch=112 global_step=43850 loss=4.01056 loss_avg=4.03864 acc=0.60156 acc_top1_avg=0.60803 acc_top5_avg=0.87769 lr=0.00010 gn=42.19884 time=61.93it/s +epoch=112 global_step=43900 loss=4.70993 loss_avg=4.01344 acc=0.53906 acc_top1_avg=0.61003 acc_top5_avg=0.87869 lr=0.00010 gn=32.20087 time=57.88it/s +epoch=112 global_step=43950 loss=4.29003 loss_avg=4.00434 acc=0.57031 acc_top1_avg=0.61130 acc_top5_avg=0.87802 lr=0.00010 gn=24.49027 time=61.73it/s +epoch=112 global_step=44000 loss=3.77767 loss_avg=4.00787 acc=0.65625 acc_top1_avg=0.61129 acc_top5_avg=0.87774 lr=0.00010 gn=45.67066 time=56.44it/s +epoch=112 global_step=44050 loss=4.26096 loss_avg=4.00268 acc=0.59375 acc_top1_avg=0.61228 acc_top5_avg=0.87888 lr=0.00010 gn=34.34648 time=60.08it/s +epoch=112 global_step=44100 loss=4.07924 loss_avg=3.99911 acc=0.60938 acc_top1_avg=0.61244 acc_top5_avg=0.87827 lr=0.00010 gn=35.54845 time=61.45it/s +epoch=112 global_step=44150 loss=4.12926 loss_avg=4.00720 acc=0.60938 acc_top1_avg=0.61167 acc_top5_avg=0.87782 lr=0.00010 gn=35.13278 time=61.56it/s +====================Eval==================== +epoch=112 global_step=44183 loss=6.64859 test_loss_avg=6.72364 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.86719 time=199.74it/s +epoch=112 global_step=44183 loss=6.64825 test_loss_avg=4.99864 acc=0.00000 test_acc_avg=0.11028 test_acc_top5_avg=0.76728 time=259.32it/s +epoch=112 global_step=44183 loss=5.09130 test_loss_avg=4.15603 acc=0.00000 test_acc_avg=0.24239 test_acc_top5_avg=0.81873 time=878.76it/s +curr_acc 0.2424 +BEST_ACC 0.2705 +curr_acc_top5 0.8187 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=113 global_step=44200 loss=3.94938 loss_avg=3.96285 acc=0.64062 acc_top1_avg=0.61673 acc_top5_avg=0.87868 lr=0.00010 gn=35.52171 time=65.44it/s +epoch=113 global_step=44250 loss=3.47232 loss_avg=4.08155 acc=0.67188 acc_top1_avg=0.60331 acc_top5_avg=0.87232 lr=0.00010 gn=32.62176 time=65.00it/s +epoch=113 global_step=44300 loss=4.23243 loss_avg=4.04359 acc=0.58594 acc_top1_avg=0.60811 acc_top5_avg=0.87366 lr=0.00010 gn=35.10523 time=64.97it/s +epoch=113 global_step=44350 loss=3.83400 loss_avg=4.04853 acc=0.63281 acc_top1_avg=0.60792 acc_top5_avg=0.87561 lr=0.00010 gn=36.54007 time=65.72it/s +epoch=113 global_step=44400 loss=4.42735 loss_avg=4.02848 acc=0.57812 acc_top1_avg=0.60941 acc_top5_avg=0.87687 lr=0.00010 gn=29.56513 time=63.91it/s +epoch=113 global_step=44450 loss=4.43079 loss_avg=4.02153 acc=0.56250 acc_top1_avg=0.61034 acc_top5_avg=0.87708 lr=0.00010 gn=35.12472 time=63.13it/s +epoch=113 global_step=44500 loss=3.96504 loss_avg=4.01787 acc=0.61719 acc_top1_avg=0.61112 acc_top5_avg=0.87744 lr=0.00010 gn=27.47671 time=65.47it/s +epoch=113 global_step=44550 loss=4.21451 loss_avg=4.00844 acc=0.59375 acc_top1_avg=0.61204 acc_top5_avg=0.87743 lr=0.00010 gn=37.21360 time=65.83it/s +====================Eval==================== +epoch=113 global_step=44574 loss=4.58340 test_loss_avg=4.62510 acc=0.00000 test_acc_avg=0.14029 test_acc_top5_avg=0.91236 time=248.07it/s +epoch=113 global_step=44574 loss=4.84560 test_loss_avg=4.07830 acc=0.00781 test_acc_avg=0.25706 test_acc_top5_avg=0.82598 time=246.14it/s +epoch=113 global_step=44574 loss=5.10202 test_loss_avg=4.15241 acc=0.00000 test_acc_avg=0.23823 test_acc_top5_avg=0.82506 time=920.01it/s +curr_acc 0.2382 +BEST_ACC 0.2705 +curr_acc_top5 0.8251 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=114 global_step=44600 loss=3.96814 loss_avg=4.01919 acc=0.60938 acc_top1_avg=0.60817 acc_top5_avg=0.88401 lr=0.00010 gn=35.48445 time=60.12it/s +epoch=114 global_step=44650 loss=3.74580 loss_avg=4.02629 acc=0.64062 acc_top1_avg=0.61009 acc_top5_avg=0.87675 lr=0.00010 gn=32.91634 time=60.99it/s +epoch=114 global_step=44700 loss=3.12108 loss_avg=4.02169 acc=0.71094 acc_top1_avg=0.61080 acc_top5_avg=0.87841 lr=0.00010 gn=27.30404 time=59.87it/s +epoch=114 global_step=44750 loss=4.07440 loss_avg=4.02734 acc=0.60156 acc_top1_avg=0.61040 acc_top5_avg=0.87678 lr=0.00010 gn=33.36257 time=59.68it/s +epoch=114 global_step=44800 loss=3.72518 loss_avg=4.01056 acc=0.64844 acc_top1_avg=0.61242 acc_top5_avg=0.87825 lr=0.00010 gn=31.28793 time=51.35it/s +epoch=114 global_step=44850 loss=3.95622 loss_avg=4.00618 acc=0.64062 acc_top1_avg=0.61266 acc_top5_avg=0.87874 lr=0.00010 gn=38.76374 time=60.17it/s +epoch=114 global_step=44900 loss=4.51682 loss_avg=3.98824 acc=0.56250 acc_top1_avg=0.61431 acc_top5_avg=0.87840 lr=0.00010 gn=36.89659 time=61.22it/s +epoch=114 global_step=44950 loss=4.92912 loss_avg=3.99657 acc=0.50781 acc_top1_avg=0.61318 acc_top5_avg=0.87724 lr=0.00010 gn=30.48854 time=59.80it/s +====================Eval==================== +epoch=114 global_step=44965 loss=3.11629 test_loss_avg=4.94840 acc=0.30469 test_acc_avg=0.10281 test_acc_top5_avg=0.73260 time=262.82it/s +epoch=114 global_step=44965 loss=5.04739 test_loss_avg=4.14075 acc=0.00000 test_acc_avg=0.23675 test_acc_top5_avg=0.82506 time=900.45it/s +curr_acc 0.2367 +BEST_ACC 0.2705 +curr_acc_top5 0.8251 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=115 global_step=45000 loss=4.65671 loss_avg=4.03620 acc=0.53906 acc_top1_avg=0.60893 acc_top5_avg=0.88013 lr=0.00010 gn=25.89286 time=58.72it/s +epoch=115 global_step=45050 loss=4.76895 loss_avg=4.05346 acc=0.53125 acc_top1_avg=0.60744 acc_top5_avg=0.87610 lr=0.00010 gn=37.76713 time=60.25it/s +epoch=115 global_step=45100 loss=3.52828 loss_avg=4.03585 acc=0.66406 acc_top1_avg=0.60932 acc_top5_avg=0.87685 lr=0.00010 gn=26.92955 time=58.47it/s +epoch=115 global_step=45150 loss=4.10290 loss_avg=4.01958 acc=0.59375 acc_top1_avg=0.61102 acc_top5_avg=0.87838 lr=0.00010 gn=34.85742 time=63.47it/s +epoch=115 global_step=45200 loss=4.49468 loss_avg=4.00324 acc=0.55469 acc_top1_avg=0.61237 acc_top5_avg=0.87916 lr=0.00010 gn=35.40487 time=58.06it/s +epoch=115 global_step=45250 loss=3.96630 loss_avg=4.00344 acc=0.63281 acc_top1_avg=0.61258 acc_top5_avg=0.87928 lr=0.00010 gn=34.04626 time=60.20it/s +epoch=115 global_step=45300 loss=3.32135 loss_avg=3.98709 acc=0.67969 acc_top1_avg=0.61472 acc_top5_avg=0.87922 lr=0.00010 gn=31.44895 time=61.63it/s +epoch=115 global_step=45350 loss=3.76481 loss_avg=3.99203 acc=0.63281 acc_top1_avg=0.61400 acc_top5_avg=0.87857 lr=0.00010 gn=36.55762 time=61.96it/s +====================Eval==================== +epoch=115 global_step=45356 loss=2.71300 test_loss_avg=4.73133 acc=0.41406 test_acc_avg=0.18177 test_acc_top5_avg=0.88229 time=243.71it/s +epoch=115 global_step=45356 loss=0.18575 test_loss_avg=4.38707 acc=0.94531 test_acc_avg=0.20841 test_acc_top5_avg=0.80733 time=262.11it/s +epoch=115 global_step=45356 loss=4.95724 test_loss_avg=4.16380 acc=0.00000 test_acc_avg=0.23527 test_acc_top5_avg=0.82021 time=907.27it/s +curr_acc 0.2353 +BEST_ACC 0.2705 +curr_acc_top5 0.8202 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=116 global_step=45400 loss=3.76253 loss_avg=3.94752 acc=0.63281 acc_top1_avg=0.61754 acc_top5_avg=0.87908 lr=0.00010 gn=34.57474 time=61.90it/s +epoch=116 global_step=45450 loss=3.92575 loss_avg=3.96239 acc=0.61719 acc_top1_avg=0.61702 acc_top5_avg=0.88107 lr=0.00010 gn=35.09692 time=60.25it/s +epoch=116 global_step=45500 loss=3.54886 loss_avg=3.97657 acc=0.66406 acc_top1_avg=0.61534 acc_top5_avg=0.88081 lr=0.00010 gn=28.55629 time=61.33it/s +epoch=116 global_step=45550 loss=3.49595 loss_avg=3.99517 acc=0.67969 acc_top1_avg=0.61356 acc_top5_avg=0.87798 lr=0.00010 gn=36.68211 time=58.10it/s +epoch=116 global_step=45600 loss=3.35111 loss_avg=3.98570 acc=0.69531 acc_top1_avg=0.61459 acc_top5_avg=0.87718 lr=0.00010 gn=40.76977 time=61.67it/s +epoch=116 global_step=45650 loss=4.13535 loss_avg=3.98925 acc=0.60156 acc_top1_avg=0.61416 acc_top5_avg=0.87779 lr=0.00010 gn=35.47641 time=56.77it/s +epoch=116 global_step=45700 loss=4.01439 loss_avg=3.99553 acc=0.61719 acc_top1_avg=0.61328 acc_top5_avg=0.87779 lr=0.00010 gn=39.66323 time=60.53it/s +====================Eval==================== +epoch=116 global_step=45747 loss=6.20009 test_loss_avg=5.03666 acc=0.00000 test_acc_avg=0.09766 test_acc_top5_avg=0.74067 time=261.72it/s +epoch=116 global_step=45747 loss=5.00068 test_loss_avg=4.10441 acc=0.00000 test_acc_avg=0.24624 test_acc_top5_avg=0.82496 time=885.81it/s +curr_acc 0.2462 +BEST_ACC 0.2705 +curr_acc_top5 0.8250 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=117 global_step=45750 loss=4.22706 loss_avg=3.94674 acc=0.58594 acc_top1_avg=0.61458 acc_top5_avg=0.87240 lr=0.00010 gn=33.76038 time=58.24it/s +epoch=117 global_step=45800 loss=3.78547 loss_avg=4.01630 acc=0.64062 acc_top1_avg=0.61262 acc_top5_avg=0.87544 lr=0.00010 gn=34.35204 time=64.17it/s +epoch=117 global_step=45850 loss=4.65629 loss_avg=3.99669 acc=0.54688 acc_top1_avg=0.61408 acc_top5_avg=0.87454 lr=0.00010 gn=29.53752 time=59.88it/s +epoch=117 global_step=45900 loss=3.95717 loss_avg=3.96983 acc=0.62500 acc_top1_avg=0.61652 acc_top5_avg=0.87699 lr=0.00010 gn=24.65868 time=60.64it/s +epoch=117 global_step=45950 loss=4.06492 loss_avg=3.96682 acc=0.61719 acc_top1_avg=0.61680 acc_top5_avg=0.87950 lr=0.00010 gn=38.76542 time=53.59it/s +epoch=117 global_step=46000 loss=3.54507 loss_avg=3.97512 acc=0.65625 acc_top1_avg=0.61595 acc_top5_avg=0.87942 lr=0.00010 gn=32.66429 time=54.03it/s +epoch=117 global_step=46050 loss=2.93128 loss_avg=3.98365 acc=0.72656 acc_top1_avg=0.61497 acc_top5_avg=0.87871 lr=0.00010 gn=31.54231 time=60.36it/s +epoch=117 global_step=46100 loss=4.36729 loss_avg=3.97739 acc=0.58594 acc_top1_avg=0.61564 acc_top5_avg=0.87890 lr=0.00010 gn=41.09780 time=61.83it/s +====================Eval==================== +epoch=117 global_step=46138 loss=6.29180 test_loss_avg=6.44529 acc=0.00000 test_acc_avg=0.00000 test_acc_top5_avg=0.81696 time=251.04it/s +epoch=117 global_step=46138 loss=0.65811 test_loss_avg=4.87602 acc=0.78125 test_acc_avg=0.12569 test_acc_top5_avg=0.77810 time=252.82it/s +epoch=117 global_step=46138 loss=5.04225 test_loss_avg=4.13731 acc=0.00000 test_acc_avg=0.23843 test_acc_top5_avg=0.82180 time=899.29it/s +curr_acc 0.2384 +BEST_ACC 0.2705 +curr_acc_top5 0.8218 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=118 global_step=46150 loss=4.40175 loss_avg=3.97591 acc=0.57812 acc_top1_avg=0.61523 acc_top5_avg=0.88086 lr=0.00010 gn=39.97420 time=61.55it/s +epoch=118 global_step=46200 loss=4.22378 loss_avg=3.94072 acc=0.58594 acc_top1_avg=0.61983 acc_top5_avg=0.88130 lr=0.00010 gn=25.60333 time=61.69it/s +epoch=118 global_step=46250 loss=4.04266 loss_avg=3.92920 acc=0.59375 acc_top1_avg=0.62012 acc_top5_avg=0.87905 lr=0.00010 gn=30.61296 time=62.59it/s +epoch=118 global_step=46300 loss=3.63545 loss_avg=3.95561 acc=0.66406 acc_top1_avg=0.61704 acc_top5_avg=0.87828 lr=0.00010 gn=42.26916 time=56.37it/s +epoch=118 global_step=46350 loss=3.32329 loss_avg=3.96444 acc=0.68750 acc_top1_avg=0.61627 acc_top5_avg=0.87703 lr=0.00010 gn=32.67698 time=55.38it/s +epoch=118 global_step=46400 loss=3.73974 loss_avg=3.97203 acc=0.64844 acc_top1_avg=0.61561 acc_top5_avg=0.87625 lr=0.00010 gn=36.70713 time=61.57it/s +epoch=118 global_step=46450 loss=3.70869 loss_avg=3.97747 acc=0.64844 acc_top1_avg=0.61564 acc_top5_avg=0.87610 lr=0.00010 gn=32.39606 time=64.88it/s +epoch=118 global_step=46500 loss=4.71454 loss_avg=3.96998 acc=0.54688 acc_top1_avg=0.61665 acc_top5_avg=0.87686 lr=0.00010 gn=34.06896 time=65.54it/s +====================Eval==================== +epoch=118 global_step=46529 loss=5.93492 test_loss_avg=4.89874 acc=0.00000 test_acc_avg=0.10993 test_acc_top5_avg=0.84989 time=254.37it/s +epoch=118 global_step=46529 loss=4.93426 test_loss_avg=4.13688 acc=0.01562 test_acc_avg=0.24139 test_acc_top5_avg=0.81801 time=263.16it/s +epoch=118 global_step=46529 loss=5.01851 test_loss_avg=4.14804 acc=0.00000 test_acc_avg=0.23833 test_acc_top5_avg=0.81952 time=864.45it/s +curr_acc 0.2383 +BEST_ACC 0.2705 +curr_acc_top5 0.8195 +BEST_ACC_top5 0.8710 +Model Saved! + +====================Training==================== +epoch=119 global_step=46550 loss=4.08406 loss_avg=3.96280 acc=0.60938 acc_top1_avg=0.61905 acc_top5_avg=0.87723 lr=0.00010 gn=39.06379 time=58.80it/s +epoch=119 global_step=46600 loss=3.15153 loss_avg=3.90205 acc=0.69531 acc_top1_avg=0.62291 acc_top5_avg=0.88182 lr=0.00010 gn=37.21027 time=58.27it/s +epoch=119 global_step=46650 loss=3.68984 loss_avg=3.93401 acc=0.65625 acc_top1_avg=0.62035 acc_top5_avg=0.87849 lr=0.00010 gn=37.64401 time=58.85it/s +epoch=119 global_step=46700 loss=4.02575 loss_avg=3.96073 acc=0.60938 acc_top1_avg=0.61691 acc_top5_avg=0.87788 lr=0.00010 gn=27.55517 time=60.52it/s +epoch=119 global_step=46750 loss=4.39201 loss_avg=3.95495 acc=0.57031 acc_top1_avg=0.61765 acc_top5_avg=0.87776 lr=0.00010 gn=31.71968 time=62.04it/s +epoch=119 global_step=46800 loss=4.41513 loss_avg=3.96345 acc=0.57812 acc_top1_avg=0.61670 acc_top5_avg=0.87800 lr=0.00010 gn=44.62899 time=58.48it/s +epoch=119 global_step=46850 loss=4.50761 loss_avg=3.96783 acc=0.55469 acc_top1_avg=0.61638 acc_top5_avg=0.87685 lr=0.00010 gn=33.84546 time=58.21it/s +epoch=119 global_step=46900 loss=4.53928 loss_avg=3.96799 acc=0.56250 acc_top1_avg=0.61635 acc_top5_avg=0.87788 lr=0.00010 gn=25.97112 time=53.30it/s +====================Eval==================== +epoch=119 global_step=46920 loss=6.48996 test_loss_avg=4.90034 acc=0.00000 test_acc_avg=0.11193 test_acc_top5_avg=0.75255 time=256.34it/s +epoch=119 global_step=46920 loss=5.06386 test_loss_avg=4.13266 acc=0.00000 test_acc_avg=0.24021 test_acc_top5_avg=0.82150 time=895.84it/s +curr_acc 0.2402 +BEST_ACC 0.2705 +curr_acc_top5 0.8215 +BEST_ACC_top5 0.8710 +Model Saved! +